release 2014.05.12

[FunnyOrDie] fix thumbnails + add test (fixes #2856)

.gitignore

@@ -23,6 +23,8 @@ updates_key.pem
 *.vtt
 *.flv
 *.mp4
+*.m4a
+*.m4v
 *.part
 test/testdata
 .tox

MANIFEST.in

@@ -3,3 +3,4 @@ include test/*.py
 include test/*.json
 include youtube-dl.bash-completion
 include youtube-dl.1
+recursive-include docs Makefile conf.py *.rst

Makefile

@@ -72,8 +72,9 @@ youtube-dl.tar.gz: youtube-dl README.md README.txt youtube-dl.1 youtube-dl.bash-
 		--exclude '__pycache' \
 		--exclude '.git' \
 		--exclude 'testdata' \
+		--exclude 'docs/_build' \
 		-- \
-		bin devscripts test youtube_dl \
+		bin devscripts test youtube_dl docs \
 		CHANGELOG LICENSE README.md README.txt \
 		Makefile MANIFEST.in youtube-dl.1 youtube-dl.bash-completion setup.py \
 		youtube-dl

README.md

@@ -14,187 +14,254 @@ your Unix box, on Windows or on Mac OS X. It is released to the public domain,
 which means you can modify it, redistribute it or use it however you like.
 
 # OPTIONS
-    -h, --help                 print this help text and exit
-    --version                  print program version and exit
-    -U, --update               update this program to latest version. Make sure
-                               that you have sufficient permissions (run with
-                               sudo if needed)
-    -i, --ignore-errors        continue on download errors, for example to to
-                               skip unavailable videos in a playlist
-    --abort-on-error           Abort downloading of further videos (in the
-                               playlist or the command line) if an error occurs
-    --dump-user-agent          display the current browser identification
-    --user-agent UA            specify a custom user agent
-    --referer REF              specify a custom referer, use if the video access
-                               is restricted to one domain
-    --list-extractors          List all supported extractors and the URLs they
-                               would handle
-    --extractor-descriptions   Output descriptions of all supported extractors
-    --proxy URL                Use the specified HTTP/HTTPS proxy. Pass in an
-                               empty string (--proxy "") for direct connection
-    --no-check-certificate     Suppress HTTPS certificate validation.
-    --cache-dir DIR            Location in the filesystem where youtube-dl can
-                               store downloaded information permanently. By
-                               default $XDG_CACHE_HOME/youtube-dl or ~/.cache
-                               /youtube-dl .
-    --no-cache-dir             Disable filesystem caching
-    --bidi-workaround          Work around terminals that lack bidirectional
-                               text support. Requires bidiv or fribidi
-                               executable in PATH
+    -h, --help                       print this help text and exit
+    --version                        print program version and exit
+    -U, --update                     update this program to latest version. Make
+                                     sure that you have sufficient permissions
+                                     (run with sudo if needed)
+    -i, --ignore-errors              continue on download errors, for example to
+                                     skip unavailable videos in a playlist
+    --abort-on-error                 Abort downloading of further videos (in the
+                                     playlist or the command line) if an error
+                                     occurs
+    --dump-user-agent                display the current browser identification
+    --user-agent UA                  specify a custom user agent
+    --referer REF                    specify a custom referer, use if the video
+                                     access is restricted to one domain
+    --add-header FIELD:VALUE         specify a custom HTTP header and its value,
+                                     separated by a colon ':'. You can use this
+                                     option multiple times
+    --list-extractors                List all supported extractors and the URLs
+                                     they would handle
+    --extractor-descriptions         Output descriptions of all supported
+                                     extractors
+    --proxy URL                      Use the specified HTTP/HTTPS proxy. Pass in
+                                     an empty string (--proxy "") for direct
+                                     connection
+    --no-check-certificate           Suppress HTTPS certificate validation.
+    --prefer-insecure                Use an unencrypted connection to retrieve
+                                     information about the video. (Currently
+                                     supported only for YouTube)
+    --cache-dir DIR                  Location in the filesystem where youtube-dl
+                                     can store some downloaded information
+                                     permanently. By default $XDG_CACHE_HOME
+                                     /youtube-dl or ~/.cache/youtube-dl . At the
+                                     moment, only YouTube player files (for
+                                     videos with obfuscated signatures) are
+                                     cached, but that may change.
+    --no-cache-dir                   Disable filesystem caching
+    --socket-timeout None            Time to wait before giving up, in seconds
+    --bidi-workaround                Work around terminals that lack
+                                     bidirectional text support. Requires bidiv
+                                     or fribidi executable in PATH
+    --default-search PREFIX          Use this prefix for unqualified URLs. For
+                                     example "gvsearch2:" downloads two videos
+                                     from google videos for  youtube-dl "large
+                                     apple". By default (with value "auto")
+                                     youtube-dl guesses.
+    --ignore-config                  Do not read configuration files. When given
+                                     in the global configuration file /etc
+                                     /youtube-dl.conf: do not read the user
+                                     configuration in ~/.config/youtube-dl.conf
+                                     (%APPDATA%/youtube-dl/config.txt on
+                                     Windows)
+    --encoding ENCODING              Force the specified encoding (experimental)
 
 ## Video Selection:
-    --playlist-start NUMBER    playlist video to start at (default is 1)
-    --playlist-end NUMBER      playlist video to end at (default is last)
-    --match-title REGEX        download only matching titles (regex or caseless
-                               sub-string)
-    --reject-title REGEX       skip download for matching titles (regex or
-                               caseless sub-string)
-    --max-downloads NUMBER     Abort after downloading NUMBER files
-    --min-filesize SIZE        Do not download any videos smaller than SIZE
-                               (e.g. 50k or 44.6m)
-    --max-filesize SIZE        Do not download any videos larger than SIZE (e.g.
-                               50k or 44.6m)
-    --date DATE                download only videos uploaded in this date
-    --datebefore DATE          download only videos uploaded before this date
-    --dateafter DATE           download only videos uploaded after this date
-    --min-views COUNT          Do not download any videos with less than COUNT
-                               views
-    --max-views COUNT          Do not download any videos with more than COUNT
-                               views
-    --no-playlist              download only the currently playing video
-    --age-limit YEARS          download only videos suitable for the given age
-    --download-archive FILE    Download only videos not listed in the archive
-                               file. Record the IDs of all downloaded videos in
-                               it.
+    --playlist-start NUMBER          playlist video to start at (default is 1)
+    --playlist-end NUMBER            playlist video to end at (default is last)
+    --match-title REGEX              download only matching titles (regex or
+                                     caseless sub-string)
+    --reject-title REGEX             skip download for matching titles (regex or
+                                     caseless sub-string)
+    --max-downloads NUMBER           Abort after downloading NUMBER files
+    --min-filesize SIZE              Do not download any videos smaller than
+                                     SIZE (e.g. 50k or 44.6m)
+    --max-filesize SIZE              Do not download any videos larger than SIZE
+                                     (e.g. 50k or 44.6m)
+    --date DATE                      download only videos uploaded in this date
+    --datebefore DATE                download only videos uploaded on or before
+                                     this date (i.e. inclusive)
+    --dateafter DATE                 download only videos uploaded on or after
+                                     this date (i.e. inclusive)
+    --min-views COUNT                Do not download any videos with less than
+                                     COUNT views
+    --max-views COUNT                Do not download any videos with more than
+                                     COUNT views
+    --no-playlist                    download only the currently playing video
+    --age-limit YEARS                download only videos suitable for the given
+                                     age
+    --download-archive FILE          Download only videos not listed in the
+                                     archive file. Record the IDs of all
+                                     downloaded videos in it.
+    --include-ads                    Download advertisements as well
+                                     (experimental)
+    --youtube-include-dash-manifest  Try to download the DASH manifest on
+                                     YouTube videos (experimental)
 
 ## Download Options:
-    -r, --rate-limit LIMIT     maximum download rate in bytes per second (e.g.
-                               50K or 4.2M)
-    -R, --retries RETRIES      number of retries (default is 10)
-    --buffer-size SIZE         size of download buffer (e.g. 1024 or 16K)
-                               (default is 1024)
-    --no-resize-buffer         do not automatically adjust the buffer size. By
-                               default, the buffer size is automatically resized
-                               from an initial value of SIZE.
+    -r, --rate-limit LIMIT           maximum download rate in bytes per second
+                                     (e.g. 50K or 4.2M)
+    -R, --retries RETRIES            number of retries (default is 10)
+    --buffer-size SIZE               size of download buffer (e.g. 1024 or 16K)
+                                     (default is 1024)
+    --no-resize-buffer               do not automatically adjust the buffer
+                                     size. By default, the buffer size is
+                                     automatically resized from an initial value
+                                     of SIZE.
 
 ## Filesystem Options:
-    -t, --title                use title in file name (default)
-    --id                       use only video ID in file name
-    -l, --literal              [deprecated] alias of --title
-    -A, --auto-number          number downloaded files starting from 00000
-    -o, --output TEMPLATE      output filename template. Use %(title)s to get
-                               the title, %(uploader)s for the uploader name,
-                               %(uploader_id)s for the uploader nickname if
-                               different, %(autonumber)s to get an automatically
-                               incremented number, %(ext)s for the filename
-                               extension, %(format)s for the format description
-                               (like "22 - 1280x720" or "HD"),%(format_id)s for
-                               the unique id of the format (like Youtube's
-                               itags: "137"),%(upload_date)s for the upload date
-                               (YYYYMMDD), %(extractor)s for the provider
-                               (youtube, metacafe, etc), %(id)s for the video id
-                               , %(playlist)s for the playlist the video is in,
-                               %(playlist_index)s for the position in the
-                               playlist and %% for a literal percent. Use - to
-                               output to stdout. Can also be used to download to
-                               a different directory, for example with -o '/my/d
-                               ownloads/%(uploader)s/%(title)s-%(id)s.%(ext)s' .
-    --autonumber-size NUMBER   Specifies the number of digits in %(autonumber)s
-                               when it is present in output filename template or
-                               --auto-number option is given
-    --restrict-filenames       Restrict filenames to only ASCII characters, and
-                               avoid "&" and spaces in filenames
-    -a, --batch-file FILE      file containing URLs to download ('-' for stdin)
-    --load-info FILE           json file containing the video information
-                               (created with the "--write-json" option
-    -w, --no-overwrites        do not overwrite files
-    -c, --continue             force resume of partially downloaded files. By
-                               default, youtube-dl will resume downloads if
-                               possible.
-    --no-continue              do not resume partially downloaded files (restart
-                               from beginning)
-    --cookies FILE             file to read cookies from and dump cookie jar in
-    --no-part                  do not use .part files
-    --no-mtime                 do not use the Last-modified header to set the
-                               file modification time
-    --write-description        write video description to a .description file
-    --write-info-json          write video metadata to a .info.json file
-    --write-annotations        write video annotations to a .annotation file
-    --write-thumbnail          write thumbnail image to disk
+    -t, --title                      use title in file name (default)
+    --id                             use only video ID in file name
+    -l, --literal                    [deprecated] alias of --title
+    -A, --auto-number                number downloaded files starting from 00000
+    -o, --output TEMPLATE            output filename template. Use %(title)s to
+                                     get the title, %(uploader)s for the
+                                     uploader name, %(uploader_id)s for the
+                                     uploader nickname if different,
+                                     %(autonumber)s to get an automatically
+                                     incremented number, %(ext)s for the
+                                     filename extension, %(format)s for the
+                                     format description (like "22 - 1280x720" or
+                                     "HD"), %(format_id)s for the unique id of
+                                     the format (like Youtube's itags: "137"),
+                                     %(upload_date)s for the upload date
+                                     (YYYYMMDD), %(extractor)s for the provider
+                                     (youtube, metacafe, etc), %(id)s for the
+                                     video id, %(playlist)s for the playlist the
+                                     video is in, %(playlist_index)s for the
+                                     position in the playlist and %% for a
+                                     literal percent. %(height)s and %(width)s
+                                     for the width and height of the video
+                                     format. %(resolution)s for a textual
+                                     description of the resolution of the video
+                                     format. Use - to output to stdout. Can also
+                                     be used to download to a different
+                                     directory, for example with -o '/my/downloa
+                                     ds/%(uploader)s/%(title)s-%(id)s.%(ext)s' .
+    --autonumber-size NUMBER         Specifies the number of digits in
+                                     %(autonumber)s when it is present in output
+                                     filename template or --auto-number option
+                                     is given
+    --restrict-filenames             Restrict filenames to only ASCII
+                                     characters, and avoid "&" and spaces in
+                                     filenames
+    -a, --batch-file FILE            file containing URLs to download ('-' for
+                                     stdin)
+    --load-info FILE                 json file containing the video information
+                                     (created with the "--write-json" option)
+    -w, --no-overwrites              do not overwrite files
+    -c, --continue                   force resume of partially downloaded files.
+                                     By default, youtube-dl will resume
+                                     downloads if possible.
+    --no-continue                    do not resume partially downloaded files
+                                     (restart from beginning)
+    --cookies FILE                   file to read cookies from and dump cookie
+                                     jar in
+    --no-part                        do not use .part files
+    --no-mtime                       do not use the Last-modified header to set
+                                     the file modification time
+    --write-description              write video description to a .description
+                                     file
+    --write-info-json                write video metadata to a .info.json file
+    --write-annotations              write video annotations to a .annotation
+                                     file
+    --write-thumbnail                write thumbnail image to disk
 
 ## Verbosity / Simulation Options:
-    -q, --quiet                activates quiet mode
-    -s, --simulate             do not download the video and do not write
-                               anything to disk
-    --skip-download            do not download the video
-    -g, --get-url              simulate, quiet but print URL
-    -e, --get-title            simulate, quiet but print title
-    --get-id                   simulate, quiet but print id
-    --get-thumbnail            simulate, quiet but print thumbnail URL
-    --get-description          simulate, quiet but print video description
-    --get-duration             simulate, quiet but print video length
-    --get-filename             simulate, quiet but print output filename
-    --get-format               simulate, quiet but print output format
-    -j, --dump-json            simulate, quiet but print JSON information
-    --newline                  output progress bar as new lines
-    --no-progress              do not print progress bar
-    --console-title            display progress in console titlebar
-    -v, --verbose              print various debugging information
-    --dump-intermediate-pages  print downloaded pages to debug problems(very
-                               verbose)
-    --write-pages              Write downloaded intermediary pages to files in
-                               the current directory to debug problems
+    -q, --quiet                      activates quiet mode
+    --no-warnings                    Ignore warnings
+    -s, --simulate                   do not download the video and do not write
+                                     anything to disk
+    --skip-download                  do not download the video
+    -g, --get-url                    simulate, quiet but print URL
+    -e, --get-title                  simulate, quiet but print title
+    --get-id                         simulate, quiet but print id
+    --get-thumbnail                  simulate, quiet but print thumbnail URL
+    --get-description                simulate, quiet but print video description
+    --get-duration                   simulate, quiet but print video length
+    --get-filename                   simulate, quiet but print output filename
+    --get-format                     simulate, quiet but print output format
+    -j, --dump-json                  simulate, quiet but print JSON information.
+                                     See --output for a description of available
+                                     keys.
+    --newline                        output progress bar as new lines
+    --no-progress                    do not print progress bar
+    --console-title                  display progress in console titlebar
+    -v, --verbose                    print various debugging information
+    --dump-intermediate-pages        print downloaded pages to debug problems
+                                     (very verbose)
+    --write-pages                    Write downloaded intermediary pages to
+                                     files in the current directory to debug
+                                     problems
+    --print-traffic                  Display sent and read HTTP traffic
 
 ## Video Format Options:
-    -f, --format FORMAT        video format code, specify the order of
-                               preference using slashes: "-f 22/17/18". "-f mp4"
-                               and "-f flv" are also supported
-    --all-formats              download all available video formats
-    --prefer-free-formats      prefer free video formats unless a specific one
-                               is requested
-    --max-quality FORMAT       highest quality format to download
-    -F, --list-formats         list all available formats (currently youtube
-                               only)
+    -f, --format FORMAT              video format code, specify the order of
+                                     preference using slashes: "-f 22/17/18".
+                                     "-f mp4" and "-f flv" are also supported.
+                                     You can also use the special names "best",
+                                     "bestvideo", "bestaudio", "worst",
+                                     "worstvideo" and "worstaudio". By default,
+                                     youtube-dl will pick the best quality.
+    --all-formats                    download all available video formats
+    --prefer-free-formats            prefer free video formats unless a specific
+                                     one is requested
+    --max-quality FORMAT             highest quality format to download
+    -F, --list-formats               list all available formats
 
 ## Subtitle Options:
-    --write-sub                write subtitle file
-    --write-auto-sub           write automatic subtitle file (youtube only)
-    --all-subs                 downloads all the available subtitles of the
-                               video
-    --list-subs                lists all available subtitles for the video
-    --sub-format FORMAT        subtitle format (default=srt) ([sbv/vtt] youtube
-                               only)
-    --sub-lang LANGS           languages of the subtitles to download (optional)
-                               separated by commas, use IETF language tags like
-                               'en,pt'
+    --write-sub                      write subtitle file
+    --write-auto-sub                 write automatic subtitle file (youtube
+                                     only)
+    --all-subs                       downloads all the available subtitles of
+                                     the video
+    --list-subs                      lists all available subtitles for the video
+    --sub-format FORMAT              subtitle format (default=srt) ([sbv/vtt]
+                                     youtube only)
+    --sub-lang LANGS                 languages of the subtitles to download
+                                     (optional) separated by commas, use IETF
+                                     language tags like 'en,pt'
 
 ## Authentication Options:
-    -u, --username USERNAME    account username
-    -p, --password PASSWORD    account password
-    -n, --netrc                use .netrc authentication data
-    --video-password PASSWORD  video password (vimeo only)
+    -u, --username USERNAME          account username
+    -p, --password PASSWORD          account password
+    -n, --netrc                      use .netrc authentication data
+    --video-password PASSWORD        video password (vimeo, smotri)
 
 ## Post-processing Options:
-    -x, --extract-audio        convert video files to audio-only files (requires
-                               ffmpeg or avconv and ffprobe or avprobe)
-    --audio-format FORMAT      "best", "aac", "vorbis", "mp3", "m4a", "opus", or
-                               "wav"; best by default
-    --audio-quality QUALITY    ffmpeg/avconv audio quality specification, insert
-                               a value between 0 (better) and 9 (worse) for VBR
-                               or a specific bitrate like 128K (default 5)
-    --recode-video FORMAT      Encode the video to another format if necessary
-                               (currently supported: mp4|flv|ogg|webm)
-    -k, --keep-video           keeps the video file on disk after the post-
-                               processing; the video is erased by default
-    --no-post-overwrites       do not overwrite post-processed files; the post-
-                               processed files are overwritten by default
-    --embed-subs               embed subtitles in the video (only for mp4
-                               videos)
-    --add-metadata             add metadata to the files
+    -x, --extract-audio              convert video files to audio-only files
+                                     (requires ffmpeg or avconv and ffprobe or
+                                     avprobe)
+    --audio-format FORMAT            "best", "aac", "vorbis", "mp3", "m4a",
+                                     "opus", or "wav"; best by default
+    --audio-quality QUALITY          ffmpeg/avconv audio quality specification,
+                                     insert a value between 0 (better) and 9
+                                     (worse) for VBR or a specific bitrate like
+                                     128K (default 5)
+    --recode-video FORMAT            Encode the video to another format if
+                                     necessary (currently supported:
+                                     mp4|flv|ogg|webm)
+    -k, --keep-video                 keeps the video file on disk after the
+                                     post-processing; the video is erased by
+                                     default
+    --no-post-overwrites             do not overwrite post-processed files; the
+                                     post-processed files are overwritten by
+                                     default
+    --embed-subs                     embed subtitles in the video (only for mp4
+                                     videos)
+    --embed-thumbnail                embed thumbnail in the audio as cover art
+    --add-metadata                   write metadata to the video file
+    --xattrs                         write metadata to the video file's xattrs
+                                     (using dublin core and xdg standards)
+    --prefer-avconv                  Prefer avconv over ffmpeg for running the
+                                     postprocessors (default)
+    --prefer-ffmpeg                  Prefer ffmpeg over avconv for running the
+                                     postprocessors
 
 # CONFIGURATION
 
-You can configure youtube-dl by placing default arguments (such as `--extract-audio --no-mtime` to always extract the audio and not copy the mtime) into `/etc/youtube-dl.conf` and/or `~/.config/youtube-dl.conf`. On Windows, the configuration file locations are `%APPDATA%\youtube-dl\config.txt` and `C:\Users\<Yourname>\youtube-dl.conf`.
+You can configure youtube-dl by placing default arguments (such as `--extract-audio --no-mtime` to always extract the audio and not copy the mtime) into `/etc/youtube-dl.conf` and/or `~/.config/youtube-dl/config`. On Windows, the configuration file locations are `%APPDATA%\youtube-dl\config.txt` and `C:\Users\<Yourname>\youtube-dl.conf`.
 
 # OUTPUT TEMPLATE
 
@@ -229,9 +296,14 @@ Videos can be filtered by their upload date using the options `--date`, `--dateb
  
 Examples:
 
-	$ youtube-dl --dateafter now-6months #will only download the videos uploaded in the last 6 months
-	$ youtube-dl --date 19700101 #will only download the videos uploaded in January 1, 1970
-	$ youtube-dl --dateafter 20000101 --datebefore 20100101 #will only download the videos uploaded between 2000 and 2010
+    # Download only the videos uploaded in the last 6 months
+    $ youtube-dl --dateafter now-6months
+
+    # Download only the videos uploaded on January 1, 1970
+    $ youtube-dl --date 19700101
+
+    $ # will only download the videos uploaded in the 200x decade
+    $ youtube-dl --dateafter 20000101 --datebefore 20091231
 
 # FAQ
 
@@ -276,11 +348,91 @@ Since June 2012 (#342) youtube-dl is packed as an executable zipfile, simply unz
 
 To run the exe you need to install first the [Microsoft Visual C++ 2008 Redistributable Package](http://www.microsoft.com/en-us/download/details.aspx?id=29).
 
-# COPYRIGHT
+# DEVELOPER INSTRUCTIONS
 
-youtube-dl is released into the public domain by the copyright holders.
+Most users do not need to build youtube-dl and can [download the builds](http://rg3.github.io/youtube-dl/download.html) or get them from their distribution.
 
-This README file was originally written by Daniel Bolton (<https://github.com/dbbolton>) and is likewise released into the public domain.
+To run youtube-dl as a developer, you don't need to build anything either. Simply execute
+
+    python -m youtube_dl
+
+To run the test, simply invoke your favorite test runner, or execute a test file directly; any of the following work:
+
+    python -m unittest discover
+    python test/test_download.py
+    nosetests
+
+If you want to create a build of youtube-dl yourself, you'll need
+
+* python
+* make
+* pandoc
+* zip
+* nosetests
+
+### Adding support for a new site
+
+If you want to add support for a new site, you can follow this quick list (assuming your service is called `yourextractor`):
+
+1. [Fork this repository](https://github.com/rg3/youtube-dl/fork)
+2. Check out the source code with `git clone git@github.com:YOUR_GITHUB_USERNAME/youtube-dl.git`
+3. Start a new git branch with `cd youtube-dl; git checkout -b yourextractor`
+4. Start with this simple template and save it to `youtube_dl/extractor/yourextractor.py`:
+
+        # coding: utf-8
+        from __future__ import unicode_literals
+
+        import re
+
+        from .common import InfoExtractor
+        
+        
+        class YourExtractorIE(InfoExtractor):
+            _VALID_URL = r'https?://(?:www\.)?yourextractor\.com/watch/(?P<id>[0-9]+)'
+            _TEST = {
+                'url': 'http://yourextractor.com/watch/42',
+                'md5': 'TODO: md5 sum of the first 10KiB of the video file',
+                'info_dict': {
+                    'id': '42',
+                    'ext': 'mp4',
+                    'title': 'Video title goes here',
+                    # TODO more properties, either as:
+                    # * A value
+                    # * MD5 checksum; start the string with md5:
+                    # * A regular expression; start the string with re:
+                    # * Any Python type (for example int or float)
+                }
+            }
+
+            def _real_extract(self, url):
+                mobj = re.match(self._VALID_URL, url)
+                video_id = mobj.group('id')
+
+                # TODO more code goes here, for example ...
+                webpage = self._download_webpage(url, video_id)
+                title = self._html_search_regex(r'<h1>(.*?)</h1>', webpage, 'title')
+
+                return {
+                    'id': video_id,
+                    'title': title,
+                    # TODO more properties (see youtube_dl/extractor/common.py)
+                }
+
+
+5. Add an import in [`youtube_dl/extractor/__init__.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/__init__.py).
+6. Run `python test/test_download.py TestDownload.test_YourExtractor`. This *should fail* at first, but you can continually re-run it until you're done.
+7. Have a look at [`youtube_dl/common/extractor/common.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py) for possible helper methods and a [detailed description of what your extractor should return](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L38). Add tests and code for as many as you want.
+8. If you can, check the code with [pyflakes](https://pypi.python.org/pypi/pyflakes) (a good idea) and [pep8](https://pypi.python.org/pypi/pep8) (optional, ignore E501).
+9. When the tests pass, [add](https://www.kernel.org/pub/software/scm/git/docs/git-add.html) the new files and [commit](https://www.kernel.org/pub/software/scm/git/docs/git-commit.html) them and [push](https://www.kernel.org/pub/software/scm/git/docs/git-push.html) the result, like this:
+
+        $ git add youtube_dl/extractor/__init__.py
+        $ git add youtube_dl/extractor/yourextractor.py
+        $ git commit -m '[yourextractor] Add new extractor'
+        $ git push origin yourextractor
+
+10. Finally, [create a pull request](https://help.github.com/articles/creating-a-pull-request). We'll then review and merge it.
+
+In any case, thank you very much for your contributions!
 
 # BUGS
 
@@ -310,7 +462,7 @@ Site support requests must contain an example URL. An example URL is a URL you m
 
 ###  Are you using the latest version?
 
-Before reporting any issue, type youtube-dl -U. This should report that you're up-to-date. Ábout 20% of the reports we receive are already fixed, but people are using outdated versions. This goes for feature requests as well.
+Before reporting any issue, type youtube-dl -U. This should report that you're up-to-date. About 20% of the reports we receive are already fixed, but people are using outdated versions. This goes for feature requests as well.
 
 ###  Is the issue already documented?
 
@@ -335,3 +487,13 @@ In particular, every site support request issue should only pertain to services
 ###  Is anyone going to need the feature?
 
 Only post features that you (or an incapicated friend you can personally talk to) require. Do not post features because they seem like a good idea. If they are really useful, they will be requested by someone who requires them.
+
+###  Is your question about youtube-dl?
+
+It may sound strange, but some bug reports we receive are completely unrelated to youtube-dl and relate to a different or even the reporter's own application. Please make sure that you are actually using youtube-dl. If you are using a UI for youtube-dl, report the bug to the maintainer of the actual application providing the UI. On the other hand, if your UI for youtube-dl fails in some way you believe is related to youtube-dl, by all means, go ahead and report the bug.
+
+# COPYRIGHT
+
+youtube-dl is released into the public domain by the copyright holders.
+
+This README file was originally written by Daniel Bolton (<https://github.com/dbbolton>) and is likewise released into the public domain.

devscripts/bash-completion.in

@@ -6,7 +6,7 @@ __youtube_dl()
     prev="${COMP_WORDS[COMP_CWORD-1]}"
     opts="{{flags}}"
     keywords=":ytfavorites :ytrecommended :ytsubscriptions :ytwatchlater :ythistory"
-    fileopts="-a|--batch-file|--download-archive|--cookies"
+    fileopts="-a|--batch-file|--download-archive|--cookies|--load-info"
     diropts="--cache-dir"
 
     if [[ ${prev} =~ ${fileopts} ]]; then

devscripts/check-porn.py

@@ -3,6 +3,9 @@
 """
 This script employs a VERY basic heuristic ('porn' in webpage.lower()) to check
 if we are not 'age_limit' tagging some porn site
+
+A second approach implemented relies on a list of porn domains, to activate it
+pass the list filename as the only argument
 """
 
 # Allow direct execution
@@ -11,25 +14,42 @@ import sys
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from test.helper import get_testcases
+from youtube_dl.utils import compat_urllib_parse_urlparse
 from youtube_dl.utils import compat_urllib_request
 
+if len(sys.argv) > 1:
+    METHOD = 'LIST'
+    LIST = open(sys.argv[1]).read().decode('utf8').strip()
+else:
+    METHOD = 'EURISTIC'
+
 for test in get_testcases():
-    try:
-        webpage = compat_urllib_request.urlopen(test['url'], timeout=10).read()
-    except:
-        print('\nFail: {0}'.format(test['name']))
-        continue
+    if METHOD == 'EURISTIC':
+        try:
+            webpage = compat_urllib_request.urlopen(test['url'], timeout=10).read()
+        except:
+            print('\nFail: {0}'.format(test['name']))
+            continue
 
-    webpage = webpage.decode('utf8', 'replace')
+        webpage = webpage.decode('utf8', 'replace')
 
-    if 'porn' in webpage.lower() and ('info_dict' not in test
-                                      or 'age_limit' not in test['info_dict']
-                                      or test['info_dict']['age_limit'] != 18):
+        RESULT = 'porn' in webpage.lower()
+
+    elif METHOD == 'LIST':
+        domain = compat_urllib_parse_urlparse(test['url']).netloc
+        if not domain:
+            print('\nFail: {0}'.format(test['name']))
+            continue
+        domain = '.'.join(domain.split('.')[-2:])
+
+        RESULT = ('.' + domain + '\n' in LIST or '\n' + domain + '\n' in LIST)
+
+    if RESULT and ('info_dict' not in test or 'age_limit' not in test['info_dict']
+                   or test['info_dict']['age_limit'] != 18):
         print('\nPotential missing age_limit check: {0}'.format(test['name']))
 
-    elif 'porn' not in webpage.lower() and ('info_dict' in test and
-                                            'age_limit' in test['info_dict'] and
-                                            test['info_dict']['age_limit'] == 18):
+    elif not RESULT and ('info_dict' in test and 'age_limit' in test['info_dict']
+                         and test['info_dict']['age_limit'] == 18):
         print('\nPotential false negative: {0}'.format(test['name']))
 
     else:

devscripts/gh-pages/update-feed.py

@@ -1,56 +1,76 @@
 #!/usr/bin/env python3
 
 import datetime
-
+import io
+import json
 import textwrap
 
-import json
 
-atom_template=textwrap.dedent("""\
-								<?xml version='1.0' encoding='utf-8'?>
-								<atom:feed xmlns:atom="http://www.w3.org/2005/Atom">
-									<atom:title>youtube-dl releases</atom:title>
-									<atom:id>youtube-dl-updates-feed</atom:id>
-									<atom:updated>@TIMESTAMP@</atom:updated>
-									@ENTRIES@
-								</atom:feed>""")
+atom_template = textwrap.dedent("""\
+    <?xml version="1.0" encoding="utf-8"?>
+    <feed xmlns="http://www.w3.org/2005/Atom">
+        <link rel="self" href="http://rg3.github.io/youtube-dl/update/releases.atom" />
+        <title>youtube-dl releases</title>
+        <id>https://yt-dl.org/feed/youtube-dl-updates-feed</id>
+        <updated>@TIMESTAMP@</updated>
+        @ENTRIES@
+    </feed>""")
 
-entry_template=textwrap.dedent("""
-								<atom:entry>
-									<atom:id>youtube-dl-@VERSION@</atom:id>
-									<atom:title>New version @VERSION@</atom:title>
-									<atom:link href="http://rg3.github.io/youtube-dl" />
-									<atom:content type="xhtml">
-										<div xmlns="http://www.w3.org/1999/xhtml">
-											Downloads available at <a href="https://yt-dl.org/downloads/@VERSION@/">https://yt-dl.org/downloads/@VERSION@/</a>
-										</div>
-									</atom:content>
-									<atom:author>
-										<atom:name>The youtube-dl maintainers</atom:name>
-									</atom:author>
-									<atom:updated>@TIMESTAMP@</atom:updated>
-								</atom:entry>
-								""")
+entry_template = textwrap.dedent("""
+    <entry>
+        <id>https://yt-dl.org/feed/youtube-dl-updates-feed/youtube-dl-@VERSION@</id>
+        <title>New version @VERSION@</title>
+        <link href="http://rg3.github.io/youtube-dl" />
+        <content type="xhtml">
+            <div xmlns="http://www.w3.org/1999/xhtml">
+                Downloads available at <a href="https://yt-dl.org/downloads/@VERSION@/">https://yt-dl.org/downloads/@VERSION@/</a>
+            </div>
+        </content>
+        <author>
+            <name>The youtube-dl maintainers</name>
+        </author>
+        <updated>@TIMESTAMP@</updated>
+    </entry>
+    """)
 
 now = datetime.datetime.now()
-now_iso = now.isoformat()
+now_iso = now.isoformat() + 'Z'
 
-atom_template = atom_template.replace('@TIMESTAMP@',now_iso)
-
-entries=[]
+atom_template = atom_template.replace('@TIMESTAMP@', now_iso)
 
 versions_info = json.load(open('update/versions.json'))
 versions = list(versions_info['versions'].keys())
 versions.sort()
 
+entries = []
 for v in versions:
-	entry = entry_template.replace('@TIMESTAMP@',v.replace('.','-'))
-	entry = entry.replace('@VERSION@',v)
-	entries.append(entry)
+    fields = v.split('.')
+    year, month, day = map(int, fields[:3])
+    faked = 0
+    patchlevel = 0
+    while True:
+        try:
+            datetime.date(year, month, day)
+        except ValueError:
+            day -= 1
+            faked += 1
+            assert day > 0
+            continue
+        break
+    if len(fields) >= 4:
+        try:
+            patchlevel = int(fields[3])
+        except ValueError:
+            patchlevel = 1
+    timestamp = '%04d-%02d-%02dT00:%02d:%02dZ' % (year, month, day, faked, patchlevel)
+
+    entry = entry_template.replace('@TIMESTAMP@', timestamp)
+    entry = entry.replace('@VERSION@', v)
+    entries.append(entry)
 
 entries_str = textwrap.indent(''.join(entries), '\t')
 atom_template = atom_template.replace('@ENTRIES@', entries_str)
 
-with open('update/releases.atom','w',encoding='utf-8') as atom_file:
-	atom_file.write(atom_template)
+with io.open('update/releases.atom', 'w', encoding='utf-8') as atom_file:
+    atom_file.write(atom_template)
 

devscripts/make_readme.py

@@ -1,20 +1,24 @@
+import io
 import sys
 import re
 
 README_FILE = 'README.md'
 helptext = sys.stdin.read()
 
-with open(README_FILE) as f:
+if isinstance(helptext, bytes):
+    helptext = helptext.decode('utf-8')
+
+with io.open(README_FILE, encoding='utf-8') as f:
     oldreadme = f.read()
 
 header = oldreadme[:oldreadme.index('# OPTIONS')]
 footer = oldreadme[oldreadme.index('# CONFIGURATION'):]
 
-options = helptext[helptext.index('  General Options:')+19:]
+options = helptext[helptext.index('  General Options:') + 19:]
 options = re.sub(r'^  (\w.+)$', r'## \1', options, flags=re.M)
 options = '# OPTIONS\n' + options + '\n'
 
-with open(README_FILE, 'w') as f:
+with io.open(README_FILE, 'w', encoding='utf-8') as f:
     f.write(header)
     f.write(options)
     f.write(footer)

devscripts/release.sh

@@ -14,16 +14,24 @@
 
 set -e
 
-skip_tests=false
-if [ "$1" = '--skip-test' ]; then
-    skip_tests=true
+skip_tests=true
+if [ "$1" = '--run-tests' ]; then
+    skip_tests=false
     shift
 fi
 
 if [ -z "$1" ]; then echo "ERROR: specify version number like this: $0 1994.09.06"; exit 1; fi
 version="$1"
+major_version=$(echo "$version" | sed -n 's#^\([0-9]*\.[0-9]*\.[0-9]*\).*#\1#p')
+if test "$major_version" '!=' "$(date '+%Y.%m.%d')"; then
+    echo "$version does not start with today's date!"
+    exit 1
+fi
+
 if [ ! -z "`git tag | grep "$version"`" ]; then echo 'ERROR: version already present'; exit 1; fi
 if [ ! -z "`git status --porcelain | grep -v CHANGELOG`" ]; then echo 'ERROR: the working directory is not clean; commit or stash changes'; exit 1; fi
+useless_files=$(find youtube_dl -type f -not -name '*.py')
+if [ ! -z "$useless_files" ]; then echo "ERROR: Non-.py files in youtube_dl: $useless_files"; exit 1; fi
 if [ ! -f "updates_key.pem" ]; then echo 'ERROR: updates_key.pem missing'; exit 1; fi
 
 /bin/echo -e "\n### First of all, testing..."
@@ -68,7 +76,7 @@ RELEASE_FILES="youtube-dl youtube-dl.exe youtube-dl-$version.tar.gz"
 git checkout HEAD -- youtube-dl youtube-dl.exe
 
 /bin/echo -e "\n### Signing and uploading the new binaries to yt-dl.org ..."
-for f in $RELEASE_FILES; do gpg --detach-sig "build/$version/$f"; done
+for f in $RELEASE_FILES; do gpg --passphrase-repeat 5 --detach-sig "build/$version/$f"; done
 scp -r "build/$version" ytdl@yt-dl.org:html/tmp/
 ssh ytdl@yt-dl.org "mv html/tmp/$version html/downloads/"
 ssh ytdl@yt-dl.org "sh html/update_latest.sh $version"
@@ -95,7 +103,7 @@ rm -rf build
 
 make pypi-files
 echo "Uploading to PyPi ..."
-python setup.py sdist upload
+python setup.py sdist bdist_wheel upload
 make clean
 
 /bin/echo -e "\n### DONE!"

docs/.gitignore

@@ -0,0 +1 @@
+_build/

docs/Makefile

@@ -0,0 +1,177 @@
+# Makefile for Sphinx documentation
+#
+
+# You can set these variables from the command line.
+SPHINXOPTS    =
+SPHINXBUILD   = sphinx-build
+PAPER         =
+BUILDDIR      = _build
+
+# User-friendly check for sphinx-build
+ifeq ($(shell which $(SPHINXBUILD) >/dev/null 2>&1; echo $$?), 1)
+$(error The '$(SPHINXBUILD)' command was not found. Make sure you have Sphinx installed, then set the SPHINXBUILD environment variable to point to the full path of the '$(SPHINXBUILD)' executable. Alternatively you can add the directory with the executable to your PATH. If you don't have Sphinx installed, grab it from http://sphinx-doc.org/)
+endif
+
+# Internal variables.
+PAPEROPT_a4     = -D latex_paper_size=a4
+PAPEROPT_letter = -D latex_paper_size=letter
+ALLSPHINXOPTS   = -d $(BUILDDIR)/doctrees $(PAPEROPT_$(PAPER)) $(SPHINXOPTS) .
+# the i18n builder cannot share the environment and doctrees with the others
+I18NSPHINXOPTS  = $(PAPEROPT_$(PAPER)) $(SPHINXOPTS) .
+
+.PHONY: help clean html dirhtml singlehtml pickle json htmlhelp qthelp devhelp epub latex latexpdf text man changes linkcheck doctest gettext
+
+help:
+	@echo "Please use \`make <target>' where <target> is one of"
+	@echo "  html       to make standalone HTML files"
+	@echo "  dirhtml    to make HTML files named index.html in directories"
+	@echo "  singlehtml to make a single large HTML file"
+	@echo "  pickle     to make pickle files"
+	@echo "  json       to make JSON files"
+	@echo "  htmlhelp   to make HTML files and a HTML help project"
+	@echo "  qthelp     to make HTML files and a qthelp project"
+	@echo "  devhelp    to make HTML files and a Devhelp project"
+	@echo "  epub       to make an epub"
+	@echo "  latex      to make LaTeX files, you can set PAPER=a4 or PAPER=letter"
+	@echo "  latexpdf   to make LaTeX files and run them through pdflatex"
+	@echo "  latexpdfja to make LaTeX files and run them through platex/dvipdfmx"
+	@echo "  text       to make text files"
+	@echo "  man        to make manual pages"
+	@echo "  texinfo    to make Texinfo files"
+	@echo "  info       to make Texinfo files and run them through makeinfo"
+	@echo "  gettext    to make PO message catalogs"
+	@echo "  changes    to make an overview of all changed/added/deprecated items"
+	@echo "  xml        to make Docutils-native XML files"
+	@echo "  pseudoxml  to make pseudoxml-XML files for display purposes"
+	@echo "  linkcheck  to check all external links for integrity"
+	@echo "  doctest    to run all doctests embedded in the documentation (if enabled)"
+
+clean:
+	rm -rf $(BUILDDIR)/*
+
+html:
+	$(SPHINXBUILD) -b html $(ALLSPHINXOPTS) $(BUILDDIR)/html
+	@echo
+	@echo "Build finished. The HTML pages are in $(BUILDDIR)/html."
+
+dirhtml:
+	$(SPHINXBUILD) -b dirhtml $(ALLSPHINXOPTS) $(BUILDDIR)/dirhtml
+	@echo
+	@echo "Build finished. The HTML pages are in $(BUILDDIR)/dirhtml."
+
+singlehtml:
+	$(SPHINXBUILD) -b singlehtml $(ALLSPHINXOPTS) $(BUILDDIR)/singlehtml
+	@echo
+	@echo "Build finished. The HTML page is in $(BUILDDIR)/singlehtml."
+
+pickle:
+	$(SPHINXBUILD) -b pickle $(ALLSPHINXOPTS) $(BUILDDIR)/pickle
+	@echo
+	@echo "Build finished; now you can process the pickle files."
+
+json:
+	$(SPHINXBUILD) -b json $(ALLSPHINXOPTS) $(BUILDDIR)/json
+	@echo
+	@echo "Build finished; now you can process the JSON files."
+
+htmlhelp:
+	$(SPHINXBUILD) -b htmlhelp $(ALLSPHINXOPTS) $(BUILDDIR)/htmlhelp
+	@echo
+	@echo "Build finished; now you can run HTML Help Workshop with the" \
+	      ".hhp project file in $(BUILDDIR)/htmlhelp."
+
+qthelp:
+	$(SPHINXBUILD) -b qthelp $(ALLSPHINXOPTS) $(BUILDDIR)/qthelp
+	@echo
+	@echo "Build finished; now you can run "qcollectiongenerator" with the" \
+	      ".qhcp project file in $(BUILDDIR)/qthelp, like this:"
+	@echo "# qcollectiongenerator $(BUILDDIR)/qthelp/youtube-dl.qhcp"
+	@echo "To view the help file:"
+	@echo "# assistant -collectionFile $(BUILDDIR)/qthelp/youtube-dl.qhc"
+
+devhelp:
+	$(SPHINXBUILD) -b devhelp $(ALLSPHINXOPTS) $(BUILDDIR)/devhelp
+	@echo
+	@echo "Build finished."
+	@echo "To view the help file:"
+	@echo "# mkdir -p $$HOME/.local/share/devhelp/youtube-dl"
+	@echo "# ln -s $(BUILDDIR)/devhelp $$HOME/.local/share/devhelp/youtube-dl"
+	@echo "# devhelp"
+
+epub:
+	$(SPHINXBUILD) -b epub $(ALLSPHINXOPTS) $(BUILDDIR)/epub
+	@echo
+	@echo "Build finished. The epub file is in $(BUILDDIR)/epub."
+
+latex:
+	$(SPHINXBUILD) -b latex $(ALLSPHINXOPTS) $(BUILDDIR)/latex
+	@echo
+	@echo "Build finished; the LaTeX files are in $(BUILDDIR)/latex."
+	@echo "Run \`make' in that directory to run these through (pdf)latex" \
+	      "(use \`make latexpdf' here to do that automatically)."
+
+latexpdf:
+	$(SPHINXBUILD) -b latex $(ALLSPHINXOPTS) $(BUILDDIR)/latex
+	@echo "Running LaTeX files through pdflatex..."
+	$(MAKE) -C $(BUILDDIR)/latex all-pdf
+	@echo "pdflatex finished; the PDF files are in $(BUILDDIR)/latex."
+
+latexpdfja:
+	$(SPHINXBUILD) -b latex $(ALLSPHINXOPTS) $(BUILDDIR)/latex
+	@echo "Running LaTeX files through platex and dvipdfmx..."
+	$(MAKE) -C $(BUILDDIR)/latex all-pdf-ja
+	@echo "pdflatex finished; the PDF files are in $(BUILDDIR)/latex."
+
+text:
+	$(SPHINXBUILD) -b text $(ALLSPHINXOPTS) $(BUILDDIR)/text
+	@echo
+	@echo "Build finished. The text files are in $(BUILDDIR)/text."
+
+man:
+	$(SPHINXBUILD) -b man $(ALLSPHINXOPTS) $(BUILDDIR)/man
+	@echo
+	@echo "Build finished. The manual pages are in $(BUILDDIR)/man."
+
+texinfo:
+	$(SPHINXBUILD) -b texinfo $(ALLSPHINXOPTS) $(BUILDDIR)/texinfo
+	@echo
+	@echo "Build finished. The Texinfo files are in $(BUILDDIR)/texinfo."
+	@echo "Run \`make' in that directory to run these through makeinfo" \
+	      "(use \`make info' here to do that automatically)."
+
+info:
+	$(SPHINXBUILD) -b texinfo $(ALLSPHINXOPTS) $(BUILDDIR)/texinfo
+	@echo "Running Texinfo files through makeinfo..."
+	make -C $(BUILDDIR)/texinfo info
+	@echo "makeinfo finished; the Info files are in $(BUILDDIR)/texinfo."
+
+gettext:
+	$(SPHINXBUILD) -b gettext $(I18NSPHINXOPTS) $(BUILDDIR)/locale
+	@echo
+	@echo "Build finished. The message catalogs are in $(BUILDDIR)/locale."
+
+changes:
+	$(SPHINXBUILD) -b changes $(ALLSPHINXOPTS) $(BUILDDIR)/changes
+	@echo
+	@echo "The overview file is in $(BUILDDIR)/changes."
+
+linkcheck:
+	$(SPHINXBUILD) -b linkcheck $(ALLSPHINXOPTS) $(BUILDDIR)/linkcheck
+	@echo
+	@echo "Link check complete; look for any errors in the above output " \
+	      "or in $(BUILDDIR)/linkcheck/output.txt."
+
+doctest:
+	$(SPHINXBUILD) -b doctest $(ALLSPHINXOPTS) $(BUILDDIR)/doctest
+	@echo "Testing of doctests in the sources finished, look at the " \
+	      "results in $(BUILDDIR)/doctest/output.txt."
+
+xml:
+	$(SPHINXBUILD) -b xml $(ALLSPHINXOPTS) $(BUILDDIR)/xml
+	@echo
+	@echo "Build finished. The XML files are in $(BUILDDIR)/xml."
+
+pseudoxml:
+	$(SPHINXBUILD) -b pseudoxml $(ALLSPHINXOPTS) $(BUILDDIR)/pseudoxml
+	@echo
+	@echo "Build finished. The pseudo-XML files are in $(BUILDDIR)/pseudoxml."

docs/conf.py

@@ -0,0 +1,71 @@
+# -*- coding: utf-8 -*-
+#
+# youtube-dl documentation build configuration file, created by
+# sphinx-quickstart on Fri Mar 14 21:05:43 2014.
+#
+# This file is execfile()d with the current directory set to its
+# containing dir.
+#
+# Note that not all possible configuration values are present in this
+# autogenerated file.
+#
+# All configuration values have a default; values that are commented out
+# serve to show the default.
+
+import sys
+import os
+# Allows to import youtube_dl
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+# -- General configuration ------------------------------------------------
+
+# Add any Sphinx extension module names here, as strings. They can be
+# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
+# ones.
+extensions = [
+    'sphinx.ext.autodoc',
+]
+
+# Add any paths that contain templates here, relative to this directory.
+templates_path = ['_templates']
+
+# The suffix of source filenames.
+source_suffix = '.rst'
+
+# The master toctree document.
+master_doc = 'index'
+
+# General information about the project.
+project = u'youtube-dl'
+copyright = u'2014, Ricardo Garcia Gonzalez'
+
+# The version info for the project you're documenting, acts as replacement for
+# |version| and |release|, also used in various other places throughout the
+# built documents.
+#
+# The short X.Y version.
+import youtube_dl
+version = youtube_dl.__version__
+# The full version, including alpha/beta/rc tags.
+release = version
+
+# List of patterns, relative to source directory, that match files and
+# directories to ignore when looking for source files.
+exclude_patterns = ['_build']
+
+# The name of the Pygments (syntax highlighting) style to use.
+pygments_style = 'sphinx'
+
+# -- Options for HTML output ----------------------------------------------
+
+# The theme to use for HTML and HTML Help pages.  See the documentation for
+# a list of builtin themes.
+html_theme = 'default'
+
+# Add any paths that contain custom static files (such as style sheets) here,
+# relative to this directory. They are copied after the builtin static files,
+# so a file named "default.css" will overwrite the builtin "default.css".
+html_static_path = ['_static']
+
+# Output file base name for HTML help builder.
+htmlhelp_basename = 'youtube-dldoc'

docs/index.rst

@@ -0,0 +1,23 @@
+Welcome to youtube-dl's documentation!
+======================================
+
+*youtube-dl* is a command-line program to download videos from YouTube.com and more sites.
+It can also be used in Python code.
+
+Developer guide
+---------------
+
+This section contains information for using *youtube-dl* from Python programs.
+
+.. toctree::
+    :maxdepth: 2
+
+    module_guide
+
+Indices and tables
+==================
+
+* :ref:`genindex`
+* :ref:`modindex`
+* :ref:`search`
+

docs/module_guide.rst

@@ -0,0 +1,67 @@
+Using the ``youtube_dl`` module
+===============================
+
+When using the ``youtube_dl`` module, you start by creating an instance of :class:`YoutubeDL` and adding all the available extractors:
+
+.. code-block:: python
+
+    >>> from youtube_dl import YoutubeDL
+    >>> ydl = YoutubeDL()
+    >>> ydl.add_default_info_extractors()
+
+Extracting video information
+----------------------------
+
+You use the :meth:`YoutubeDL.extract_info` method for getting the video information, which returns a dictionary:
+
+.. code-block:: python
+
+    >>> info = ydl.extract_info('http://www.youtube.com/watch?v=BaW_jenozKc', download=False)
+    [youtube] Setting language
+    [youtube] BaW_jenozKc: Downloading webpage
+    [youtube] BaW_jenozKc: Downloading video info webpage
+    [youtube] BaW_jenozKc: Extracting video information
+    >>> info['title']
+    'youtube-dl test video "\'/\\ä↭𝕐'
+    >>> info['height'], info['width']
+    (720, 1280)
+
+If you want to download or play the video you can get its url:
+
+.. code-block:: python
+
+    >>> info['url']
+    'https://...'
+
+Extracting playlist information
+-------------------------------
+
+The playlist information is extracted in a similar way, but the dictionary is a bit different:
+
+.. code-block:: python
+
+    >>> playlist = ydl.extract_info('http://www.ted.com/playlists/13/open_source_open_world', download=False)
+    [TED] open_source_open_world: Downloading playlist webpage
+    ...
+    >>> playlist['title']
+    'Open-source, open world'
+
+
+
+You can access the videos in the playlist with the ``entries`` field:
+
+.. code-block:: python
+
+    >>> for video in playlist['entries']:
+    ...     print('Video #%d: %s' % (video['playlist_index'], video['title']))
+
+    Video #1: How Arduino is open-sourcing imagination
+    Video #2: The year open data went worldwide
+    Video #3: Massive-scale online collaboration
+    Video #4: The art of asking
+    Video #5: How cognitive surplus will change the world
+    Video #6: The birth of Wikipedia
+    Video #7: Coding a better government
+    Video #8: The era of open innovation
+    Video #9: The currency of the new economy is trust
+

setup.cfg

@@ -0,0 +1,2 @@
+[wheel]
+universal = True

setup.py

@@ -3,7 +3,9 @@
 
 from __future__ import print_function
 
+import os.path
 import pkg_resources
+import warnings
 import sys
 
 try:
@@ -44,12 +46,24 @@ py2exe_params = {
 if len(sys.argv) >= 2 and sys.argv[1] == 'py2exe':
     params = py2exe_params
 else:
+    files_spec = [
+        ('etc/bash_completion.d', ['youtube-dl.bash-completion']),
+        ('share/doc/youtube_dl', ['README.txt']),
+        ('share/man/man1', ['youtube-dl.1'])
+    ]
+    root = os.path.dirname(os.path.abspath(__file__))
+    data_files = []
+    for dirname, files in files_spec:
+        resfiles = []
+        for fn in files:
+            if not os.path.exists(fn):
+                warnings.warn('Skipping file %s since it is not present. Type  make  to build all automatically generated files.' % fn)
+            else:
+                resfiles.append(fn)
+        data_files.append((dirname, resfiles))
+
     params = {
-        'data_files': [  # Installing system-wide would require sudo...
-            ('etc/bash_completion.d', ['youtube-dl.bash-completion']),
-            ('share/doc/youtube_dl', ['README.txt']),
-            ('share/man/man1', ['youtube-dl.1'])
-        ]
+        'data_files': data_files,
     }
     if setuptools_available:
         params['entry_points'] = {'console_scripts': ['youtube-dl = youtube_dl:main']}
@@ -71,7 +85,10 @@ setup(
     author_email='ytdl@yt-dl.org',
     maintainer='Philipp Hagemeister',
     maintainer_email='phihag@phihag.de',
-    packages=['youtube_dl', 'youtube_dl.extractor'],
+    packages=[
+        'youtube_dl',
+        'youtube_dl.extractor', 'youtube_dl.downloader',
+        'youtube_dl.postprocessor'],
 
     # Provokes warning on most systems (why?!)
     # test_suite = 'nose.collector',

test/helper.py

@@ -9,7 +9,10 @@ import sys
 
 import youtube_dl.extractor
 from youtube_dl import YoutubeDL
-from youtube_dl.utils import preferredencoding
+from youtube_dl.utils import (
+    compat_str,
+    preferredencoding,
+)
 
 
 def get_params(override=None):
@@ -71,15 +74,77 @@ class FakeYDL(YoutubeDL):
             old_report_warning(message)
         self.report_warning = types.MethodType(report_warning, self)
 
-def get_testcases():
+
+def gettestcases(include_onlymatching=False):
     for ie in youtube_dl.extractor.gen_extractors():
         t = getattr(ie, '_TEST', None)
         if t:
-            t['name'] = type(ie).__name__[:-len('IE')]
-            yield t
-        for t in getattr(ie, '_TESTS', []):
+            assert not hasattr(ie, '_TESTS'), \
+                '%s has _TEST and _TESTS' % type(ie).__name__
+            tests = [t]
+        else:
+            tests = getattr(ie, '_TESTS', [])
+        for t in tests:
+            if not include_onlymatching and t.get('only_matching', False):
+                continue
             t['name'] = type(ie).__name__[:-len('IE')]
             yield t
 
 
 md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
+
+
+def expect_info_dict(self, expected_dict, got_dict):
+    for info_field, expected in expected_dict.items():
+        if isinstance(expected, compat_str) and expected.startswith('re:'):
+            got = got_dict.get(info_field)
+            match_str = expected[len('re:'):]
+            match_rex = re.compile(match_str)
+
+            self.assertTrue(
+                isinstance(got, compat_str) and match_rex.match(got),
+                u'field %s (value: %r) should match %r' % (info_field, got, match_str))
+        elif isinstance(expected, type):
+            got = got_dict.get(info_field)
+            self.assertTrue(isinstance(got, expected),
+                u'Expected type %r, but got value %r of type %r' % (expected, got, type(got)))
+        else:
+            if isinstance(expected, compat_str) and expected.startswith('md5:'):
+                got = 'md5:' + md5(got_dict.get(info_field))
+            else:
+                got = got_dict.get(info_field)
+            self.assertEqual(expected, got,
+                u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
+
+    # Check for the presence of mandatory fields
+    for key in ('id', 'url', 'title', 'ext'):
+        self.assertTrue(got_dict.get(key), 'Missing mandatory field %s' % key)
+    # Check for mandatory fields that are automatically set by YoutubeDL
+    for key in ['webpage_url', 'extractor', 'extractor_key']:
+        self.assertTrue(got_dict.get(key), u'Missing field: %s' % key)
+
+    # Are checkable fields missing from the test case definition?
+    test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))
+        for key, value in got_dict.items()
+        if value and key in ('title', 'description', 'uploader', 'upload_date', 'timestamp', 'uploader_id', 'location'))
+    missing_keys = set(test_info_dict.keys()) - set(expected_dict.keys())
+    if missing_keys:
+        sys.stderr.write(u'\n"info_dict": ' + json.dumps(test_info_dict, ensure_ascii=False, indent=4) + u'\n')
+        self.assertFalse(
+            missing_keys,
+            'Missing keys in test definition: %s' % (
+                ', '.join(sorted(missing_keys))))
+
+
+def assertRegexpMatches(self, text, regexp, msg=None):
+    if hasattr(self, 'assertRegexpMatches'):
+        return self.assertRegexpMatches(text, regexp, msg)
+    else:
+        m = re.match(regexp, text)
+        if not m:
+            note = 'Regexp didn\'t match: %r not found in %r' % (regexp, text)
+            if msg is None:
+                msg = note
+            else:
+                msg = note + ', ' + msg
+            self.assertTrue(m, msg)

test/test_InfoExtractor.py

@@ -0,0 +1,44 @@
+#!/usr/bin/env python
+
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import FakeYDL
+from youtube_dl.extractor.common import InfoExtractor
+from youtube_dl.extractor import YoutubeIE, get_info_extractor
+
+
+class TestIE(InfoExtractor):
+    pass
+
+
+class TestInfoExtractor(unittest.TestCase):
+    def setUp(self):
+        self.ie = TestIE(FakeYDL())
+
+    def test_ie_key(self):
+        self.assertEqual(get_info_extractor(YoutubeIE.ie_key()), YoutubeIE)
+
+    def test_html_search_regex(self):
+        html = '<p id="foo">Watch this <a href="http://www.youtube.com/watch?v=BaW_jenozKc">video</a></p>'
+        search = lambda re, *args: self.ie._html_search_regex(re, html, *args)
+        self.assertEqual(search(r'<p id="foo">(.+?)</p>', 'foo'), 'Watch this video')
+
+    def test_opengraph(self):
+        ie = self.ie
+        html = '''
+            <meta name="og:title" content='Foo'/>
+            <meta content="Some video's description " name="og:description"/>
+            <meta property='og:image' content='http://domain.com/pic.jpg?key1=val1&amp;key2=val2'/>
+            '''
+        self.assertEqual(ie._og_search_title(html), 'Foo')
+        self.assertEqual(ie._og_search_description(html), 'Some video\'s description ')
+        self.assertEqual(ie._og_search_thumbnail(html), 'http://domain.com/pic.jpg?key1=val1&key2=val2')
+
+if __name__ == '__main__':
+    unittest.main()

test/test_YoutubeDL.py

@@ -1,13 +1,16 @@
 #!/usr/bin/env python
 
+from __future__ import unicode_literals
+
 # Allow direct execution
 import os
 import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import FakeYDL
+from test.helper import FakeYDL, assertRegexpMatches
 from youtube_dl import YoutubeDL
+from youtube_dl.extractor import YoutubeIE
 
 
 class YDL(FakeYDL):
@@ -23,111 +26,227 @@ class YDL(FakeYDL):
         self.msgs.append(msg)
 
 
+def _make_result(formats, **kwargs):
+    res = {
+        'formats': formats,
+        'id': 'testid',
+        'title': 'testttitle',
+        'extractor': 'testex',
+    }
+    res.update(**kwargs)
+    return res
+
+
 class TestFormatSelection(unittest.TestCase):
     def test_prefer_free_formats(self):
         # Same resolution => download webm
         ydl = YDL()
         ydl.params['prefer_free_formats'] = True
         formats = [
-            {u'ext': u'webm', u'height': 460},
-            {u'ext': u'mp4',  u'height': 460},
+            {'ext': 'webm', 'height': 460, 'url': 'x'},
+            {'ext': 'mp4', 'height': 460, 'url': 'y'},
         ]
-        info_dict = {u'formats': formats, u'extractor': u'test'}
+        info_dict = _make_result(formats)
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'ext'], u'webm')
+        self.assertEqual(downloaded['ext'], 'webm')
 
         # Different resolution => download best quality (mp4)
         ydl = YDL()
         ydl.params['prefer_free_formats'] = True
         formats = [
-            {u'ext': u'webm', u'height': 720},
-            {u'ext': u'mp4', u'height': 1080},
+            {'ext': 'webm', 'height': 720, 'url': 'a'},
+            {'ext': 'mp4', 'height': 1080, 'url': 'b'},
         ]
-        info_dict[u'formats'] = formats
+        info_dict['formats'] = formats
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'ext'], u'mp4')
+        self.assertEqual(downloaded['ext'], 'mp4')
 
-        # No prefer_free_formats => keep original formats order
+        # No prefer_free_formats => prefer mp4 and flv for greater compatibilty
         ydl = YDL()
         ydl.params['prefer_free_formats'] = False
         formats = [
-            {u'ext': u'webm', u'height': 720},
-            {u'ext': u'flv', u'height': 720},
+            {'ext': 'webm', 'height': 720, 'url': '_'},
+            {'ext': 'mp4', 'height': 720, 'url': '_'},
+            {'ext': 'flv', 'height': 720, 'url': '_'},
         ]
-        info_dict[u'formats'] = formats
+        info_dict['formats'] = formats
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'ext'], u'flv')
+        self.assertEqual(downloaded['ext'], 'mp4')
+
+        ydl = YDL()
+        ydl.params['prefer_free_formats'] = False
+        formats = [
+            {'ext': 'flv', 'height': 720, 'url': '_'},
+            {'ext': 'webm', 'height': 720, 'url': '_'},
+        ]
+        info_dict['formats'] = formats
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['ext'], 'flv')
 
     def test_format_limit(self):
         formats = [
-            {u'format_id': u'meh', u'url': u'http://example.com/meh'},
-            {u'format_id': u'good', u'url': u'http://example.com/good'},
-            {u'format_id': u'great', u'url': u'http://example.com/great'},
-            {u'format_id': u'excellent', u'url': u'http://example.com/exc'},
+            {'format_id': 'meh', 'url': 'http://example.com/meh', 'preference': 1},
+            {'format_id': 'good', 'url': 'http://example.com/good', 'preference': 2},
+            {'format_id': 'great', 'url': 'http://example.com/great', 'preference': 3},
+            {'format_id': 'excellent', 'url': 'http://example.com/exc', 'preference': 4},
         ]
-        info_dict = {
-            u'formats': formats, u'extractor': u'test', 'id': 'testvid'}
+        info_dict = _make_result(formats)
 
         ydl = YDL()
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'format_id'], u'excellent')
+        self.assertEqual(downloaded['format_id'], 'excellent')
 
         ydl = YDL({'format_limit': 'good'})
         assert ydl.params['format_limit'] == 'good'
-        ydl.process_ie_result(info_dict)
+        ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'format_id'], u'good')
+        self.assertEqual(downloaded['format_id'], 'good')
 
         ydl = YDL({'format_limit': 'great', 'format': 'all'})
-        ydl.process_ie_result(info_dict)
-        self.assertEqual(ydl.downloaded_info_dicts[0][u'format_id'], u'meh')
-        self.assertEqual(ydl.downloaded_info_dicts[1][u'format_id'], u'good')
-        self.assertEqual(ydl.downloaded_info_dicts[2][u'format_id'], u'great')
+        ydl.process_ie_result(info_dict.copy())
+        self.assertEqual(ydl.downloaded_info_dicts[0]['format_id'], 'meh')
+        self.assertEqual(ydl.downloaded_info_dicts[1]['format_id'], 'good')
+        self.assertEqual(ydl.downloaded_info_dicts[2]['format_id'], 'great')
         self.assertTrue('3' in ydl.msgs[0])
 
         ydl = YDL()
         ydl.params['format_limit'] = 'excellent'
-        ydl.process_ie_result(info_dict)
+        ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'format_id'], u'excellent')
+        self.assertEqual(downloaded['format_id'], 'excellent')
 
     def test_format_selection(self):
         formats = [
-            {u'format_id': u'35', u'ext': u'mp4'},
-            {u'format_id': u'45', u'ext': u'webm'},
-            {u'format_id': u'47', u'ext': u'webm'},
-            {u'format_id': u'2', u'ext': u'flv'},
+            {'format_id': '35', 'ext': 'mp4', 'preference': 1, 'url': '_'},
+            {'format_id': '45', 'ext': 'webm', 'preference': 2, 'url': '_'},
+            {'format_id': '47', 'ext': 'webm', 'preference': 3, 'url': '_'},
+            {'format_id': '2', 'ext': 'flv', 'preference': 4, 'url': '_'},
         ]
-        info_dict = {u'formats': formats, u'extractor': u'test'}
+        info_dict = _make_result(formats)
 
-        ydl = YDL({'format': u'20/47'})
-        ydl.process_ie_result(info_dict)
+        ydl = YDL({'format': '20/47'})
+        ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'47')
+        self.assertEqual(downloaded['format_id'], '47')
 
-        ydl = YDL({'format': u'20/71/worst'})
-        ydl.process_ie_result(info_dict)
+        ydl = YDL({'format': '20/71/worst'})
+        ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'35')
+        self.assertEqual(downloaded['format_id'], '35')
 
         ydl = YDL()
-        ydl.process_ie_result(info_dict)
+        ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'2')
+        self.assertEqual(downloaded['format_id'], '2')
 
-        ydl = YDL({'format': u'webm/mp4'})
-        ydl.process_ie_result(info_dict)
+        ydl = YDL({'format': 'webm/mp4'})
+        ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'47')
+        self.assertEqual(downloaded['format_id'], '47')
 
-        ydl = YDL({'format': u'3gp/40/mp4'})
-        ydl.process_ie_result(info_dict)
+        ydl = YDL({'format': '3gp/40/mp4'})
+        ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'35')
+        self.assertEqual(downloaded['format_id'], '35')
+
+    def test_format_selection_audio(self):
+        formats = [
+            {'format_id': 'audio-low', 'ext': 'webm', 'preference': 1, 'vcodec': 'none', 'url': '_'},
+            {'format_id': 'audio-mid', 'ext': 'webm', 'preference': 2, 'vcodec': 'none', 'url': '_'},
+            {'format_id': 'audio-high', 'ext': 'flv', 'preference': 3, 'vcodec': 'none', 'url': '_'},
+            {'format_id': 'vid', 'ext': 'mp4', 'preference': 4, 'url': '_'},
+        ]
+        info_dict = _make_result(formats)
+
+        ydl = YDL({'format': 'bestaudio'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'audio-high')
+
+        ydl = YDL({'format': 'worstaudio'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'audio-low')
+
+        formats = [
+            {'format_id': 'vid-low', 'ext': 'mp4', 'preference': 1, 'url': '_'},
+            {'format_id': 'vid-high', 'ext': 'mp4', 'preference': 2, 'url': '_'},
+        ]
+        info_dict = _make_result(formats)
+
+        ydl = YDL({'format': 'bestaudio/worstaudio/best'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'vid-high')
+
+    def test_format_selection_video(self):
+        formats = [
+            {'format_id': 'dash-video-low', 'ext': 'mp4', 'preference': 1, 'acodec': 'none', 'url': '_'},
+            {'format_id': 'dash-video-high', 'ext': 'mp4', 'preference': 2, 'acodec': 'none', 'url': '_'},
+            {'format_id': 'vid', 'ext': 'mp4', 'preference': 3, 'url': '_'},
+        ]
+        info_dict = _make_result(formats)
+
+        ydl = YDL({'format': 'bestvideo'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'dash-video-high')
+
+        ydl = YDL({'format': 'worstvideo'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'dash-video-low')
+
+    def test_youtube_format_selection(self):
+        order = [
+            '38', '37', '46', '22', '45', '35', '44', '18', '34', '43', '6', '5', '36', '17', '13',
+            # Apple HTTP Live Streaming
+            '96', '95', '94', '93', '92', '132', '151',
+            # 3D
+            '85', '84', '102', '83', '101', '82', '100',
+            # Dash video
+            '138', '137', '248', '136', '247', '135', '246',
+            '245', '244', '134', '243', '133', '242', '160',
+            # Dash audio
+            '141', '172', '140', '139', '171',
+        ]
+
+        for f1id, f2id in zip(order, order[1:]):
+            f1 = YoutubeIE._formats[f1id].copy()
+            f1['format_id'] = f1id
+            f1['url'] = 'url:' + f1id
+            f2 = YoutubeIE._formats[f2id].copy()
+            f2['format_id'] = f2id
+            f2['url'] = 'url:' + f2id
+
+            info_dict = _make_result([f1, f2], extractor='youtube')
+            ydl = YDL()
+            yie = YoutubeIE(ydl)
+            yie._sort_formats(info_dict['formats'])
+            ydl.process_ie_result(info_dict)
+            downloaded = ydl.downloaded_info_dicts[0]
+            self.assertEqual(downloaded['format_id'], f1id)
+
+            info_dict = _make_result([f2, f1], extractor='youtube')
+            ydl = YDL()
+            yie = YoutubeIE(ydl)
+            yie._sort_formats(info_dict['formats'])
+            ydl.process_ie_result(info_dict)
+            downloaded = ydl.downloaded_info_dicts[0]
+            self.assertEqual(downloaded['format_id'], f1id)
 
     def test_add_extra_info(self):
         test_dict = {
@@ -143,18 +262,24 @@ class TestFormatSelection(unittest.TestCase):
 
     def test_prepare_filename(self):
         info = {
-            u'id': u'1234',
-            u'ext': u'mp4',
-            u'width': None,
+            'id': '1234',
+            'ext': 'mp4',
+            'width': None,
         }
         def fname(templ):
             ydl = YoutubeDL({'outtmpl': templ})
             return ydl.prepare_filename(info)
-        self.assertEqual(fname(u'%(id)s.%(ext)s'), u'1234.mp4')
-        self.assertEqual(fname(u'%(id)s-%(width)s.%(ext)s'), u'1234-NA.mp4')
+        self.assertEqual(fname('%(id)s.%(ext)s'), '1234.mp4')
+        self.assertEqual(fname('%(id)s-%(width)s.%(ext)s'), '1234-NA.mp4')
         # Replace missing fields with 'NA'
-        self.assertEqual(fname(u'%(uploader_date)s-%(id)s.%(ext)s'), u'NA-1234.mp4')
+        self.assertEqual(fname('%(uploader_date)s-%(id)s.%(ext)s'), 'NA-1234.mp4')
 
+    def test_format_note(self):
+        ydl = YoutubeDL()
+        self.assertEqual(ydl._format_note({}), '')
+        assertRegexpMatches(self, ydl._format_note({
+            'vbr': 10,
+        }), '^x\s*10k$')
 
 if __name__ == '__main__':
     unittest.main()

test/test_all_urls.py

@@ -1,5 +1,7 @@
 #!/usr/bin/env python
 
+from __future__ import unicode_literals
+
 # Allow direct execution
 import os
 import sys
@@ -7,12 +9,13 @@ import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 
-from test.helper import get_testcases
+from test.helper import gettestcases
 
 from youtube_dl.extractor import (
     FacebookIE,
     gen_extractors,
     JustinTVIE,
+    PBSIE,
     YoutubeIE,
 )
 
@@ -29,21 +32,24 @@ class TestAllURLsMatching(unittest.TestCase):
 
     def test_youtube_playlist_matching(self):
         assertPlaylist = lambda url: self.assertMatch(url, ['youtube:playlist'])
-        assertPlaylist(u'ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
-        assertPlaylist(u'UUBABnxM4Ar9ten8Mdjj1j0Q') #585
-        assertPlaylist(u'PL63F0C78739B09958')
-        assertPlaylist(u'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')
-        assertPlaylist(u'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
-        assertPlaylist(u'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
-        assertPlaylist(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012') #668
-        self.assertFalse('youtube:playlist' in self.matching_ies(u'PLtS2H6bU1M'))
+        assertPlaylist('ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
+        assertPlaylist('UUBABnxM4Ar9ten8Mdjj1j0Q') #585
+        assertPlaylist('PL63F0C78739B09958')
+        assertPlaylist('https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')
+        assertPlaylist('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
+        assertPlaylist('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
+        assertPlaylist('https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012') #668
+        self.assertFalse('youtube:playlist' in self.matching_ies('PLtS2H6bU1M'))
+        # Top tracks
+        assertPlaylist('https://www.youtube.com/playlist?list=MCUS.20142101')
 
     def test_youtube_matching(self):
-        self.assertTrue(YoutubeIE.suitable(u'PLtS2H6bU1M'))
-        self.assertFalse(YoutubeIE.suitable(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012')) #668
+        self.assertTrue(YoutubeIE.suitable('PLtS2H6bU1M'))
+        self.assertFalse(YoutubeIE.suitable('https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012')) #668
         self.assertMatch('http://youtu.be/BaW_jenozKc', ['youtube'])
         self.assertMatch('http://www.youtube.com/v/BaW_jenozKc', ['youtube'])
         self.assertMatch('https://youtube.googleapis.com/v/BaW_jenozKc', ['youtube'])
+        self.assertMatch('http://www.cleanvideosearch.com/media/action/yt/watch?videoId=8v_4O44sfjM', ['youtube'])
 
     def test_youtube_channel_matching(self):
         assertChannel = lambda url: self.assertMatch(url, ['youtube:channel'])
@@ -63,24 +69,31 @@ class TestAllURLsMatching(unittest.TestCase):
     def test_youtube_show_matching(self):
         self.assertMatch('http://www.youtube.com/show/airdisasters', ['youtube:show'])
 
+    def test_youtube_truncated(self):
+        self.assertMatch('http://www.youtube.com/watch?', ['youtube:truncated_url'])
+
+    def test_youtube_search_matching(self):
+        self.assertMatch('http://www.youtube.com/results?search_query=making+mustard', ['youtube:search_url'])
+        self.assertMatch('https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video', ['youtube:search_url'])
+
     def test_justin_tv_channelid_matching(self):
-        self.assertTrue(JustinTVIE.suitable(u"justin.tv/vanillatv"))
-        self.assertTrue(JustinTVIE.suitable(u"twitch.tv/vanillatv"))
-        self.assertTrue(JustinTVIE.suitable(u"www.justin.tv/vanillatv"))
-        self.assertTrue(JustinTVIE.suitable(u"www.twitch.tv/vanillatv"))
-        self.assertTrue(JustinTVIE.suitable(u"http://www.justin.tv/vanillatv"))
-        self.assertTrue(JustinTVIE.suitable(u"http://www.twitch.tv/vanillatv"))
-        self.assertTrue(JustinTVIE.suitable(u"http://www.justin.tv/vanillatv/"))
-        self.assertTrue(JustinTVIE.suitable(u"http://www.twitch.tv/vanillatv/"))
+        self.assertTrue(JustinTVIE.suitable('justin.tv/vanillatv'))
+        self.assertTrue(JustinTVIE.suitable('twitch.tv/vanillatv'))
+        self.assertTrue(JustinTVIE.suitable('www.justin.tv/vanillatv'))
+        self.assertTrue(JustinTVIE.suitable('www.twitch.tv/vanillatv'))
+        self.assertTrue(JustinTVIE.suitable('http://www.justin.tv/vanillatv'))
+        self.assertTrue(JustinTVIE.suitable('http://www.twitch.tv/vanillatv'))
+        self.assertTrue(JustinTVIE.suitable('http://www.justin.tv/vanillatv/'))
+        self.assertTrue(JustinTVIE.suitable('http://www.twitch.tv/vanillatv/'))
 
     def test_justintv_videoid_matching(self):
-        self.assertTrue(JustinTVIE.suitable(u"http://www.twitch.tv/vanillatv/b/328087483"))
+        self.assertTrue(JustinTVIE.suitable('http://www.twitch.tv/vanillatv/b/328087483'))
 
     def test_justin_tv_chapterid_matching(self):
-        self.assertTrue(JustinTVIE.suitable(u"http://www.twitch.tv/tsm_theoddone/c/2349361"))
+        self.assertTrue(JustinTVIE.suitable('http://www.twitch.tv/tsm_theoddone/c/2349361'))
 
     def test_youtube_extract(self):
-        assertExtractId = lambda url, id: self.assertEqual(YoutubeIE()._extract_id(url), id)
+        assertExtractId = lambda url, id: self.assertEqual(YoutubeIE.extract_id(url), id)
         assertExtractId('http://www.youtube.com/watch?&v=BaW_jenozKc', 'BaW_jenozKc')
         assertExtractId('https://www.youtube.com/watch?&v=BaW_jenozKc', 'BaW_jenozKc')
         assertExtractId('https://www.youtube.com/watch?feature=player_embedded&v=BaW_jenozKc', 'BaW_jenozKc')
@@ -89,11 +102,11 @@ class TestAllURLsMatching(unittest.TestCase):
         assertExtractId('BaW_jenozKc', 'BaW_jenozKc')
 
     def test_facebook_matching(self):
-        self.assertTrue(FacebookIE.suitable(u'https://www.facebook.com/Shiniknoh#!/photo.php?v=10153317450565268'))
+        self.assertTrue(FacebookIE.suitable('https://www.facebook.com/Shiniknoh#!/photo.php?v=10153317450565268'))
 
     def test_no_duplicates(self):
         ies = gen_extractors()
-        for tc in get_testcases():
+        for tc in gettestcases(include_onlymatching=True):
             url = tc['url']
             for ie in ies:
                 if type(ie).__name__ in ('GenericIE', tc['name'] + 'IE'):
@@ -112,11 +125,57 @@ class TestAllURLsMatching(unittest.TestCase):
 
     def test_vimeo_matching(self):
         self.assertMatch('http://vimeo.com/channels/tributes', ['vimeo:channel'])
+        self.assertMatch('http://vimeo.com/channels/31259', ['vimeo:channel'])
+        self.assertMatch('http://vimeo.com/channels/31259/53576664', ['vimeo'])
         self.assertMatch('http://vimeo.com/user7108434', ['vimeo:user'])
+        self.assertMatch('http://vimeo.com/user7108434/videos', ['vimeo:user'])
+        self.assertMatch('https://vimeo.com/user21297594/review/75524534/3c257a1b5d', ['vimeo:review'])
 
     # https://github.com/rg3/youtube-dl/issues/1930
     def test_soundcloud_not_matching_sets(self):
         self.assertMatch('http://soundcloud.com/floex/sets/gone-ep', ['soundcloud:set'])
 
+    def test_tumblr(self):
+        self.assertMatch('http://tatianamaslanydaily.tumblr.com/post/54196191430/orphan-black-dvd-extra-behind-the-scenes', ['Tumblr'])
+        self.assertMatch('http://tatianamaslanydaily.tumblr.com/post/54196191430', ['Tumblr'])
+
+    def test_pbs(self):
+        # https://github.com/rg3/youtube-dl/issues/2350
+        self.assertMatch('http://video.pbs.org/viralplayer/2365173446/', ['PBS'])
+        self.assertMatch('http://video.pbs.org/widget/partnerplayer/980042464/', ['PBS'])
+
+    def test_ComedyCentralShows(self):
+        self.assertMatch(
+            'http://thedailyshow.cc.com/extended-interviews/xm3fnq/andrew-napolitano-extended-interview',
+            ['ComedyCentralShows'])
+        self.assertMatch(
+            'http://thecolbertreport.cc.com/videos/29w6fx/-realhumanpraise-for-fox-news',
+            ['ComedyCentralShows'])
+        self.assertMatch(
+            'http://thecolbertreport.cc.com/videos/gh6urb/neil-degrasse-tyson-pt--1?xrs=eml_col_031114',
+            ['ComedyCentralShows'])
+        self.assertMatch(
+            'http://thedailyshow.cc.com/guests/michael-lewis/3efna8/exclusive---michael-lewis-extended-interview-pt--3',
+            ['ComedyCentralShows'])
+        self.assertMatch(
+            'http://thedailyshow.cc.com/episodes/sy7yv0/april-8--2014---denis-leary',
+            ['ComedyCentralShows'])
+        self.assertMatch(
+            'http://thecolbertreport.cc.com/episodes/8ase07/april-8--2014---jane-goodall',
+            ['ComedyCentralShows'])
+        self.assertMatch(
+            'http://thedailyshow.cc.com/video-playlists/npde3s/the-daily-show-19088-highlights',
+            ['ComedyCentralShows'])
+        self.assertMatch(
+            'http://thedailyshow.cc.com/special-editions/2l8fdb/special-edition---a-look-back-at-food',
+            ['ComedyCentralShows'])
+
+    def test_yahoo_https(self):
+        # https://github.com/rg3/youtube-dl/issues/2701
+        self.assertMatch(
+            'https://screen.yahoo.com/smartwatches-latest-wearable-gadgets-163745379-cbs.html',
+            ['Yahoo'])
+
+
 if __name__ == '__main__':
     unittest.main()

test/test_download.py

@@ -8,10 +8,11 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from test.helper import (
     get_params,
-    get_testcases,
-    try_rm,
+    gettestcases,
+    expect_info_dict,
     md5,
-    report_warning
+    try_rm,
+    report_warning,
 )
 
 
@@ -22,6 +23,7 @@ import socket
 
 import youtube_dl.YoutubeDL
 from youtube_dl.utils import (
+    compat_http_client,
     compat_str,
     compat_urllib_error,
     compat_HTTPError,
@@ -49,7 +51,7 @@ def _file_md5(fn):
     with open(fn, 'rb') as f:
         return hashlib.md5(f.read()).hexdigest()
 
-defs = get_testcases()
+defs = gettestcases()
 
 
 class TestDownload(unittest.TestCase):
@@ -71,9 +73,7 @@ def generator(test_case):
         if 'playlist' not in test_case:
             info_dict = test_case.get('info_dict', {})
             if not test_case.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
-                print_skipping('The output file cannot be know, the "file" '
-                    'key is missing or the info_dict is incomplete')
-                return
+                raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?')
         if 'skip' in test_case:
             print_skipping(test_case['skip'])
             return
@@ -90,7 +90,7 @@ def generator(test_case):
         def _hook(status):
             if status['status'] == 'finished':
                 finished_hook_called.add(status['filename'])
-        ydl.fd.add_progress_hook(_hook)
+        ydl.add_progress_hook(_hook)
 
         def get_tc_filename(tc):
             return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
@@ -110,7 +110,7 @@ def generator(test_case):
                     ydl.download([test_case['url']])
                 except (DownloadError, ExtractorError) as err:
                     # Check if the exception is not a network related one
-                    if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503):
+                    if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503):
                         raise
 
                     if try_num == RETRIES:
@@ -135,27 +135,8 @@ def generator(test_case):
                     self.assertEqual(md5_for_file, tc['md5'])
                 with io.open(info_json_fn, encoding='utf-8') as infof:
                     info_dict = json.load(infof)
-                for (info_field, expected) in tc.get('info_dict', {}).items():
-                    if isinstance(expected, compat_str) and expected.startswith('md5:'):
-                        got = 'md5:' + md5(info_dict.get(info_field))
-                    else:
-                        got = info_dict.get(info_field)
-                    self.assertEqual(expected, got,
-                        u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
 
-                # If checkable fields are missing from the test case, print the info_dict
-                test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))
-                    for key, value in info_dict.items()
-                    if value and key in ('title', 'description', 'uploader', 'upload_date', 'uploader_id', 'location'))
-                if not all(key in tc.get('info_dict', {}).keys() for key in test_info_dict.keys()):
-                    sys.stderr.write(u'\n"info_dict": ' + json.dumps(test_info_dict, ensure_ascii=False, indent=2) + u'\n')
-
-                # Check for the presence of mandatory fields
-                for key in ('id', 'url', 'title', 'ext'):
-                    self.assertTrue(key in info_dict.keys() and info_dict[key])
-                # Check for mandatory fields that are automatically set by YoutubeDL
-                for key in ['webpage_url', 'extractor', 'extractor_key']:
-                    self.assertTrue(info_dict.get(key), u'Missing field: %s' % key)
+                expect_info_dict(self, tc.get('info_dict', {}), info_dict)
         finally:
             try_rm_tcs_files()
 

test/test_playlists.py

@@ -1,6 +1,7 @@
 #!/usr/bin/env python
 # encoding: utf-8
 
+from __future__ import unicode_literals
 
 # Allow direct execution
 import os
@@ -8,8 +9,11 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from test.helper import FakeYDL
-
+from test.helper import (
+    assertRegexpMatches,
+    expect_info_dict,
+    FakeYDL,
+)
 
 from youtube_dl.extractor import (
     AcademicEarthCourseIE,
@@ -22,13 +26,26 @@ from youtube_dl.extractor import (
     UstreamChannelIE,
     SoundcloudSetIE,
     SoundcloudUserIE,
+    SoundcloudPlaylistIE,
     LivestreamIE,
     NHLVideocenterIE,
     BambuserChannelIE,
     BandcampAlbumIE,
     SmotriCommunityIE,
     SmotriUserIE,
-    IviCompilationIE
+    IviCompilationIE,
+    ImdbListIE,
+    KhanAcademyIE,
+    EveryonesMixtapeIE,
+    RutubeChannelIE,
+    GoogleSearchIE,
+    GenericIE,
+    TEDIE,
+    ToypicsUserIE,
+    XTubeUserIE,
+    InstagramUserIE,
+    CSpanIE,
+    AolIE,
 )
 
 
@@ -42,23 +59,23 @@ class TestPlaylists(unittest.TestCase):
         ie = DailymotionPlaylistIE(dl)
         result = ie.extract('http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'SPORT')
+        self.assertEqual(result['title'], 'SPORT')
         self.assertTrue(len(result['entries']) > 20)
 
     def test_dailymotion_user(self):
         dl = FakeYDL()
         ie = DailymotionUserIE(dl)
-        result = ie.extract('http://www.dailymotion.com/user/generation-quoi/')
+        result = ie.extract('https://www.dailymotion.com/user/nqtv')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'Génération Quoi')
-        self.assertTrue(len(result['entries']) >= 26)
+        self.assertEqual(result['title'], 'Rémi Gaillard')
+        self.assertTrue(len(result['entries']) >= 100)
 
     def test_vimeo_channel(self):
         dl = FakeYDL()
         ie = VimeoChannelIE(dl)
         result = ie.extract('http://vimeo.com/channels/tributes')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'Vimeo Tributes')
+        self.assertEqual(result['title'], 'Vimeo Tributes')
         self.assertTrue(len(result['entries']) > 24)
 
     def test_vimeo_user(self):
@@ -66,7 +83,7 @@ class TestPlaylists(unittest.TestCase):
         ie = VimeoUserIE(dl)
         result = ie.extract('http://vimeo.com/nkistudio/videos')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'Nki')
+        self.assertEqual(result['title'], 'Nki')
         self.assertTrue(len(result['entries']) > 65)
 
     def test_vimeo_album(self):
@@ -74,7 +91,7 @@ class TestPlaylists(unittest.TestCase):
         ie = VimeoAlbumIE(dl)
         result = ie.extract('http://vimeo.com/album/2632481')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'Staff Favorites: November 2013')
+        self.assertEqual(result['title'], 'Staff Favorites: November 2013')
         self.assertTrue(len(result['entries']) > 12)
 
     def test_vimeo_groups(self):
@@ -82,7 +99,7 @@ class TestPlaylists(unittest.TestCase):
         ie = VimeoGroupsIE(dl)
         result = ie.extract('http://vimeo.com/groups/rolexawards')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'Rolex Awards for Enterprise')
+        self.assertEqual(result['title'], 'Rolex Awards for Enterprise')
         self.assertTrue(len(result['entries']) > 72)
 
     def test_ustream_channel(self):
@@ -90,15 +107,15 @@ class TestPlaylists(unittest.TestCase):
         ie = UstreamChannelIE(dl)
         result = ie.extract('http://www.ustream.tv/channel/young-americans-for-liberty')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], u'5124905')
-        self.assertTrue(len(result['entries']) >= 11)
+        self.assertEqual(result['id'], '5124905')
+        self.assertTrue(len(result['entries']) >= 6)
 
     def test_soundcloud_set(self):
         dl = FakeYDL()
         ie = SoundcloudSetIE(dl)
         result = ie.extract('https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'The Royal Concept EP')
+        self.assertEqual(result['title'], 'The Royal Concept EP')
         self.assertTrue(len(result['entries']) >= 6)
 
     def test_soundcloud_user(self):
@@ -106,15 +123,26 @@ class TestPlaylists(unittest.TestCase):
         ie = SoundcloudUserIE(dl)
         result = ie.extract('https://soundcloud.com/the-concept-band')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], u'9615865')
+        self.assertEqual(result['id'], '9615865')
         self.assertTrue(len(result['entries']) >= 12)
 
+    def test_soundcloud_playlist(self):
+        dl = FakeYDL()
+        ie = SoundcloudPlaylistIE(dl)
+        result = ie.extract('http://api.soundcloud.com/playlists/4110309')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], '4110309')
+        self.assertEqual(result['title'], 'TILT Brass - Bowery Poetry Club, August \'03 [Non-Site SCR 02]')
+        assertRegexpMatches(
+            self, result['description'], r'TILT Brass - Bowery Poetry Club')
+        self.assertEqual(len(result['entries']), 6)
+
     def test_livestream_event(self):
         dl = FakeYDL()
         ie = LivestreamIE(dl)
         result = ie.extract('http://new.livestream.com/tedx/cityenglish')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'TEDCity2.0 (English)')
+        self.assertEqual(result['title'], 'TEDCity2.0 (English)')
         self.assertTrue(len(result['entries']) >= 4)
 
     def test_nhl_videocenter(self):
@@ -122,8 +150,8 @@ class TestPlaylists(unittest.TestCase):
         ie = NHLVideocenterIE(dl)
         result = ie.extract('http://video.canucks.nhl.com/videocenter/console?catid=999')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], u'999')
-        self.assertEqual(result['title'], u'Highlights')
+        self.assertEqual(result['id'], '999')
+        self.assertEqual(result['title'], 'Highlights')
         self.assertEqual(len(result['entries']), 12)
 
     def test_bambuser_channel(self):
@@ -131,7 +159,7 @@ class TestPlaylists(unittest.TestCase):
         ie = BambuserChannelIE(dl)
         result = ie.extract('http://bambuser.com/channel/pixelversity')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'pixelversity')
+        self.assertEqual(result['title'], 'pixelversity')
         self.assertTrue(len(result['entries']) >= 60)
 
     def test_bandcamp_album(self):
@@ -139,7 +167,7 @@ class TestPlaylists(unittest.TestCase):
         ie = BandcampAlbumIE(dl)
         result = ie.extract('http://mpallante.bandcamp.com/album/nightmare-night-ep')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'Nightmare Night EP')
+        self.assertEqual(result['title'], 'Nightmare Night EP')
         self.assertTrue(len(result['entries']) >= 4)
         
     def test_smotri_community(self):
@@ -147,8 +175,8 @@ class TestPlaylists(unittest.TestCase):
         ie = SmotriCommunityIE(dl)
         result = ie.extract('http://smotri.com/community/video/kommuna')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], u'kommuna')
-        self.assertEqual(result['title'], u'КПРФ')
+        self.assertEqual(result['id'], 'kommuna')
+        self.assertEqual(result['title'], 'КПРФ')
         self.assertTrue(len(result['entries']) >= 4)
         
     def test_smotri_user(self):
@@ -156,38 +184,173 @@ class TestPlaylists(unittest.TestCase):
         ie = SmotriUserIE(dl)
         result = ie.extract('http://smotri.com/user/inspector')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], u'inspector')
-        self.assertEqual(result['title'], u'Inspector')
+        self.assertEqual(result['id'], 'inspector')
+        self.assertEqual(result['title'], 'Inspector')
         self.assertTrue(len(result['entries']) >= 9)
 
     def test_AcademicEarthCourse(self):
         dl = FakeYDL()
         ie = AcademicEarthCourseIE(dl)
-        result = ie.extract(u'http://academicearth.org/courses/building-dynamic-websites/')
+        result = ie.extract('http://academicearth.org/playlists/laws-of-nature/')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], u'building-dynamic-websites')
-        self.assertEqual(result['title'], u'Building Dynamic Websites')
-        self.assertEqual(result['description'], u"Today's websites are increasingly dynamic. Pages are no longer static HTML files but instead generated by scripts and database calls. User interfaces are more seamless, with technologies like Ajax replacing traditional page reloads. This course teaches students how to build dynamic websites with Ajax and with Linux, Apache, MySQL, and PHP (LAMP), one of today's most popular frameworks. Students learn how to set up domain names with DNS, how to structure pages with XHTML and CSS, how to program in JavaScript and PHP, how to configure Apache and MySQL, how to design and query databases with SQL, how to use Ajax with both XML and JSON, and how to build mashups. The course explores issues of security, scalability, and cross-browser support and also discusses enterprise-level deployments of websites, including third-party hosting, virtualization, colocation in data centers, firewalling, and load-balancing.")
-        self.assertEqual(len(result['entries']), 10)
+        self.assertEqual(result['id'], 'laws-of-nature')
+        self.assertEqual(result['title'], 'Laws of Nature')
+        self.assertEqual(result['description'],u'Introduce yourself to the laws of nature with these free online college lectures from Yale, Harvard, and MIT.')# u"Today's websites are increasingly dynamic. Pages are no longer static HTML files but instead generated by scripts and database calls. User interfaces are more seamless, with technologies like Ajax replacing traditional page reloads. This course teaches students how to build dynamic websites with Ajax and with Linux, Apache, MySQL, and PHP (LAMP), one of today's most popular frameworks. Students learn how to set up domain names with DNS, how to structure pages with XHTML and CSS, how to program in JavaScript and PHP, how to configure Apache and MySQL, how to design and query databases with SQL, how to use Ajax with both XML and JSON, and how to build mashups. The course explores issues of security, scalability, and cross-browser support and also discusses enterprise-level deployments of websites, including third-party hosting, virtualization, colocation in data centers, firewalling, and load-balancing.")
+        self.assertEqual(len(result['entries']), 4)
         
     def test_ivi_compilation(self):
         dl = FakeYDL()
         ie = IviCompilationIE(dl)
         result = ie.extract('http://www.ivi.ru/watch/dezhurnyi_angel')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], u'dezhurnyi_angel')
-        self.assertEqual(result['title'], u'Дежурный ангел (2010 - 2012)')
-        self.assertTrue(len(result['entries']) >= 36)
-        
+        self.assertEqual(result['id'], 'dezhurnyi_angel')
+        self.assertEqual(result['title'], 'Дежурный ангел (2010 - 2012)')
+        self.assertTrue(len(result['entries']) >= 23)
+
     def test_ivi_compilation_season(self):
         dl = FakeYDL()
         ie = IviCompilationIE(dl)
         result = ie.extract('http://www.ivi.ru/watch/dezhurnyi_angel/season2')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], u'dezhurnyi_angel/season2')
-        self.assertEqual(result['title'], u'Дежурный ангел (2010 - 2012) 2 сезон')
-        self.assertTrue(len(result['entries']) >= 20)
+        self.assertEqual(result['id'], 'dezhurnyi_angel/season2')
+        self.assertEqual(result['title'], 'Дежурный ангел (2010 - 2012) 2 сезон')
+        self.assertTrue(len(result['entries']) >= 7)
+        
+    def test_imdb_list(self):
+        dl = FakeYDL()
+        ie = ImdbListIE(dl)
+        result = ie.extract('http://www.imdb.com/list/JFs9NWw6XI0')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'JFs9NWw6XI0')
+        self.assertEqual(result['title'], 'March 23, 2012 Releases')
+        self.assertEqual(len(result['entries']), 7)
 
+    def test_khanacademy_topic(self):
+        dl = FakeYDL()
+        ie = KhanAcademyIE(dl)
+        result = ie.extract('https://www.khanacademy.org/math/applied-math/cryptography')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'cryptography')
+        self.assertEqual(result['title'], 'Journey into cryptography')
+        self.assertEqual(result['description'], 'How have humans protected their secret messages through history? What has changed today?')
+        self.assertTrue(len(result['entries']) >= 3)
+
+    def test_EveryonesMixtape(self):
+        dl = FakeYDL()
+        ie = EveryonesMixtapeIE(dl)
+        result = ie.extract('http://everyonesmixtape.com/#/mix/m7m0jJAbMQi')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'm7m0jJAbMQi')
+        self.assertEqual(result['title'], 'Driving')
+        self.assertEqual(len(result['entries']), 24)
+        
+    def test_rutube_channel(self):
+        dl = FakeYDL()
+        ie = RutubeChannelIE(dl)
+        result = ie.extract('http://rutube.ru/tags/video/1409')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], '1409')
+        self.assertTrue(len(result['entries']) >= 34)
+
+    def test_multiple_brightcove_videos(self):
+        # https://github.com/rg3/youtube-dl/issues/2283
+        dl = FakeYDL()
+        ie = GenericIE(dl)
+        result = ie.extract('http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'always-never-nuclear-command-and-control')
+        self.assertEqual(result['title'], 'Always/Never: A Little-Seen Movie About Nuclear Command and Control : The New Yorker')
+        self.assertEqual(len(result['entries']), 3)
+
+    def test_GoogleSearch(self):
+        dl = FakeYDL()
+        ie = GoogleSearchIE(dl)
+        result = ie.extract('gvsearch15:python language')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'python language')
+        self.assertEqual(result['title'], 'python language')
+        self.assertEqual(len(result['entries']), 15)
+
+    def test_generic_rss_feed(self):
+        dl = FakeYDL()
+        ie = GenericIE(dl)
+        result = ie.extract('http://phihag.de/2014/youtube-dl/rss.xml')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'http://phihag.de/2014/youtube-dl/rss.xml')
+        self.assertEqual(result['title'], 'Zero Punctuation')
+        self.assertTrue(len(result['entries']) > 10)
+
+    def test_ted_playlist(self):
+        dl = FakeYDL()
+        ie = TEDIE(dl)
+        result = ie.extract('http://www.ted.com/playlists/who_are_the_hackers')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], '10')
+        self.assertEqual(result['title'], 'Who are the hackers?')
+        self.assertTrue(len(result['entries']) >= 6)
+
+    def test_toypics_user(self):
+        dl = FakeYDL()
+        ie = ToypicsUserIE(dl)
+        result = ie.extract('http://videos.toypics.net/Mikey')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'Mikey')
+        self.assertTrue(len(result['entries']) >= 17)
+
+    def test_xtube_user(self):
+        dl = FakeYDL()
+        ie = XTubeUserIE(dl)
+        result = ie.extract('http://www.xtube.com/community/profile.php?user=greenshowers')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'greenshowers')
+        self.assertTrue(len(result['entries']) >= 155)
+
+    def test_InstagramUser(self):
+        dl = FakeYDL()
+        ie = InstagramUserIE(dl)
+        result = ie.extract('http://instagram.com/porsche')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'porsche')
+        self.assertTrue(len(result['entries']) >= 2)
+        test_video = next(
+            e for e in result['entries']
+            if e['id'] == '614605558512799803_462752227')
+        dl.add_default_extra_info(test_video, ie, '(irrelevant URL)')
+        dl.process_video_result(test_video, download=False)
+        EXPECTED = {
+            'id': '614605558512799803_462752227',
+            'ext': 'mp4',
+            'title': '#Porsche Intelligent Performance.',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'uploader': 'Porsche',
+            'uploader_id': 'porsche',
+            'timestamp': 1387486713,
+            'upload_date': '20131219',
+        }
+        expect_info_dict(self, EXPECTED, test_video)
+
+    def test_CSpan_playlist(self):
+        dl = FakeYDL()
+        ie = CSpanIE(dl)
+        result = ie.extract(
+            'http://www.c-span.org/video/?318608-1/gm-ignition-switch-recall')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], '342759')
+        self.assertEqual(
+            result['title'], 'General Motors Ignition Switch Recall')
+        whole_duration = sum(e['duration'] for e in result['entries'])
+        self.assertEqual(whole_duration, 14855)
+
+    def test_aol_playlist(self):
+        dl = FakeYDL()
+        ie = AolIE(dl)
+        result = ie.extract(
+            'http://on.aol.com/playlist/brace-yourself---todays-weirdest-news-152147?icid=OnHomepageC4_Omg_Img#_videoid=518184316')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], '152147')
+        self.assertEqual(
+            result['title'], 'Brace Yourself - Today\'s Weirdest News')
+        self.assertTrue(len(result['entries']) >= 10)
 
 if __name__ == '__main__':
     unittest.main()

test/test_subtitles.py

@@ -10,9 +10,11 @@ from test.helper import FakeYDL, md5
 
 
 from youtube_dl.extractor import (
+    BlipTVIE,
     YoutubeIE,
     DailymotionIE,
     TEDIE,
+    VimeoIE,
 )
 
 
@@ -36,10 +38,6 @@ class TestYoutubeSubtitles(BaseTestSubtitles):
     url = 'QRS8MkLhQmM'
     IE = YoutubeIE
 
-    def getSubtitles(self):
-        info_dict = self.getInfoDict()
-        return info_dict[0]['subtitles']
-
     def test_youtube_no_writesubtitles(self):
         self.DL.params['writesubtitles'] = False
         subtitles = self.getSubtitles()
@@ -171,19 +169,19 @@ class TestTedSubtitles(BaseTestSubtitles):
     def test_subtitles(self):
         self.DL.params['writesubtitles'] = True
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '2154f31ff9b9f89a0aa671537559c21d')
+        self.assertEqual(md5(subtitles['en']), '4262c1665ff928a2dada178f62cb8d14')
 
     def test_subtitles_lang(self):
         self.DL.params['writesubtitles'] = True
         self.DL.params['subtitleslangs'] = ['fr']
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['fr']), '7616cbc6df20ec2c1204083c83871cf6')
+        self.assertEqual(md5(subtitles['fr']), '66a63f7f42c97a50f8c0e90bc7797bb5')
 
     def test_allsubtitles(self):
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
-        self.assertEqual(len(subtitles.keys()), 28)
+        self.assertTrue(len(subtitles.keys()) >= 28)
 
     def test_list_subtitles(self):
         self.DL.expect_warning(u'Automatic Captions not supported by this server')
@@ -206,5 +204,80 @@ class TestTedSubtitles(BaseTestSubtitles):
         for lang in langs:
             self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
 
+
+class TestBlipTVSubtitles(BaseTestSubtitles):
+    url = 'http://blip.tv/a/a-6603250'
+    IE = BlipTVIE
+
+    def test_list_subtitles(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['listsubtitles'] = True
+        info_dict = self.getInfoDict()
+        self.assertEqual(info_dict, None)
+
+    def test_allsubtitles(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['en']))
+        self.assertEqual(md5(subtitles['en']), '5b75c300af65fe4476dff79478bb93e4')
+
+
+class TestVimeoSubtitles(BaseTestSubtitles):
+    url = 'http://vimeo.com/76979871'
+    IE = VimeoIE
+
+    def test_no_writesubtitles(self):
+        subtitles = self.getSubtitles()
+        self.assertEqual(subtitles, None)
+
+    def test_subtitles(self):
+        self.DL.params['writesubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['en']), '8062383cf4dec168fc40a088aa6d5888')
+
+    def test_subtitles_lang(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['subtitleslangs'] = ['fr']
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['fr']), 'b6191146a6c5d3a452244d853fde6dc8')
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['de', 'en', 'es', 'fr']))
+
+    def test_list_subtitles(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['listsubtitles'] = True
+        info_dict = self.getInfoDict()
+        self.assertEqual(info_dict, None)
+
+    def test_automatic_captions(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['writeautomaticsub'] = True
+        self.DL.params['subtitleslang'] = ['en']
+        subtitles = self.getSubtitles()
+        self.assertTrue(len(subtitles.keys()) == 0)
+
+    def test_nosubtitles(self):
+        self.DL.expect_warning(u'video doesn\'t have subtitles')
+        self.url = 'http://vimeo.com/56015672'
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(len(subtitles), 0)
+
+    def test_multiple_langs(self):
+        self.DL.params['writesubtitles'] = True
+        langs = ['es', 'fr', 'de']
+        self.DL.params['subtitleslangs'] = langs
+        subtitles = self.getSubtitles()
+        for lang in langs:
+            self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
+
+
 if __name__ == '__main__':
     unittest.main()

test/test_unicode_literals.py

@@ -0,0 +1,47 @@
+from __future__ import unicode_literals
+
+import io
+import os
+import re
+import unittest
+
+rootDir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+
+IGNORED_FILES = [
+    'setup.py',  # http://bugs.python.org/issue13943
+]
+
+
+class TestUnicodeLiterals(unittest.TestCase):
+    def test_all_files(self):
+        print('Skipping this test (not yet fully implemented)')
+        return
+
+        for dirpath, _, filenames in os.walk(rootDir):
+            for basename in filenames:
+                if not basename.endswith('.py'):
+                    continue
+                if basename in IGNORED_FILES:
+                    continue
+
+                fn = os.path.join(dirpath, basename)
+                with io.open(fn, encoding='utf-8') as inf:
+                    code = inf.read()
+
+                if "'" not in code and '"' not in code:
+                    continue
+                imps = 'from __future__ import unicode_literals'
+                self.assertTrue(
+                    imps in code,
+                    ' %s  missing in %s' % (imps, fn))
+
+                m = re.search(r'(?<=\s)u[\'"](?!\)|,|$)', code)
+                if m is not None:
+                    self.assertTrue(
+                        m is None,
+                        'u present in %s, around %s' % (
+                            fn, code[m.start() - 10:m.end() + 10]))
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_utils.py

@@ -9,6 +9,8 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 
 # Various small unit tests
+import io
+import json
 import xml.etree.ElementTree
 
 #from youtube_dl.utils import htmlentity_transform
@@ -16,18 +18,27 @@ from youtube_dl.utils import (
     DateRange,
     encodeFilename,
     find_xpath_attr,
+    fix_xml_ampersands,
     get_meta_content,
     orderedSet,
+    PagedList,
+    parse_duration,
+    read_batch_urls,
     sanitize_filename,
     shell_quote,
     smuggle_url,
     str_to_int,
+    struct_unpack,
     timeconvert,
     unescapeHTML,
     unified_strdate,
     unsmuggle_url,
     url_basename,
+    urlencode_postdata,
     xpath_with_ns,
+    parse_iso8601,
+    strip_jsonp,
+    uppercase_escape,
 )
 
 if sys.version_info < (3, 0):
@@ -124,6 +135,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(unified_strdate('8/7/2009'), '20090708')
         self.assertEqual(unified_strdate('Dec 14, 2012'), '20121214')
         self.assertEqual(unified_strdate('2012/10/11 01:56:38 +0000'), '20121011')
+        self.assertEqual(unified_strdate('1968-12-10'), '19681210')
 
     def test_find_xpath_attr(self):
         testxml = u'''<root>
@@ -192,5 +204,85 @@ class TestUtil(unittest.TestCase):
             url_basename(u'http://media.w3.org/2010/05/sintel/trailer.mp4'),
             u'trailer.mp4')
 
+    def test_parse_duration(self):
+        self.assertEqual(parse_duration(None), None)
+        self.assertEqual(parse_duration('1'), 1)
+        self.assertEqual(parse_duration('1337:12'), 80232)
+        self.assertEqual(parse_duration('9:12:43'), 33163)
+        self.assertEqual(parse_duration('12:00'), 720)
+        self.assertEqual(parse_duration('00:01:01'), 61)
+        self.assertEqual(parse_duration('x:y'), None)
+        self.assertEqual(parse_duration('3h11m53s'), 11513)
+        self.assertEqual(parse_duration('62m45s'), 3765)
+        self.assertEqual(parse_duration('6m59s'), 419)
+        self.assertEqual(parse_duration('49s'), 49)
+        self.assertEqual(parse_duration('0h0m0s'), 0)
+        self.assertEqual(parse_duration('0m0s'), 0)
+        self.assertEqual(parse_duration('0s'), 0)
+
+    def test_fix_xml_ampersands(self):
+        self.assertEqual(
+            fix_xml_ampersands('"&x=y&z=a'), '"&amp;x=y&amp;z=a')
+        self.assertEqual(
+            fix_xml_ampersands('"&amp;x=y&wrong;&z=a'),
+            '"&amp;x=y&amp;wrong;&amp;z=a')
+        self.assertEqual(
+            fix_xml_ampersands('&amp;&apos;&gt;&lt;&quot;'),
+            '&amp;&apos;&gt;&lt;&quot;')
+        self.assertEqual(
+            fix_xml_ampersands('&#1234;&#x1abC;'), '&#1234;&#x1abC;')
+        self.assertEqual(fix_xml_ampersands('&#&#'), '&amp;#&amp;#')
+
+    def test_paged_list(self):
+        def testPL(size, pagesize, sliceargs, expected):
+            def get_page(pagenum):
+                firstid = pagenum * pagesize
+                upto = min(size, pagenum * pagesize + pagesize)
+                for i in range(firstid, upto):
+                    yield i
+
+            pl = PagedList(get_page, pagesize)
+            got = pl.getslice(*sliceargs)
+            self.assertEqual(got, expected)
+
+        testPL(5, 2, (), [0, 1, 2, 3, 4])
+        testPL(5, 2, (1,), [1, 2, 3, 4])
+        testPL(5, 2, (2,), [2, 3, 4])
+        testPL(5, 2, (4,), [4])
+        testPL(5, 2, (0, 3), [0, 1, 2])
+        testPL(5, 2, (1, 4), [1, 2, 3])
+        testPL(5, 2, (2, 99), [2, 3, 4])
+        testPL(5, 2, (20, 99), [])
+
+    def test_struct_unpack(self):
+        self.assertEqual(struct_unpack(u'!B', b'\x00'), (0,))
+
+    def test_read_batch_urls(self):
+        f = io.StringIO(u'''\xef\xbb\xbf foo
+            bar\r
+            baz
+            # More after this line\r
+            ; or after this
+            bam''')
+        self.assertEqual(read_batch_urls(f), [u'foo', u'bar', u'baz', u'bam'])
+
+    def test_urlencode_postdata(self):
+        data = urlencode_postdata({'username': 'foo@bar.com', 'password': '1234'})
+        self.assertTrue(isinstance(data, bytes))
+
+    def test_parse_iso8601(self):
+        self.assertEqual(parse_iso8601('2014-03-23T23:04:26+0100'), 1395612266)
+        self.assertEqual(parse_iso8601('2014-03-23T22:04:26+0000'), 1395612266)
+        self.assertEqual(parse_iso8601('2014-03-23T22:04:26Z'), 1395612266)
+
+    def test_strip_jsonp(self):
+        stripped = strip_jsonp('cb ([ {"id":"532cb",\n\n\n"x":\n3}\n]\n);')
+        d = json.loads(stripped)
+        self.assertEqual(d, [{"id": "532cb", "x": 3}])
+
+    def test_uppercase_escpae(self):
+        self.assertEqual(uppercase_escape(u'aä'), u'aä')
+        self.assertEqual(uppercase_escape(u'\\U0001d550'), u'𝕐')
+
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_lists.py

@@ -16,6 +16,7 @@ from youtube_dl.extractor import (
     YoutubeChannelIE,
     YoutubeShowIE,
     YoutubeTopListIE,
+    YoutubeSearchURLIE,
 )
 
 
@@ -30,7 +31,7 @@ class TestYoutubeLists(unittest.TestCase):
         result = ie.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
         self.assertIsPlaylist(result)
         self.assertEqual(result['title'], 'ytdl test PL')
-        ytie_results = [YoutubeIE()._extract_id(url['url']) for url in result['entries']]
+        ytie_results = [YoutubeIE().extract_id(url['url']) for url in result['entries']]
         self.assertEqual(ytie_results, [ 'bV9L5Ht9LgY', 'FXxLjLQi3Fg', 'tU3Bgo5qJZE'])
 
     def test_youtube_playlist_noplaylist(self):
@@ -39,7 +40,7 @@ class TestYoutubeLists(unittest.TestCase):
         ie = YoutubePlaylistIE(dl)
         result = ie.extract('https://www.youtube.com/watch?v=FXxLjLQi3Fg&list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
         self.assertEqual(result['_type'], 'url')
-        self.assertEqual(YoutubeIE()._extract_id(result['url']), 'FXxLjLQi3Fg')
+        self.assertEqual(YoutubeIE().extract_id(result['url']), 'FXxLjLQi3Fg')
 
     def test_issue_673(self):
         dl = FakeYDL()
@@ -59,7 +60,7 @@ class TestYoutubeLists(unittest.TestCase):
         dl = FakeYDL()
         ie = YoutubePlaylistIE(dl)
         result = ie.extract('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
-        ytie_results = [YoutubeIE()._extract_id(url['url']) for url in result['entries']]
+        ytie_results = [YoutubeIE().extract_id(url['url']) for url in result['entries']]
         self.assertFalse('pElCt5oNDuI' in ytie_results)
         self.assertFalse('KdPEApIVdWM' in ytie_results)
         
@@ -76,9 +77,9 @@ class TestYoutubeLists(unittest.TestCase):
         # TODO find a > 100 (paginating?) videos course
         result = ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
         entries = result['entries']
-        self.assertEqual(YoutubeIE()._extract_id(entries[0]['url']), 'j9WZyLZCBzs')
+        self.assertEqual(YoutubeIE().extract_id(entries[0]['url']), 'j9WZyLZCBzs')
         self.assertEqual(len(entries), 25)
-        self.assertEqual(YoutubeIE()._extract_id(entries[-1]['url']), 'rYefUsYuEp0')
+        self.assertEqual(YoutubeIE().extract_id(entries[-1]['url']), 'rYefUsYuEp0')
 
     def test_youtube_channel(self):
         dl = FakeYDL()
@@ -117,12 +118,30 @@ class TestYoutubeLists(unittest.TestCase):
         original_video = entries[0]
         self.assertEqual(original_video['id'], 'rjFaenf1T-Y')
 
+    def test_youtube_toptracks(self):
+        print('Skipping: The playlist page gives error 500')
+        return
+        dl = FakeYDL()
+        ie = YoutubePlaylistIE(dl)
+        result = ie.extract('https://www.youtube.com/playlist?list=MCUS')
+        entries = result['entries']
+        self.assertEqual(len(entries), 100)
+
     def test_youtube_toplist(self):
         dl = FakeYDL()
         ie = YoutubeTopListIE(dl)
-        result = ie.extract('yttoplist:music:Top Tracks')
+        result = ie.extract('yttoplist:music:Trending')
         entries = result['entries']
         self.assertTrue(len(entries) >= 5)
 
+    def test_youtube_search_url(self):
+        dl = FakeYDL()
+        ie = YoutubeSearchURLIE(dl)
+        result = ie.extract('https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video')
+        entries = result['entries']
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['title'], 'youtube-dl test video')
+        self.assertTrue(len(entries) >= 5)
+
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_signature.py

@@ -28,10 +28,10 @@ _TESTS = [
         u'3456789a0cdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRS[UVWXYZ!"#$%&\'()*+,-./:;<=>?@',
     ),
     (
-        u'https://s.ytimg.com/yts/swfbin/watch_as3-vflg5GhxU.swf',
-        u'swf',
-        82,
-        u':/.-,+*)=\'&%$#"!ZYX0VUTSRQPONMLKJIHGFEDCBAzyxw>utsrqponmlkjihgfedcba987654321'
+        u'https://s.ytimg.com/yts/jsbin/html5player-vfle-mVwz.js',
+        u'js',
+        90,
+        u']\\[@?>=<;:/.-,+*)(\'&%$#"hZYXWVUTSRQPONMLKJIHGFEDCBAzyxwvutsrqponmlkjiagfedcb39876',
     ),
 ]
 

youtube-dl.plugin.zsh

@@ -0,0 +1,24 @@
+# This allows the youtube-dl command to be installed in ZSH using antigen.
+# Antigen is a bundle manager. It allows you to enhance the functionality of
+# your zsh session by installing bundles and themes easily.
+
+# Antigen documentation:
+# http://antigen.sharats.me/
+# https://github.com/zsh-users/antigen
+
+# Install youtube-dl:
+# antigen bundle rg3/youtube-dl
+# Bundles installed by antigen are available for use immediately.
+
+# Update youtube-dl (and all other antigen bundles):
+# antigen update
+
+# The antigen command will download the git repository to a folder and then
+# execute an enabling script (this file). The complete process for loading the
+# code is documented here:
+# https://github.com/zsh-users/antigen#notes-on-writing-plugins
+
+# This specific script just aliases youtube-dl to the python script that this
+# library provides. This requires updating the PYTHONPATH to ensure that the
+# full set of code can be located.
+alias youtube-dl="PYTHONPATH=$(dirname $0) $(dirname $0)/bin/youtube-dl"

youtube_dl/FileDownloader.py

@@ -1,724 +1,12 @@
-import os
-import re
-import subprocess
-import sys
-import time
-
-from .utils import (
-    compat_urllib_error,
-    compat_urllib_request,
-    ContentTooShortError,
-    determine_ext,
-    encodeFilename,
-    format_bytes,
-    sanitize_open,
-    timeconvert,
-)
-
-
-class FileDownloader(object):
-    """File Downloader class.
-
-    File downloader objects are the ones responsible of downloading the
-    actual video file and writing it to disk.
-
-    File downloaders accept a lot of parameters. In order not to saturate
-    the object constructor with arguments, it receives a dictionary of
-    options instead.
-
-    Available options:
-
-    verbose:           Print additional info to stdout.
-    quiet:             Do not print messages to stdout.
-    ratelimit:         Download speed limit, in bytes/sec.
-    retries:           Number of times to retry for HTTP error 5xx
-    buffersize:        Size of download buffer in bytes.
-    noresizebuffer:    Do not automatically resize the download buffer.
-    continuedl:        Try to continue downloads if possible.
-    noprogress:        Do not print the progress bar.
-    logtostderr:       Log messages to stderr instead of stdout.
-    consoletitle:      Display progress in console window's titlebar.
-    nopart:            Do not use temporary .part files.
-    updatetime:        Use the Last-modified header to set output file timestamps.
-    test:              Download only first bytes to test the downloader.
-    min_filesize:      Skip files smaller than this size
-    max_filesize:      Skip files larger than this size
-    """
-
-    params = None
-
-    def __init__(self, ydl, params):
-        """Create a FileDownloader object with the given options."""
-        self.ydl = ydl
-        self._progress_hooks = []
-        self.params = params
-
-    @staticmethod
-    def format_seconds(seconds):
-        (mins, secs) = divmod(seconds, 60)
-        (hours, mins) = divmod(mins, 60)
-        if hours > 99:
-            return '--:--:--'
-        if hours == 0:
-            return '%02d:%02d' % (mins, secs)
-        else:
-            return '%02d:%02d:%02d' % (hours, mins, secs)
-
-    @staticmethod
-    def calc_percent(byte_counter, data_len):
-        if data_len is None:
-            return None
-        return float(byte_counter) / float(data_len) * 100.0
-
-    @staticmethod
-    def format_percent(percent):
-        if percent is None:
-            return '---.-%'
-        return '%6s' % ('%3.1f%%' % percent)
-
-    @staticmethod
-    def calc_eta(start, now, total, current):
-        if total is None:
-            return None
-        dif = now - start
-        if current == 0 or dif < 0.001: # One millisecond
-            return None
-        rate = float(current) / dif
-        return int((float(total) - float(current)) / rate)
-
-    @staticmethod
-    def format_eta(eta):
-        if eta is None:
-            return '--:--'
-        return FileDownloader.format_seconds(eta)
-
-    @staticmethod
-    def calc_speed(start, now, bytes):
-        dif = now - start
-        if bytes == 0 or dif < 0.001: # One millisecond
-            return None
-        return float(bytes) / dif
-
-    @staticmethod
-    def format_speed(speed):
-        if speed is None:
-            return '%10s' % '---b/s'
-        return '%10s' % ('%s/s' % format_bytes(speed))
-
-    @staticmethod
-    def best_block_size(elapsed_time, bytes):
-        new_min = max(bytes / 2.0, 1.0)
-        new_max = min(max(bytes * 2.0, 1.0), 4194304) # Do not surpass 4 MB
-        if elapsed_time < 0.001:
-            return int(new_max)
-        rate = bytes / elapsed_time
-        if rate > new_max:
-            return int(new_max)
-        if rate < new_min:
-            return int(new_min)
-        return int(rate)
-
-    @staticmethod
-    def parse_bytes(bytestr):
-        """Parse a string indicating a byte quantity into an integer."""
-        matchobj = re.match(r'(?i)^(\d+(?:\.\d+)?)([kMGTPEZY]?)$', bytestr)
-        if matchobj is None:
-            return None
-        number = float(matchobj.group(1))
-        multiplier = 1024.0 ** 'bkmgtpezy'.index(matchobj.group(2).lower())
-        return int(round(number * multiplier))
-
-    def to_screen(self, *args, **kargs):
-        self.ydl.to_screen(*args, **kargs)
-
-    def to_stderr(self, message):
-        self.ydl.to_screen(message)
-
-    def to_console_title(self, message):
-        self.ydl.to_console_title(message)
-
-    def trouble(self, *args, **kargs):
-        self.ydl.trouble(*args, **kargs)
-
-    def report_warning(self, *args, **kargs):
-        self.ydl.report_warning(*args, **kargs)
-
-    def report_error(self, *args, **kargs):
-        self.ydl.report_error(*args, **kargs)
-
-    def slow_down(self, start_time, byte_counter):
-        """Sleep if the download speed is over the rate limit."""
-        rate_limit = self.params.get('ratelimit', None)
-        if rate_limit is None or byte_counter == 0:
-            return
-        now = time.time()
-        elapsed = now - start_time
-        if elapsed <= 0.0:
-            return
-        speed = float(byte_counter) / elapsed
-        if speed > rate_limit:
-            time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit)
-
-    def temp_name(self, filename):
-        """Returns a temporary filename for the given filename."""
-        if self.params.get('nopart', False) or filename == u'-' or \
-                (os.path.exists(encodeFilename(filename)) and not os.path.isfile(encodeFilename(filename))):
-            return filename
-        return filename + u'.part'
-
-    def undo_temp_name(self, filename):
-        if filename.endswith(u'.part'):
-            return filename[:-len(u'.part')]
-        return filename
-
-    def try_rename(self, old_filename, new_filename):
-        try:
-            if old_filename == new_filename:
-                return
-            os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
-        except (IOError, OSError):
-            self.report_error(u'unable to rename file')
-
-    def try_utime(self, filename, last_modified_hdr):
-        """Try to set the last-modified time of the given file."""
-        if last_modified_hdr is None:
-            return
-        if not os.path.isfile(encodeFilename(filename)):
-            return
-        timestr = last_modified_hdr
-        if timestr is None:
-            return
-        filetime = timeconvert(timestr)
-        if filetime is None:
-            return filetime
-        # Ignore obviously invalid dates
-        if filetime == 0:
-            return
-        try:
-            os.utime(filename, (time.time(), filetime))
-        except:
-            pass
-        return filetime
-
-    def report_destination(self, filename):
-        """Report destination filename."""
-        self.to_screen(u'[download] Destination: ' + filename)
-
-    def _report_progress_status(self, msg, is_last_line=False):
-        fullmsg = u'[download] ' + msg
-        if self.params.get('progress_with_newline', False):
-            self.to_screen(fullmsg)
-        else:
-            if os.name == 'nt':
-                prev_len = getattr(self, '_report_progress_prev_line_length',
-                                   0)
-                if prev_len > len(fullmsg):
-                    fullmsg += u' ' * (prev_len - len(fullmsg))
-                self._report_progress_prev_line_length = len(fullmsg)
-                clear_line = u'\r'
-            else:
-                clear_line = (u'\r\x1b[K' if sys.stderr.isatty() else u'\r')
-            self.to_screen(clear_line + fullmsg, skip_eol=not is_last_line)
-        self.to_console_title(u'youtube-dl ' + msg)
-
-    def report_progress(self, percent, data_len_str, speed, eta):
-        """Report download progress."""
-        if self.params.get('noprogress', False):
-            return
-        if eta is not None:
-            eta_str = self.format_eta(eta)
-        else:
-            eta_str = 'Unknown ETA'
-        if percent is not None:
-            percent_str = self.format_percent(percent)
-        else:
-            percent_str = 'Unknown %'
-        speed_str = self.format_speed(speed)
-
-        msg = (u'%s of %s at %s ETA %s' %
-               (percent_str, data_len_str, speed_str, eta_str))
-        self._report_progress_status(msg)
-
-    def report_progress_live_stream(self, downloaded_data_len, speed, elapsed):
-        if self.params.get('noprogress', False):
-            return
-        downloaded_str = format_bytes(downloaded_data_len)
-        speed_str = self.format_speed(speed)
-        elapsed_str = FileDownloader.format_seconds(elapsed)
-        msg = u'%s at %s (%s)' % (downloaded_str, speed_str, elapsed_str)
-        self._report_progress_status(msg)
-
-    def report_finish(self, data_len_str, tot_time):
-        """Report download finished."""
-        if self.params.get('noprogress', False):
-            self.to_screen(u'[download] Download completed')
-        else:
-            self._report_progress_status(
-                (u'100%% of %s in %s' %
-                 (data_len_str, self.format_seconds(tot_time))),
-                is_last_line=True)
-
-    def report_resuming_byte(self, resume_len):
-        """Report attempt to resume at given byte."""
-        self.to_screen(u'[download] Resuming download at byte %s' % resume_len)
-
-    def report_retry(self, count, retries):
-        """Report retry in case of HTTP error 5xx"""
-        self.to_screen(u'[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count, retries))
-
-    def report_file_already_downloaded(self, file_name):
-        """Report file has already been fully downloaded."""
-        try:
-            self.to_screen(u'[download] %s has already been downloaded' % file_name)
-        except UnicodeEncodeError:
-            self.to_screen(u'[download] The file has already been downloaded')
-
-    def report_unable_to_resume(self):
-        """Report it was impossible to resume download."""
-        self.to_screen(u'[download] Unable to resume')
-
-    def _download_with_rtmpdump(self, filename, url, player_url, page_url, play_path, tc_url, live, conn):
-        def run_rtmpdump(args):
-            start = time.time()
-            resume_percent = None
-            resume_downloaded_data_len = None
-            proc = subprocess.Popen(args, stderr=subprocess.PIPE)
-            cursor_in_new_line = True
-            proc_stderr_closed = False
-            while not proc_stderr_closed:
-                # read line from stderr
-                line = u''
-                while True:
-                    char = proc.stderr.read(1)
-                    if not char:
-                        proc_stderr_closed = True
-                        break
-                    if char in [b'\r', b'\n']:
-                        break
-                    line += char.decode('ascii', 'replace')
-                if not line:
-                    # proc_stderr_closed is True
-                    continue
-                mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec \(([0-9]{1,2}\.[0-9])%\)', line)
-                if mobj:
-                    downloaded_data_len = int(float(mobj.group(1))*1024)
-                    percent = float(mobj.group(2))
-                    if not resume_percent:
-                        resume_percent = percent
-                        resume_downloaded_data_len = downloaded_data_len
-                    eta = self.calc_eta(start, time.time(), 100-resume_percent, percent-resume_percent)
-                    speed = self.calc_speed(start, time.time(), downloaded_data_len-resume_downloaded_data_len)
-                    data_len = None
-                    if percent > 0:
-                        data_len = int(downloaded_data_len * 100 / percent)
-                    data_len_str = u'~' + format_bytes(data_len)
-                    self.report_progress(percent, data_len_str, speed, eta)
-                    cursor_in_new_line = False
-                    self._hook_progress({
-                        'downloaded_bytes': downloaded_data_len,
-                        'total_bytes': data_len,
-                        'tmpfilename': tmpfilename,
-                        'filename': filename,
-                        'status': 'downloading',
-                        'eta': eta,
-                        'speed': speed,
-                    })
-                else:
-                    # no percent for live streams
-                    mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec', line)
-                    if mobj:
-                        downloaded_data_len = int(float(mobj.group(1))*1024)
-                        time_now = time.time()
-                        speed = self.calc_speed(start, time_now, downloaded_data_len)
-                        self.report_progress_live_stream(downloaded_data_len, speed, time_now - start)
-                        cursor_in_new_line = False
-                        self._hook_progress({
-                            'downloaded_bytes': downloaded_data_len,
-                            'tmpfilename': tmpfilename,
-                            'filename': filename,
-                            'status': 'downloading',
-                            'speed': speed,
-                        })
-                    elif self.params.get('verbose', False):
-                        if not cursor_in_new_line:
-                            self.to_screen(u'')
-                        cursor_in_new_line = True
-                        self.to_screen(u'[rtmpdump] '+line)
-            proc.wait()
-            if not cursor_in_new_line:
-                self.to_screen(u'')
-            return proc.returncode
-
-        self.report_destination(filename)
-        tmpfilename = self.temp_name(filename)
-        test = self.params.get('test', False)
-
-        # Check for rtmpdump first
-        try:
-            subprocess.call(['rtmpdump', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
-        except (OSError, IOError):
-            self.report_error(u'RTMP download detected but "rtmpdump" could not be run')
-            return False
-
-        # Download using rtmpdump. rtmpdump returns exit code 2 when
-        # the connection was interrumpted and resuming appears to be
-        # possible. This is part of rtmpdump's normal usage, AFAIK.
-        basic_args = ['rtmpdump', '--verbose', '-r', url, '-o', tmpfilename]
-        if player_url is not None:
-            basic_args += ['--swfVfy', player_url]
-        if page_url is not None:
-            basic_args += ['--pageUrl', page_url]
-        if play_path is not None:
-            basic_args += ['--playpath', play_path]
-        if tc_url is not None:
-            basic_args += ['--tcUrl', url]
-        if test:
-            basic_args += ['--stop', '1']
-        if live:
-            basic_args += ['--live']
-        if conn:
-            basic_args += ['--conn', conn]
-        args = basic_args + [[], ['--resume', '--skip', '1']][self.params.get('continuedl', False)]
-
-        if sys.platform == 'win32' and sys.version_info < (3, 0):
-            # Windows subprocess module does not actually support Unicode
-            # on Python 2.x
-            # See http://stackoverflow.com/a/9951851/35070
-            subprocess_encoding = sys.getfilesystemencoding()
-            args = [a.encode(subprocess_encoding, 'ignore') for a in args]
-        else:
-            subprocess_encoding = None
-
-        if self.params.get('verbose', False):
-            if subprocess_encoding:
-                str_args = [
-                    a.decode(subprocess_encoding) if isinstance(a, bytes) else a
-                    for a in args]
-            else:
-                str_args = args
-            try:
-                import pipes
-                shell_quote = lambda args: ' '.join(map(pipes.quote, str_args))
-            except ImportError:
-                shell_quote = repr
-            self.to_screen(u'[debug] rtmpdump command line: ' + shell_quote(str_args))
-
-        retval = run_rtmpdump(args)
-
-        while (retval == 2 or retval == 1) and not test:
-            prevsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen(u'[rtmpdump] %s bytes' % prevsize)
-            time.sleep(5.0) # This seems to be needed
-            retval = run_rtmpdump(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1])
-            cursize = os.path.getsize(encodeFilename(tmpfilename))
-            if prevsize == cursize and retval == 1:
-                break
-             # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those
-            if prevsize == cursize and retval == 2 and cursize > 1024:
-                self.to_screen(u'[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
-                retval = 0
-                break
-        if retval == 0 or (test and retval == 2):
-            fsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen(u'[rtmpdump] %s bytes' % fsize)
-            self.try_rename(tmpfilename, filename)
-            self._hook_progress({
-                'downloaded_bytes': fsize,
-                'total_bytes': fsize,
-                'filename': filename,
-                'status': 'finished',
-            })
-            return True
-        else:
-            self.to_stderr(u"\n")
-            self.report_error(u'rtmpdump exited with code %d' % retval)
-            return False
-
-    def _download_with_mplayer(self, filename, url):
-        self.report_destination(filename)
-        tmpfilename = self.temp_name(filename)
-
-        args = ['mplayer', '-really-quiet', '-vo', 'null', '-vc', 'dummy', '-dumpstream', '-dumpfile', tmpfilename, url]
-        # Check for mplayer first
-        try:
-            subprocess.call(['mplayer', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
-        except (OSError, IOError):
-            self.report_error(u'MMS or RTSP download detected but "%s" could not be run' % args[0] )
-            return False
-
-        # Download using mplayer. 
-        retval = subprocess.call(args)
-        if retval == 0:
-            fsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen(u'\r[%s] %s bytes' % (args[0], fsize))
-            self.try_rename(tmpfilename, filename)
-            self._hook_progress({
-                'downloaded_bytes': fsize,
-                'total_bytes': fsize,
-                'filename': filename,
-                'status': 'finished',
-            })
-            return True
-        else:
-            self.to_stderr(u"\n")
-            self.report_error(u'mplayer exited with code %d' % retval)
-            return False
-
-    def _download_m3u8_with_ffmpeg(self, filename, url):
-        self.report_destination(filename)
-        tmpfilename = self.temp_name(filename)
-
-        args = ['-y', '-i', url, '-f', 'mp4', '-c', 'copy',
-            '-bsf:a', 'aac_adtstoasc', tmpfilename]
-
-        for program in ['avconv', 'ffmpeg']:
-            try:
-                subprocess.call([program, '-version'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
-                break
-            except (OSError, IOError):
-                pass
-        else:
-            self.report_error(u'm3u8 download detected but ffmpeg or avconv could not be found')
-        cmd = [program] + args
-
-        retval = subprocess.call(cmd)
-        if retval == 0:
-            fsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen(u'\r[%s] %s bytes' % (args[0], fsize))
-            self.try_rename(tmpfilename, filename)
-            self._hook_progress({
-                'downloaded_bytes': fsize,
-                'total_bytes': fsize,
-                'filename': filename,
-                'status': 'finished',
-            })
-            return True
-        else:
-            self.to_stderr(u"\n")
-            self.report_error(u'ffmpeg exited with code %d' % retval)
-            return False
+# Legacy file for backwards compatibility, use youtube_dl.downloader instead!
+from .downloader import FileDownloader as RealFileDownloader
+from .downloader import get_suitable_downloader
 
 
+# This class reproduces the old behaviour of FileDownloader
+class FileDownloader(RealFileDownloader):
     def _do_download(self, filename, info_dict):
-        url = info_dict['url']
-
-        # Check file already present
-        if self.params.get('continuedl', False) and os.path.isfile(encodeFilename(filename)) and not self.params.get('nopart', False):
-            self.report_file_already_downloaded(filename)
-            self._hook_progress({
-                'filename': filename,
-                'status': 'finished',
-                'total_bytes': os.path.getsize(encodeFilename(filename)),
-            })
-            return True
-
-        # Attempt to download using rtmpdump
-        if url.startswith('rtmp'):
-            return self._download_with_rtmpdump(filename, url,
-                                                info_dict.get('player_url', None),
-                                                info_dict.get('page_url', None),
-                                                info_dict.get('play_path', None),
-                                                info_dict.get('tc_url', None),
-                                                info_dict.get('rtmp_live', False),
-                                                info_dict.get('rtmp_conn', None))
-
-        # Attempt to download using mplayer
-        if url.startswith('mms') or url.startswith('rtsp'):
-            return self._download_with_mplayer(filename, url)
-
-        # m3u8 manifest are downloaded with ffmpeg
-        if determine_ext(url) == u'm3u8':
-            return self._download_m3u8_with_ffmpeg(filename, url)
-
-        tmpfilename = self.temp_name(filename)
-        stream = None
-
-        # Do not include the Accept-Encoding header
-        headers = {'Youtubedl-no-compression': 'True'}
-        if 'user_agent' in info_dict:
-            headers['Youtubedl-user-agent'] = info_dict['user_agent']
-        basic_request = compat_urllib_request.Request(url, None, headers)
-        request = compat_urllib_request.Request(url, None, headers)
-
-        if self.params.get('test', False):
-            request.add_header('Range','bytes=0-10240')
-
-        # Establish possible resume length
-        if os.path.isfile(encodeFilename(tmpfilename)):
-            resume_len = os.path.getsize(encodeFilename(tmpfilename))
-        else:
-            resume_len = 0
-
-        open_mode = 'wb'
-        if resume_len != 0:
-            if self.params.get('continuedl', False):
-                self.report_resuming_byte(resume_len)
-                request.add_header('Range','bytes=%d-' % resume_len)
-                open_mode = 'ab'
-            else:
-                resume_len = 0
-
-        count = 0
-        retries = self.params.get('retries', 0)
-        while count <= retries:
-            # Establish connection
-            try:
-                if count == 0 and 'urlhandle' in info_dict:
-                    data = info_dict['urlhandle']
-                data = compat_urllib_request.urlopen(request)
-                break
-            except (compat_urllib_error.HTTPError, ) as err:
-                if (err.code < 500 or err.code >= 600) and err.code != 416:
-                    # Unexpected HTTP error
-                    raise
-                elif err.code == 416:
-                    # Unable to resume (requested range not satisfiable)
-                    try:
-                        # Open the connection again without the range header
-                        data = compat_urllib_request.urlopen(basic_request)
-                        content_length = data.info()['Content-Length']
-                    except (compat_urllib_error.HTTPError, ) as err:
-                        if err.code < 500 or err.code >= 600:
-                            raise
-                    else:
-                        # Examine the reported length
-                        if (content_length is not None and
-                                (resume_len - 100 < int(content_length) < resume_len + 100)):
-                            # The file had already been fully downloaded.
-                            # Explanation to the above condition: in issue #175 it was revealed that
-                            # YouTube sometimes adds or removes a few bytes from the end of the file,
-                            # changing the file size slightly and causing problems for some users. So
-                            # I decided to implement a suggested change and consider the file
-                            # completely downloaded if the file size differs less than 100 bytes from
-                            # the one in the hard drive.
-                            self.report_file_already_downloaded(filename)
-                            self.try_rename(tmpfilename, filename)
-                            self._hook_progress({
-                                'filename': filename,
-                                'status': 'finished',
-                            })
-                            return True
-                        else:
-                            # The length does not match, we start the download over
-                            self.report_unable_to_resume()
-                            open_mode = 'wb'
-                            break
-            # Retry
-            count += 1
-            if count <= retries:
-                self.report_retry(count, retries)
-
-        if count > retries:
-            self.report_error(u'giving up after %s retries' % retries)
-            return False
-
-        data_len = data.info().get('Content-length', None)
-        if data_len is not None:
-            data_len = int(data_len) + resume_len
-            min_data_len = self.params.get("min_filesize", None)
-            max_data_len =  self.params.get("max_filesize", None)
-            if min_data_len is not None and data_len < min_data_len:
-                self.to_screen(u'\r[download] File is smaller than min-filesize (%s bytes < %s bytes). Aborting.' % (data_len, min_data_len))
-                return False
-            if max_data_len is not None and data_len > max_data_len:
-                self.to_screen(u'\r[download] File is larger than max-filesize (%s bytes > %s bytes). Aborting.' % (data_len, max_data_len))
-                return False
-
-        data_len_str = format_bytes(data_len)
-        byte_counter = 0 + resume_len
-        block_size = self.params.get('buffersize', 1024)
-        start = time.time()
-        while True:
-            # Download and write
-            before = time.time()
-            data_block = data.read(block_size)
-            after = time.time()
-            if len(data_block) == 0:
-                break
-            byte_counter += len(data_block)
-
-            # Open file just in time
-            if stream is None:
-                try:
-                    (stream, tmpfilename) = sanitize_open(tmpfilename, open_mode)
-                    assert stream is not None
-                    filename = self.undo_temp_name(tmpfilename)
-                    self.report_destination(filename)
-                except (OSError, IOError) as err:
-                    self.report_error(u'unable to open for writing: %s' % str(err))
-                    return False
-            try:
-                stream.write(data_block)
-            except (IOError, OSError) as err:
-                self.to_stderr(u"\n")
-                self.report_error(u'unable to write data: %s' % str(err))
-                return False
-            if not self.params.get('noresizebuffer', False):
-                block_size = self.best_block_size(after - before, len(data_block))
-
-            # Progress message
-            speed = self.calc_speed(start, time.time(), byte_counter - resume_len)
-            if data_len is None:
-                eta = percent = None
-            else:
-                percent = self.calc_percent(byte_counter, data_len)
-                eta = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len)
-            self.report_progress(percent, data_len_str, speed, eta)
-
-            self._hook_progress({
-                'downloaded_bytes': byte_counter,
-                'total_bytes': data_len,
-                'tmpfilename': tmpfilename,
-                'filename': filename,
-                'status': 'downloading',
-                'eta': eta,
-                'speed': speed,
-            })
-
-            # Apply rate limit
-            self.slow_down(start, byte_counter - resume_len)
-
-        if stream is None:
-            self.to_stderr(u"\n")
-            self.report_error(u'Did not get any data blocks')
-            return False
-        stream.close()
-        self.report_finish(data_len_str, (time.time() - start))
-        if data_len is not None and byte_counter != data_len:
-            raise ContentTooShortError(byte_counter, int(data_len))
-        self.try_rename(tmpfilename, filename)
-
-        # Update file modification time
-        if self.params.get('updatetime', True):
-            info_dict['filetime'] = self.try_utime(filename, data.info().get('last-modified', None))
-
-        self._hook_progress({
-            'downloaded_bytes': byte_counter,
-            'total_bytes': byte_counter,
-            'filename': filename,
-            'status': 'finished',
-        })
-
-        return True
-
-    def _hook_progress(self, status):
+        real_fd = get_suitable_downloader(info_dict)(self.ydl, self.params)
         for ph in self._progress_hooks:
-            ph(status)
-
-    def add_progress_hook(self, ph):
-        """ ph gets called on download progress, with a dictionary with the entries
-        * filename: The final filename
-        * status: One of "downloading" and "finished"
-
-        It can also have some of the following entries:
-
-        * downloaded_bytes: Bytes on disks
-        * total_bytes: Total bytes, None if unknown
-        * tmpfilename: The filename we're currently writing to
-        * eta: The estimated time in seconds, None if unknown
-        * speed: The download speed in bytes/second, None if unknown
-
-        Hooks are guaranteed to be called at least once (with status "finished")
-        if the download is successful.
-        """
-        self._progress_hooks.append(ph)
+            real_fd.add_progress_hook(ph)
+        return real_fd.download(filename, info_dict)

youtube_dl/InfoExtractors.py

@@ -1,4 +0,0 @@
-# Legacy file for backwards compatibility, use youtube_dl.extractor instead!
-
-from .extractor.common import InfoExtractor, SearchInfoExtractor
-from .extractor import gen_extractors, get_info_extractor

youtube_dl/__init__.py

@@ -38,12 +38,29 @@ __authors__  = (
     'Takuya Tsuchida',
     'Sergey M.',
     'Michael Orlitzky',
+    'Chris Gahan',
+    'Saimadhav Heblikar',
+    'Mike Col',
+    'Oleg Prutz',
+    'pulpe',
+    'Andreas Schmitz',
+    'Michael Kaiser',
+    'Niklas Laxström',
+    'David Triendl',
+    'Anthony Weems',
+    'David Wagner',
+    'Juan C. Olivares',
+    'Mattias Harrysson',
+    'phaer',
+    'Sainyam Kapoor',
+    'Nicolas Évrard',
 )
 
 __license__ = 'Public Domain'
 
 import codecs
-import getpass
+import io
+import locale
 import optparse
 import os
 import random
@@ -53,14 +70,17 @@ import sys
 
 
 from .utils import (
+    compat_getpass,
     compat_print,
     DateRange,
+    DEFAULT_OUTTMPL,
     decodeOption,
     get_term_width,
     DownloadError,
     get_cachedir,
     MaxDownloadsReached,
     preferredencoding,
+    read_batch_urls,
     SameFileError,
     setproctitle,
     std_headers,
@@ -73,11 +93,14 @@ from .FileDownloader import (
 from .extractor import gen_extractors
 from .version import __version__
 from .YoutubeDL import YoutubeDL
-from .PostProcessor import (
+from .postprocessor import (
+    AtomicParsleyPP,
+    FFmpegAudioFixPP,
     FFmpegMetadataPP,
     FFmpegVideoConvertor,
     FFmpegExtractAudioPP,
     FFmpegEmbedSubtitlePP,
+    XAttrMetadataPP,
 )
 
 
@@ -95,6 +118,43 @@ def parseOpts(overrideArguments=None):
             optionf.close()
         return res
 
+    def _readUserConf():
+        xdg_config_home = os.environ.get('XDG_CONFIG_HOME')
+        if xdg_config_home:
+            userConfFile = os.path.join(xdg_config_home, 'youtube-dl', 'config')
+            if not os.path.isfile(userConfFile):
+                userConfFile = os.path.join(xdg_config_home, 'youtube-dl.conf')
+        else:
+            userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl', 'config')
+            if not os.path.isfile(userConfFile):
+                userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl.conf')
+        userConf = _readOptions(userConfFile, None)
+
+        if userConf is None:
+            appdata_dir = os.environ.get('appdata')
+            if appdata_dir:
+                userConf = _readOptions(
+                    os.path.join(appdata_dir, 'youtube-dl', 'config'),
+                    default=None)
+                if userConf is None:
+                    userConf = _readOptions(
+                        os.path.join(appdata_dir, 'youtube-dl', 'config.txt'),
+                        default=None)
+
+        if userConf is None:
+            userConf = _readOptions(
+                os.path.join(os.path.expanduser('~'), 'youtube-dl.conf'),
+                default=None)
+        if userConf is None:
+            userConf = _readOptions(
+                os.path.join(os.path.expanduser('~'), 'youtube-dl.conf.txt'),
+                default=None)
+
+        if userConf is None:
+            userConf = []
+
+        return userConf
+
     def _format_option_string(option):
         ''' ('-o', '--option') -> -o, --format METAVAR'''
 
@@ -161,7 +221,7 @@ def parseOpts(overrideArguments=None):
     general.add_option('-U', '--update',
             action='store_true', dest='update_self', help='update this program to latest version. Make sure that you have sufficient permissions (run with sudo if needed)')
     general.add_option('-i', '--ignore-errors',
-            action='store_true', dest='ignoreerrors', help='continue on download errors, for example to to skip unavailable videos in a playlist', default=False)
+            action='store_true', dest='ignoreerrors', help='continue on download errors, for example to skip unavailable videos in a playlist', default=False)
     general.add_option('--abort-on-error',
             action='store_false', dest='ignoreerrors',
             help='Abort downloading of further videos (in the playlist or the command line) if an error occurs')
@@ -173,6 +233,9 @@ def parseOpts(overrideArguments=None):
     general.add_option('--referer',
             dest='referer', help='specify a custom referer, use if the video access is restricted to one domain',
             metavar='REF', default=None)
+    general.add_option('--add-header',
+            dest='headers', help='specify a custom HTTP header and its value, separated by a colon \':\'. You can use this option multiple times', action="append",
+            metavar='FIELD:VALUE')
     general.add_option('--list-extractors',
             action='store_true', dest='list_extractors',
             help='List all supported extractors and the URLs they would handle', default=False)
@@ -183,19 +246,32 @@ def parseOpts(overrideArguments=None):
         '--proxy', dest='proxy', default=None, metavar='URL',
         help='Use the specified HTTP/HTTPS proxy. Pass in an empty string (--proxy "") for direct connection')
     general.add_option('--no-check-certificate', action='store_true', dest='no_check_certificate', default=False, help='Suppress HTTPS certificate validation.')
+    general.add_option(
+        '--prefer-insecure', '--prefer-unsecure', action='store_true', dest='prefer_insecure',
+        help='Use an unencrypted connection to retrieve information about the video. (Currently supported only for YouTube)')
     general.add_option(
         '--cache-dir', dest='cachedir', default=get_cachedir(), metavar='DIR',
-        help='Location in the filesystem where youtube-dl can store downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl .')
+        help='Location in the filesystem where youtube-dl can store some downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl . At the moment, only YouTube player files (for videos with obfuscated signatures) are cached, but that may change.')
     general.add_option(
         '--no-cache-dir', action='store_const', const=None, dest='cachedir',
         help='Disable filesystem caching')
     general.add_option(
         '--socket-timeout', dest='socket_timeout',
-        type=float, default=None, help=optparse.SUPPRESS_HELP)
+        type=float, default=None, help=u'Time to wait before giving up, in seconds')
     general.add_option(
         '--bidi-workaround', dest='bidi_workaround', action='store_true',
         help=u'Work around terminals that lack bidirectional text support. Requires bidiv or fribidi executable in PATH')
-
+    general.add_option(
+        '--default-search',
+        dest='default_search', metavar='PREFIX',
+        help='Use this prefix for unqualified URLs. For example "gvsearch2:" downloads two videos from google videos for  youtube-dl "large apple". By default (with value "auto") youtube-dl guesses.')
+    general.add_option(
+        '--ignore-config',
+        action='store_true',
+        help='Do not read configuration files. When given in the global configuration file /etc/youtube-dl.conf: do not read the user configuration in ~/.config/youtube-dl.conf (%APPDATA%/youtube-dl/config.txt on Windows)')
+    general.add_option(
+        '--encoding', dest='encoding', metavar='ENCODING',
+        help='Force the specified encoding (experimental)')
 
     selection.add_option(
         '--playlist-start',
@@ -213,8 +289,12 @@ def parseOpts(overrideArguments=None):
     selection.add_option('--min-filesize', metavar='SIZE', dest='min_filesize', help="Do not download any videos smaller than SIZE (e.g. 50k or 44.6m)", default=None)
     selection.add_option('--max-filesize', metavar='SIZE', dest='max_filesize', help="Do not download any videos larger than SIZE (e.g. 50k or 44.6m)", default=None)
     selection.add_option('--date', metavar='DATE', dest='date', help='download only videos uploaded in this date', default=None)
-    selection.add_option('--datebefore', metavar='DATE', dest='datebefore', help='download only videos uploaded before this date', default=None)
-    selection.add_option('--dateafter', metavar='DATE', dest='dateafter', help='download only videos uploaded after this date', default=None)
+    selection.add_option(
+        '--datebefore', metavar='DATE', dest='datebefore', default=None,
+        help='download only videos uploaded on or before this date (i.e. inclusive)')
+    selection.add_option(
+        '--dateafter', metavar='DATE', dest='dateafter', default=None,
+        help='download only videos uploaded on or after this date (i.e. inclusive)')
     selection.add_option(
         '--min-views', metavar='COUNT', dest='min_views',
         default=None, type=int,
@@ -230,7 +310,14 @@ def parseOpts(overrideArguments=None):
     selection.add_option('--download-archive', metavar='FILE',
                          dest='download_archive',
                          help='Download only videos not listed in the archive file. Record the IDs of all downloaded videos in it.')
-
+    selection.add_option(
+        '--include-ads', dest='include_ads',
+        action='store_true',
+        help='Download advertisements as well (experimental)')
+    selection.add_option(
+        '--youtube-include-dash-manifest', action='store_true',
+        dest='youtube_include_dash_manifest', default=False,
+        help='Try to download the DASH manifest on YouTube videos (experimental)')
 
     authentication.add_option('-u', '--username',
             dest='username', metavar='USERNAME', help='account username')
@@ -239,12 +326,12 @@ def parseOpts(overrideArguments=None):
     authentication.add_option('-n', '--netrc',
             action='store_true', dest='usenetrc', help='use .netrc authentication data', default=False)
     authentication.add_option('--video-password',
-            dest='videopassword', metavar='PASSWORD', help='video password (vimeo only)')
+            dest='videopassword', metavar='PASSWORD', help='video password (vimeo, smotri)')
 
 
     video_format.add_option('-f', '--format',
-            action='store', dest='format', metavar='FORMAT', default='best',
-            help='video format code, specify the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported')
+            action='store', dest='format', metavar='FORMAT', default=None,
+            help='video format code, specify the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported. You can also use the special names "best", "bestvideo", "bestaudio", "worst", "worstvideo" and "worstaudio". By default, youtube-dl will pick the best quality.')
     video_format.add_option('--all-formats',
             action='store_const', dest='format', help='download all available video formats', const='all')
     video_format.add_option('--prefer-free-formats',
@@ -252,7 +339,7 @@ def parseOpts(overrideArguments=None):
     video_format.add_option('--max-quality',
             action='store', dest='format_limit', metavar='FORMAT', help='highest quality format to download')
     video_format.add_option('-F', '--list-formats',
-            action='store_true', dest='listformats', help='list all available formats (currently youtube only)')
+            action='store_true', dest='listformats', help='list all available formats')
 
     subtitles.add_option('--write-sub', '--write-srt',
             action='store_true', dest='writesubtitles',
@@ -287,6 +374,10 @@ def parseOpts(overrideArguments=None):
 
     verbosity.add_option('-q', '--quiet',
             action='store_true', dest='quiet', help='activates quiet mode', default=False)
+    verbosity.add_option(
+        '--no-warnings',
+        dest='no_warnings', action='store_true', default=False,
+        help='Ignore warnings')
     verbosity.add_option('-s', '--simulate',
             action='store_true', dest='simulate', help='do not download the video and do not write anything to disk', default=False)
     verbosity.add_option('--skip-download',
@@ -314,7 +405,7 @@ def parseOpts(overrideArguments=None):
             help='simulate, quiet but print output format', default=False)
     verbosity.add_option('-j', '--dump-json',
             action='store_true', dest='dumpjson',
-            help='simulate, quiet but print JSON information', default=False)
+            help='simulate, quiet but print JSON information. See --output for a description of available keys.', default=False)
     verbosity.add_option('--newline',
             action='store_true', dest='progress_with_newline', help='output progress bar as new lines', default=False)
     verbosity.add_option('--no-progress',
@@ -326,13 +417,16 @@ def parseOpts(overrideArguments=None):
             action='store_true', dest='verbose', help='print various debugging information', default=False)
     verbosity.add_option('--dump-intermediate-pages',
             action='store_true', dest='dump_intermediate_pages', default=False,
-            help='print downloaded pages to debug problems(very verbose)')
+            help='print downloaded pages to debug problems (very verbose)')
     verbosity.add_option('--write-pages',
             action='store_true', dest='write_pages', default=False,
             help='Write downloaded intermediary pages to files in the current directory to debug problems')
     verbosity.add_option('--youtube-print-sig-code',
             action='store_true', dest='youtube_print_sig_code', default=False,
             help=optparse.SUPPRESS_HELP)
+    verbosity.add_option('--print-traffic',
+            dest='debug_printtraffic', action='store_true', default=False,
+            help='Display sent and read HTTP traffic')
 
 
     filesystem.add_option('-t', '--title',
@@ -350,12 +444,14 @@ def parseOpts(overrideArguments=None):
                   '%(uploader)s for the uploader name, %(uploader_id)s for the uploader nickname if different, '
                   '%(autonumber)s to get an automatically incremented number, '
                   '%(ext)s for the filename extension, '
-                  '%(format)s for the format description (like "22 - 1280x720" or "HD"),'
-                  '%(format_id)s for the unique id of the format (like Youtube\'s itags: "137"),'
+                  '%(format)s for the format description (like "22 - 1280x720" or "HD"), '
+                  '%(format_id)s for the unique id of the format (like Youtube\'s itags: "137"), '
                   '%(upload_date)s for the upload date (YYYYMMDD), '
                   '%(extractor)s for the provider (youtube, metacafe, etc), '
-                  '%(id)s for the video id , %(playlist)s for the playlist the video is in, '
+                  '%(id)s for the video id, %(playlist)s for the playlist the video is in, '
                   '%(playlist_index)s for the position in the playlist and %% for a literal percent. '
+                  '%(height)s and %(width)s for the width and height of the video format. '
+                  '%(resolution)s for a textual description of the resolution of the video format. '
                   'Use - to output to stdout. Can also be used to download to a different directory, '
                   'for example with -o \'/my/downloads/%(uploader)s/%(title)s-%(id)s.%(ext)s\' .'))
     filesystem.add_option('--autonumber-size',
@@ -368,7 +464,7 @@ def parseOpts(overrideArguments=None):
             dest='batchfile', metavar='FILE', help='file containing URLs to download (\'-\' for stdin)')
     filesystem.add_option('--load-info',
             dest='load_info_filename', metavar='FILE',
-            help='json file containing the video information (created with the "--write-json" option')
+            help='json file containing the video information (created with the "--write-json" option)')
     filesystem.add_option('-w', '--no-overwrites',
             action='store_true', dest='nooverwrites', help='do not overwrite files', default=False)
     filesystem.add_option('-c', '--continue',
@@ -411,8 +507,16 @@ def parseOpts(overrideArguments=None):
             help='do not overwrite post-processed files; the post-processed files are overwritten by default')
     postproc.add_option('--embed-subs', action='store_true', dest='embedsubtitles', default=False,
             help='embed subtitles in the video (only for mp4 videos)')
+    postproc.add_option('--embed-thumbnail', action='store_true', dest='embedthumbnail', default=False,
+            help='embed thumbnail in the audio as cover art')
     postproc.add_option('--add-metadata', action='store_true', dest='addmetadata', default=False,
-            help='add metadata to the files')
+            help='write metadata to the video file')
+    postproc.add_option('--xattrs', action='store_true', dest='xattrs', default=False,
+            help='write metadata to the video file\'s xattrs (using dublin core and xdg standards)')
+    postproc.add_option('--prefer-avconv', action='store_false', dest='prefer_ffmpeg',
+        help='Prefer avconv over ffmpeg for running the postprocessors (default)')
+    postproc.add_option('--prefer-ffmpeg', action='store_true', dest='prefer_ffmpeg',
+        help='Prefer ffmpeg over avconv for running the postprocessors')
 
 
     parser.add_option_group(general)
@@ -430,44 +534,18 @@ def parseOpts(overrideArguments=None):
         if opts.verbose:
             write_string(u'[debug] Override config: ' + repr(overrideArguments) + '\n')
     else:
-        systemConf = _readOptions('/etc/youtube-dl.conf')
-
-        xdg_config_home = os.environ.get('XDG_CONFIG_HOME')
-        if xdg_config_home:
-            userConfFile = os.path.join(xdg_config_home, 'youtube-dl', 'config')
-            if not os.path.isfile(userConfFile):
-                userConfFile = os.path.join(xdg_config_home, 'youtube-dl.conf')
-        else:
-            userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl', 'config')
-            if not os.path.isfile(userConfFile):
-                userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl.conf')
-        userConf = _readOptions(userConfFile, None)
-
-        if userConf is None:
-            appdata_dir = os.environ.get('appdata')
-            if appdata_dir:
-                userConf = _readOptions(
-                    os.path.join(appdata_dir, 'youtube-dl', 'config'),
-                    default=None)
-                if userConf is None:
-                    userConf = _readOptions(
-                        os.path.join(appdata_dir, 'youtube-dl', 'config.txt'),
-                        default=None)
-
-        if userConf is None:
-            userConf = _readOptions(
-                os.path.join(os.path.expanduser('~'), 'youtube-dl.conf'),
-                default=None)
-        if userConf is None:
-            userConf = _readOptions(
-                os.path.join(os.path.expanduser('~'), 'youtube-dl.conf.txt'),
-                default=None)
-
-        if userConf is None:
-            userConf = []
-
         commandLineConf = sys.argv[1:]
+        if '--ignore-config' in commandLineConf:
+            systemConf = []
+            userConf = []
+        else:
+            systemConf = _readOptions('/etc/youtube-dl.conf')
+            if '--ignore-config' in systemConf:
+                userConf = []
+            else:
+                userConf = _readUserConf()
         argv = systemConf + userConf + commandLineConf
+
         opts, args = parser.parse_args(argv)
         if opts.verbose:
             write_string(u'[debug] System config: ' + repr(_hide_login_info(systemConf)) + '\n')
@@ -495,28 +573,38 @@ def _real_main(argv=None):
     if opts.referer is not None:
         std_headers['Referer'] = opts.referer
 
+    # Custom HTTP headers
+    if opts.headers is not None:
+        for h in opts.headers:
+            if h.find(':', 1) < 0:
+                parser.error(u'wrong header formatting, it should be key:value, not "%s"'%h)
+            key, value = h.split(':', 2)
+            if opts.verbose:
+                write_string(u'[debug] Adding header from command line option %s:%s\n'%(key, value))
+            std_headers[key] = value
+
     # Dump user agent
     if opts.dump_user_agent:
         compat_print(std_headers['User-Agent'])
         sys.exit(0)
 
     # Batch file verification
-    batchurls = []
+    batch_urls = []
     if opts.batchfile is not None:
         try:
             if opts.batchfile == '-':
                 batchfd = sys.stdin
             else:
-                batchfd = open(opts.batchfile, 'r')
-            batchurls = batchfd.readlines()
-            batchurls = [x.strip() for x in batchurls]
-            batchurls = [x for x in batchurls if len(x) > 0 and not re.search(r'^[#/;]', x)]
+                batchfd = io.open(opts.batchfile, 'r', encoding='utf-8', errors='ignore')
+            batch_urls = read_batch_urls(batchfd)
             if opts.verbose:
-                write_string(u'[debug] Batch file urls: ' + repr(batchurls) + u'\n')
+                write_string(u'[debug] Batch file urls: ' + repr(batch_urls) + u'\n')
         except IOError:
             sys.exit(u'ERROR: batch file could not be read')
-    all_urls = batchurls + args
+    all_urls = batch_urls + args
     all_urls = [url.strip() for url in all_urls]
+    _enc = preferredencoding()
+    all_urls = [url.decode(_enc, 'ignore') if isinstance(url, bytes) else url for url in all_urls]
 
     extractors = gen_extractors()
 
@@ -546,13 +634,13 @@ def _real_main(argv=None):
     if opts.usenetrc and (opts.username is not None or opts.password is not None):
         parser.error(u'using .netrc conflicts with giving username/password')
     if opts.password is not None and opts.username is None:
-        parser.error(u' account username missing\n')
+        parser.error(u'account username missing\n')
     if opts.outtmpl is not None and (opts.usetitle or opts.autonumber or opts.useid):
         parser.error(u'using output template conflicts with using title, video ID or auto number')
     if opts.usetitle and opts.useid:
         parser.error(u'using title conflicts with using video ID')
     if opts.username is not None and opts.password is None:
-        opts.password = getpass.getpass(u'Type account password and press return:')
+        opts.password = compat_getpass(u'Type account password and press [Return]: ')
     if opts.ratelimit is not None:
         numeric_limit = FileDownloader.parse_bytes(opts.ratelimit)
         if numeric_limit is None:
@@ -590,12 +678,18 @@ def _real_main(argv=None):
         if not opts.audioquality.isdigit():
             parser.error(u'invalid audio quality specified')
     if opts.recodevideo is not None:
-        if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg']:
+        if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg', 'mkv']:
             parser.error(u'invalid video recode format specified')
     if opts.date is not None:
         date = DateRange.day(opts.date)
     else:
         date = DateRange(opts.dateafter, opts.datebefore)
+    if opts.default_search not in ('auto', 'auto_warning', None) and ':' not in opts.default_search:
+        parser.error(u'--default-search invalid; did you forget a colon (:) at the end?')
+
+    # Do not download videos when there are audio-only formats
+    if opts.extractaudio and not opts.keepvideo and opts.format is None:
+        opts.format = 'bestaudio/best'
 
     # --all-sub automatically sets --write-sub if --write-auto-sub is not given
     # this was the old behaviour if only --all-sub was given.
@@ -613,13 +707,14 @@ def _real_main(argv=None):
             or (opts.usetitle and u'%(title)s-%(id)s.%(ext)s')
             or (opts.useid and u'%(id)s.%(ext)s')
             or (opts.autonumber and u'%(autonumber)s-%(id)s.%(ext)s')
-            or u'%(title)s-%(id)s.%(ext)s')
+            or DEFAULT_OUTTMPL)
     if not os.path.splitext(outtmpl)[1] and opts.extractaudio:
         parser.error(u'Cannot download a video and extract audio into the same'
                      u' file! Use "{0}.%(ext)s" instead of "{0}" as the output'
                      u' template'.format(outtmpl))
 
     any_printing = opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson
+    download_archive_fn = os.path.expanduser(opts.download_archive) if opts.download_archive is not None else opts.download_archive
 
     ydl_opts = {
         'usenetrc': opts.usenetrc,
@@ -627,6 +722,7 @@ def _real_main(argv=None):
         'password': opts.password,
         'videopassword': opts.videopassword,
         'quiet': (opts.quiet or any_printing),
+        'no_warnings': opts.no_warnings,
         'forceurl': opts.geturl,
         'forcetitle': opts.gettitle,
         'forceid': opts.getid,
@@ -687,12 +783,19 @@ def _real_main(argv=None):
         'cachedir': opts.cachedir,
         'youtube_print_sig_code': opts.youtube_print_sig_code,
         'age_limit': opts.age_limit,
-        'download_archive': opts.download_archive,
+        'download_archive': download_archive_fn,
         'cookiefile': opts.cookiefile,
         'nocheckcertificate': opts.no_check_certificate,
+        'prefer_insecure': opts.prefer_insecure,
         'proxy': opts.proxy,
         'socket_timeout': opts.socket_timeout,
         'bidi_workaround': opts.bidi_workaround,
+        'debug_printtraffic': opts.debug_printtraffic,
+        'prefer_ffmpeg': opts.prefer_ffmpeg,
+        'include_ads': opts.include_ads,
+        'default_search': opts.default_search,
+        'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
+        'encoding': opts.encoding,
     }
 
     with YoutubeDL(ydl_opts) as ydl:
@@ -709,6 +812,12 @@ def _real_main(argv=None):
             ydl.add_post_processor(FFmpegVideoConvertor(preferedformat=opts.recodevideo))
         if opts.embedsubtitles:
             ydl.add_post_processor(FFmpegEmbedSubtitlePP(subtitlesformat=opts.subtitlesformat))
+        if opts.xattrs:
+            ydl.add_post_processor(XAttrMetadataPP())
+        if opts.embedthumbnail:
+            if not opts.addmetadata:
+                ydl.add_post_processor(FFmpegAudioFixPP())
+            ydl.add_post_processor(AtomicParsleyPP())
 
         # Update version
         if opts.update_self:

youtube_dl/downloader/__init__.py

@@ -0,0 +1,29 @@
+from __future__ import unicode_literals
+
+from .common import FileDownloader
+from .hls import HlsFD
+from .http import HttpFD
+from .mplayer import MplayerFD
+from .rtmp import RtmpFD
+from .f4m import F4mFD
+
+from ..utils import (
+    determine_ext,
+)
+
+
+def get_suitable_downloader(info_dict):
+    """Get the downloader class that can handle the info dict."""
+    url = info_dict['url']
+    protocol = info_dict.get('protocol')
+
+    if url.startswith('rtmp'):
+        return RtmpFD
+    if (protocol == 'm3u8') or (protocol is None and determine_ext(url) == 'm3u8'):
+        return HlsFD
+    if url.startswith('mms') or url.startswith('rtsp'):
+        return MplayerFD
+    if determine_ext(url) == 'f4m':
+        return F4mFD
+    else:
+        return HttpFD

youtube_dl/downloader/common.py

@@ -0,0 +1,317 @@
+import os
+import re
+import sys
+import time
+
+from ..utils import (
+    compat_str,
+    encodeFilename,
+    format_bytes,
+    timeconvert,
+)
+
+
+class FileDownloader(object):
+    """File Downloader class.
+
+    File downloader objects are the ones responsible of downloading the
+    actual video file and writing it to disk.
+
+    File downloaders accept a lot of parameters. In order not to saturate
+    the object constructor with arguments, it receives a dictionary of
+    options instead.
+
+    Available options:
+
+    verbose:           Print additional info to stdout.
+    quiet:             Do not print messages to stdout.
+    ratelimit:         Download speed limit, in bytes/sec.
+    retries:           Number of times to retry for HTTP error 5xx
+    buffersize:        Size of download buffer in bytes.
+    noresizebuffer:    Do not automatically resize the download buffer.
+    continuedl:        Try to continue downloads if possible.
+    noprogress:        Do not print the progress bar.
+    logtostderr:       Log messages to stderr instead of stdout.
+    consoletitle:      Display progress in console window's titlebar.
+    nopart:            Do not use temporary .part files.
+    updatetime:        Use the Last-modified header to set output file timestamps.
+    test:              Download only first bytes to test the downloader.
+    min_filesize:      Skip files smaller than this size
+    max_filesize:      Skip files larger than this size
+
+    Subclasses of this one must re-define the real_download method.
+    """
+
+    params = None
+
+    def __init__(self, ydl, params):
+        """Create a FileDownloader object with the given options."""
+        self.ydl = ydl
+        self._progress_hooks = []
+        self.params = params
+
+    @staticmethod
+    def format_seconds(seconds):
+        (mins, secs) = divmod(seconds, 60)
+        (hours, mins) = divmod(mins, 60)
+        if hours > 99:
+            return '--:--:--'
+        if hours == 0:
+            return '%02d:%02d' % (mins, secs)
+        else:
+            return '%02d:%02d:%02d' % (hours, mins, secs)
+
+    @staticmethod
+    def calc_percent(byte_counter, data_len):
+        if data_len is None:
+            return None
+        return float(byte_counter) / float(data_len) * 100.0
+
+    @staticmethod
+    def format_percent(percent):
+        if percent is None:
+            return '---.-%'
+        return '%6s' % ('%3.1f%%' % percent)
+
+    @staticmethod
+    def calc_eta(start, now, total, current):
+        if total is None:
+            return None
+        dif = now - start
+        if current == 0 or dif < 0.001: # One millisecond
+            return None
+        rate = float(current) / dif
+        return int((float(total) - float(current)) / rate)
+
+    @staticmethod
+    def format_eta(eta):
+        if eta is None:
+            return '--:--'
+        return FileDownloader.format_seconds(eta)
+
+    @staticmethod
+    def calc_speed(start, now, bytes):
+        dif = now - start
+        if bytes == 0 or dif < 0.001: # One millisecond
+            return None
+        return float(bytes) / dif
+
+    @staticmethod
+    def format_speed(speed):
+        if speed is None:
+            return '%10s' % '---b/s'
+        return '%10s' % ('%s/s' % format_bytes(speed))
+
+    @staticmethod
+    def best_block_size(elapsed_time, bytes):
+        new_min = max(bytes / 2.0, 1.0)
+        new_max = min(max(bytes * 2.0, 1.0), 4194304) # Do not surpass 4 MB
+        if elapsed_time < 0.001:
+            return int(new_max)
+        rate = bytes / elapsed_time
+        if rate > new_max:
+            return int(new_max)
+        if rate < new_min:
+            return int(new_min)
+        return int(rate)
+
+    @staticmethod
+    def parse_bytes(bytestr):
+        """Parse a string indicating a byte quantity into an integer."""
+        matchobj = re.match(r'(?i)^(\d+(?:\.\d+)?)([kMGTPEZY]?)$', bytestr)
+        if matchobj is None:
+            return None
+        number = float(matchobj.group(1))
+        multiplier = 1024.0 ** 'bkmgtpezy'.index(matchobj.group(2).lower())
+        return int(round(number * multiplier))
+
+    def to_screen(self, *args, **kargs):
+        self.ydl.to_screen(*args, **kargs)
+
+    def to_stderr(self, message):
+        self.ydl.to_screen(message)
+
+    def to_console_title(self, message):
+        self.ydl.to_console_title(message)
+
+    def trouble(self, *args, **kargs):
+        self.ydl.trouble(*args, **kargs)
+
+    def report_warning(self, *args, **kargs):
+        self.ydl.report_warning(*args, **kargs)
+
+    def report_error(self, *args, **kargs):
+        self.ydl.report_error(*args, **kargs)
+
+    def slow_down(self, start_time, byte_counter):
+        """Sleep if the download speed is over the rate limit."""
+        rate_limit = self.params.get('ratelimit', None)
+        if rate_limit is None or byte_counter == 0:
+            return
+        now = time.time()
+        elapsed = now - start_time
+        if elapsed <= 0.0:
+            return
+        speed = float(byte_counter) / elapsed
+        if speed > rate_limit:
+            time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit)
+
+    def temp_name(self, filename):
+        """Returns a temporary filename for the given filename."""
+        if self.params.get('nopart', False) or filename == u'-' or \
+                (os.path.exists(encodeFilename(filename)) and not os.path.isfile(encodeFilename(filename))):
+            return filename
+        return filename + u'.part'
+
+    def undo_temp_name(self, filename):
+        if filename.endswith(u'.part'):
+            return filename[:-len(u'.part')]
+        return filename
+
+    def try_rename(self, old_filename, new_filename):
+        try:
+            if old_filename == new_filename:
+                return
+            os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
+        except (IOError, OSError) as err:
+            self.report_error(u'unable to rename file: %s' % compat_str(err))
+
+    def try_utime(self, filename, last_modified_hdr):
+        """Try to set the last-modified time of the given file."""
+        if last_modified_hdr is None:
+            return
+        if not os.path.isfile(encodeFilename(filename)):
+            return
+        timestr = last_modified_hdr
+        if timestr is None:
+            return
+        filetime = timeconvert(timestr)
+        if filetime is None:
+            return filetime
+        # Ignore obviously invalid dates
+        if filetime == 0:
+            return
+        try:
+            os.utime(filename, (time.time(), filetime))
+        except:
+            pass
+        return filetime
+
+    def report_destination(self, filename):
+        """Report destination filename."""
+        self.to_screen(u'[download] Destination: ' + filename)
+
+    def _report_progress_status(self, msg, is_last_line=False):
+        fullmsg = u'[download] ' + msg
+        if self.params.get('progress_with_newline', False):
+            self.to_screen(fullmsg)
+        else:
+            if os.name == 'nt':
+                prev_len = getattr(self, '_report_progress_prev_line_length',
+                                   0)
+                if prev_len > len(fullmsg):
+                    fullmsg += u' ' * (prev_len - len(fullmsg))
+                self._report_progress_prev_line_length = len(fullmsg)
+                clear_line = u'\r'
+            else:
+                clear_line = (u'\r\x1b[K' if sys.stderr.isatty() else u'\r')
+            self.to_screen(clear_line + fullmsg, skip_eol=not is_last_line)
+        self.to_console_title(u'youtube-dl ' + msg)
+
+    def report_progress(self, percent, data_len_str, speed, eta):
+        """Report download progress."""
+        if self.params.get('noprogress', False):
+            return
+        if eta is not None:
+            eta_str = self.format_eta(eta)
+        else:
+            eta_str = 'Unknown ETA'
+        if percent is not None:
+            percent_str = self.format_percent(percent)
+        else:
+            percent_str = 'Unknown %'
+        speed_str = self.format_speed(speed)
+
+        msg = (u'%s of %s at %s ETA %s' %
+               (percent_str, data_len_str, speed_str, eta_str))
+        self._report_progress_status(msg)
+
+    def report_progress_live_stream(self, downloaded_data_len, speed, elapsed):
+        if self.params.get('noprogress', False):
+            return
+        downloaded_str = format_bytes(downloaded_data_len)
+        speed_str = self.format_speed(speed)
+        elapsed_str = FileDownloader.format_seconds(elapsed)
+        msg = u'%s at %s (%s)' % (downloaded_str, speed_str, elapsed_str)
+        self._report_progress_status(msg)
+
+    def report_finish(self, data_len_str, tot_time):
+        """Report download finished."""
+        if self.params.get('noprogress', False):
+            self.to_screen(u'[download] Download completed')
+        else:
+            self._report_progress_status(
+                (u'100%% of %s in %s' %
+                 (data_len_str, self.format_seconds(tot_time))),
+                is_last_line=True)
+
+    def report_resuming_byte(self, resume_len):
+        """Report attempt to resume at given byte."""
+        self.to_screen(u'[download] Resuming download at byte %s' % resume_len)
+
+    def report_retry(self, count, retries):
+        """Report retry in case of HTTP error 5xx"""
+        self.to_screen(u'[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count, retries))
+
+    def report_file_already_downloaded(self, file_name):
+        """Report file has already been fully downloaded."""
+        try:
+            self.to_screen(u'[download] %s has already been downloaded' % file_name)
+        except UnicodeEncodeError:
+            self.to_screen(u'[download] The file has already been downloaded')
+
+    def report_unable_to_resume(self):
+        """Report it was impossible to resume download."""
+        self.to_screen(u'[download] Unable to resume')
+
+    def download(self, filename, info_dict):
+        """Download to a filename using the info from info_dict
+        Return True on success and False otherwise
+        """
+        # Check file already present
+        if self.params.get('continuedl', False) and os.path.isfile(encodeFilename(filename)) and not self.params.get('nopart', False):
+            self.report_file_already_downloaded(filename)
+            self._hook_progress({
+                'filename': filename,
+                'status': 'finished',
+                'total_bytes': os.path.getsize(encodeFilename(filename)),
+            })
+            return True
+
+        return self.real_download(filename, info_dict)
+
+    def real_download(self, filename, info_dict):
+        """Real download process. Redefine in subclasses."""
+        raise NotImplementedError(u'This method must be implemented by sublcasses')
+
+    def _hook_progress(self, status):
+        for ph in self._progress_hooks:
+            ph(status)
+
+    def add_progress_hook(self, ph):
+        """ ph gets called on download progress, with a dictionary with the entries
+        * filename: The final filename
+        * status: One of "downloading" and "finished"
+
+        It can also have some of the following entries:
+
+        * downloaded_bytes: Bytes on disks
+        * total_bytes: Total bytes, None if unknown
+        * tmpfilename: The filename we're currently writing to
+        * eta: The estimated time in seconds, None if unknown
+        * speed: The download speed in bytes/second, None if unknown
+
+        Hooks are guaranteed to be called at least once (with status "finished")
+        if the download is successful.
+        """
+        self._progress_hooks.append(ph)

youtube_dl/downloader/f4m.py

@@ -0,0 +1,315 @@
+from __future__ import unicode_literals
+
+import base64
+import io
+import itertools
+import os
+import time
+import xml.etree.ElementTree as etree
+
+from .common import FileDownloader
+from .http import HttpFD
+from ..utils import (
+    struct_pack,
+    struct_unpack,
+    compat_urlparse,
+    format_bytes,
+    encodeFilename,
+    sanitize_open,
+)
+
+
+class FlvReader(io.BytesIO):
+    """
+    Reader for Flv files
+    The file format is documented in https://www.adobe.com/devnet/f4v.html
+    """
+
+    # Utility functions for reading numbers and strings
+    def read_unsigned_long_long(self):
+        return struct_unpack('!Q', self.read(8))[0]
+
+    def read_unsigned_int(self):
+        return struct_unpack('!I', self.read(4))[0]
+
+    def read_unsigned_char(self):
+        return struct_unpack('!B', self.read(1))[0]
+
+    def read_string(self):
+        res = b''
+        while True:
+            char = self.read(1)
+            if char == b'\x00':
+                break
+            res += char
+        return res
+
+    def read_box_info(self):
+        """
+        Read a box and return the info as a tuple: (box_size, box_type, box_data)
+        """
+        real_size = size = self.read_unsigned_int()
+        box_type = self.read(4)
+        header_end = 8
+        if size == 1:
+            real_size = self.read_unsigned_long_long()
+            header_end = 16
+        return real_size, box_type, self.read(real_size-header_end)
+
+    def read_asrt(self):
+        # version
+        self.read_unsigned_char()
+        # flags
+        self.read(3)
+        quality_entry_count = self.read_unsigned_char()
+        # QualityEntryCount
+        for i in range(quality_entry_count):
+            self.read_string()
+
+        segment_run_count = self.read_unsigned_int()
+        segments = []
+        for i in range(segment_run_count):
+            first_segment = self.read_unsigned_int()
+            fragments_per_segment = self.read_unsigned_int()
+            segments.append((first_segment, fragments_per_segment))
+
+        return {
+            'segment_run': segments,
+        }
+
+    def read_afrt(self):
+        # version
+        self.read_unsigned_char()
+        # flags
+        self.read(3)
+        # time scale
+        self.read_unsigned_int()
+
+        quality_entry_count = self.read_unsigned_char()
+        # QualitySegmentUrlModifiers
+        for i in range(quality_entry_count):
+            self.read_string()
+
+        fragments_count = self.read_unsigned_int()
+        fragments = []
+        for i in range(fragments_count):
+            first = self.read_unsigned_int()
+            first_ts = self.read_unsigned_long_long()
+            duration = self.read_unsigned_int()
+            if duration == 0:
+                discontinuity_indicator = self.read_unsigned_char()
+            else:
+                discontinuity_indicator = None
+            fragments.append({
+                'first': first,
+                'ts': first_ts,
+                'duration': duration,
+                'discontinuity_indicator': discontinuity_indicator,
+            })
+
+        return {
+            'fragments': fragments,
+        }
+
+    def read_abst(self):
+        # version
+        self.read_unsigned_char()
+        # flags
+        self.read(3)
+
+        self.read_unsigned_int()  # BootstrapinfoVersion
+        # Profile,Live,Update,Reserved
+        self.read(1)
+        # time scale
+        self.read_unsigned_int()
+        # CurrentMediaTime
+        self.read_unsigned_long_long()
+        # SmpteTimeCodeOffset
+        self.read_unsigned_long_long()
+
+        self.read_string()  # MovieIdentifier
+        server_count = self.read_unsigned_char()
+        # ServerEntryTable
+        for i in range(server_count):
+            self.read_string()
+        quality_count = self.read_unsigned_char()
+        # QualityEntryTable
+        for i in range(quality_count):
+            self.read_string()
+        # DrmData
+        self.read_string()
+        # MetaData
+        self.read_string()
+
+        segments_count = self.read_unsigned_char()
+        segments = []
+        for i in range(segments_count):
+            box_size, box_type, box_data = self.read_box_info()
+            assert box_type == b'asrt'
+            segment = FlvReader(box_data).read_asrt()
+            segments.append(segment)
+        fragments_run_count = self.read_unsigned_char()
+        fragments = []
+        for i in range(fragments_run_count):
+            box_size, box_type, box_data = self.read_box_info()
+            assert box_type == b'afrt'
+            fragments.append(FlvReader(box_data).read_afrt())
+
+        return {
+            'segments': segments,
+            'fragments': fragments,
+        }
+
+    def read_bootstrap_info(self):
+        total_size, box_type, box_data = self.read_box_info()
+        assert box_type == b'abst'
+        return FlvReader(box_data).read_abst()
+
+
+def read_bootstrap_info(bootstrap_bytes):
+    return FlvReader(bootstrap_bytes).read_bootstrap_info()
+
+
+def build_fragments_list(boot_info):
+    """ Return a list of (segment, fragment) for each fragment in the video """
+    res = []
+    segment_run_table = boot_info['segments'][0]
+    # I've only found videos with one segment
+    segment_run_entry = segment_run_table['segment_run'][0]
+    n_frags = segment_run_entry[1]
+    fragment_run_entry_table = boot_info['fragments'][0]['fragments']
+    first_frag_number = fragment_run_entry_table[0]['first']
+    for (i, frag_number) in zip(range(1, n_frags+1), itertools.count(first_frag_number)):
+        res.append((1, frag_number))
+    return res
+
+
+def write_flv_header(stream, metadata):
+    """Writes the FLV header and the metadata to stream"""
+    # FLV header
+    stream.write(b'FLV\x01')
+    stream.write(b'\x05')
+    stream.write(b'\x00\x00\x00\x09')
+    # FLV File body
+    stream.write(b'\x00\x00\x00\x00')
+    # FLVTAG
+    # Script data
+    stream.write(b'\x12')
+    # Size of the metadata with 3 bytes
+    stream.write(struct_pack('!L', len(metadata))[1:])
+    stream.write(b'\x00\x00\x00\x00\x00\x00\x00')
+    stream.write(metadata)
+    # Magic numbers extracted from the output files produced by AdobeHDS.php
+    #(https://github.com/K-S-V/Scripts)
+    stream.write(b'\x00\x00\x01\x73')
+
+
+def _add_ns(prop):
+    return '{http://ns.adobe.com/f4m/1.0}%s' % prop
+
+
+class HttpQuietDownloader(HttpFD):
+    def to_screen(self, *args, **kargs):
+        pass
+
+
+class F4mFD(FileDownloader):
+    """
+    A downloader for f4m manifests or AdobeHDS.
+    """
+
+    def real_download(self, filename, info_dict):
+        man_url = info_dict['url']
+        self.to_screen('[download] Downloading f4m manifest')
+        manifest = self.ydl.urlopen(man_url).read()
+        self.report_destination(filename)
+        http_dl = HttpQuietDownloader(self.ydl,
+            {
+                'continuedl': True,
+                'quiet': True,
+                'noprogress': True,
+                'test': self.params.get('test', False),
+            })
+
+        doc = etree.fromstring(manifest)
+        formats = [(int(f.attrib.get('bitrate', -1)), f) for f in doc.findall(_add_ns('media'))]
+        formats = sorted(formats, key=lambda f: f[0])
+        rate, media = formats[-1]
+        base_url = compat_urlparse.urljoin(man_url, media.attrib['url'])
+        bootstrap = base64.b64decode(doc.find(_add_ns('bootstrapInfo')).text)
+        metadata = base64.b64decode(media.find(_add_ns('metadata')).text)
+        boot_info = read_bootstrap_info(bootstrap)
+        fragments_list = build_fragments_list(boot_info)
+        if self.params.get('test', False):
+            # We only download the first fragment
+            fragments_list = fragments_list[:1]
+        total_frags = len(fragments_list)
+
+        tmpfilename = self.temp_name(filename)
+        (dest_stream, tmpfilename) = sanitize_open(tmpfilename, 'wb')
+        write_flv_header(dest_stream, metadata)
+
+        # This dict stores the download progress, it's updated by the progress
+        # hook
+        state = {
+            'downloaded_bytes': 0,
+            'frag_counter': 0,
+        }
+        start = time.time()
+
+        def frag_progress_hook(status):
+            frag_total_bytes = status.get('total_bytes', 0)
+            estimated_size = (state['downloaded_bytes'] +
+                (total_frags - state['frag_counter']) * frag_total_bytes)
+            if status['status'] == 'finished':
+                state['downloaded_bytes'] += frag_total_bytes
+                state['frag_counter'] += 1
+                progress = self.calc_percent(state['frag_counter'], total_frags)
+                byte_counter = state['downloaded_bytes']
+            else:
+                frag_downloaded_bytes = status['downloaded_bytes']
+                byte_counter = state['downloaded_bytes'] + frag_downloaded_bytes
+                frag_progress = self.calc_percent(frag_downloaded_bytes,
+                    frag_total_bytes)
+                progress = self.calc_percent(state['frag_counter'], total_frags)
+                progress += frag_progress / float(total_frags)
+
+            eta = self.calc_eta(start, time.time(), estimated_size, byte_counter)
+            self.report_progress(progress, format_bytes(estimated_size),
+                status.get('speed'), eta)
+        http_dl.add_progress_hook(frag_progress_hook)
+
+        frags_filenames = []
+        for (seg_i, frag_i) in fragments_list:
+            name = 'Seg%d-Frag%d' % (seg_i, frag_i)
+            url = base_url + name
+            frag_filename = '%s-%s' % (tmpfilename, name)
+            success = http_dl.download(frag_filename, {'url': url})
+            if not success:
+                return False
+            with open(frag_filename, 'rb') as down:
+                down_data = down.read()
+                reader = FlvReader(down_data)
+                while True:
+                    _, box_type, box_data = reader.read_box_info()
+                    if box_type == b'mdat':
+                        dest_stream.write(box_data)
+                        break
+            frags_filenames.append(frag_filename)
+
+        dest_stream.close()
+        self.report_finish(format_bytes(state['downloaded_bytes']), time.time() - start)
+
+        self.try_rename(tmpfilename, filename)
+        for frag_file in frags_filenames:
+            os.remove(frag_file)
+
+        fsize = os.path.getsize(encodeFilename(filename))
+        self._hook_progress({
+            'downloaded_bytes': fsize,
+            'total_bytes': fsize,
+            'filename': filename,
+            'status': 'finished',
+        })
+
+        return True

youtube_dl/downloader/hls.py

@@ -0,0 +1,46 @@
+import os
+import subprocess
+
+from .common import FileDownloader
+from ..utils import (
+    encodeFilename,
+)
+
+
+class HlsFD(FileDownloader):
+    def real_download(self, filename, info_dict):
+        url = info_dict['url']
+        self.report_destination(filename)
+        tmpfilename = self.temp_name(filename)
+
+        args = [
+            '-y', '-i', url, '-f', 'mp4', '-c', 'copy',
+            '-bsf:a', 'aac_adtstoasc',
+            encodeFilename(tmpfilename, for_subprocess=True)]
+
+        for program in ['avconv', 'ffmpeg']:
+            try:
+                subprocess.call([program, '-version'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
+                break
+            except (OSError, IOError):
+                pass
+        else:
+            self.report_error(u'm3u8 download detected but ffmpeg or avconv could not be found')
+        cmd = [program] + args
+
+        retval = subprocess.call(cmd)
+        if retval == 0:
+            fsize = os.path.getsize(encodeFilename(tmpfilename))
+            self.to_screen(u'\r[%s] %s bytes' % (cmd[0], fsize))
+            self.try_rename(tmpfilename, filename)
+            self._hook_progress({
+                'downloaded_bytes': fsize,
+                'total_bytes': fsize,
+                'filename': filename,
+                'status': 'finished',
+            })
+            return True
+        else:
+            self.to_stderr(u"\n")
+            self.report_error(u'ffmpeg exited with code %d' % retval)
+            return False

youtube_dl/downloader/http.py

@@ -0,0 +1,205 @@
+import os
+import time
+
+from .common import FileDownloader
+from ..utils import (
+    compat_urllib_request,
+    compat_urllib_error,
+    ContentTooShortError,
+
+    encodeFilename,
+    sanitize_open,
+    format_bytes,
+)
+
+
+class HttpFD(FileDownloader):
+    _TEST_FILE_SIZE = 10241
+
+    def real_download(self, filename, info_dict):
+        url = info_dict['url']
+        tmpfilename = self.temp_name(filename)
+        stream = None
+
+        # Do not include the Accept-Encoding header
+        headers = {'Youtubedl-no-compression': 'True'}
+        if 'user_agent' in info_dict:
+            headers['Youtubedl-user-agent'] = info_dict['user_agent']
+        if 'http_referer' in info_dict:
+            headers['Referer'] = info_dict['http_referer']
+        basic_request = compat_urllib_request.Request(url, None, headers)
+        request = compat_urllib_request.Request(url, None, headers)
+
+        is_test = self.params.get('test', False)
+
+        if is_test:
+            request.add_header('Range', 'bytes=0-%s' % str(self._TEST_FILE_SIZE - 1))
+
+        # Establish possible resume length
+        if os.path.isfile(encodeFilename(tmpfilename)):
+            resume_len = os.path.getsize(encodeFilename(tmpfilename))
+        else:
+            resume_len = 0
+
+        open_mode = 'wb'
+        if resume_len != 0:
+            if self.params.get('continuedl', False):
+                self.report_resuming_byte(resume_len)
+                request.add_header('Range', 'bytes=%d-' % resume_len)
+                open_mode = 'ab'
+            else:
+                resume_len = 0
+
+        count = 0
+        retries = self.params.get('retries', 0)
+        while count <= retries:
+            # Establish connection
+            try:
+                data = self.ydl.urlopen(request)
+                break
+            except (compat_urllib_error.HTTPError, ) as err:
+                if (err.code < 500 or err.code >= 600) and err.code != 416:
+                    # Unexpected HTTP error
+                    raise
+                elif err.code == 416:
+                    # Unable to resume (requested range not satisfiable)
+                    try:
+                        # Open the connection again without the range header
+                        data = self.ydl.urlopen(basic_request)
+                        content_length = data.info()['Content-Length']
+                    except (compat_urllib_error.HTTPError, ) as err:
+                        if err.code < 500 or err.code >= 600:
+                            raise
+                    else:
+                        # Examine the reported length
+                        if (content_length is not None and
+                                (resume_len - 100 < int(content_length) < resume_len + 100)):
+                            # The file had already been fully downloaded.
+                            # Explanation to the above condition: in issue #175 it was revealed that
+                            # YouTube sometimes adds or removes a few bytes from the end of the file,
+                            # changing the file size slightly and causing problems for some users. So
+                            # I decided to implement a suggested change and consider the file
+                            # completely downloaded if the file size differs less than 100 bytes from
+                            # the one in the hard drive.
+                            self.report_file_already_downloaded(filename)
+                            self.try_rename(tmpfilename, filename)
+                            self._hook_progress({
+                                'filename': filename,
+                                'status': 'finished',
+                            })
+                            return True
+                        else:
+                            # The length does not match, we start the download over
+                            self.report_unable_to_resume()
+                            resume_len = 0
+                            open_mode = 'wb'
+                            break
+            # Retry
+            count += 1
+            if count <= retries:
+                self.report_retry(count, retries)
+
+        if count > retries:
+            self.report_error(u'giving up after %s retries' % retries)
+            return False
+
+        data_len = data.info().get('Content-length', None)
+
+        # Range HTTP header may be ignored/unsupported by a webserver
+        # (e.g. extractor/scivee.py, extractor/bambuser.py).
+        # However, for a test we still would like to download just a piece of a file.
+        # To achieve this we limit data_len to _TEST_FILE_SIZE and manually control
+        # block size when downloading a file.
+        if is_test and (data_len is None or int(data_len) > self._TEST_FILE_SIZE):
+            data_len = self._TEST_FILE_SIZE
+
+        if data_len is not None:
+            data_len = int(data_len) + resume_len
+            min_data_len = self.params.get("min_filesize", None)
+            max_data_len = self.params.get("max_filesize", None)
+            if min_data_len is not None and data_len < min_data_len:
+                self.to_screen(u'\r[download] File is smaller than min-filesize (%s bytes < %s bytes). Aborting.' % (data_len, min_data_len))
+                return False
+            if max_data_len is not None and data_len > max_data_len:
+                self.to_screen(u'\r[download] File is larger than max-filesize (%s bytes > %s bytes). Aborting.' % (data_len, max_data_len))
+                return False
+
+        data_len_str = format_bytes(data_len)
+        byte_counter = 0 + resume_len
+        block_size = self.params.get('buffersize', 1024)
+        start = time.time()
+        while True:
+            # Download and write
+            before = time.time()
+            data_block = data.read(block_size if not is_test else min(block_size, data_len - byte_counter))
+            after = time.time()
+            if len(data_block) == 0:
+                break
+            byte_counter += len(data_block)
+
+            # Open file just in time
+            if stream is None:
+                try:
+                    (stream, tmpfilename) = sanitize_open(tmpfilename, open_mode)
+                    assert stream is not None
+                    filename = self.undo_temp_name(tmpfilename)
+                    self.report_destination(filename)
+                except (OSError, IOError) as err:
+                    self.report_error(u'unable to open for writing: %s' % str(err))
+                    return False
+            try:
+                stream.write(data_block)
+            except (IOError, OSError) as err:
+                self.to_stderr(u"\n")
+                self.report_error(u'unable to write data: %s' % str(err))
+                return False
+            if not self.params.get('noresizebuffer', False):
+                block_size = self.best_block_size(after - before, len(data_block))
+
+            # Progress message
+            speed = self.calc_speed(start, time.time(), byte_counter - resume_len)
+            if data_len is None:
+                eta = percent = None
+            else:
+                percent = self.calc_percent(byte_counter, data_len)
+                eta = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len)
+            self.report_progress(percent, data_len_str, speed, eta)
+
+            self._hook_progress({
+                'downloaded_bytes': byte_counter,
+                'total_bytes': data_len,
+                'tmpfilename': tmpfilename,
+                'filename': filename,
+                'status': 'downloading',
+                'eta': eta,
+                'speed': speed,
+            })
+
+            if is_test and byte_counter == data_len:
+                break
+
+            # Apply rate limit
+            self.slow_down(start, byte_counter - resume_len)
+
+        if stream is None:
+            self.to_stderr(u"\n")
+            self.report_error(u'Did not get any data blocks')
+            return False
+        stream.close()
+        self.report_finish(data_len_str, (time.time() - start))
+        if data_len is not None and byte_counter != data_len:
+            raise ContentTooShortError(byte_counter, int(data_len))
+        self.try_rename(tmpfilename, filename)
+
+        # Update file modification time
+        if self.params.get('updatetime', True):
+            info_dict['filetime'] = self.try_utime(filename, data.info().get('last-modified', None))
+
+        self._hook_progress({
+            'downloaded_bytes': byte_counter,
+            'total_bytes': byte_counter,
+            'filename': filename,
+            'status': 'finished',
+        })
+
+        return True

youtube_dl/downloader/mplayer.py

@@ -0,0 +1,40 @@
+import os
+import subprocess
+
+from .common import FileDownloader
+from ..utils import (
+    encodeFilename,
+)
+
+
+class MplayerFD(FileDownloader):
+    def real_download(self, filename, info_dict):
+        url = info_dict['url']
+        self.report_destination(filename)
+        tmpfilename = self.temp_name(filename)
+
+        args = ['mplayer', '-really-quiet', '-vo', 'null', '-vc', 'dummy', '-dumpstream', '-dumpfile', tmpfilename, url]
+        # Check for mplayer first
+        try:
+            subprocess.call(['mplayer', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
+        except (OSError, IOError):
+            self.report_error(u'MMS or RTSP download detected but "%s" could not be run' % args[0])
+            return False
+
+        # Download using mplayer.
+        retval = subprocess.call(args)
+        if retval == 0:
+            fsize = os.path.getsize(encodeFilename(tmpfilename))
+            self.to_screen(u'\r[%s] %s bytes' % (args[0], fsize))
+            self.try_rename(tmpfilename, filename)
+            self._hook_progress({
+                'downloaded_bytes': fsize,
+                'total_bytes': fsize,
+                'filename': filename,
+                'status': 'finished',
+            })
+            return True
+        else:
+            self.to_stderr(u"\n")
+            self.report_error(u'mplayer exited with code %d' % retval)
+            return False

youtube_dl/downloader/rtmp.py

@@ -0,0 +1,199 @@
+from __future__ import unicode_literals
+
+import os
+import re
+import subprocess
+import sys
+import time
+
+from .common import FileDownloader
+from ..utils import (
+    encodeFilename,
+    format_bytes,
+    compat_str,
+)
+
+
+class RtmpFD(FileDownloader):
+    def real_download(self, filename, info_dict):
+        def run_rtmpdump(args):
+            start = time.time()
+            resume_percent = None
+            resume_downloaded_data_len = None
+            proc = subprocess.Popen(args, stderr=subprocess.PIPE)
+            cursor_in_new_line = True
+            proc_stderr_closed = False
+            while not proc_stderr_closed:
+                # read line from stderr
+                line = ''
+                while True:
+                    char = proc.stderr.read(1)
+                    if not char:
+                        proc_stderr_closed = True
+                        break
+                    if char in [b'\r', b'\n']:
+                        break
+                    line += char.decode('ascii', 'replace')
+                if not line:
+                    # proc_stderr_closed is True
+                    continue
+                mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec \(([0-9]{1,2}\.[0-9])%\)', line)
+                if mobj:
+                    downloaded_data_len = int(float(mobj.group(1))*1024)
+                    percent = float(mobj.group(2))
+                    if not resume_percent:
+                        resume_percent = percent
+                        resume_downloaded_data_len = downloaded_data_len
+                    eta = self.calc_eta(start, time.time(), 100-resume_percent, percent-resume_percent)
+                    speed = self.calc_speed(start, time.time(), downloaded_data_len-resume_downloaded_data_len)
+                    data_len = None
+                    if percent > 0:
+                        data_len = int(downloaded_data_len * 100 / percent)
+                    data_len_str = '~' + format_bytes(data_len)
+                    self.report_progress(percent, data_len_str, speed, eta)
+                    cursor_in_new_line = False
+                    self._hook_progress({
+                        'downloaded_bytes': downloaded_data_len,
+                        'total_bytes': data_len,
+                        'tmpfilename': tmpfilename,
+                        'filename': filename,
+                        'status': 'downloading',
+                        'eta': eta,
+                        'speed': speed,
+                    })
+                else:
+                    # no percent for live streams
+                    mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec', line)
+                    if mobj:
+                        downloaded_data_len = int(float(mobj.group(1))*1024)
+                        time_now = time.time()
+                        speed = self.calc_speed(start, time_now, downloaded_data_len)
+                        self.report_progress_live_stream(downloaded_data_len, speed, time_now - start)
+                        cursor_in_new_line = False
+                        self._hook_progress({
+                            'downloaded_bytes': downloaded_data_len,
+                            'tmpfilename': tmpfilename,
+                            'filename': filename,
+                            'status': 'downloading',
+                            'speed': speed,
+                        })
+                    elif self.params.get('verbose', False):
+                        if not cursor_in_new_line:
+                            self.to_screen('')
+                        cursor_in_new_line = True
+                        self.to_screen('[rtmpdump] '+line)
+            proc.wait()
+            if not cursor_in_new_line:
+                self.to_screen('')
+            return proc.returncode
+
+        url = info_dict['url']
+        player_url = info_dict.get('player_url', None)
+        page_url = info_dict.get('page_url', None)
+        app = info_dict.get('app', None)
+        play_path = info_dict.get('play_path', None)
+        tc_url = info_dict.get('tc_url', None)
+        flash_version = info_dict.get('flash_version', None)
+        live = info_dict.get('rtmp_live', False)
+        conn = info_dict.get('rtmp_conn', None)
+
+        self.report_destination(filename)
+        tmpfilename = self.temp_name(filename)
+        test = self.params.get('test', False)
+
+        # Check for rtmpdump first
+        try:
+            subprocess.call(['rtmpdump', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
+        except (OSError, IOError):
+            self.report_error('RTMP download detected but "rtmpdump" could not be run')
+            return False
+
+        # Download using rtmpdump. rtmpdump returns exit code 2 when
+        # the connection was interrumpted and resuming appears to be
+        # possible. This is part of rtmpdump's normal usage, AFAIK.
+        basic_args = ['rtmpdump', '--verbose', '-r', url, '-o', tmpfilename]
+        if player_url is not None:
+            basic_args += ['--swfVfy', player_url]
+        if page_url is not None:
+            basic_args += ['--pageUrl', page_url]
+        if app is not None:
+            basic_args += ['--app', app]
+        if play_path is not None:
+            basic_args += ['--playpath', play_path]
+        if tc_url is not None:
+            basic_args += ['--tcUrl', url]
+        if test:
+            basic_args += ['--stop', '1']
+        if flash_version is not None:
+            basic_args += ['--flashVer', flash_version]
+        if live:
+            basic_args += ['--live']
+        if isinstance(conn, list):
+            for entry in conn:
+                basic_args += ['--conn', entry]
+        elif isinstance(conn, compat_str):
+            basic_args += ['--conn', conn]
+        args = basic_args + [[], ['--resume', '--skip', '1']][not live and self.params.get('continuedl', False)]
+
+        if sys.platform == 'win32' and sys.version_info < (3, 0):
+            # Windows subprocess module does not actually support Unicode
+            # on Python 2.x
+            # See http://stackoverflow.com/a/9951851/35070
+            subprocess_encoding = sys.getfilesystemencoding()
+            args = [a.encode(subprocess_encoding, 'ignore') for a in args]
+        else:
+            subprocess_encoding = None
+
+        if self.params.get('verbose', False):
+            if subprocess_encoding:
+                str_args = [
+                    a.decode(subprocess_encoding) if isinstance(a, bytes) else a
+                    for a in args]
+            else:
+                str_args = args
+            try:
+                import pipes
+                shell_quote = lambda args: ' '.join(map(pipes.quote, str_args))
+            except ImportError:
+                shell_quote = repr
+            self.to_screen('[debug] rtmpdump command line: ' + shell_quote(str_args))
+
+        RD_SUCCESS = 0
+        RD_FAILED = 1
+        RD_INCOMPLETE = 2
+        RD_NO_CONNECT = 3
+
+        retval = run_rtmpdump(args)
+
+        if retval == RD_NO_CONNECT:
+            self.report_error('[rtmpdump] Could not connect to RTMP server.')
+            return False
+
+        while (retval == RD_INCOMPLETE or retval == RD_FAILED) and not test and not live:
+            prevsize = os.path.getsize(encodeFilename(tmpfilename))
+            self.to_screen('[rtmpdump] %s bytes' % prevsize)
+            time.sleep(5.0) # This seems to be needed
+            retval = run_rtmpdump(basic_args + ['-e'] + [[], ['-k', '1']][retval == RD_FAILED])
+            cursize = os.path.getsize(encodeFilename(tmpfilename))
+            if prevsize == cursize and retval == RD_FAILED:
+                break
+             # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those
+            if prevsize == cursize and retval == RD_INCOMPLETE and cursize > 1024:
+                self.to_screen('[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
+                retval = RD_SUCCESS
+                break
+        if retval == RD_SUCCESS or (test and retval == RD_INCOMPLETE):
+            fsize = os.path.getsize(encodeFilename(tmpfilename))
+            self.to_screen('[rtmpdump] %s bytes' % fsize)
+            self.try_rename(tmpfilename, filename)
+            self._hook_progress({
+                'downloaded_bytes': fsize,
+                'total_bytes': fsize,
+                'filename': filename,
+                'status': 'finished',
+            })
+            return True
+        else:
+            self.to_stderr('\n')
+            self.report_error('rtmpdump exited with code %d' % retval)
+            return False

youtube_dl/extractor/__init__.py

@@ -1,6 +1,8 @@
 from .academicearth import AcademicEarthCourseIE
 from .addanime import AddAnimeIE
+from .aftonbladet import AftonbladetIE
 from .anitube import AnitubeIE
+from .aol import AolIE
 from .aparat import AparatIE
 from .appletrailers import AppleTrailersIE
 from .archiveorg import ArchiveOrgIE
@@ -9,26 +11,43 @@ from .arte import (
     ArteTvIE,
     ArteTVPlus7IE,
     ArteTVCreativeIE,
+    ArteTVConcertIE,
     ArteTVFutureIE,
     ArteTVDDCIE,
+    ArteTVEmbedIE,
 )
 from .auengine import AUEngineIE
 from .bambuser import BambuserIE, BambuserChannelIE
 from .bandcamp import BandcampIE, BandcampAlbumIE
+from .bbccouk import BBCCoUkIE
+from .bilibili import BiliBiliIE
 from .blinkx import BlinkxIE
 from .bliptv import BlipTVIE, BlipTVUserIE
 from .bloomberg import BloombergIE
+from .br import BRIE
 from .breakcom import BreakIE
 from .brightcove import BrightcoveIE
+from .byutv import BYUtvIE
 from .c56 import C56IE
+from .canal13cl import Canal13clIE
 from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
 from .cbs import CBSIE
+from .cbsnews import CBSNewsIE
+from .ceskatelevize import CeskaTelevizeIE
 from .channel9 import Channel9IE
+from .chilloutzone import ChilloutzoneIE
 from .cinemassacre import CinemassacreIE
 from .clipfish import ClipfishIE
+from .cliphunter import CliphunterIE
 from .clipsyndicate import ClipsyndicateIE
-from .cnn import CNNIE
+from .clubic import ClubicIE
+from .cmt import CMTIE
+from .cnet import CNETIE
+from .cnn import (
+    CNNIE,
+    CNNBlogsIE,
+)
 from .collegehumor import CollegeHumorIE
 from .comedycentral import ComedyCentralIE, ComedyCentralShowsIE
 from .condenast import CondeNastIE
@@ -42,89 +61,151 @@ from .dailymotion import (
     DailymotionUserIE,
 )
 from .daum import DaumIE
-from .depositfiles import DepositFilesIE
 from .dotsub import DotsubIE
 from .dreisat import DreiSatIE
 from .defense import DefenseGouvFrIE
+from .discovery import DiscoveryIE
+from .divxstage import DivxStageIE
+from .dropbox import DropboxIE
 from .ebaumsworld import EbaumsWorldIE
 from .ehow import EHowIE
 from .eighttracks import EightTracksIE
 from .eitb import EitbIE
+from .elpais import ElPaisIE
+from .engadget import EngadgetIE
 from .escapist import EscapistIE
+from .everyonesmixtape import EveryonesMixtapeIE
 from .exfm import ExfmIE
 from .extremetube import ExtremeTubeIE
 from .facebook import FacebookIE
 from .faz import FazIE
+from .firstpost import FirstpostIE
+from .firsttv import FirstTVIE
+from .fivemin import FiveMinIE
 from .fktv import (
     FKTVIE,
     FKTVPosteckeIE,
 )
 from .flickr import FlickrIE
+from .fourtube import FourTubeIE
+from .franceculture import FranceCultureIE
+from .franceinter import FranceInterIE
 from .francetv import (
     PluzzIE,
     FranceTvInfoIE,
     FranceTVIE,
-    GenerationQuoiIE
+    GenerationQuoiIE,
+    CultureboxIE,
 )
 from .freesound import FreesoundIE
+from .freespeech import FreespeechIE
 from .funnyordie import FunnyOrDieIE
 from .gamekings import GamekingsIE
 from .gamespot import GameSpotIE
 from .gametrailers import GametrailersIE
+from .gdcvault import GDCVaultIE
 from .generic import GenericIE
 from .googleplus import GooglePlusIE
 from .googlesearch import GoogleSearchIE
 from .hark import HarkIE
+from .helsinki import HelsinkiIE
 from .hotnewhiphop import HotNewHipHopIE
 from .howcast import HowcastIE
+from .huffpost import HuffPostIE
 from .hypem import HypemIE
+from .iconosquare import IconosquareIE
 from .ign import IGNIE, OneUPIE
-from .imdb import ImdbIE
+from .imdb import (
+    ImdbIE,
+    ImdbListIE
+)
 from .ina import InaIE
 from .infoq import InfoQIE
-from .instagram import InstagramIE
+from .instagram import InstagramIE, InstagramUserIE
 from .internetvideoarchive import InternetVideoArchiveIE
+from .iprima import IPrimaIE
 from .ivi import (
     IviIE,
     IviCompilationIE
 )
+from .jadorecettepub import JadoreCettePubIE
 from .jeuxvideo import JeuxVideoIE
 from .jukebox import JukeboxIE
 from .justintv import JustinTVIE
+from .jpopsukitv import JpopsukiIE
 from .kankan import KankanIE
 from .keezmovies import KeezMoviesIE
+from .khanacademy import KhanAcademyIE
 from .kickstarter import KickStarterIE
 from .keek import KeekIE
+from .kontrtube import KontrTubeIE
+from .la7 import LA7IE
+from .lifenews import LifeNewsIE
 from .liveleak import LiveLeakIE
 from .livestream import LivestreamIE, LivestreamOriginalIE
+from .lynda import (
+    LyndaIE,
+    LyndaCourseIE
+)
+from .m6 import M6IE
+from .macgamestore import MacGameStoreIE
+from .mailru import MailRuIE
+from .malemotion import MalemotionIE
 from .mdr import MDRIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
-from .mit import TechTVMITIE, MITIE
+from .mit import TechTVMITIE, MITIE, OCWMITIE
 from .mixcloud import MixcloudIE
+from .mpora import MporaIE
 from .mofosex import MofosexIE
-from .mtv import MTVIE
+from .mooshare import MooshareIE
+from .morningstar import MorningstarIE
+from .motorsport import MotorsportIE
+from .moviezine import MoviezineIE
+from .movshare import MovShareIE
+from .mtv import (
+    MTVIE,
+    MTVIggyIE,
+)
+from .musicplayon import MusicPlayOnIE
 from .muzu import MuzuTVIE
 from .myspace import MySpaceIE
 from .myspass import MySpassIE
 from .myvideo import MyVideoIE
 from .naver import NaverIE
 from .nba import NBAIE
-from .nbc import NBCNewsIE
+from .nbc import (
+    NBCIE,
+    NBCNewsIE,
+)
+from .ndr import NDRIE
 from .ndtv import NDTVIE
 from .newgrounds import NewgroundsIE
+from .newstube import NewstubeIE
+from .nfb import NFBIE
 from .nhl import NHLIE, NHLVideocenterIE
 from .niconico import NiconicoIE
 from .ninegag import NineGagIE
+from .noco import NocoIE
+from .normalboots import NormalbootsIE
+from .novamov import NovaMovIE
+from .nowness import NownessIE
 from .nowvideo import NowVideoIE
+from .nrk import NRKIE
+from .ntv import NTVIE
+from .nytimes import NYTimesIE
+from .oe1 import OE1IE
 from .ooyala import OoyalaIE
 from .orf import ORFIE
+from .parliamentliveuk import ParliamentLiveUKIE
 from .pbs import PBSIE
 from .photobucket import PhotobucketIE
+from .playvid import PlayvidIE
 from .podomatic import PodomaticIE
 from .pornhd import PornHdIE
 from .pornhub import PornHubIE
 from .pornotube import PornotubeIE
+from .prosiebensat1 import ProSiebenSat1IE
 from .pyvideo import PyvideoIE
 from .radiofrance import RadioFranceIE
 from .rbmaradio import RBMARadioIE
@@ -133,10 +214,21 @@ from .ringtv import RingTVIE
 from .ro220 import Ro220IE
 from .rottentomatoes import RottenTomatoesIE
 from .roxwel import RoxwelIE
+from .rtbf import RTBFIE
 from .rtlnow import RTLnowIE
-from .rutube import RutubeIE
+from .rts import RTSIE
+from .rtve import RTVEALaCartaIE
+from .rutube import (
+    RutubeIE,
+    RutubeChannelIE,
+    RutubeMovieIE,
+    RutubePersonIE,
+)
+from .rutv import RUTVIE
+from .savefrom import SaveFromIE
+from .scivee import SciVeeIE
+from .servingsys import ServingSysIE
 from .sina import SinaIE
-from .slashdot import SlashdotIE
 from .slideshare import SlideshareIE
 from .smotri import (
     SmotriIE,
@@ -145,7 +237,12 @@ from .smotri import (
     SmotriBroadcastIE,
 )
 from .sohu import SohuIE
-from .soundcloud import SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE
+from .soundcloud import (
+    SoundcloudIE,
+    SoundcloudSetIE,
+    SoundcloudUserIE,
+    SoundcloudPlaylistIE
+)
 from .southparkstudios import (
     SouthParkStudiosIE,
     SouthparkDeIE,
@@ -153,56 +250,82 @@ from .southparkstudios import (
 from .space import SpaceIE
 from .spankwire import SpankwireIE
 from .spiegel import SpiegelIE
+from .spike import SpikeIE
 from .stanfordoc import StanfordOpenClassroomIE
-from .statigram import StatigramIE
 from .steam import SteamIE
 from .streamcloud import StreamcloudIE
+from .streamcz import StreamCZIE
+from .syfy import SyfyIE
 from .sztvhu import SztvHuIE
 from .teamcoco import TeamcocoIE
 from .techtalks import TechTalksIE
 from .ted import TEDIE
+from .testurl import TestURLIE
 from .tf1 import TF1IE
 from .theplatform import ThePlatformIE
 from .thisav import ThisAVIE
+from .tinypic import TinyPicIE
+from .tlc import TlcIE, TlcDeIE
 from .toutv import TouTvIE
+from .toypics import ToypicsUserIE, ToypicsIE
 from .traileraddict import TrailerAddictIE
 from .trilulilu import TriluliluIE
+from .trutube import TruTubeIE
 from .tube8 import Tube8IE
 from .tudou import TudouIE
 from .tumblr import TumblrIE
 from .tutv import TutvIE
+from .tvigle import TvigleIE
 from .tvp import TvpIE
+from .udemy import (
+    UdemyIE,
+    UdemyCourseIE
+)
 from .unistra import UnistraIE
+from .urort import UrortIE
 from .ustream import UstreamIE, UstreamChannelIE
 from .vbox7 import Vbox7IE
 from .veehd import VeeHDIE
 from .veoh import VeohIE
+from .vesti import VestiIE
 from .vevo import VevoIE
-from .vice import ViceIE
 from .viddler import ViddlerIE
+from .videobam import VideoBamIE
 from .videodetective import VideoDetectiveIE
+from .videolecturesnet import VideoLecturesNetIE
 from .videofyme import VideofyMeIE
 from .videopremium import VideoPremiumIE
+from .videott import VideoTtIE
+from .videoweed import VideoWeedIE
 from .vimeo import (
     VimeoIE,
     VimeoChannelIE,
     VimeoUserIE,
     VimeoAlbumIE,
     VimeoGroupsIE,
+    VimeoReviewIE,
+    VimeoWatchLaterIE,
 )
 from .vine import VineIE
 from .viki import VikiIE
 from .vk import VKIE
+from .vube import VubeIE
+from .vuclip import VuClipIE
+from .washingtonpost import WashingtonPostIE
 from .wat import WatIE
-from .websurg import WeBSurgIE
+from .wdr import (
+    WDRIE,
+    WDRMausIE,
+)
 from .weibo import WeiboIE
 from .wimp import WimpIE
 from .wistia import WistiaIE
 from .worldstarhiphop import WorldStarHipHopIE
+from .xbef import XBefIE
 from .xhamster import XHamsterIE
 from .xnxx import XNXXIE
 from .xvideos import XVideosIE
-from .xtube import XTubeIE
+from .xtube import XTubeUserIE, XTubeIE
 from .yahoo import (
     YahooIE,
     YahooNewsIE,
@@ -213,19 +336,20 @@ from .youku import YoukuIE
 from .youporn import YouPornIE
 from .youtube import (
     YoutubeIE,
-    YoutubePlaylistIE,
-    YoutubeSearchIE,
-    YoutubeSearchDateIE,
-    YoutubeUserIE,
     YoutubeChannelIE,
-    YoutubeShowIE,
-    YoutubeSubscriptionsIE,
-    YoutubeRecommendedIE,
-    YoutubeTruncatedURLIE,
-    YoutubeWatchLaterIE,
     YoutubeFavouritesIE,
     YoutubeHistoryIE,
+    YoutubePlaylistIE,
+    YoutubeRecommendedIE,
+    YoutubeSearchDateIE,
+    YoutubeSearchIE,
+    YoutubeSearchURLIE,
+    YoutubeShowIE,
+    YoutubeSubscriptionsIE,
     YoutubeTopListIE,
+    YoutubeTruncatedURLIE,
+    YoutubeUserIE,
+    YoutubeWatchLaterIE,
 )
 from .zdf import ZDFIE
 

youtube_dl/extractor/academicearth.py

@@ -1,11 +1,12 @@
+from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 
 
 class AcademicEarthCourseIE(InfoExtractor):
-    _VALID_URL = r'^https?://(?:www\.)?academicearth\.org/(?:courses|playlists)/(?P<id>[^?#/]+)'
-    IE_NAME = u'AcademicEarth:Course'
+    _VALID_URL = r'^https?://(?:www\.)?academicearth\.org/playlists/(?P<id>[^?#/]+)'
+    IE_NAME = 'AcademicEarth:Course'
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url)
@@ -13,12 +14,12 @@ class AcademicEarthCourseIE(InfoExtractor):
 
         webpage = self._download_webpage(url, playlist_id)
         title = self._html_search_regex(
-            r'<h1 class="playlist-name">(.*?)</h1>', webpage, u'title')
+            r'<h1 class="playlist-name"[^>]*?>(.*?)</h1>', webpage, u'title')
         description = self._html_search_regex(
-            r'<p class="excerpt">(.*?)</p>',
+            r'<p class="excerpt"[^>]*?>(.*?)</p>',
             webpage, u'description', fatal=False)
         urls = re.findall(
-            r'<h3 class="lecture-title"><a target="_blank" href="([^"]+)">',
+            r'<li class="lecture-preview">\s*?<a target="_blank" href="([^"]+)">',
             webpage)
         entries = [self.url_result(u) for u in urls]
 

youtube_dl/extractor/addanime.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -14,14 +16,14 @@ from ..utils import (
 class AddAnimeIE(InfoExtractor):
 
     _VALID_URL = r'^http://(?:\w+\.)?add-anime\.net/watch_video\.php\?(?:.*?)v=(?P<video_id>[\w_]+)(?:.*)'
-    IE_NAME = u'AddAnime'
     _TEST = {
-        u'url': u'http://www.add-anime.net/watch_video.php?v=24MR3YO5SAS9',
-        u'file': u'24MR3YO5SAS9.mp4',
-        u'md5': u'72954ea10bc979ab5e2eb288b21425a0',
-        u'info_dict': {
-            u"description": u"One Piece 606",
-            u"title": u"One Piece 606"
+        'url': 'http://www.add-anime.net/watch_video.php?v=24MR3YO5SAS9',
+        'md5': '72954ea10bc979ab5e2eb288b21425a0',
+        'info_dict': {
+            'id': '24MR3YO5SAS9',
+            'ext': 'mp4',
+            'description': 'One Piece 606',
+            'title': 'One Piece 606',
         }
     }
 
@@ -38,10 +40,10 @@ class AddAnimeIE(InfoExtractor):
             redir_webpage = ee.cause.read().decode('utf-8')
             action = self._search_regex(
                 r'<form id="challenge-form" action="([^"]+)"',
-                redir_webpage, u'Redirect form')
+                redir_webpage, 'Redirect form')
             vc = self._search_regex(
                 r'<input type="hidden" name="jschl_vc" value="([^"]+)"/>',
-                redir_webpage, u'redirect vc value')
+                redir_webpage, 'redirect vc value')
             av = re.search(
                 r'a\.value = ([0-9]+)[+]([0-9]+)[*]([0-9]+);',
                 redir_webpage)
@@ -52,19 +54,19 @@ class AddAnimeIE(InfoExtractor):
             parsed_url = compat_urllib_parse_urlparse(url)
             av_val = av_res + len(parsed_url.netloc)
             confirm_url = (
-                parsed_url.scheme + u'://' + parsed_url.netloc +
+                parsed_url.scheme + '://' + parsed_url.netloc +
                 action + '?' +
                 compat_urllib_parse.urlencode({
                     'jschl_vc': vc, 'jschl_answer': compat_str(av_val)}))
             self._download_webpage(
                 confirm_url, video_id,
-                note=u'Confirming after redirect')
+                note='Confirming after redirect')
             webpage = self._download_webpage(url, video_id)
 
         formats = []
         for format_id in ('normal', 'hq'):
             rex = r"var %s_video_file = '(.*?)';" % re.escape(format_id)
-            video_url = self._search_regex(rex, webpage, u'video file URLx',
+            video_url = self._search_regex(rex, webpage, 'video file URLx',
                                            fatal=False)
             if not video_url:
                 continue
@@ -72,14 +74,13 @@ class AddAnimeIE(InfoExtractor):
                 'format_id': format_id,
                 'url': video_url,
             })
-        if not formats:
-            raise ExtractorError(u'Cannot find any video format!')
+        self._sort_formats(formats)
         video_title = self._og_search_title(webpage)
         video_description = self._og_search_description(webpage)
 
         return {
             '_type': 'video',
-            'id':  video_id,
+            'id': video_id,
             'formats': formats,
             'title': video_title,
             'description': video_description

youtube_dl/extractor/aftonbladet.py

@@ -0,0 +1,69 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import datetime
+import re
+
+from .common import InfoExtractor
+
+
+class AftonbladetIE(InfoExtractor):
+    _VALID_URL = r'^http://tv\.aftonbladet\.se/webbtv.+?(?P<video_id>article[0-9]+)\.ab(?:$|[?#])'
+    _TEST = {
+        'url': 'http://tv.aftonbladet.se/webbtv/nyheter/vetenskap/rymden/article36015.ab',
+        'info_dict': {
+            'id': 'article36015',
+            'ext': 'mp4',
+            'title': 'Vulkanutbrott i rymden - nu släpper NASA bilderna',
+            'description': 'Jupiters måne mest aktiv av alla himlakroppar',
+            'upload_date': '20140306',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.search(self._VALID_URL, url)
+
+        video_id = mobj.group('video_id')
+        webpage = self._download_webpage(url, video_id)
+
+        # find internal video meta data
+        META_URL = 'http://aftonbladet-play.drlib.aptoma.no/video/%s.json'
+        internal_meta_id = self._html_search_regex(
+            r'data-aptomaId="([\w\d]+)"', webpage, 'internal_meta_id')
+        internal_meta_url = META_URL % internal_meta_id
+        internal_meta_json = self._download_json(
+            internal_meta_url, video_id, 'Downloading video meta data')
+
+        # find internal video formats
+        FORMATS_URL = 'http://aftonbladet-play.videodata.drvideo.aptoma.no/actions/video/?id=%s'
+        internal_video_id = internal_meta_json['videoId']
+        internal_formats_url = FORMATS_URL % internal_video_id
+        internal_formats_json = self._download_json(
+            internal_formats_url, video_id, 'Downloading video formats')
+
+        formats = []
+        for fmt in internal_formats_json['formats']['http']['pseudostreaming']['mp4']:
+            p = fmt['paths'][0]
+            formats.append({
+                'url': 'http://%s:%d/%s/%s' % (p['address'], p['port'], p['path'], p['filename']),
+                'ext': 'mp4',
+                'width': fmt['width'],
+                'height': fmt['height'],
+                'tbr': fmt['bitrate'],
+                'protocol': 'http',
+            })
+        self._sort_formats(formats)
+
+        timestamp = datetime.datetime.fromtimestamp(internal_meta_json['timePublished'])
+        upload_date = timestamp.strftime('%Y%m%d')
+
+        return {
+            'id': video_id,
+            'title': internal_meta_json['title'],
+            'formats': formats,
+            'thumbnail': internal_meta_json['imageUrl'],
+            'description': internal_meta_json['shortPreamble'],
+            'upload_date': upload_date,
+            'duration': internal_meta_json['duration'],
+            'view_count': internal_meta_json['views'],
+        }

youtube_dl/extractor/aol.py

@@ -0,0 +1,65 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from .fivemin import FiveMinIE
+
+
+class AolIE(InfoExtractor):
+    IE_NAME = 'on.aol.com'
+    _VALID_URL = r'''(?x)
+        (?:
+            aol-video:|
+            http://on\.aol\.com/
+            (?:
+                video/.*-|
+                playlist/(?P<playlist_display_id>[^/?#]+?)-(?P<playlist_id>[0-9]+)[?#].*_videoid=
+            )
+        )
+        (?P<id>[0-9]+)
+        (?:$|\?)
+    '''
+
+    _TEST = {
+        'url': 'http://on.aol.com/video/u-s--official-warns-of-largest-ever-irs-phone-scam-518167793?icid=OnHomepageC2Wide_MustSee_Img',
+        'md5': '18ef68f48740e86ae94b98da815eec42',
+        'info_dict': {
+            'id': '518167793',
+            'ext': 'mp4',
+            'title': 'U.S. Official Warns Of \'Largest Ever\' IRS Phone Scam',
+        },
+        'add_ie': ['FiveMin'],
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        playlist_id = mobj.group('playlist_id')
+        if playlist_id and not self._downloader.params.get('noplaylist'):
+            self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
+
+            webpage = self._download_webpage(url, playlist_id)
+            title = self._html_search_regex(
+                r'<h1 class="video-title[^"]*">(.+?)</h1>', webpage, 'title')
+            playlist_html = self._search_regex(
+                r"(?s)<ul\s+class='video-related[^']*'>(.*?)</ul>", webpage,
+                'playlist HTML')
+            entries = [{
+                '_type': 'url',
+                'url': 'aol-video:%s' % m.group('id'),
+                'ie_key': 'Aol',
+            } for m in re.finditer(
+                r"<a\s+href='.*videoid=(?P<id>[0-9]+)'\s+class='video-thumb'>",
+                playlist_html)]
+
+            return {
+                '_type': 'playlist',
+                'id': playlist_id,
+                'display_id': mobj.group('playlist_display_id'),
+                'title': title,
+                'entries': entries,
+            }
+
+        return FiveMinIE._build_result(video_id)

youtube_dl/extractor/appletrailers.py

@@ -1,58 +1,63 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
 from .common import InfoExtractor
 from ..utils import (
     compat_urlparse,
-    determine_ext,
 )
 
 
 class AppleTrailersIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/trailers/(?P<company>[^/]+)/(?P<movie>[^/]+)'
     _TEST = {
-        u"url": u"http://trailers.apple.com/trailers/wb/manofsteel/",
-        u"playlist": [
+        "url": "http://trailers.apple.com/trailers/wb/manofsteel/",
+        "playlist": [
             {
-                u"file": u"manofsteel-trailer4.mov",
-                u"md5": u"d97a8e575432dbcb81b7c3acb741f8a8",
-                u"info_dict": {
-                    u"duration": 111,
-                    u"title": u"Trailer 4",
-                    u"upload_date": u"20130523",
-                    u"uploader_id": u"wb",
+                "md5": "d97a8e575432dbcb81b7c3acb741f8a8",
+                "info_dict": {
+                    "id": "manofsteel-trailer4",
+                    "ext": "mov",
+                    "duration": 111,
+                    "title": "Trailer 4",
+                    "upload_date": "20130523",
+                    "uploader_id": "wb",
                 },
             },
             {
-                u"file": u"manofsteel-trailer3.mov",
-                u"md5": u"b8017b7131b721fb4e8d6f49e1df908c",
-                u"info_dict": {
-                    u"duration": 182,
-                    u"title": u"Trailer 3",
-                    u"upload_date": u"20130417",
-                    u"uploader_id": u"wb",
+                "md5": "b8017b7131b721fb4e8d6f49e1df908c",
+                "info_dict": {
+                    "id": "manofsteel-trailer3",
+                    "ext": "mov",
+                    "duration": 182,
+                    "title": "Trailer 3",
+                    "upload_date": "20130417",
+                    "uploader_id": "wb",
                 },
             },
             {
-                u"file": u"manofsteel-trailer.mov",
-                u"md5": u"d0f1e1150989b9924679b441f3404d48",
-                u"info_dict": {
-                    u"duration": 148,
-                    u"title": u"Trailer",
-                    u"upload_date": u"20121212",
-                    u"uploader_id": u"wb",
+                "md5": "d0f1e1150989b9924679b441f3404d48",
+                "info_dict": {
+                    "id": "manofsteel-trailer",
+                    "ext": "mov",
+                    "duration": 148,
+                    "title": "Trailer",
+                    "upload_date": "20121212",
+                    "uploader_id": "wb",
                 },
             },
             {
-                u"file": u"manofsteel-teaser.mov",
-                u"md5": u"5fe08795b943eb2e757fa95cb6def1cb",
-                u"info_dict": {
-                    u"duration": 93,
-                    u"title": u"Teaser",
-                    u"upload_date": u"20120721",
-                    u"uploader_id": u"wb",
+                "md5": "5fe08795b943eb2e757fa95cb6def1cb",
+                "info_dict": {
+                    "id": "manofsteel-teaser",
+                    "ext": "mov",
+                    "duration": 93,
+                    "title": "Teaser",
+                    "upload_date": "20120721",
+                    "uploader_id": "wb",
                 },
-            }
+            },
         ]
     }
 
@@ -63,16 +68,16 @@ class AppleTrailersIE(InfoExtractor):
         movie = mobj.group('movie')
         uploader_id = mobj.group('company')
 
-        playlist_url = compat_urlparse.urljoin(url, u'includes/playlists/itunes.inc')
+        playlist_url = compat_urlparse.urljoin(url, 'includes/playlists/itunes.inc')
         def fix_html(s):
-            s = re.sub(r'(?s)<script[^<]*?>.*?</script>', u'', s)
+            s = re.sub(r'(?s)<script[^<]*?>.*?</script>', '', s)
             s = re.sub(r'<img ([^<]*?)>', r'<img \1/>', s)
             # The ' in the onClick attributes are not escaped, it couldn't be parsed
             # like: http://trailers.apple.com/trailers/wb/gravity/
             def _clean_json(m):
-                return u'iTunes.playURL(%s);' % m.group(1).replace('\'', '&#39;')
+                return 'iTunes.playURL(%s);' % m.group(1).replace('\'', '&#39;')
             s = re.sub(self._JSON_RE, _clean_json, s)
-            s = u'<html>' + s + u'</html>'
+            s = '<html>' + s + u'</html>'
             return s
         doc = self._download_xml(playlist_url, movie, transform_source=fix_html)
 
@@ -80,7 +85,7 @@ class AppleTrailersIE(InfoExtractor):
         for li in doc.findall('./div/ul/li'):
             on_click = li.find('.//a').attrib['onClick']
             trailer_info_json = self._search_regex(self._JSON_RE,
-                on_click, u'trailer info')
+                on_click, 'trailer info')
             trailer_info = json.loads(trailer_info_json)
             title = trailer_info['title']
             video_id = movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', title).lower()
@@ -96,8 +101,7 @@ class AppleTrailersIE(InfoExtractor):
             first_url = trailer_info['url']
             trailer_id = first_url.split('/')[-1].rpartition('_')[0].lower()
             settings_json_url = compat_urlparse.urljoin(url, 'includes/settings/%s.json' % trailer_id)
-            settings_json = self._download_webpage(settings_json_url, trailer_id, u'Downloading settings json')
-            settings = json.loads(settings_json)
+            settings = self._download_json(settings_json_url, trailer_id, 'Downloading settings json')
 
             formats = []
             for format in settings['metadata']['sizes']:
@@ -105,12 +109,12 @@ class AppleTrailersIE(InfoExtractor):
                 format_url = re.sub(r'_(\d*p.mov)', r'_h\1', format['src'])
                 formats.append({
                     'url': format_url,
-                    'ext': determine_ext(format_url),
                     'format': format['type'],
                     'width': format['width'],
                     'height': int(format['height']),
                 })
-            formats = sorted(formats, key=lambda f: (f['height'], f['width']))
+
+            self._sort_formats(formats)
 
             playlist.append({
                 '_type': 'video',

youtube_dl/extractor/archiveorg.py

@@ -1,9 +1,10 @@
+from __future__ import unicode_literals
+
 import json
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
     unified_strdate,
 )
 
@@ -13,23 +14,22 @@ class ArchiveOrgIE(InfoExtractor):
     IE_DESC = 'archive.org videos'
     _VALID_URL = r'(?:https?://)?(?:www\.)?archive\.org/details/(?P<id>[^?/]+)(?:[?].*)?$'
     _TEST = {
-        u"url": u"http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect",
-        u'file': u'XD300-23_68HighlightsAResearchCntAugHumanIntellect.ogv',
-        u'md5': u'8af1d4cf447933ed3c7f4871162602db',
-        u'info_dict': {
-            u"title": u"1968 Demo - FJCC Conference Presentation Reel #1",
-            u"description": u"Reel 1 of 3: Also known as the \"Mother of All Demos\", Doug Engelbart's presentation at the Fall Joint Computer Conference in San Francisco, December 9, 1968 titled \"A Research Center for Augmenting Human Intellect.\" For this presentation, Doug and his team astonished the audience by not only relating their research, but demonstrating it live. This was the debut of the mouse, interactive computing, hypermedia, computer supported software engineering, video teleconferencing, etc. See also <a href=\"http://dougengelbart.org/firsts/dougs-1968-demo.html\" rel=\"nofollow\">Doug's 1968 Demo page</a> for more background, highlights, links, and the detailed paper published in this conference proceedings. Filmed on 3 reels: Reel 1 | <a href=\"http://www.archive.org/details/XD300-24_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 2</a> | <a href=\"http://www.archive.org/details/XD300-25_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 3</a>",
-            u"upload_date": u"19681210",
-            u"uploader": u"SRI International"
+        "url": "http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect",
+        'file': 'XD300-23_68HighlightsAResearchCntAugHumanIntellect.ogv',
+        'md5': '8af1d4cf447933ed3c7f4871162602db',
+        'info_dict': {
+            "title": "1968 Demo - FJCC Conference Presentation Reel #1",
+            "description": "Reel 1 of 3: Also known as the \"Mother of All Demos\", Doug Engelbart's presentation at the Fall Joint Computer Conference in San Francisco, December 9, 1968 titled \"A Research Center for Augmenting Human Intellect.\" For this presentation, Doug and his team astonished the audience by not only relating their research, but demonstrating it live. This was the debut of the mouse, interactive computing, hypermedia, computer supported software engineering, video teleconferencing, etc. See also <a href=\"http://dougengelbart.org/firsts/dougs-1968-demo.html\" rel=\"nofollow\">Doug's 1968 Demo page</a> for more background, highlights, links, and the detailed paper published in this conference proceedings. Filmed on 3 reels: Reel 1 | <a href=\"http://www.archive.org/details/XD300-24_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 2</a> | <a href=\"http://www.archive.org/details/XD300-25_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 3</a>",
+            "upload_date": "19681210",
+            "uploader": "SRI International"
         }
     }
 
-
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
 
-        json_url = url + (u'?' if u'?' in url else '&') + u'output=json'
+        json_url = url + ('?' if '?' in url else '&') + 'output=json'
         json_data = self._download_webpage(json_url, video_id)
         data = json.loads(json_data)
 
@@ -38,16 +38,16 @@ class ArchiveOrgIE(InfoExtractor):
         uploader = data['metadata']['creator'][0]
         upload_date = unified_strdate(data['metadata']['date'][0])
 
-        formats = [{
+        formats = [
+            {
                 'format': fdata['format'],
                 'url': 'http://' + data['server'] + data['dir'] + fn,
                 'file_size': int(fdata['size']),
             }
-            for fn,fdata in data['files'].items()
+            for fn, fdata in data['files'].items()
             if 'Video' in fdata['format']]
-        formats.sort(key=lambda fdata: fdata['file_size'])
-        for f in formats:
-            f['ext'] = determine_ext(f['url'])
+
+        self._sort_formats(formats)
 
         return {
             '_type': 'video',

youtube_dl/extractor/ard.py

@@ -1,22 +1,28 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
+    determine_ext,
     ExtractorError,
 )
 
+
 class ARDIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?'
-    _TITLE = r'<h1(?: class="boxTopHeadline")?>(?P<title>.*)</h1>'
-    _MEDIA_STREAM = r'mediaCollection\.addMediaStream\((?P<media_type>\d+), (?P<quality>\d+), "(?P<rtmp_url>[^"]*)", "(?P<video_url>[^"]*)", "[^"]*"\)'
+    _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?'
+
     _TEST = {
-        u'url': u'http://www.ardmediathek.de/das-erste/tagesschau-in-100-sek?documentId=14077640',
-        u'file': u'14077640.mp4',
-        u'md5': u'6ca8824255460c787376353f9e20bbd8',
-        u'info_dict': {
-            u"title": u"11.04.2013 09:23 Uhr - Tagesschau in 100 Sekunden"
+        'url': 'http://www.ardmediathek.de/das-erste/guenther-jauch/edward-snowden-im-interview-held-oder-verraeter?documentId=19288786',
+        'file': '19288786.mp4',
+        'md5': '515bf47ce209fb3f5a61b7aad364634c',
+        'info_dict': {
+            'title': 'Edward Snowden im Interview - Held oder Verräter?',
+            'description': 'Edward Snowden hat alles aufs Spiel gesetzt, um die weltweite \xdcberwachung durch die Geheimdienste zu enttarnen. Nun stellt sich der ehemalige NSA-Mitarbeiter erstmals weltweit in einem TV-Interview den Fragen eines NDR-Journalisten. Die Sendung vom Sonntagabend.',
+            'thumbnail': 'http://www.ardmediathek.de/ard/servlet/contentblob/19/28/87/90/19288790/bild/2250037',
         },
-        u'skip': u'Requires rtmpdump'
+        'skip': 'Blocked outside of Germany',
     }
 
     def _real_extract(self, url):
@@ -29,26 +35,49 @@ class ARDIE(InfoExtractor):
         else:
             video_id = m.group('video_id')
 
-        # determine title and media streams from webpage
-        html = self._download_webpage(url, video_id)
-        title = re.search(self._TITLE, html).group('title')
-        streams = [mo.groupdict() for mo in re.finditer(self._MEDIA_STREAM, html)]
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._html_search_regex(
+            r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>', webpage, 'title')
+        description = self._html_search_meta(
+            'dcterms.abstract', webpage, 'description')
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        streams = [
+            mo.groupdict()
+            for mo in re.finditer(
+                r'mediaCollection\.addMediaStream\((?P<media_type>\d+), (?P<quality>\d+), "(?P<rtmp_url>[^"]*)", "(?P<video_url>[^"]*)", "[^"]*"\)', webpage)]
         if not streams:
-            assert '"fsk"' in html
-            raise ExtractorError(u'This video is only available after 8:00 pm')
+            if '"fsk"' in webpage:
+                raise ExtractorError('This video is only available after 20:00')
 
-        # choose default media type and highest quality for now
-        stream = max([s for s in streams if int(s["media_type"]) == 0],
-                     key=lambda s: int(s["quality"]))
+        formats = []
+        for s in streams:
+            format = {
+                'quality': int(s['quality']),
+            }
+            if s.get('rtmp_url'):
+                format['protocol'] = 'rtmp'
+                format['url'] = s['rtmp_url']
+                format['playpath'] = s['video_url']
+            else:
+                format['url'] = s['video_url']
 
-        # there's two possibilities: RTMP stream or HTTP download
-        info = {'id': video_id, 'title': title, 'ext': 'mp4'}
-        if stream['rtmp_url']:
-            self.to_screen(u'RTMP download detected')
-            assert stream['video_url'].startswith('mp4:')
-            info["url"] = stream["rtmp_url"]
-            info["play_path"] = stream['video_url']
-        else:
-            assert stream["video_url"].endswith('.mp4')
-            info["url"] = stream["video_url"]
-        return [info]
+            quality_name = self._search_regex(
+                r'[,.]([a-zA-Z0-9_-]+),?\.mp4', format['url'],
+                'quality name', default='NA')
+            format['format_id'] = '%s-%s-%s-%s' % (
+                determine_ext(format['url']), quality_name, s['media_type'],
+                s['quality'])
+
+            formats.append(format)
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'formats': formats,
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/arte.py

@@ -1,6 +1,7 @@
 # encoding: utf-8
+from __future__ import unicode_literals
+
 import re
-import json
 
 from .common import InfoExtractor
 from ..utils import (
@@ -17,115 +18,46 @@ from ..utils import (
 # is different for each one. The videos usually expire in 7 days, so we can't
 # add tests.
 
+
 class ArteTvIE(InfoExtractor):
-    _VIDEOS_URL = r'(?:http://)?videos\.arte\.tv/(?P<lang>fr|de)/.*-(?P<id>.*?)\.html'
-    _LIVEWEB_URL = r'(?:http://)?liveweb\.arte\.tv/(?P<lang>fr|de)/(?P<subpage>.+?)/(?P<name>.+)'
-    _LIVE_URL = r'index-[0-9]+\.html$'
-
-    IE_NAME = u'arte.tv'
-
-    @classmethod
-    def suitable(cls, url):
-        return any(re.match(regex, url) for regex in (cls._VIDEOS_URL, cls._LIVEWEB_URL))
-
-    # TODO implement Live Stream
-    # from ..utils import compat_urllib_parse
-    # def extractLiveStream(self, url):
-    #     video_lang = url.split('/')[-4]
-    #     info = self.grep_webpage(
-    #         url,
-    #         r'src="(.*?/videothek_js.*?\.js)',
-    #         0,
-    #         [
-    #             (1, 'url', u'Invalid URL: %s' % url)
-    #         ]
-    #     )
-    #     http_host = url.split('/')[2]
-    #     next_url = 'http://%s%s' % (http_host, compat_urllib_parse.unquote(info.get('url')))
-    #     info = self.grep_webpage(
-    #         next_url,
-    #         r'(s_artestras_scst_geoFRDE_' + video_lang + '.*?)\'.*?' +
-    #             '(http://.*?\.swf).*?' +
-    #             '(rtmp://.*?)\'',
-    #         re.DOTALL,
-    #         [
-    #             (1, 'path',   u'could not extract video path: %s' % url),
-    #             (2, 'player', u'could not extract video player: %s' % url),
-    #             (3, 'url',    u'could not extract video url: %s' % url)
-    #         ]
-    #     )
-    #     video_url = u'%s/%s' % (info.get('url'), info.get('path'))
+    _VALID_URL = r'http://videos\.arte\.tv/(?P<lang>fr|de)/.*-(?P<id>.*?)\.html'
+    IE_NAME = 'arte.tv'
 
     def _real_extract(self, url):
-        mobj = re.match(self._VIDEOS_URL, url)
-        if mobj is not None:
-            id = mobj.group('id')
-            lang = mobj.group('lang')
-            return self._extract_video(url, id, lang)
+        mobj = re.match(self._VALID_URL, url)
+        lang = mobj.group('lang')
+        video_id = mobj.group('id')
 
-        mobj = re.match(self._LIVEWEB_URL, url)
-        if mobj is not None:
-            name = mobj.group('name')
-            lang = mobj.group('lang')
-            return self._extract_liveweb(url, name, lang)
-
-        if re.search(self._LIVE_URL, url) is not None:
-            raise ExtractorError(u'Arte live streams are not yet supported, sorry')
-            # self.extractLiveStream(url)
-            # return
-
-    def _extract_video(self, url, video_id, lang):
-        """Extract from videos.arte.tv"""
         ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
         ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
-        ref_xml_doc = self._download_xml(ref_xml_url, video_id, note=u'Downloading metadata')
+        ref_xml_doc = self._download_xml(
+            ref_xml_url, video_id, note='Downloading metadata')
         config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
         config_xml_url = config_node.attrib['ref']
-        config_xml = self._download_webpage(config_xml_url, video_id, note=u'Downloading configuration')
+        config = self._download_xml(
+            config_xml_url, video_id, note='Downloading configuration')
 
-        video_urls = list(re.finditer(r'<url quality="(?P<quality>.*?)">(?P<url>.*?)</url>', config_xml))
-        def _key(m):
-            quality = m.group('quality')
-            if quality == 'hd':
-                return 2
-            else:
-                return 1
-        # We pick the best quality
-        video_urls = sorted(video_urls, key=_key)
-        video_url = list(video_urls)[-1].group('url')
-        
-        title = self._html_search_regex(r'<name>(.*?)</name>', config_xml, 'title')
-        thumbnail = self._html_search_regex(r'<firstThumbnailUrl>(.*?)</firstThumbnailUrl>',
-                                            config_xml, 'thumbnail')
-        return {'id': video_id,
-                'title': title,
-                'thumbnail': thumbnail,
-                'url': video_url,
-                'ext': 'flv',
-                }
+        formats = [{
+            'forma_id': q.attrib['quality'],
+            'url': q.text,
+            'ext': 'flv',
+            'quality': 2 if q.attrib['quality'] == 'hd' else 1,
+        } for q in config.findall('./urls/url')]
+        self._sort_formats(formats)
 
-    def _extract_liveweb(self, url, name, lang):
-        """Extract form http://liveweb.arte.tv/"""
-        webpage = self._download_webpage(url, name)
-        video_id = self._search_regex(r'eventId=(\d+?)("|&)', webpage, u'event id')
-        config_doc = self._download_xml('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id,
-                                            video_id, u'Downloading information')
-        event_doc = config_doc.find('event')
-        url_node = event_doc.find('video').find('urlHd')
-        if url_node is None:
-            url_node = event_doc.find('urlSd')
-
-        return {'id': video_id,
-                'title': event_doc.find('name%s' % lang.capitalize()).text,
-                'url': url_node.text.replace('MP4', 'mp4'),
-                'ext': 'flv',
-                'thumbnail': self._og_search_thumbnail(webpage),
-                }
+        title = config.find('.//name').text
+        thumbnail = config.find('.//firstThumbnailUrl').text
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'formats': formats,
+        }
 
 
 class ArteTVPlus7IE(InfoExtractor):
-    IE_NAME = u'arte.tv:+7'
-    _VALID_URL = r'https?://www\.arte.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
+    IE_NAME = 'arte.tv:+7'
+    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
 
     @classmethod
     def _extract_url_info(cls, url):
@@ -142,13 +74,12 @@ class ArteTVPlus7IE(InfoExtractor):
         return self._extract_from_webpage(webpage, video_id, lang)
 
     def _extract_from_webpage(self, webpage, video_id, lang):
-        json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')
+        json_url = self._html_search_regex(
+            r'arte_vp_url="(.*?)"', webpage, 'json vp url')
         return self._extract_from_json_url(json_url, video_id, lang)
 
     def _extract_from_json_url(self, json_url, video_id, lang):
-        json_info = self._download_webpage(json_url, video_id, 'Downloading info json')
-        self.report_extraction(video_id)
-        info = json.loads(json_info)
+        info = self._download_json(json_url, video_id)
         player_info = info['videoJsonPlayer']
 
         info_dict = {
@@ -170,6 +101,8 @@ class ArteTVPlus7IE(InfoExtractor):
                 l = 'F'
             elif lang == 'de':
                 l = 'A'
+            else:
+                l = lang
             regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l]
             return any(re.match(r, f['versionCode']) for r in regexes)
         # Some formats may not be in the same language as the url
@@ -188,14 +121,19 @@ class ArteTVPlus7IE(InfoExtractor):
                 return ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])
         else:
             def sort_key(f):
+                versionCode = f.get('versionCode')
+                if versionCode is None:
+                    versionCode = ''
                 return (
                     # Sort first by quality
-                    int(f.get('height',-1)),
-                    int(f.get('bitrate',-1)),
+                    int(f.get('height', -1)),
+                    int(f.get('bitrate', -1)),
                     # The original version with subtitles has lower relevance
-                    re.match(r'VO-ST(F|A)', f.get('versionCode', '')) is None,
+                    re.match(r'VO-ST(F|A)', versionCode) is None,
                     # The version with sourds/mal subtitles has also lower relevance
-                    re.match(r'VO?(F|A)-STM\1', f.get('versionCode', '')) is None,
+                    re.match(r'VO?(F|A)-STM\1', versionCode) is None,
+                    # Prefer http downloads over m3u8
+                    0 if f['url'].endswith('m3u8') else 1,
                 )
         formats = sorted(formats, key=sort_key)
         def _format(format_info):
@@ -207,7 +145,7 @@ class ArteTVPlus7IE(InfoExtractor):
             if bitrate is not None:
                 quality += '-%d' % bitrate
             if format_info.get('versionCode') is not None:
-                format_id = u'%s-%s' % (quality, format_info['versionCode'])
+                format_id = '%s-%s' % (quality, format_info['versionCode'])
             else:
                 format_id = quality
             info = {
@@ -216,7 +154,7 @@ class ArteTVPlus7IE(InfoExtractor):
                 'width': format_info.get('width'),
                 'height': height,
             }
-            if format_info['mediaType'] == u'rtmp':
+            if format_info['mediaType'] == 'rtmp':
                 info['url'] = format_info['streamer']
                 info['play_path'] = 'mp4:' + format_info['url']
                 info['ext'] = 'flv'
@@ -231,27 +169,29 @@ class ArteTVPlus7IE(InfoExtractor):
 
 # It also uses the arte_vp_url url from the webpage to extract the information
 class ArteTVCreativeIE(ArteTVPlus7IE):
-    IE_NAME = u'arte.tv:creative'
+    IE_NAME = 'arte.tv:creative'
     _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/magazine?/(?P<id>.+)'
 
     _TEST = {
-        u'url': u'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
-        u'file': u'050489-002.mp4',
-        u'info_dict': {
-            u'title': u'Agentur Amateur / Agence Amateur #2 : Corporate Design',
+        'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
+        'info_dict': {
+            'id': '050489-002',
+            'ext': 'mp4',
+            'title': 'Agentur Amateur / Agence Amateur #2 : Corporate Design',
         },
     }
 
 
 class ArteTVFutureIE(ArteTVPlus7IE):
-    IE_NAME = u'arte.tv:future'
+    IE_NAME = 'arte.tv:future'
     _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(thema|sujet)/.*?#article-anchor-(?P<id>\d+)'
 
     _TEST = {
-        u'url': u'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081',
-        u'file': u'050940-003.mp4',
-        u'info_dict': {
-            u'title': u'Les champignons au secours de la planète',
+        'url': 'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081',
+        'info_dict': {
+            'id': '050940-003',
+            'ext': 'mp4',
+            'title': 'Les champignons au secours de la planète',
         },
     }
 
@@ -263,8 +203,8 @@ class ArteTVFutureIE(ArteTVPlus7IE):
 
 
 class ArteTVDDCIE(ArteTVPlus7IE):
-    IE_NAME = u'arte.tv:ddc'
-    _VALID_URL = r'http?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)'
+    IE_NAME = 'arte.tv:ddc'
+    _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)'
 
     def _real_extract(self, url):
         video_id, lang = self._extract_url_info(url)
@@ -278,3 +218,39 @@ class ArteTVDDCIE(ArteTVPlus7IE):
         javascriptPlayerGenerator = self._download_webpage(script_url, video_id, 'Download javascript player generator')
         json_url = self._search_regex(r"json_url=(.*)&rendering_place.*", javascriptPlayerGenerator, 'json url')
         return self._extract_from_json_url(json_url, video_id, lang)
+
+
+class ArteTVConcertIE(ArteTVPlus7IE):
+    IE_NAME = 'arte.tv:concert'
+    _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>de|fr)/(?P<id>.+)'
+
+    _TEST = {
+        'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde',
+        'md5': '9ea035b7bd69696b67aa2ccaaa218161',
+        'info_dict': {
+            'id': '186',
+            'ext': 'mp4',
+            'title': 'The Notwist im Pariser Konzertclub "Divan du Monde"',
+            'upload_date': '20140128',
+            'description': 'md5:486eb08f991552ade77439fe6d82c305',
+        },
+    }
+
+
+class ArteTVEmbedIE(ArteTVPlus7IE):
+    IE_NAME = 'arte.tv:embed'
+    _VALID_URL = r'''(?x)
+        http://www\.arte\.tv
+        /playerv2/embed\.php\?json_url=
+        (?P<json_url>
+            http://arte\.tv/papi/tvguide/videos/stream/player/
+            (?P<lang>[^/]+)/(?P<id>[^/]+)[^&]*
+        )
+    '''
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        lang = mobj.group('lang')
+        json_url = mobj.group('json_url')
+        return self._extract_from_json_url(json_url, video_id, lang)

youtube_dl/extractor/auengine.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -7,23 +9,26 @@ from ..utils import (
     ExtractorError,
 )
 
+
 class AUEngineIE(InfoExtractor):
+    _VALID_URL = r'http://(?:www\.)?auengine\.com/embed\.php\?.*?file=(?P<id>[^&]+).*?'
+
     _TEST = {
-        u'url': u'http://auengine.com/embed.php?file=lfvlytY6&w=650&h=370',
-        u'file': u'lfvlytY6.mp4',
-        u'md5': u'48972bdbcf1a3a2f5533e62425b41d4f',
-        u'info_dict': {
-            u"title": u"[Commie]The Legend of the Legendary Heroes - 03 - Replication Eye (Alpha Stigma)[F9410F5A]"
+        'url': 'http://auengine.com/embed.php?file=lfvlytY6&w=650&h=370',
+        'md5': '48972bdbcf1a3a2f5533e62425b41d4f',
+        'info_dict': {
+            'id': 'lfvlytY6',
+            'ext': 'mp4',
+            'title': '[Commie]The Legend of the Legendary Heroes - 03 - Replication Eye (Alpha Stigma)[F9410F5A]'
         }
     }
-    _VALID_URL = r'(?:http://)?(?:www\.)?auengine\.com/embed\.php\?.*?file=([^&]+).*?'
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1)
+        video_id = mobj.group('id')
+
         webpage = self._download_webpage(url, video_id)
-        title = self._html_search_regex(r'<title>(?P<title>.+?)</title>',
-                webpage, u'title')
+        title = self._html_search_regex(r'<title>(?P<title>.+?)</title>', webpage, 'title')
         title = title.strip()
         links = re.findall(r'\s(?:file|url):\s*["\']([^\'"]+)["\']', webpage)
         links = map(compat_urllib_parse.unquote, links)
@@ -36,14 +41,15 @@ class AUEngineIE(InfoExtractor):
             elif '/videos/' in link:
                 video_url = link
         if not video_url:
-            raise ExtractorError(u'Could not find video URL')
-        ext = u'.' + determine_ext(video_url)
+            raise ExtractorError('Could not find video URL')
+        ext = '.' + determine_ext(video_url)
         if ext == title[-len(ext):]:
             title = title[:-len(ext)]
 
         return {
-            'id':        video_id,
-            'url':       video_url,
-            'title':     title,
+            'id': video_id,
+            'url': video_url,
+            'title': title,
             'thumbnail': thumbnail,
+            'http_referer': 'http://www.auengine.com/flowplayer/flowplayer.commercial-3.2.14.swf',
         }

youtube_dl/extractor/bambuser.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 import json
 import itertools
@@ -9,26 +11,26 @@ from ..utils import (
 
 
 class BambuserIE(InfoExtractor):
-    IE_NAME = u'bambuser'
+    IE_NAME = 'bambuser'
     _VALID_URL = r'https?://bambuser\.com/v/(?P<id>\d+)'
     _API_KEY = '005f64509e19a868399060af746a00aa'
 
     _TEST = {
-        u'url': u'http://bambuser.com/v/4050584',
+        'url': 'http://bambuser.com/v/4050584',
         # MD5 seems to be flaky, see https://travis-ci.org/rg3/youtube-dl/jobs/14051016#L388
-        #u'md5': u'fba8f7693e48fd4e8641b3fd5539a641',
-        u'info_dict': {
-            u'id': u'4050584',
-            u'ext': u'flv',
-            u'title': u'Education engineering days - lightning talks',
-            u'duration': 3741,
-            u'uploader': u'pixelversity',
-            u'uploader_id': u'344706',
+        #u'md5': 'fba8f7693e48fd4e8641b3fd5539a641',
+        'info_dict': {
+            'id': '4050584',
+            'ext': 'flv',
+            'title': 'Education engineering days - lightning talks',
+            'duration': 3741,
+            'uploader': 'pixelversity',
+            'uploader_id': '344706',
         },
-        u'params': {
+        'params': {
             # It doesn't respect the 'Range' header, it would download the whole video
             # caused the travis builds to fail: https://travis-ci.org/rg3/youtube-dl/jobs/14493845#L59
-            u'skip_download': True,
+            'skip_download': True,
         },
     }
 
@@ -53,7 +55,7 @@ class BambuserIE(InfoExtractor):
 
 
 class BambuserChannelIE(InfoExtractor):
-    IE_NAME = u'bambuser:channel'
+    IE_NAME = 'bambuser:channel'
     _VALID_URL = r'https?://bambuser\.com/channel/(?P<user>.*?)(?:/|#|\?|$)'
     # The maximum number we can get with each request
     _STEP = 50
@@ -72,7 +74,7 @@ class BambuserChannelIE(InfoExtractor):
             # Without setting this header, we wouldn't get any result
             req.add_header('Referer', 'http://bambuser.com/channel/%s' % user)
             info_json = self._download_webpage(req, user,
-                u'Downloading page %d' % i)
+                'Downloading page %d' % i)
             results = json.loads(info_json)['result']
             if len(results) == 0:
                 break

youtube_dl/extractor/bandcamp.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import json
 import re
 
@@ -10,16 +12,16 @@ from ..utils import (
 
 
 class BandcampIE(InfoExtractor):
-    IE_NAME = u'Bandcamp'
-    _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P<title>.*)'
+    _VALID_URL = r'https?://.*?\.bandcamp\.com/track/(?P<title>.*)'
     _TESTS = [{
-        u'url': u'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
-        u'file': u'1812978515.mp3',
-        u'md5': u'cdeb30cdae1921719a3cbcab696ef53c',
-        u'info_dict': {
-            u"title": u"youtube-dl test song \"'/\\\u00e4\u21ad"
+        'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
+        'file': '1812978515.mp3',
+        'md5': 'c557841d5e50261777a6585648adf439',
+        'info_dict': {
+            "title": "youtube-dl  \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad",
+            "duration": 10,
         },
-        u'skip': u'There is a limit of 200 free downloads / month for the test song'
+        '_skip': 'There is a limit of 200 free downloads / month for the test song'
     }]
 
     def _real_extract(self, url):
@@ -30,100 +32,117 @@ class BandcampIE(InfoExtractor):
         m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
         if m_download is None:
             m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage)
-        if m_trackinfo:
-            json_code = m_trackinfo.group(1)
-            data = json.loads(json_code)
+            if m_trackinfo:
+                json_code = m_trackinfo.group(1)
+                data = json.loads(json_code)
+                d = data[0]
+
+                duration = int(round(d['duration']))
+                formats = []
+                for format_id, format_url in d['file'].items():
+                    ext, _, abr_str = format_id.partition('-')
+
+                    formats.append({
+                        'format_id': format_id,
+                        'url': format_url,
+                        'ext': format_id.partition('-')[0],
+                        'vcodec': 'none',
+                        'acodec': format_id.partition('-')[0],
+                        'abr': int(format_id.partition('-')[2]),
+                    })
+
+                self._sort_formats(formats)
 
-            for d in data:
-                formats = [{
-                    'format_id': 'format_id',
-                    'url': format_url,
-                    'ext': format_id.partition('-')[0]
-                } for format_id, format_url in sorted(d['file'].items())]
                 return {
                     'id': compat_str(d['id']),
                     'title': d['title'],
                     'formats': formats,
+                    'duration': duration,
                 }
-        else:
-            raise ExtractorError(u'No free songs found')
+            else:
+                raise ExtractorError('No free songs found')
 
         download_link = m_download.group(1)
-        id = re.search(r'var TralbumData = {(.*?)id: (?P<id>\d*?)$', 
-                       webpage, re.MULTILINE|re.DOTALL).group('id')
+        video_id = re.search(
+            r'var TralbumData = {(.*?)id: (?P<id>\d*?)$',
+            webpage, re.MULTILINE | re.DOTALL).group('id')
 
-        download_webpage = self._download_webpage(download_link, id,
+        download_webpage = self._download_webpage(download_link, video_id,
                                                   'Downloading free downloads page')
         # We get the dictionary of the track from some javascrip code
         info = re.search(r'items: (.*?),$',
                          download_webpage, re.MULTILINE).group(1)
         info = json.loads(info)[0]
         # We pick mp3-320 for now, until format selection can be easily implemented.
-        mp3_info = info[u'downloads'][u'mp3-320']
+        mp3_info = info['downloads']['mp3-320']
         # If we try to use this url it says the link has expired
-        initial_url = mp3_info[u'url']
+        initial_url = mp3_info['url']
         re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
         m_url = re.match(re_url, initial_url)
         #We build the url we will use to get the final track url
         # This url is build in Bandcamp in the script download_bunde_*.js
-        request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), id, m_url.group('ts'))
-        final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url')
+        request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), video_id, m_url.group('ts'))
+        final_url_webpage = self._download_webpage(request_url, video_id, 'Requesting download url')
         # If we could correctly generate the .rand field the url would be
         #in the "download_url" key
         final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
 
-        track_info = {'id':id,
-                      'title' : info[u'title'],
-                      'ext' :   'mp3',
-                      'url' :   final_url,
-                      'thumbnail' : info[u'thumb_url'],
-                      'uploader' :  info[u'artist']
-                      }
-
-        return [track_info]
+        return {
+            'id': video_id,
+            'title': info['title'],
+            'ext': 'mp3',
+            'vcodec': 'none',
+            'url': final_url,
+            'thumbnail': info.get('thumb_url'),
+            'uploader': info.get('artist'),
+        }
 
 
 class BandcampAlbumIE(InfoExtractor):
-    IE_NAME = u'Bandcamp:album'
-    _VALID_URL = r'http://.*?\.bandcamp\.com/album/(?P<title>.*)'
+    IE_NAME = 'Bandcamp:album'
+    _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<title>[^?#]+))?'
 
     _TEST = {
-        u'url': u'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
-        u'playlist': [
+        'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
+        'playlist': [
             {
-                u'file': u'1353101989.mp3',
-                u'md5': u'39bc1eded3476e927c724321ddf116cf',
-                u'info_dict': {
-                    u'title': u'Intro',
+                'file': '1353101989.mp3',
+                'md5': '39bc1eded3476e927c724321ddf116cf',
+                'info_dict': {
+                    'title': 'Intro',
                 }
             },
             {
-                u'file': u'38097443.mp3',
-                u'md5': u'1a2c32e2691474643e912cc6cd4bffaa',
-                u'info_dict': {
-                    u'title': u'Kero One - Keep It Alive (Blazo remix)',
+                'file': '38097443.mp3',
+                'md5': '1a2c32e2691474643e912cc6cd4bffaa',
+                'info_dict': {
+                    'title': 'Kero One - Keep It Alive (Blazo remix)',
                 }
             },
         ],
-        u'params': {
-            u'playlistend': 2
+        'params': {
+            'playlistend': 2
         },
-        u'skip': u'Bancamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test'
+        'skip': 'Bancamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test'
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('subdomain')
         title = mobj.group('title')
-        webpage = self._download_webpage(url, title)
+        display_id = title or playlist_id
+        webpage = self._download_webpage(url, display_id)
         tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
         if not tracks_paths:
-            raise ExtractorError(u'The page doesn\'t contain any track')
+            raise ExtractorError('The page doesn\'t contain any tracks')
         entries = [
             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
             for t_path in tracks_paths]
-        title = self._search_regex(r'album_title : "(.*?)"', webpage, u'title')
+        title = self._search_regex(r'album_title : "(.*?)"', webpage, 'title')
         return {
             '_type': 'playlist',
+            'id': playlist_id,
+            'display_id': display_id,
             'title': title,
             'entries': entries,
         }

youtube_dl/extractor/bbccouk.py

@@ -0,0 +1,223 @@
+from __future__ import unicode_literals
+
+import re
+
+from .subtitles import SubtitlesInfoExtractor
+from ..utils import ExtractorError
+
+
+class BBCCoUkIE(SubtitlesInfoExtractor):
+    IE_NAME = 'bbc.co.uk'
+    IE_DESC = 'BBC iPlayer'
+    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:programmes|iplayer/episode)/(?P<id>[\da-z]{8})'
+
+    _TESTS = [
+        {
+            'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
+            'info_dict': {
+                'id': 'b039d07m',
+                'ext': 'flv',
+                'title': 'Kaleidoscope: Leonard Cohen',
+                'description': 'md5:db4755d7a665ae72343779f7dacb402c',
+                'duration': 1740,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            }
+        },
+        {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
+            'info_dict': {
+                'id': 'b00yng1d',
+                'ext': 'flv',
+                'title': 'The Man in Black: Series 3: The Printed Name',
+                'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
+                'duration': 1800,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Episode is no longer available on BBC iPlayer Radio',
+        },
+        {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
+            'info_dict': {
+                'id': 'b00yng1d',
+                'ext': 'flv',
+                'title': 'The Voice UK: Series 3: Blind Auditions 5',
+                'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.",
+                'duration': 5100,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
+        }
+    ]
+
+    def _extract_asx_playlist(self, connection, programme_id):
+        asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
+        return [ref.get('href') for ref in asx.findall('./Entry/ref')]
+
+    def _extract_connection(self, connection, programme_id):
+        formats = []
+        protocol = connection.get('protocol')
+        supplier = connection.get('supplier')
+        if protocol == 'http':
+            href = connection.get('href')
+            # ASX playlist
+            if supplier == 'asx':
+                for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
+                    formats.append({
+                        'url': ref,
+                        'format_id': 'ref%s_%s' % (i, supplier),
+                    })
+            # Direct link
+            else:
+                formats.append({
+                    'url': href,
+                    'format_id': supplier,
+                })
+        elif protocol == 'rtmp':
+            application = connection.get('application', 'ondemand')
+            auth_string = connection.get('authString')
+            identifier = connection.get('identifier')
+            server = connection.get('server')
+            formats.append({
+                'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
+                'play_path': identifier,
+                'app': '%s?%s' % (application, auth_string),
+                'page_url': 'http://www.bbc.co.uk',
+                'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
+                'rtmp_live': False,
+                'ext': 'flv',
+                'format_id': supplier,
+            })
+        return formats
+
+    def _extract_items(self, playlist):
+        return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
+
+    def _extract_medias(self, media_selection):
+        return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
+
+    def _extract_connections(self, media):
+        return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
+
+    def _extract_video(self, media, programme_id):
+        formats = []
+        vbr = int(media.get('bitrate'))
+        vcodec = media.get('encoding')
+        service = media.get('service')
+        width = int(media.get('width'))
+        height = int(media.get('height'))
+        file_size = int(media.get('media_file_size'))
+        for connection in self._extract_connections(media):
+            conn_formats = self._extract_connection(connection, programme_id)
+            for format in conn_formats:
+                format.update({
+                    'format_id': '%s_%s' % (service, format['format_id']),
+                    'width': width,
+                    'height': height,
+                    'vbr': vbr,
+                    'vcodec': vcodec,
+                    'filesize': file_size,
+                })
+            formats.extend(conn_formats)
+        return formats
+
+    def _extract_audio(self, media, programme_id):
+        formats = []
+        abr = int(media.get('bitrate'))
+        acodec = media.get('encoding')
+        service = media.get('service')
+        for connection in self._extract_connections(media):
+            conn_formats = self._extract_connection(connection, programme_id)
+            for format in conn_formats:
+                format.update({
+                    'format_id': '%s_%s' % (service, format['format_id']),
+                    'abr': abr,
+                    'acodec': acodec,
+                })
+            formats.extend(conn_formats)
+        return formats
+
+    def _extract_captions(self, media, programme_id):
+        subtitles = {}
+        for connection in self._extract_connections(media):
+            captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
+            lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
+            ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
+            srt = ''
+            for pos, p in enumerate(ps):
+                srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'),
+                                                          p.text.strip() if p.text is not None else '')
+            subtitles[lang] = srt
+        return subtitles
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        group_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, group_id, 'Downloading video page')
+        if re.search(r'id="emp-error" class="notinuk">', webpage):
+            raise ExtractorError('Currently BBC iPlayer TV programmes are available to play in the UK only',
+                expected=True)
+
+        playlist = self._download_xml('http://www.bbc.co.uk/iplayer/playlist/%s' % group_id, group_id,
+            'Downloading playlist XML')
+
+        no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
+        if no_items is not None:
+            reason = no_items.get('reason')
+            if reason == 'preAvailability':
+                msg = 'Episode %s is not yet available' % group_id
+            elif reason == 'postAvailability':
+                msg = 'Episode %s is no longer available' % group_id
+            else:
+                msg = 'Episode %s is not available: %s' % (group_id, reason)
+            raise ExtractorError(msg, expected=True)
+
+        formats = []
+        subtitles = None
+
+        for item in self._extract_items(playlist):
+            kind = item.get('kind')
+            if kind != 'programme' and kind != 'radioProgramme':
+                continue
+            title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
+            description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
+
+            programme_id = item.get('identifier')
+            duration = int(item.get('duration'))
+
+            media_selection = self._download_xml(
+                'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s'  % programme_id,
+                programme_id, 'Downloading media selection XML')
+
+            for media in self._extract_medias(media_selection):
+                kind = media.get('kind')
+                if kind == 'audio':
+                    formats.extend(self._extract_audio(media, programme_id))
+                elif kind == 'video':
+                    formats.extend(self._extract_video(media, programme_id))
+                elif kind == 'captions':
+                    subtitles = self._extract_captions(media, programme_id)
+
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(programme_id, subtitles)
+            return
+
+        self._sort_formats(formats)
+
+        return {
+            'id': programme_id,
+            'title': title,
+            'description': description,
+            'duration': duration,
+            'formats': formats,
+            'subtitles': subtitles,
+        }

youtube_dl/extractor/bilibili.py

@@ -0,0 +1,106 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_parse_qs,
+    ExtractorError,
+    int_or_none,
+    unified_strdate,
+)
+
+
+class BiliBiliIE(InfoExtractor):
+    _VALID_URL = r'http://www\.bilibili\.tv/video/av(?P<id>[0-9]+)/'
+
+    _TEST = {
+        'url': 'http://www.bilibili.tv/video/av1074402/',
+        'md5': '2c301e4dab317596e837c3e7633e7d86',
+        'info_dict': {
+            'id': '1074402',
+            'ext': 'flv',
+            'title': '【金坷垃】金泡沫',
+            'duration': 308,
+            'upload_date': '20140420',
+            'thumbnail': 're:^https?://.+\.jpg',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        video_code = self._search_regex(
+            r'(?s)<div itemprop="video".*?>(.*?)</div>', webpage, 'video code')
+
+        title = self._html_search_meta(
+            'media:title', video_code, 'title', fatal=True)
+        duration_str = self._html_search_meta(
+            'duration', video_code, 'duration')
+        if duration_str is None:
+            duration = None
+        else:
+            duration_mobj = re.match(
+                r'^T(?:(?P<hours>[0-9]+)H)?(?P<minutes>[0-9]+)M(?P<seconds>[0-9]+)S$',
+                duration_str)
+            duration = (
+                int_or_none(duration_mobj.group('hours'), default=0) * 3600 +
+                int(duration_mobj.group('minutes')) * 60 +
+                int(duration_mobj.group('seconds')))
+        upload_date = unified_strdate(self._html_search_meta(
+            'uploadDate', video_code, fatal=False))
+        thumbnail = self._html_search_meta(
+            'thumbnailUrl', video_code, 'thumbnail', fatal=False)
+
+        player_params = compat_parse_qs(self._html_search_regex(
+            r'<iframe .*?class="player" src="https://secure.bilibili.tv/secure,([^"]+)"',
+            webpage, 'player params'))
+
+        if 'cid' in player_params:
+            cid = player_params['cid'][0]
+
+            lq_doc = self._download_xml(
+                'http://interface.bilibili.cn/v_cdn_play?cid=%s' % cid,
+                video_id,
+                note='Downloading LQ video info'
+            )
+            lq_durl = lq_doc.find('.//durl')
+            formats = [{
+                'format_id': 'lq',
+                'quality': 1,
+                'url': lq_durl.find('./url').text,
+                'filesize': int_or_none(
+                    lq_durl.find('./size'), get_attr='text'),
+            }]
+
+            hq_doc = self._download_xml(
+                'http://interface.bilibili.cn/playurl?cid=%s' % cid,
+                video_id,
+                note='Downloading HQ video info',
+                fatal=False,
+            )
+            if hq_doc is not False:
+                hq_durl = hq_doc.find('.//durl')
+                formats.append({
+                    'format_id': 'hq',
+                    'quality': 2,
+                    'ext': 'flv',
+                    'url': hq_durl.find('./url').text,
+                    'filesize': int_or_none(
+                        hq_durl.find('./size'), get_attr='text'),
+                })
+        else:
+            raise ExtractorError('Unsupported player parameters: %r' % (player_params,))
+
+        self._sort_formats(formats)
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'duration': duration,
+            'upload_date': upload_date,
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/blinkx.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import datetime
 import json
 import re
@@ -10,19 +12,19 @@ from ..utils import (
 
 class BlinkxIE(InfoExtractor):
     _VALID_URL = r'^(?:https?://(?:www\.)blinkx\.com/#?ce/|blinkx:)(?P<id>[^?]+)'
-    _IE_NAME = u'blinkx'
+    IE_NAME = 'blinkx'
 
     _TEST = {
-        u'url': u'http://www.blinkx.com/ce/8aQUy7GVFYgFzpKhT0oqsilwOGFRVXk3R1ZGWWdGenBLaFQwb3FzaWx3OGFRVXk3R1ZGWWdGenB',
-        u'file': u'8aQUy7GV.mp4',
-        u'md5': u'2e9a07364af40163a908edbf10bb2492',
-        u'info_dict': {
-            u"title": u"Police Car Rolls Away",
-            u"uploader": u"stupidvideos.com",
-            u"upload_date": u"20131215",
-            u"description": u"A police car gently rolls away from a fight. Maybe it felt weird being around a confrontation and just had to get out of there!",
-            u"duration": 14.886,
-            u"thumbnails": [{
+        'url': 'http://www.blinkx.com/ce/8aQUy7GVFYgFzpKhT0oqsilwOGFRVXk3R1ZGWWdGenBLaFQwb3FzaWx3OGFRVXk3R1ZGWWdGenB',
+        'file': '8aQUy7GV.mp4',
+        'md5': '2e9a07364af40163a908edbf10bb2492',
+        'info_dict': {
+            "title": "Police Car Rolls Away",
+            "uploader": "stupidvideos.com",
+            "upload_date": "20131215",
+            "description": "A police car gently rolls away from a fight. Maybe it felt weird being around a confrontation and just had to get out of there!",
+            "duration": 14.886,
+            "thumbnails": [{
                 "width": 100,
                 "height": 76,
                 "url": "http://cdn.blinkx.com/stream/b/41/StupidVideos/20131215/1873969261/1873969261_tn_0.jpg",
@@ -30,17 +32,17 @@ class BlinkxIE(InfoExtractor):
         },
     }
 
-    def _real_extract(self, url):
-        m = re.match(self._VALID_URL, url)
+    def _real_extract(self, rl):
+        m = re.match(self._VALID_URL, rl)
         video_id = m.group('id')
         display_id = video_id[:8]
 
         api_url = (u'https://apib4.blinkx.com/api.php?action=play_video&' +
-                   u'video=%s' % video_id)
+                   'video=%s' % video_id)
         data_json = self._download_webpage(api_url, display_id)
         data = json.loads(data_json)['api']['results'][0]
         dt = datetime.datetime.fromtimestamp(data['pubdate_epoch'])
-        upload_date = dt.strftime('%Y%m%d')
+        pload_date = dt.strftime('%Y%m%d')
 
         duration = None
         thumbnails = []
@@ -61,9 +63,10 @@ class BlinkxIE(InfoExtractor):
             elif m['type'] in ('flv', 'mp4'):
                 vcodec = remove_start(m['vcodec'], 'ff')
                 acodec = remove_start(m['acodec'], 'ff')
+                tbr = (int(m['vbr']) + int(m['abr'])) // 1000
                 format_id = (u'%s-%sk-%s' %
                              (vcodec,
-                              (int(m['vbr']) + int(m['abr'])) // 1000,
+                              tbr,
                               m['w']))
                 formats.append({
                     'format_id': format_id,
@@ -72,10 +75,12 @@ class BlinkxIE(InfoExtractor):
                     'acodec': acodec,
                     'abr': int(m['abr']) // 1000,
                     'vbr': int(m['vbr']) // 1000,
+                    'tbr': tbr,
                     'width': int(m['w']),
                     'height': int(m['h']),
                 })
-        formats.sort(key=lambda f: (f['width'], f['vbr'], f['abr']))
+
+        self._sort_formats(formats)
 
         return {
             'id': display_id,
@@ -83,7 +88,7 @@ class BlinkxIE(InfoExtractor):
             'title': data['title'],
             'formats': formats,
             'uploader': data['channel_name'],
-            'upload_date': upload_date,
+            'upload_date': pload_date,
             'description': data.get('description'),
             'thumbnails': thumbnails,
             'duration': duration,

youtube_dl/extractor/bliptv.py

@@ -1,156 +1,145 @@
+from __future__ import unicode_literals
+
 import datetime
-import json
-import os
 import re
-import socket
 
 from .common import InfoExtractor
+from .subtitles import SubtitlesInfoExtractor
 from ..utils import (
-    compat_http_client,
-    compat_parse_qs,
     compat_str,
-    compat_urllib_error,
-    compat_urllib_parse_urlparse,
     compat_urllib_request,
 
-    ExtractorError,
     unescapeHTML,
 )
 
 
-class BlipTVIE(InfoExtractor):
+class BlipTVIE(SubtitlesInfoExtractor):
     """Information extractor for blip.tv"""
 
-    _VALID_URL = r'^(?:https?://)?(?:\w+\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(.+)$'
-    _URL_EXT = r'^.*\.([a-z0-9]+)$'
-    IE_NAME = u'blip.tv'
-    _TEST = {
-        u'url': u'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
-        u'file': u'5779306.m4v',
-        u'md5': u'80baf1ec5c3d2019037c1c707d676b9f',
-        u'info_dict': {
-            u"upload_date": u"20111205", 
-            u"description": u"md5:9bc31f227219cde65e47eeec8d2dc596", 
-            u"uploader": u"Comic Book Resources - CBR TV", 
-            u"title": u"CBR EXCLUSIVE: \"Gotham City Imposters\" Bats VS Jokerz Short 3"
-        }
-    }
+    _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(?P<presumptive_id>.+)$'
 
-    def report_direct_download(self, title):
-        """Report information extraction."""
-        self.to_screen(u'%s: Direct download detected' % title)
+    _TESTS = [{
+        'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
+        'md5': 'c6934ad0b6acf2bd920720ec888eb812',
+        'info_dict': {
+            'id': '5779306',
+            'ext': 'mov',
+            'upload_date': '20111205',
+            'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
+            'uploader': 'Comic Book Resources - CBR TV',
+            'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
+        }
+    }, {
+        # https://github.com/rg3/youtube-dl/pull/2274
+        'note': 'Video with subtitles',
+        'url': 'http://blip.tv/play/h6Uag5OEVgI.html',
+        'md5': '309f9d25b820b086ca163ffac8031806',
+        'info_dict': {
+            'id': '6586561',
+            'ext': 'mp4',
+            'uploader': 'Red vs. Blue',
+            'description': 'One-Zero-One',
+            'upload_date': '20130614',
+            'title': 'Red vs. Blue Season 11 Episode 1',
+        }
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+        presumptive_id = mobj.group('presumptive_id')
 
         # See https://github.com/rg3/youtube-dl/issues/857
-        api_mobj = re.match(r'http://a\.blip\.tv/api\.swf#(?P<video_id>[\d\w]+)', url)
-        if api_mobj is not None:
-            url = 'http://blip.tv/play/g_%s' % api_mobj.group('video_id')
-        urlp = compat_urllib_parse_urlparse(url)
-        if urlp.path.startswith('/play/'):
-            response = self._request_webpage(url, None, False)
-            redirecturl = response.geturl()
-            rurlp = compat_urllib_parse_urlparse(redirecturl)
-            file_id = compat_parse_qs(rurlp.fragment)['file'][0].rpartition('/')[2]
-            url = 'http://blip.tv/a/a-' + file_id
-            return self._real_extract(url)
-
-
-        if '?' in url:
-            cchar = '&'
-        else:
-            cchar = '?'
+        embed_mobj = re.match(r'https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', url)
+        if embed_mobj:
+            info_url = 'http://blip.tv/play/%s.x?p=1' % embed_mobj.group(1)
+            info_page = self._download_webpage(info_url, embed_mobj.group(1))
+            video_id = self._search_regex(
+                r'data-episode-id="([0-9]+)', info_page, 'video_id')
+            return self.url_result('http://blip.tv/a/a-' + video_id, 'BlipTV')
+        
+        cchar = '&' if '?' in url else '?'
         json_url = url + cchar + 'skin=json&version=2&no_wrap=1'
         request = compat_urllib_request.Request(json_url)
         request.add_header('User-Agent', 'iTunes/10.6.1')
-        self.report_extraction(mobj.group(1))
-        info = None
-        urlh = self._request_webpage(request, None, False,
-            u'unable to download video info webpage')
 
-        if urlh.headers.get('Content-Type', '').startswith('video/'): # Direct download
-            basename = url.split('/')[-1]
-            title,ext = os.path.splitext(basename)
-            title = title.decode('UTF-8')
-            ext = ext.replace('.', '')
-            self.report_direct_download(title)
-            return {
-                'id': title,
-                'url': url,
-                'uploader': None,
-                'upload_date': None,
-                'title': title,
-                'ext': ext,
-                'urlhandle': urlh
-            }
+        json_data = self._download_json(request, video_id=presumptive_id)
 
-        try:
-            json_code_bytes = urlh.read()
-            json_code = json_code_bytes.decode('utf-8')
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            raise ExtractorError(u'Unable to read video info webpage: %s' % compat_str(err))
+        if 'Post' in json_data:
+            data = json_data['Post']
+        else:
+            data = json_data
 
-        try:
-            json_data = json.loads(json_code)
-            if 'Post' in json_data:
-                data = json_data['Post']
-            else:
-                data = json_data
+        video_id = compat_str(data['item_id'])
+        upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
+        subtitles = {}
+        formats = []
+        if 'additionalMedia' in data:
+            for f in data['additionalMedia']:
+                if f.get('file_type_srt') == 1:
+                    LANGS = {
+                        'english': 'en',
+                    }
+                    lang = f['role'].rpartition('-')[-1].strip().lower()
+                    langcode = LANGS.get(lang, lang)
+                    subtitles[langcode] = f['url']
+                    continue
+                if not int(f['media_width']):  # filter m3u8
+                    continue
+                formats.append({
+                    'url': f['url'],
+                    'format_id': f['role'],
+                    'width': int(f['media_width']),
+                    'height': int(f['media_height']),
+                })
+        else:
+            formats.append({
+                'url': data['media']['url'],
+                'width': int(data['media']['width']),
+                'height': int(data['media']['height']),
+            })
+        self._sort_formats(formats)
 
-            upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
-            if 'additionalMedia' in data:
-                formats = sorted(data['additionalMedia'], key=lambda f: int(f['media_height']))
-                best_format = formats[-1]
-                video_url = best_format['url']
-            else:
-                video_url = data['media']['url']
-            umobj = re.match(self._URL_EXT, video_url)
-            if umobj is None:
-                raise ValueError('Can not determine filename extension')
-            ext = umobj.group(1)
+        # subtitles
+        video_subtitles = self.extract_subtitles(video_id, subtitles)
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(video_id, subtitles)
+            return
 
-            return {
-                'id': compat_str(data['item_id']),
-                'url': video_url,
-                'uploader': data['display_name'],
-                'upload_date': upload_date,
-                'title': data['title'],
-                'ext': ext,
-                'format': data['media']['mimeType'],
-                'thumbnail': data['thumbnailUrl'],
-                'description': data['description'],
-                'player_url': data['embedUrl'],
-                'user_agent': 'iTunes/10.6.1',
-            }
-        except (ValueError, KeyError) as err:
-            raise ExtractorError(u'Unable to parse video information: %s' % repr(err))
+        return {
+            'id': video_id,
+            'uploader': data['display_name'],
+            'upload_date': upload_date,
+            'title': data['title'],
+            'thumbnail': data['thumbnailUrl'],
+            'description': data['description'],
+            'user_agent': 'iTunes/10.6.1',
+            'formats': formats,
+            'subtitles': video_subtitles,
+        }
+
+    def _download_subtitle_url(self, sub_lang, url):
+        # For some weird reason, blip.tv serves a video instead of subtitles
+        # when we request with a common UA
+        req = compat_urllib_request.Request(url)
+        req.add_header('Youtubedl-user-agent', 'youtube-dl')
+        return self._download_webpage(req, None, note=False)
 
 
 class BlipTVUserIE(InfoExtractor):
-    """Information Extractor for blip.tv users."""
-
     _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
     _PAGE_SIZE = 12
-    IE_NAME = u'blip.tv:user'
+    IE_NAME = 'blip.tv:user'
 
     def _real_extract(self, url):
-        # Extract username
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
-
         username = mobj.group(1)
 
         page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
 
-        page = self._download_webpage(url, username, u'Downloading user page')
+        page = self._download_webpage(url, username, 'Downloading user page')
         mobj = re.search(r'data-users-id="([^"]+)"', page)
         page_base = page_base % mobj.group(1)
 
-
         # Download video ids using BlipTV Ajax calls. Result size per
         # query is limited (currently to 12 videos) so we need to query
         # page by page until there are no video ids - it means we got
@@ -161,8 +150,8 @@ class BlipTVUserIE(InfoExtractor):
 
         while True:
             url = page_base + "&page=" + str(pagenum)
-            page = self._download_webpage(url, username,
-                                          u'Downloading video ids from page %d' % pagenum)
+            page = self._download_webpage(
+                url, username, 'Downloading video ids from page %d' % pagenum)
 
             # Extract video identifiers
             ids_in_page = []
@@ -184,6 +173,6 @@ class BlipTVUserIE(InfoExtractor):
 
             pagenum += 1
 
-        urls = [u'http://blip.tv/%s' % video_id for video_id in video_ids]
+        urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
-        return [self.playlist_result(url_entries, playlist_title = username)]
+        return [self.playlist_result(url_entries, playlist_title=username)]

youtube_dl/extractor/bloomberg.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -7,15 +9,13 @@ class BloombergIE(InfoExtractor):
     _VALID_URL = r'https?://www\.bloomberg\.com/video/(?P<name>.+?)\.html'
 
     _TEST = {
-        u'url': u'http://www.bloomberg.com/video/shah-s-presentation-on-foreign-exchange-strategies-qurhIVlJSB6hzkVi229d8g.html',
-        u'file': u'12bzhqZTqQHmmlA8I-i0NpzJgcG5NNYX.mp4',
-        u'info_dict': {
-            u'title': u'Shah\'s Presentation on Foreign-Exchange Strategies',
-            u'description': u'md5:abc86e5236f9f0e4866c59ad36736686',
-        },
-        u'params': {
-            # Requires ffmpeg (m3u8 manifest)
-            u'skip_download': True,
+        'url': 'http://www.bloomberg.com/video/shah-s-presentation-on-foreign-exchange-strategies-qurhIVlJSB6hzkVi229d8g.html',
+        'md5': '7bf08858ff7c203c870e8a6190e221e5',
+        'info_dict': {
+            'id': 'qurhIVlJSB6hzkVi229d8g',
+            'ext': 'flv',
+            'title': 'Shah\'s Presentation on Foreign-Exchange Strategies',
+            'description': 'md5:0681e0d30dcdfc6abf34594961d8ea88',
         },
     }
 
@@ -23,5 +23,16 @@ class BloombergIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         name = mobj.group('name')
         webpage = self._download_webpage(url, name)
-        ooyala_url = self._og_search_video_url(webpage)
-        return self.url_result(ooyala_url, ie='Ooyala')
+        f4m_url = self._search_regex(
+            r'<source src="(https?://[^"]+\.f4m.*?)"', webpage,
+            'f4m url')
+        title = re.sub(': Video$', '', self._og_search_title(webpage))
+
+        return {
+            'id': name.split('-')[-1],
+            'title': title,
+            'url': f4m_url,
+            'ext': 'flv',
+            'description': self._og_search_description(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+        }

youtube_dl/extractor/br.py

@@ -0,0 +1,139 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+)
+
+
+class BRIE(InfoExtractor):
+    IE_DESC = 'Bayerischer Rundfunk Mediathek'
+    _VALID_URL = r'https?://(?:www\.)?br\.de/(?:[a-z0-9\-]+/)+(?P<id>[a-z0-9\-]+)\.html'
+    _BASE_URL = 'http://www.br.de'
+
+    _TESTS = [
+        {
+            'url': 'http://www.br.de/mediathek/video/anselm-gruen-114.html',
+            'md5': 'c4f83cf0f023ba5875aba0bf46860df2',
+            'info_dict': {
+                'id': '2c8d81c5-6fb7-4a74-88d4-e768e5856532',
+                'ext': 'mp4',
+                'title': 'Feiern und Verzichten',
+                'description': 'Anselm Grün: Feiern und Verzichten',
+                'uploader': 'BR/Birgit Baier',
+                'upload_date': '20140301',
+            }
+        },
+        {
+            'url': 'http://www.br.de/mediathek/video/sendungen/unter-unserem-himmel/unter-unserem-himmel-alpen-ueber-den-pass-100.html',
+            'md5': 'ab451b09d861dbed7d7cc9ab0be19ebe',
+            'info_dict': {
+                'id': '2c060e69-3a27-4e13-b0f0-668fac17d812',
+                'ext': 'mp4',
+                'title': 'Über den Pass',
+                'description': 'Die Eroberung der Alpen: Über den Pass',
+            }
+        },
+        {
+            'url': 'http://www.br.de/nachrichten/schaeuble-haushaltsentwurf-bundestag-100.html',
+            'md5': '3db0df1a9a9cd9fa0c70e6ea8aa8e820',
+            'info_dict': {
+                'id': 'c6aae3de-2cf9-43f2-957f-f17fef9afaab',
+                'ext': 'aac',
+                'title': '"Keine neuen Schulden im nächsten Jahr"',
+                'description': 'Haushaltsentwurf: "Keine neuen Schulden im nächsten Jahr"',
+            }
+        },
+        {
+            'url': 'http://www.br.de/radio/bayern1/service/team/videos/team-video-erdelt100.html',
+            'md5': 'dbab0aef2e047060ea7a21fc1ce1078a',
+            'info_dict': {
+                'id': '6ba73750-d405-45d3-861d-1ce8c524e059',
+                'ext': 'mp4',
+                'title': 'Umweltbewusster Häuslebauer',
+                'description': 'Uwe Erdelt: Umweltbewusster Häuslebauer',
+            }
+        },
+        {
+            'url': 'http://www.br.de/fernsehen/br-alpha/sendungen/kant-fuer-anfaenger/kritik-der-reinen-vernunft/kant-kritik-01-metaphysik100.html',
+            'md5': '23bca295f1650d698f94fc570977dae3',
+            'info_dict': {
+                'id': 'd982c9ce-8648-4753-b358-98abb8aec43d',
+                'ext': 'mp4',
+                'title': 'Folge 1 - Metaphysik',
+                'description': 'Kant für Anfänger: Folge 1 - Metaphysik',
+                'uploader': 'Eva Maria Steimle',
+                'upload_date': '20140117',
+            }
+        },
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('id')
+        page = self._download_webpage(url, display_id)
+        xml_url = self._search_regex(
+            r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/(?:[a-z0-9\-]+/)+[a-z0-9/~_.-]+)'}\)\);", page, 'XMLURL')
+        xml = self._download_xml(self._BASE_URL + xml_url, None)
+
+        medias = []
+
+        for xml_media in xml.findall('video') + xml.findall('audio'):
+            media = {
+                'id': xml_media.get('externalId'),
+                'title': xml_media.find('title').text,
+                'formats': self._extract_formats(xml_media.find('assets')),
+                'thumbnails': self._extract_thumbnails(xml_media.find('teaserImage/variants')),
+                'description': ' '.join(xml_media.find('shareTitle').text.splitlines()),
+                'webpage_url': xml_media.find('permalink').text
+            }
+            if xml_media.find('author').text:
+                media['uploader'] = xml_media.find('author').text
+            if xml_media.find('broadcastDate').text:
+                media['upload_date'] = ''.join(reversed(xml_media.find('broadcastDate').text.split('.')))
+            medias.append(media)
+
+        if len(medias) > 1:
+            self._downloader.report_warning(
+                'found multiple medias; please '
+                'report this with the video URL to http://yt-dl.org/bug')
+        if not medias:
+            raise ExtractorError('No media entries found')
+        return medias[0]
+
+    def _extract_formats(self, assets):
+
+        def text_or_none(asset, tag):
+            elem = asset.find(tag)
+            return None if elem is None else elem.text
+
+        formats = [{
+            'url': text_or_none(asset, 'downloadUrl'),
+            'ext': text_or_none(asset, 'mediaType'),
+            'format_id': asset.get('type'),
+            'width': int_or_none(text_or_none(asset, 'frameWidth')),
+            'height': int_or_none(text_or_none(asset, 'frameHeight')),
+            'tbr': int_or_none(text_or_none(asset, 'bitrateVideo')),
+            'abr': int_or_none(text_or_none(asset, 'bitrateAudio')),
+            'vcodec': text_or_none(asset, 'codecVideo'),
+            'acodec': text_or_none(asset, 'codecAudio'),
+            'container': text_or_none(asset, 'mediaType'),
+            'filesize': int_or_none(text_or_none(asset, 'size')),
+        } for asset in assets.findall('asset')
+            if asset.find('downloadUrl') is not None]
+
+        self._sort_formats(formats)
+        return formats
+
+    def _extract_thumbnails(self, variants):
+        thumbnails = [{
+            'url': self._BASE_URL + variant.find('url').text,
+            'width': int_or_none(variant.find('width').text),
+            'height': int_or_none(variant.find('height').text),
+        } for variant in variants.findall('variant')]
+        thumbnails.sort(key=lambda x: x['width'] * x['height'], reverse=True)
+        return thumbnails

youtube_dl/extractor/breakcom.py

@@ -1,18 +1,20 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
 from .common import InfoExtractor
-from ..utils import determine_ext
 
 
 class BreakIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?break\.com/video/([^/]+)'
+    _VALID_URL = r'http://(?:www\.)?break\.com/video/([^/]+)'
     _TEST = {
-        u'url': u'http://www.break.com/video/when-girls-act-like-guys-2468056',
-        u'file': u'2468056.mp4',
-        u'md5': u'a3513fb1547fba4fb6cfac1bffc6c46b',
-        u'info_dict': {
-            u"title": u"When Girls Act Like D-Bags"
+        'url': 'http://www.break.com/video/when-girls-act-like-guys-2468056',
+        'md5': 'a3513fb1547fba4fb6cfac1bffc6c46b',
+        'info_dict': {
+            'id': '2468056',
+            'ext': 'mp4',
+            'title': 'When Girls Act Like D-Bags',
         }
     }
 
@@ -21,18 +23,18 @@ class BreakIE(InfoExtractor):
         video_id = mobj.group(1).split("-")[-1]
         embed_url = 'http://www.break.com/embed/%s' % video_id
         webpage = self._download_webpage(embed_url, video_id)
-        info_json = self._search_regex(r'var embedVars = ({.*?});', webpage,
-                                       u'info json', flags=re.DOTALL)
+        info_json = self._search_regex(r'var embedVars = ({.*})\s*?</script>',
+            webpage, 'info json', flags=re.DOTALL)
         info = json.loads(info_json)
         video_url = info['videoUri']
-        m_youtube = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', video_url)
-        if m_youtube is not None:
-            return self.url_result(m_youtube.group(1), 'Youtube')
+        youtube_id = info.get('youtubeId')
+        if youtube_id:
+            return self.url_result(youtube_id, 'Youtube')
+
         final_url = video_url + '?' + info['AuthToken']
-        return [{
-            'id':        video_id,
-            'url':       final_url,
-            'ext':       determine_ext(final_url),
-            'title':     info['contentName'],
+        return {
+            'id': video_id,
+            'url': final_url,
+            'title': info['contentName'],
             'thumbnail': info['thumbUri'],
-        }]
+        }

youtube_dl/extractor/brightcove.py

@@ -1,4 +1,5 @@
 # encoding: utf-8
+from __future__ import unicode_literals
 
 import re
 import json
@@ -8,65 +9,68 @@ from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
     find_xpath_attr,
+    fix_xml_ampersands,
     compat_urlparse,
     compat_str,
     compat_urllib_request,
+    compat_parse_qs,
 
     ExtractorError,
+    unsmuggle_url,
+    unescapeHTML,
 )
 
 
 class BrightcoveIE(InfoExtractor):
     _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*\?(?P<query>.*)'
     _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'
-    _PLAYLIST_URL_TEMPLATE = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s'
 
     _TESTS = [
         {
             # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/
-            u'url': u'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001',
-            u'file': u'2371591881001.mp4',
-            u'md5': u'5423e113865d26e40624dce2e4b45d95',
-            u'note': u'Test Brightcove downloads and detection in GenericIE',
-            u'info_dict': {
-                u'title': u'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
-                u'uploader': u'8TV',
-                u'description': u'md5:a950cc4285c43e44d763d036710cd9cd',
+            'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001',
+            'file': '2371591881001.mp4',
+            'md5': '5423e113865d26e40624dce2e4b45d95',
+            'note': 'Test Brightcove downloads and detection in GenericIE',
+            'info_dict': {
+                'title': 'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
+                'uploader': '8TV',
+                'description': 'md5:a950cc4285c43e44d763d036710cd9cd',
             }
         },
         {
             # From http://medianetwork.oracle.com/video/player/1785452137001
-            u'url': u'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001',
-            u'file': u'1785452137001.flv',
-            u'info_dict': {
-                u'title': u'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges',
-                u'description': u'John Rose speaks at the JVM Language Summit, August 1, 2012.',
-                u'uploader': u'Oracle',
+            'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001',
+            'file': '1785452137001.flv',
+            'info_dict': {
+                'title': 'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges',
+                'description': 'John Rose speaks at the JVM Language Summit, August 1, 2012.',
+                'uploader': 'Oracle',
             },
         },
         {
             # From http://mashable.com/2013/10/26/thermoelectric-bracelet-lets-you-control-your-body-temperature/
-            u'url': u'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001',
-            u'info_dict': {
-                u'id': u'2750934548001',
-                u'ext': u'mp4',
-                u'title': u'This Bracelet Acts as a Personal Thermostat',
-                u'description': u'md5:547b78c64f4112766ccf4e151c20b6a0',
-                u'uploader': u'Mashable',
+            'url': 'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001',
+            'info_dict': {
+                'id': '2750934548001',
+                'ext': 'mp4',
+                'title': 'This Bracelet Acts as a Personal Thermostat',
+                'description': 'md5:547b78c64f4112766ccf4e151c20b6a0',
+                'uploader': 'Mashable',
             },
         },
         {
             # test that the default referer works
             # from http://national.ballet.ca/interact/video/Lost_in_Motion_II/
-            u'url': u'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001',
-            u'info_dict': {
-                u'id': u'2878862109001',
-                u'ext': u'mp4',
-                u'title': u'Lost in Motion II',
-                u'description': u'md5:363109c02998fee92ec02211bd8000df',
-                u'uploader': u'National Ballet of Canada',
+            'url': 'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001',
+            'info_dict': {
+                'id': '2878862109001',
+                'ext': 'mp4',
+                'title': 'Lost in Motion II',
+                'description': 'md5:363109c02998fee92ec02211bd8000df',
+                'uploader': 'National Ballet of Canada',
             },
-        },
+        }
     ]
 
     @classmethod
@@ -80,18 +84,34 @@ class BrightcoveIE(InfoExtractor):
         object_str = re.sub(r'(<param name="[^"]+" value="[^"]+")>',
                             lambda m: m.group(1) + '/>', object_str)
         # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608
-        object_str = object_str.replace(u'<--', u'<!--')
+        object_str = object_str.replace('<--', '<!--')
+        object_str = fix_xml_ampersands(object_str)
+
+        object_doc = xml.etree.ElementTree.fromstring(object_str.encode('utf-8'))
+
+        fv_el = find_xpath_attr(object_doc, './param', 'name', 'flashVars')
+        if fv_el is not None:
+            flashvars = dict(
+                (k, v[0])
+                for k, v in compat_parse_qs(fv_el.attrib['value']).items())
+        else:
+            flashvars = {}
 
-        object_doc = xml.etree.ElementTree.fromstring(object_str)
-        assert u'BrightcoveExperience' in object_doc.attrib['class']
-        params = {'flashID': object_doc.attrib['id'],
-                  'playerID': find_xpath_attr(object_doc, './param', 'name', 'playerID').attrib['value'],
-                  }
         def find_param(name):
+            if name in flashvars:
+                return flashvars[name]
             node = find_xpath_attr(object_doc, './param', 'name', name)
             if node is not None:
                 return node.attrib['value']
             return None
+
+        params = {}
+
+        playerID = find_param('playerID')
+        if playerID is None:
+            raise ExtractorError('Cannot find player ID')
+        params['playerID'] = playerID
+
         playerKey = find_param('playerKey')
         # Not all pages define this value
         if playerKey is not None:
@@ -108,18 +128,36 @@ class BrightcoveIE(InfoExtractor):
 
     @classmethod
     def _extract_brightcove_url(cls, webpage):
-        """Try to extract the brightcove url from the wepbage, returns None
+        """Try to extract the brightcove url from the webpage, returns None
         if it can't be found
         """
-        m_brightcove = re.search(
-            r'<object[^>]+?class=([\'"])[^>]*?BrightcoveExperience.*?\1.+?</object>',
-            webpage, re.DOTALL)
-        if m_brightcove is not None:
-            return cls._build_brighcove_url(m_brightcove.group())
-        else:
-            return None
+        urls = cls._extract_brightcove_urls(webpage)
+        return urls[0] if urls else None
+
+    @classmethod
+    def _extract_brightcove_urls(cls, webpage):
+        """Return a list of all Brightcove URLs from the webpage """
+
+        url_m = re.search(r'<meta\s+property="og:video"\s+content="(http://c.brightcove.com/[^"]+)"', webpage)
+        if url_m:
+            url = unescapeHTML(url_m.group(1))
+            # Some sites don't add it, we can't download with this url, for example:
+            # http://www.ktvu.com/videos/news/raw-video-caltrain-releases-video-of-man-almost/vCTZdY/
+            if 'playerKey' in url:
+                return [url]
+
+        matches = re.findall(
+            r'''(?sx)<object
+            (?:
+                [^>]+?class=[\'"][^>]*?BrightcoveExperience.*?[\'"] |
+                [^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/
+            ).+?</object>''',
+            webpage)
+        return [cls._build_brighcove_url(m) for m in matches]
 
     def _real_extract(self, url):
+        url, smuggled_data = unsmuggle_url(url, {})
+
         # Change the 'videoId' and others field to '@videoPlayer'
         url = re.sub(r'(?<=[?&])(videoI(d|D)|bctid)', '%40videoPlayer', url)
         # Change bckey (used by bcove.me urls) to playerKey
@@ -130,9 +168,10 @@ class BrightcoveIE(InfoExtractor):
 
         videoPlayer = query.get('@videoPlayer')
         if videoPlayer:
-            return self._get_video_info(videoPlayer[0], query_str, query,
-                # We set the original url as the default 'Referer' header
-                referer=url)
+            # We set the original url as the default 'Referer' header
+            referer = smuggled_data.get('Referer', url)
+            return self._get_video_info(
+                videoPlayer[0], query_str, query, referer=referer)
         else:
             player_key = query['playerKey']
             return self._get_playlist_info(player_key[0])
@@ -151,16 +190,18 @@ class BrightcoveIE(InfoExtractor):
         info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json')
         info = json.loads(info)['data']
         video_info = info['programmedContent']['videoPlayer']['mediaDTO']
+        video_info['_youtubedl_adServerURL'] = info.get('adServerURL')
 
         return self._extract_video_info(video_info)
 
     def _get_playlist_info(self, player_key):
-        playlist_info = self._download_webpage(self._PLAYLIST_URL_TEMPLATE % player_key,
-                                               player_key, u'Downloading playlist information')
+        info_url = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s' % player_key
+        playlist_info = self._download_webpage(
+            info_url, player_key, 'Downloading playlist information')
 
         json_data = json.loads(playlist_info)
         if 'videoList' not in json_data:
-            raise ExtractorError(u'Empty playlist')
+            raise ExtractorError('Empty playlist')
         playlist_info = json_data['videoList']
         videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
 
@@ -170,7 +211,7 @@ class BrightcoveIE(InfoExtractor):
     def _extract_video_info(self, video_info):
         info = {
             'id': compat_str(video_info['id']),
-            'title': video_info['displayName'],
+            'title': video_info['displayName'].strip(),
             'description': video_info.get('shortDescription'),
             'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
             'uploader': video_info.get('publisherName'),
@@ -188,6 +229,23 @@ class BrightcoveIE(InfoExtractor):
             info.update({
                 'url': video_info['FLVFullLengthURL'],
             })
-        else:
-            raise ExtractorError(u'Unable to extract video url for %s' % info['id'])
+
+        if self._downloader.params.get('include_ads', False):
+            adServerURL = video_info.get('_youtubedl_adServerURL')
+            if adServerURL:
+                ad_info = {
+                    '_type': 'url',
+                    'url': adServerURL,
+                }
+                if 'url' in info:
+                    return {
+                        '_type': 'playlist',
+                        'title': info['title'],
+                        'entries': [ad_info, info],
+                    }
+                else:
+                    return ad_info
+
+        if 'url' not in info and not info.get('formats'):
+            raise ExtractorError('Unable to extract video url for %s' % info['id'])
         return info

youtube_dl/extractor/byutv.py

@@ -0,0 +1,48 @@
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import ExtractorError
+
+
+class BYUtvIE(InfoExtractor):
+    _VALID_URL = r'^https?://(?:www\.)?byutv.org/watch/[0-9a-f-]+/(?P<video_id>[^/?#]+)'
+    _TEST = {
+        'url': 'http://www.byutv.org/watch/44e80f7b-e3ba-43ba-8c51-b1fd96c94a79/granite-flats-talking',
+        'info_dict': {
+            'id': 'granite-flats-talking',
+            'ext': 'mp4',
+            'description': 'md5:4e9a7ce60f209a33eca0ac65b4918e1c',
+            'title': 'Talking',
+            'thumbnail': 're:^https?://.*promo.*'
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('video_id')
+
+        webpage = self._download_webpage(url, video_id)
+        episode_code = self._search_regex(
+            r'(?s)episode:(.*?\}),\s*\n', webpage, 'episode information')
+        episode_json = re.sub(
+            r'(\n\s+)([a-zA-Z]+):\s+\'(.*?)\'', r'\1"\2": "\3"', episode_code)
+        ep = json.loads(episode_json)
+
+        if ep['providerType'] == 'Ooyala':
+            return {
+                '_type': 'url_transparent',
+                'ie_key': 'Ooyala',
+                'url': 'ooyala:%s' % ep['providerId'],
+                'id': video_id,
+                'title': ep['title'],
+                'description': ep.get('description'),
+                'thumbnail': ep.get('imageThumbnail'),
+            }
+        else:
+            raise ExtractorError('Unsupported provider %s' % ep['provider'])

youtube_dl/extractor/c56.py

@@ -1,36 +1,47 @@
 # coding: utf-8
+from __future__ import unicode_literals
 
 import re
-import json
 
 from .common import InfoExtractor
-from ..utils import determine_ext
+
 
 class C56IE(InfoExtractor):
-    _VALID_URL = r'https?://((www|player)\.)?56\.com/(.+?/)?(v_|(play_album.+-))(?P<textid>.+?)\.(html|swf)'
-    IE_NAME = u'56.com'
-
-    _TEST ={
-        u'url': u'http://www.56.com/u39/v_OTM0NDA3MTY.html',
-        u'file': u'93440716.flv',
-        u'md5': u'e59995ac63d0457783ea05f93f12a866',
-        u'info_dict': {
-            u'title': u'网事知多少 第32期：车怒',
+    _VALID_URL = r'https?://(?:(?:www|player)\.)?56\.com/(?:.+?/)?(?:v_|(?:play_album.+-))(?P<textid>.+?)\.(?:html|swf)'
+    IE_NAME = '56.com'
+    _TEST = {
+        'url': 'http://www.56.com/u39/v_OTM0NDA3MTY.html',
+        'md5': 'e59995ac63d0457783ea05f93f12a866',
+        'info_dict': {
+            'id': '93440716',
+            'ext': 'flv',
+            'title': '网事知多少 第32期：车怒',
+            'duration': 283.813,
         },
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
         text_id = mobj.group('textid')
-        info_page = self._download_webpage('http://vxml.56.com/json/%s/' % text_id,
-                                           text_id, u'Downloading video info')
-        info = json.loads(info_page)['info']
-        best_format = sorted(info['rfiles'], key=lambda f: int(f['filesize']))[-1]
-        video_url = best_format['url']
 
-        return {'id': info['vid'],
-                'title': info['Subject'],
-                'url': video_url,
-                'ext': determine_ext(video_url),
-                'thumbnail': info.get('bimg') or info.get('img'),
-                }
+        page = self._download_json(
+            'http://vxml.56.com/json/%s/' % text_id, text_id, 'Downloading video info')
+
+        info = page['info']
+
+        formats = [
+            {
+                'format_id': f['type'],
+                'filesize': int(f['filesize']),
+                'url': f['url']
+            } for f in info['rfiles']
+        ]
+        self._sort_formats(formats)
+
+        return {
+            'id': info['vid'],
+            'title': info['Subject'],
+            'duration': int(info['duration']) / 1000.0,
+            'formats': formats,
+            'thumbnail': info.get('bimg') or info.get('img'),
+        }

youtube_dl/extractor/canal13cl.py

@@ -0,0 +1,48 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class Canal13clIE(InfoExtractor):
+    _VALID_URL = r'^http://(?:www\.)?13\.cl/(?:[^/?#]+/)*(?P<id>[^/?#]+)'
+    _TEST = {
+        'url': 'http://www.13.cl/t13/nacional/el-circulo-de-hierro-de-michelle-bachelet-en-su-regreso-a-la-moneda',
+        'md5': '4cb1fa38adcad8fea88487a078831755',
+        'info_dict': {
+            'id': '1403022125',
+            'display_id': 'el-circulo-de-hierro-de-michelle-bachelet-en-su-regreso-a-la-moneda',
+            'ext': 'mp4',
+            'title': 'El "círculo de hierro" de Michelle Bachelet en su regreso a La Moneda',
+            'description': '(Foto: Agencia Uno) En nueve días más, Michelle Bachelet va a asumir por segunda vez como presidenta de la República. Entre aquellos que la acompañarán hay caras que se repiten y otras que se consolidan en su entorno de colaboradores más cercanos.',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        title = self._html_search_meta(
+            'twitter:title', webpage, 'title', fatal=True)
+        description = self._html_search_meta(
+            'twitter:description', webpage, 'description')
+        url = self._html_search_regex(
+            r'articuloVideo = \"(.*?)\"', webpage, 'url')
+        real_id = self._search_regex(
+            r'[^0-9]([0-9]{7,})[^0-9]', url, 'id', default=display_id)
+        thumbnail = self._html_search_regex(
+            r'articuloImagen = \"(.*?)\"', webpage, 'thumbnail')
+
+        return {
+            'id': real_id,
+            'display_id': display_id,
+            'url': url,
+            'title': title,
+            'description': description,
+            'ext': 'mp4',
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/canalc2.py

@@ -1,4 +1,6 @@
 # coding: utf-8
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -9,11 +11,12 @@ class Canalc2IE(InfoExtractor):
     _VALID_URL = r'http://.*?\.canalc2\.tv/video\.asp\?.*?idVideo=(?P<id>\d+)'
 
     _TEST = {
-        u'url': u'http://www.canalc2.tv/video.asp?idVideo=12163&voir=oui',
-        u'file': u'12163.mp4',
-        u'md5': u'060158428b650f896c542dfbb3d6487f',
-        u'info_dict': {
-            u'title': u'Terrasses du Numérique'
+        'url': 'http://www.canalc2.tv/video.asp?idVideo=12163&voir=oui',
+        'md5': '060158428b650f896c542dfbb3d6487f',
+        'info_dict': {
+            'id': '12163',
+            'ext': 'mp4',
+            'title': 'Terrasses du Numérique'
         }
     }
 
@@ -28,10 +31,11 @@ class Canalc2IE(InfoExtractor):
         video_url = 'http://vod-flash.u-strasbg.fr:8080/' + file_name
 
         title = self._html_search_regex(
-            r'class="evenement8">(.*?)</a>', webpage, u'title')
-        
-        return {'id': video_id,
-                'ext': 'mp4',
-                'url': video_url,
-                'title': title,
-                }
+            r'class="evenement8">(.*?)</a>', webpage, 'title')
+
+        return {
+            'id': video_id,
+            'ext': 'mp4',
+            'url': video_url,
+            'title': title,
+        }

youtube_dl/extractor/canalplus.py

@@ -1,53 +1,72 @@
 # encoding: utf-8
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
-from ..utils import unified_strdate
+from ..utils import (
+    unified_strdate,
+    url_basename,
+)
 
 
 class CanalplusIE(InfoExtractor):
-    _VALID_URL = r'https?://(www\.canalplus\.fr/.*?/(?P<path>.*)|player\.canalplus\.fr/#/(?P<id>\d+))'
+    _VALID_URL = r'https?://(?:www\.canalplus\.fr/.*?/(?P<path>.*)|player\.canalplus\.fr/#/(?P<id>[0-9]+))'
     _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/cplus/%s'
-    IE_NAME = u'canalplus.fr'
+    IE_NAME = 'canalplus.fr'
 
     _TEST = {
-        u'url': u'http://www.canalplus.fr/c-infos-documentaires/pid1830-c-zapping.html?vid=922470',
-        u'file': u'922470.flv',
-        u'info_dict': {
-            u'title': u'Zapping - 26/08/13',
-            u'description': u'Le meilleur de toutes les chaînes, tous les jours.\nEmission du 26 août 2013',
-            u'upload_date': u'20130826',
-        },
-        u'params': {
-            u'skip_download': True,
+        'url': 'http://www.canalplus.fr/c-infos-documentaires/pid1830-c-zapping.html?vid=922470',
+        'md5': '3db39fb48b9685438ecf33a1078023e4',
+        'info_dict': {
+            'id': '922470',
+            'ext': 'flv',
+            'title': 'Zapping - 26/08/13',
+            'description': 'Le meilleur de toutes les chaînes, tous les jours.\nEmission du 26 août 2013',
+            'upload_date': '20130826',
         },
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.groupdict().get('id')
+
+        # Beware, some subclasses do not define an id group
+        display_id = url_basename(mobj.group('path'))
+
         if video_id is None:
-            webpage = self._download_webpage(url, mobj.group('path'))
-            video_id = self._search_regex(r'videoId = "(\d+)";', webpage, u'video id')
+            webpage = self._download_webpage(url, display_id)
+            video_id = self._search_regex(r'<canal:player videoId="(\d+)"', webpage, 'video id')
+
         info_url = self._VIDEO_INFO_TEMPLATE % video_id
-        doc = self._download_xml(info_url,video_id, 
-                                           u'Downloading video info')
+        doc = self._download_xml(info_url, video_id, 'Downloading video XML')
 
-        self.report_extraction(video_id)
         video_info = [video for video in doc if video.find('ID').text == video_id][0]
-        infos = video_info.find('INFOS')
         media = video_info.find('MEDIA')
-        formats = [media.find('VIDEOS/%s' % format)
-            for format in ['BAS_DEBIT', 'HAUT_DEBIT', 'HD']]
-        video_url = [format.text for format in formats if format is not None][-1]
+        infos = video_info.find('INFOS')
 
-        return {'id': video_id,
-                'title': u'%s - %s' % (infos.find('TITRAGE/TITRE').text,
-                                       infos.find('TITRAGE/SOUS_TITRE').text),
-                'url': video_url,
-                'ext': 'flv',
-                'upload_date': unified_strdate(infos.find('PUBLICATION/DATE').text),
-                'thumbnail': media.find('IMAGES/GRAND').text,
-                'description': infos.find('DESCRIPTION').text,
-                'view_count': int(infos.find('NB_VUES').text),
-                }
+        preferences = ['MOBILE', 'BAS_DEBIT', 'HAUT_DEBIT', 'HD', 'HLS', 'HDS']
+
+        formats = [
+            {
+                'url': fmt.text + '?hdcore=2.11.3' if fmt.tag == 'HDS' else fmt.text,
+                'format_id': fmt.tag,
+                'ext': 'mp4' if fmt.tag == 'HLS' else 'flv',
+                'preference': preferences.index(fmt.tag) if fmt.tag in preferences else -1,
+            } for fmt in media.find('VIDEOS') if fmt.text
+        ]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': '%s - %s' % (infos.find('TITRAGE/TITRE').text,
+                                  infos.find('TITRAGE/SOUS_TITRE').text),
+            'upload_date': unified_strdate(infos.find('PUBLICATION/DATE').text),
+            'thumbnail': media.find('IMAGES/GRAND').text,
+            'description': infos.find('DESCRIPTION').text,
+            'view_count': int(infos.find('NB_VUES').text),
+            'like_count': int(infos.find('NB_LIKES').text),
+            'comment_count': int(infos.find('NB_COMMENTS').text),
+            'formats': formats,
+        }

youtube_dl/extractor/cbsnews.py

@@ -0,0 +1,87 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+
+
+class CBSNewsIE(InfoExtractor):
+    IE_DESC = 'CBS News'
+    _VALID_URL = r'http://(?:www\.)?cbsnews\.com/(?:[^/]+/)+(?P<id>[\da-z_-]+)'
+
+    _TESTS = [
+        {
+            'url': 'http://www.cbsnews.com/news/tesla-and-spacex-elon-musks-industrial-empire/',
+            'info_dict': {
+                'id': 'tesla-and-spacex-elon-musks-industrial-empire',
+                'ext': 'flv',
+                'title': 'Tesla and SpaceX: Elon Musk\'s industrial empire',
+                'thumbnail': 'http://beta.img.cbsnews.com/i/2014/03/30/60147937-2f53-4565-ad64-1bdd6eb64679/60-0330-pelley-640x360.jpg',
+                'duration': 791,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
+            'info_dict': {
+                'id': 'fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack',
+                'ext': 'flv',
+                'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
+                'thumbnail': 'http://cbsnews2.cbsistatic.com/hub/i/r/2014/04/04/0c9fbc66-576b-41ca-8069-02d122060dd2/thumbnail/140x90/6dad7a502f88875ceac38202984b6d58/en-0404-werner-replace-640x360.jpg',
+                'duration': 205,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        },
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        video_info = json.loads(self._html_search_regex(
+            r'(?:<ul class="media-list items" id="media-related-items"><li data-video-info|<div id="cbsNewsVideoPlayer" data-video-player-options)=\'({.+?})\'',
+            webpage, 'video JSON info'))
+
+        item = video_info['item'] if 'item' in video_info else video_info
+        title = item.get('articleTitle') or item.get('hed')
+        duration = item.get('duration')
+        thumbnail = item.get('mediaImage') or item.get('thumbnail')
+
+        formats = []
+        for format_id in ['RtmpMobileLow', 'RtmpMobileHigh', 'Hls', 'RtmpDesktop']:
+            uri = item.get('media' + format_id + 'URI')
+            if not uri:
+                continue
+            fmt = {
+                'url': uri,
+                'format_id': format_id,
+            }
+            if uri.startswith('rtmp'):
+                fmt.update({
+                    'app': 'ondemand?auth=cbs',
+                    'play_path': 'mp4:' + uri.split('<break>')[-1],
+                    'player_url': 'http://www.cbsnews.com/[[IMPORT]]/vidtech.cbsinteractive.com/player/3_3_0/CBSI_PLAYER_HD.swf',
+                    'page_url': 'http://www.cbsnews.com',
+                    'ext': 'flv',
+                })
+            elif uri.endswith('.m3u8'):
+                fmt['ext'] = 'mp4'
+            formats.append(fmt)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,
+        }

youtube_dl/extractor/ceskatelevize.py

@@ -0,0 +1,126 @@
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_request,
+    compat_urllib_parse,
+    compat_urllib_parse_urlparse,
+    ExtractorError,
+)
+
+
+class CeskaTelevizeIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(.+/)?(?P<id>[^?#]+)'
+
+    _TESTS = [
+        {
+            'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/213512120230004-spanelska-chripka',
+            'info_dict': {
+                'id': '213512120230004',
+                'ext': 'flv',
+                'title': 'První republika: Španělská chřipka',
+                'duration': 3107.4,
+            },
+            'params': {
+                'skip_download': True,  # requires rtmpdump
+            },
+            'skip': 'Works only from Czech Republic.',
+        },
+        {
+            'url': 'http://www.ceskatelevize.cz/ivysilani/1030584952-tsatsiki-maminka-a-policajt',
+            'info_dict': {
+                'id': '20138143440',
+                'ext': 'flv',
+                'title': 'Tsatsiki, maminka a policajt',
+                'duration': 6754.1,
+            },
+            'params': {
+                'skip_download': True,  # requires rtmpdump
+            },
+            'skip': 'Works only from Czech Republic.',
+        },
+        {
+            'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina',
+            'info_dict': {
+                'id': '14716',
+                'ext': 'flv',
+                'title': 'První republika: Zpěvačka z Dupárny Bobina',
+                'duration': 90,
+            },
+            'params': {
+                'skip_download': True,  # requires rtmpdump
+            },
+        },
+    ]
+
+    def _real_extract(self, url):
+        url = url.replace('/porady/', '/ivysilani/').replace('/video/', '')
+
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
+        if '%s</p>' % NOT_AVAILABLE_STRING in webpage:
+            raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
+
+        typ = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type')
+        episode_id = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id')
+
+        data = {
+            'playlist[0][type]': typ,
+            'playlist[0][id]': episode_id,
+            'requestUrl': compat_urllib_parse_urlparse(url).path,
+            'requestSource': 'iVysilani',
+        }
+
+        req = compat_urllib_request.Request('http://www.ceskatelevize.cz/ivysilani/ajax/get-playlist-url',
+                                            data=compat_urllib_parse.urlencode(data))
+
+        req.add_header('Content-type', 'application/x-www-form-urlencoded')
+        req.add_header('x-addr', '127.0.0.1')
+        req.add_header('X-Requested-With', 'XMLHttpRequest')
+        req.add_header('Referer', url)
+
+        playlistpage = self._download_json(req, video_id)
+
+        req = compat_urllib_request.Request(compat_urllib_parse.unquote(playlistpage['url']))
+        req.add_header('Referer', url)
+
+        playlist = self._download_xml(req, video_id)
+        
+        formats = []
+        for i in playlist.find('smilRoot/body'):
+            if 'AD' not in i.attrib['id']:
+                base_url = i.attrib['base']
+                parsedurl = compat_urllib_parse_urlparse(base_url)
+                duration = i.attrib['duration']
+
+                for video in i.findall('video'):
+                    if video.attrib['label'] != 'AD':
+                        format_id = video.attrib['label']
+                        play_path = video.attrib['src']
+                        vbr = int(video.attrib['system-bitrate'])
+
+                        formats.append({
+                            'format_id': format_id,
+                            'url': base_url,
+                            'vbr': vbr,
+                            'play_path': play_path,
+                            'app': parsedurl.path[1:] + '?' + parsedurl.query,
+                            'rtmp_live': True,
+                            'ext': 'flv',
+                        })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': episode_id,
+            'title': self._html_search_regex(r'<title>(.+?) — iVysílání — Česká televize</title>', webpage, 'title'),
+            'duration': float(duration),
+            'formats': formats,
+        }

youtube_dl/extractor/channel9.py

@@ -1,4 +1,4 @@
-# encoding: utf-8
+from __future__ import unicode_literals
 
 import re
 
@@ -11,38 +11,40 @@ class Channel9IE(InfoExtractor):
 
     The type of provided URL (video or playlist) is determined according to
     meta Search.PageType from web page HTML rather than URL itself, as it is
-    not always possible to do.    
+    not always possible to do.
     '''
-    IE_DESC = u'Channel 9'
-    IE_NAME = u'channel9'
-    _VALID_URL = r'^https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+)/?'
+    IE_DESC = 'Channel 9'
+    IE_NAME = 'channel9'
+    _VALID_URL = r'https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+)/?'
 
     _TESTS = [
         {
-            u'url': u'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002',
-            u'file': u'Events_TechEd_Australia_2013_KOS002.mp4',
-            u'md5': u'bbd75296ba47916b754e73c3a4bbdf10',
-            u'info_dict': {
-                u'title': u'Developer Kick-Off Session: Stuff We Love',
-                u'description': u'md5:c08d72240b7c87fcecafe2692f80e35f',
-                u'duration': 4576,
-                u'thumbnail': u'http://media.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg',
-                u'session_code': u'KOS002',
-                u'session_day': u'Day 1',
-                u'session_room': u'Arena 1A',
-                u'session_speakers': [ u'Ed Blankenship', u'Andrew Coates', u'Brady Gaster', u'Patrick Klug', u'Mads Kristensen' ],
+            'url': 'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002',
+            'md5': 'bbd75296ba47916b754e73c3a4bbdf10',
+            'info_dict': {
+                'id': 'Events/TechEd/Australia/2013/KOS002',
+                'ext': 'mp4',
+                'title': 'Developer Kick-Off Session: Stuff We Love',
+                'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
+                'duration': 4576,
+                'thumbnail': 'http://media.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg',
+                'session_code': 'KOS002',
+                'session_day': 'Day 1',
+                'session_room': 'Arena 1A',
+                'session_speakers': [ 'Ed Blankenship', 'Andrew Coates', 'Brady Gaster', 'Patrick Klug', 'Mads Kristensen' ],
             },
         },
         {
-            u'url': u'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing',
-            u'file': u'posts_Self-service-BI-with-Power-BI-nuclear-testing.mp4',
-            u'md5': u'b43ee4529d111bc37ba7ee4f34813e68',
-            u'info_dict': {
-                u'title': u'Self-service BI with Power BI - nuclear testing',
-                u'description': u'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
-                u'duration': 1540,
-                u'thumbnail': u'http://media.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg',
-                u'authors': [ u'Mike Wilmot' ],
+            'url': 'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing',
+            'md5': 'b43ee4529d111bc37ba7ee4f34813e68',
+            'info_dict': {
+                'id': 'posts/Self-service-BI-with-Power-BI-nuclear-testing',
+                'ext': 'mp4',
+                'title': 'Self-service BI with Power BI - nuclear testing',
+                'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
+                'duration': 1540,
+                'thumbnail': 'http://media.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg',
+                'authors': [ 'Mike Wilmot' ],
             },
         }
     ]
@@ -60,7 +62,7 @@ class Channel9IE(InfoExtractor):
             return 0
         units = m.group('units')
         try:
-            exponent = [u'B', u'KB', u'MB', u'GB', u'TB', u'PB', u'EB', u'ZB', u'YB'].index(units.upper())
+            exponent = ['B', 'KB', 'MB', 'GB', 'TB', 'PB', 'EB', 'ZB', 'YB'].index(units.upper())
         except ValueError:
             return 0
         size = float(m.group('size'))
@@ -76,21 +78,25 @@ class Channel9IE(InfoExtractor):
             </div>)?                                                # File size part may be missing
         '''
         # Extract known formats
-        formats = [{'url': x.group('url'),
-                 'format_id': x.group('quality'),
-                 'format_note': x.group('note'),
-                 'format': '%s (%s)' % (x.group('quality'), x.group('note')), 
-                 'filesize': self._restore_bytes(x.group('filesize')), # File size is approximate
-                 } for x in list(re.finditer(FORMAT_REGEX, html)) if x.group('quality') in self._known_formats]
-        # Sort according to known formats list
-        formats.sort(key=lambda fmt: self._known_formats.index(fmt['format_id']))
+        formats = [{
+            'url': x.group('url'),
+            'format_id': x.group('quality'),
+            'format_note': x.group('note'),
+            'format': '%s (%s)' % (x.group('quality'), x.group('note')),
+            'filesize': self._restore_bytes(x.group('filesize')), # File size is approximate
+            'preference': self._known_formats.index(x.group('quality')),
+            'vcodec': 'none' if x.group('note') == 'Audio only' else None,
+        } for x in list(re.finditer(FORMAT_REGEX, html)) if x.group('quality') in self._known_formats]
+
+        self._sort_formats(formats)
+
         return formats
 
     def _extract_title(self, html):
-        title = self._html_search_meta(u'title', html, u'title')
+        title = self._html_search_meta('title', html, 'title')
         if title is None:           
             title = self._og_search_title(html)
-            TITLE_SUFFIX = u' (Channel 9)'
+            TITLE_SUFFIX = ' (Channel 9)'
             if title is not None and title.endswith(TITLE_SUFFIX):
                 title = title[:-len(TITLE_SUFFIX)]
         return title
@@ -106,7 +112,7 @@ class Channel9IE(InfoExtractor):
         m = re.search(DESCRIPTION_REGEX, html)
         if m is not None:
             return m.group('description')
-        return self._html_search_meta(u'description', html, u'description')
+        return self._html_search_meta('description', html, 'description')
 
     def _extract_duration(self, html):
         m = re.search(r'data-video_duration="(?P<hours>\d{2}):(?P<minutes>\d{2}):(?P<seconds>\d{2})"', html)
@@ -168,7 +174,7 @@ class Channel9IE(InfoExtractor):
 
         # Nothing to download
         if len(formats) == 0 and slides is None and zip_ is None:
-            self._downloader.report_warning(u'None of recording, slides or zip are available for %s' % content_path)
+            self._downloader.report_warning('None of recording, slides or zip are available for %s' % content_path)
             return
 
         # Extract meta
@@ -240,7 +246,7 @@ class Channel9IE(InfoExtractor):
         return contents
 
     def _extract_list(self, content_path):
-        rss = self._download_xml(self._RSS_URL % content_path, content_path, u'Downloading RSS')
+        rss = self._download_xml(self._RSS_URL % content_path, content_path, 'Downloading RSS')
         entries = [self.url_result(session_url.text, 'Channel9')
                    for session_url in rss.findall('./channel/item/link')]
         title_text = rss.find('./channel/title').text
@@ -250,11 +256,11 @@ class Channel9IE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         content_path = mobj.group('contentpath')
 
-        webpage = self._download_webpage(url, content_path, u'Downloading web page')
+        webpage = self._download_webpage(url, content_path, 'Downloading web page')
 
         page_type_m = re.search(r'<meta name="Search.PageType" content="(?P<pagetype>[^"]+)"/>', webpage)
         if page_type_m is None:
-            raise ExtractorError(u'Search.PageType not found, don\'t know how to process this page', expected=True)
+            raise ExtractorError('Search.PageType not found, don\'t know how to process this page', expected=True)
 
         page_type = page_type_m.group('pagetype')
         if page_type == 'List':         # List page, may contain list of 'item'-like objects
@@ -264,4 +270,4 @@ class Channel9IE(InfoExtractor):
         elif page_type == 'Session':    # Event session page, may contain downloadable content
             return self._extract_session(webpage, content_path)
         else:
-            raise ExtractorError(u'Unexpected Search.PageType %s' % page_type, expected=True)
+            raise ExtractorError('Unexpected Search.PageType %s' % page_type, expected=True)

youtube_dl/extractor/chilloutzone.py

@@ -0,0 +1,97 @@
+from __future__ import unicode_literals
+
+import re
+import base64
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    ExtractorError
+)
+
+
+class ChilloutzoneIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?chilloutzone\.net/video/(?P<id>[\w|-]+)\.html'
+    _TESTS = [{
+        'url': 'http://www.chilloutzone.net/video/enemene-meck-alle-katzen-weg.html',
+        'md5': 'a76f3457e813ea0037e5244f509e66d1',
+        'info_dict': {
+            'id': 'enemene-meck-alle-katzen-weg',
+            'ext': 'mp4',
+            'title': 'Enemene Meck - Alle Katzen weg',
+            'description': 'Ist das der Umkehrschluss des Niesenden Panda-Babys?',
+        },
+    }, {
+        'note': 'Video hosted at YouTube',
+        'url': 'http://www.chilloutzone.net/video/eine-sekunde-bevor.html',
+        'info_dict': {
+            'id': '1YVQaAgHyRU',
+            'ext': 'mp4',
+            'title': '16 Photos Taken 1 Second Before Disaster',
+            'description': 'md5:58a8fcf6a459fe0a08f54140f0ad1814',
+            'uploader': 'BuzzFeedVideo',
+            'uploader_id': 'BuzzFeedVideo',
+            'upload_date': '20131105',
+        },
+    }, {
+        'note': 'Video hosted at Vimeo',
+        'url': 'http://www.chilloutzone.net/video/icon-blending.html',
+        'md5': '2645c678b8dc4fefcc0e1b60db18dac1',
+        'info_dict': {
+            'id': '85523671',
+            'ext': 'mp4',
+            'title': 'The Sunday Times - Icons',
+            'description': 'md5:3e1c0dc6047498d6728dcdaad0891762',
+            'uploader': 'Us',
+            'uploader_id': 'usfilms',
+            'upload_date': '20140131'
+        },
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        base64_video_info = self._html_search_regex(
+            r'var cozVidData = "(.+?)";', webpage, 'video data')
+        decoded_video_info = base64.b64decode(base64_video_info).decode("utf-8")
+        video_info_dict = json.loads(decoded_video_info)
+
+        # get video information from dict
+        video_url = video_info_dict['mediaUrl']
+        description = clean_html(video_info_dict.get('description'))
+        title = video_info_dict['title']
+        native_platform = video_info_dict['nativePlatform']
+        native_video_id = video_info_dict['nativeVideoId']
+        source_priority = video_info_dict['sourcePriority']
+
+        # If nativePlatform is None a fallback mechanism is used (i.e. youtube embed)
+        if native_platform is None:
+            youtube_url = self._html_search_regex(
+                r'<iframe.* src="((?:https?:)?//(?:[^.]+\.)?youtube\.com/.+?)"',
+                webpage, 'fallback video URL', default=None)
+            if youtube_url is not None:
+                return self.url_result(youtube_url, ie='Youtube')
+
+        # Non Fallback: Decide to use native source (e.g. youtube or vimeo) or
+        # the own CDN
+        if source_priority == 'native':
+            if native_platform == 'youtube':
+                return self.url_result(native_video_id, ie='Youtube')
+            if native_platform == 'vimeo':
+                return self.url_result(
+                    'http://vimeo.com/' + native_video_id, ie='Vimeo')
+
+        if not video_url:
+            raise ExtractorError('No video found')
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'ext': 'mp4',
+            'title': title,
+            'description': description,
+        }

youtube_dl/extractor/cinemassacre.py

@@ -1,4 +1,5 @@
 # encoding: utf-8
+from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
@@ -8,77 +9,71 @@ from ..utils import (
 
 
 class CinemassacreIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?(?P<url>cinemassacre\.com/(?P<date_Y>[0-9]{4})/(?P<date_m>[0-9]{2})/(?P<date_d>[0-9]{2})/.+?)(?:[/?].*)?'
-    _TESTS = [{
-        u'url': u'http://cinemassacre.com/2012/11/10/avgn-the-movie-trailer/',
-        u'file': u'19911.flv',
-        u'info_dict': {
-            u'upload_date': u'20121110',
-            u'title': u'“Angry Video Game Nerd: The Movie” – Trailer',
-            u'description': u'md5:fb87405fcb42a331742a0dce2708560b',
+    _VALID_URL = r'http://(?:www\.)?cinemassacre\.com/(?P<date_Y>[0-9]{4})/(?P<date_m>[0-9]{2})/(?P<date_d>[0-9]{2})/(?P<display_id>[^?#/]+)'
+    _TESTS = [
+        {
+            'url': 'http://cinemassacre.com/2012/11/10/avgn-the-movie-trailer/',
+            'file': '19911.mp4',
+            'md5': '782f8504ca95a0eba8fc9177c373eec7',
+            'info_dict': {
+                'upload_date': '20121110',
+                'title': '“Angry Video Game Nerd: The Movie” – Trailer',
+                'description': 'md5:fb87405fcb42a331742a0dce2708560b',
+            },
         },
-        u'params': {
-            # rtmp download
-            u'skip_download': True,
-        },
-    },
-    {
-        u'url': u'http://cinemassacre.com/2013/10/02/the-mummys-hand-1940',
-        u'file': u'521be8ef82b16.flv',
-        u'info_dict': {
-            u'upload_date': u'20131002',
-            u'title': u'The Mummy’s Hand (1940)',
-        },
-        u'params': {
-            # rtmp download
-            u'skip_download': True,
-        },
-    }]
+        {
+            'url': 'http://cinemassacre.com/2013/10/02/the-mummys-hand-1940',
+            'file': '521be8ef82b16.mp4',
+            'md5': 'dec39ee5118f8d9cc067f45f9cbe3a35',
+            'info_dict': {
+                'upload_date': '20131002',
+                'title': 'The Mummy’s Hand (1940)',
+            },
+        }
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('display_id')
 
-        webpage_url = u'http://' + mobj.group('url')
-        webpage = self._download_webpage(webpage_url, None) # Don't know video id yet
+        webpage = self._download_webpage(url, display_id)
         video_date = mobj.group('date_Y') + mobj.group('date_m') + mobj.group('date_d')
         mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?id=(?:Cinemassacre-)?(?P<video_id>.+?))"', webpage)
         if not mobj:
-            raise ExtractorError(u'Can\'t extract embed url and video id')
-        playerdata_url = mobj.group(u'embed_url')
-        video_id = mobj.group(u'video_id')
+            raise ExtractorError('Can\'t extract embed url and video id')
+        playerdata_url = mobj.group('embed_url')
+        video_id = mobj.group('video_id')
 
-        video_title = self._html_search_regex(r'<title>(?P<title>.+?)\|',
-            webpage, u'title')
-        video_description = self._html_search_regex(r'<div class="entry-content">(?P<description>.+?)</div>',
-            webpage, u'description', flags=re.DOTALL, fatal=False)
-        if len(video_description) == 0:
-            video_description = None
+        video_title = self._html_search_regex(
+            r'<title>(?P<title>.+?)\|', webpage, 'title')
+        video_description = self._html_search_regex(
+            r'<div class="entry-content">(?P<description>.+?)</div>',
+            webpage, 'description', flags=re.DOTALL, fatal=False)
 
         playerdata = self._download_webpage(playerdata_url, video_id)
-        url = self._html_search_regex(r'\'streamer\': \'(?P<url>[^\']+)\'', playerdata, u'url')
 
-        sd_file = self._html_search_regex(r'\'file\': \'(?P<sd_file>[^\']+)\'', playerdata, u'sd_file')
-        hd_file = self._html_search_regex(r'\'?file\'?: "(?P<hd_file>[^"]+)"', playerdata, u'hd_file')
-        video_thumbnail = self._html_search_regex(r'\'image\': \'(?P<thumbnail>[^\']+)\'', playerdata, u'thumbnail', fatal=False)
+        sd_url = self._html_search_regex(r'file: \'([^\']+)\', label: \'SD\'', playerdata, 'sd_file')
+        hd_url = self._html_search_regex(
+            r'file: \'([^\']+)\', label: \'HD\'', playerdata, 'hd_file',
+            default=None)
+        video_thumbnail = self._html_search_regex(r'image: \'(?P<thumbnail>[^\']+)\'', playerdata, 'thumbnail', fatal=False)
 
-        formats = [
-            {
-                'url': url,
-                'play_path': 'mp4:' + sd_file,
-                'rtmp_live': True, # workaround
-                'ext': 'flv',
-                'format': 'sd',
-                'format_id': 'sd',
-            },
-            {
-                'url': url,
-                'play_path': 'mp4:' + hd_file,
-                'rtmp_live': True, # workaround
-                'ext': 'flv',
+        formats = [{
+            'url': sd_url,
+            'ext': 'mp4',
+            'format': 'sd',
+            'format_id': 'sd',
+            'quality': 1,
+        }]
+        if hd_url:
+            formats.append({
+                'url': hd_url,
+                'ext': 'mp4',
                 'format': 'hd',
                 'format_id': 'hd',
-            },
-        ]
+                'quality': 2,
+            })
+        self._sort_formats(formats)
 
         return {
             'id': video_id,

youtube_dl/extractor/clipfish.py

@@ -1,22 +1,28 @@
+from __future__ import unicode_literals
+
 import re
 import time
 import xml.etree.ElementTree
 
 from .common import InfoExtractor
-from ..utils import ExtractorError
+from ..utils import (
+    ExtractorError,
+    parse_duration,
+)
 
 
 class ClipfishIE(InfoExtractor):
-    IE_NAME = u'clipfish'
+    IE_NAME = 'clipfish'
 
     _VALID_URL = r'^https?://(?:www\.)?clipfish\.de/.*?/video/(?P<id>[0-9]+)/'
     _TEST = {
-        u'url': u'http://www.clipfish.de/special/game-trailer/video/3966754/fifa-14-e3-2013-trailer/',
-        u'file': u'3966754.mp4',
-        u'md5': u'2521cd644e862936cf2e698206e47385',
-        u'info_dict': {
-            u'title': u'FIFA 14 - E3 2013 Trailer',
-            u'duration': 82,
+        'url': 'http://www.clipfish.de/special/game-trailer/video/3966754/fifa-14-e3-2013-trailer/',
+        'md5': '2521cd644e862936cf2e698206e47385',
+        'info_dict': {
+            'id': '3966754',
+            'ext': 'mp4',
+            'title': 'FIFA 14 - E3 2013 Trailer',
+            'duration': 82,
         },
         u'skip': 'Blocked in the US'
     }
@@ -33,21 +39,10 @@ class ClipfishIE(InfoExtractor):
         video_url = doc.find('filename').text
         if video_url is None:
             xml_bytes = xml.etree.ElementTree.tostring(doc)
-            raise ExtractorError(u'Cannot find video URL in document %r' %
+            raise ExtractorError('Cannot find video URL in document %r' %
                                  xml_bytes)
         thumbnail = doc.find('imageurl').text
-        duration_str = doc.find('duration').text
-        m = re.match(
-            r'^(?P<hours>[0-9]+):(?P<minutes>[0-9]{2}):(?P<seconds>[0-9]{2}):(?P<ms>[0-9]*)$',
-            duration_str)
-        if m:
-            duration = (
-                (int(m.group('hours')) * 60 * 60) +
-                (int(m.group('minutes')) * 60) +
-                (int(m.group('seconds')))
-            )
-        else:
-            duration = None
+        duration = parse_duration(doc.find('duration').text)
 
         return {
             'id': video_id,

youtube_dl/extractor/cliphunter.py

@@ -0,0 +1,56 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+translation_table = {
+    'a': 'h', 'd': 'e', 'e': 'v', 'f': 'o', 'g': 'f', 'i': 'd', 'l': 'n',
+    'm': 'a', 'n': 'm', 'p': 'u', 'q': 't', 'r': 's', 'v': 'p', 'x': 'r',
+    'y': 'l', 'z': 'i',
+    '$': ':', '&': '.', '(': '=', '^': '&', '=': '/',
+}
+
+
+class CliphunterIE(InfoExtractor):
+    IE_NAME = 'cliphunter'
+
+    _VALID_URL = r'''(?x)http://(?:www\.)?cliphunter\.com/w/
+        (?P<id>[0-9]+)/
+        (?P<seo>.+?)(?:$|[#\?])
+    '''
+    _TEST = {
+        'url': 'http://www.cliphunter.com/w/1012420/Fun_Jynx_Maze_solo',
+        'file': '1012420.flv',
+        'md5': '15e7740f30428abf70f4223478dc1225',
+        'info_dict': {
+            'title': 'Fun Jynx Maze solo',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        pl_fiji = self._search_regex(
+            r'pl_fiji = \'([^\']+)\'', webpage, 'video data')
+        pl_c_qual = self._search_regex(
+            r'pl_c_qual = "(.)"', webpage, 'video quality')
+        video_title = self._search_regex(
+            r'mediaTitle = "([^"]+)"', webpage, 'title')
+
+        video_url = ''.join(translation_table.get(c, c) for c in pl_fiji)
+
+        formats = [{
+            'url': video_url,
+            'format_id': pl_c_qual,
+        }]
+
+        return {
+            'id': video_id,
+            'title': video_title,
+            'formats': formats,
+        }

youtube_dl/extractor/clipsyndicate.py

@@ -1,9 +1,11 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
     find_xpath_attr,
-    fix_xml_all_ampersand,
+    fix_xml_ampersands
 )
 
 
@@ -11,13 +13,14 @@ class ClipsyndicateIE(InfoExtractor):
     _VALID_URL = r'http://www\.clipsyndicate\.com/video/play(list/\d+)?/(?P<id>\d+)'
 
     _TEST = {
-        u'url': u'http://www.clipsyndicate.com/video/play/4629301/brick_briscoe',
-        u'md5': u'4d7d549451bad625e0ff3d7bd56d776c',
-        u'info_dict': {
-            u'id': u'4629301',
-            u'ext': u'mp4',
-            u'title': u'Brick Briscoe',
-            u'duration': 612,
+        'url': 'http://www.clipsyndicate.com/video/play/4629301/brick_briscoe',
+        'md5': '4d7d549451bad625e0ff3d7bd56d776c',
+        'info_dict': {
+            'id': '4629301',
+            'ext': 'mp4',
+            'title': 'Brick Briscoe',
+            'duration': 612,
+            'thumbnail': 're:^https?://.+\.jpg',
         },
     }
 
@@ -26,14 +29,14 @@ class ClipsyndicateIE(InfoExtractor):
         video_id = mobj.group('id')
         js_player = self._download_webpage(
             'http://eplayer.clipsyndicate.com/embed/player.js?va_id=%s' % video_id,
-            video_id, u'Downlaoding player')
+            video_id, 'Downlaoding player')
         # it includes a required token
-        flvars = self._search_regex(r'flvars: "(.*?)"', js_player, u'flvars')
+        flvars = self._search_regex(r'flvars: "(.*?)"', js_player, 'flvars')
 
         pdoc = self._download_xml(
             'http://eplayer.clipsyndicate.com/osmf/playlist?%s' % flvars,
-            video_id, u'Downloading video info',
-            transform_source=fix_xml_all_ampersand) 
+            video_id, 'Downloading video info',
+            transform_source=fix_xml_ampersands)
 
         track_doc = pdoc.find('trackList/track')
         def find_param(name):

youtube_dl/extractor/clubic.py

@@ -0,0 +1,58 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    qualities,
+)
+
+
+class ClubicIE(InfoExtractor):
+    _VALID_URL = r'http://(?:www\.)?clubic\.com/video/[^/]+/video.*-(?P<id>[0-9]+)\.html'
+
+    _TEST = {
+        'url': 'http://www.clubic.com/video/clubic-week/video-clubic-week-2-0-le-fbi-se-lance-dans-la-photo-d-identite-448474.html',
+        'md5': '1592b694ba586036efac1776b0b43cd3',
+        'info_dict': {
+            'id': '448474',
+            'ext': 'mp4',
+            'title': 'Clubic Week 2.0 : le FBI se lance dans la photo d\u0092identité',
+            'description': 're:Gueule de bois chez Nokia. Le constructeur a indiqué cette.*',
+            'thumbnail': 're:^http://img\.clubic\.com/.*\.jpg$',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        player_url = 'http://player.m6web.fr/v1/player/clubic/%s.html' % video_id
+        player_page = self._download_webpage(player_url, video_id)
+
+        config_json = self._search_regex(
+            r'(?m)M6\.Player\.config\s*=\s*(\{.+?\});$', player_page,
+            'configuration')
+        config = json.loads(config_json)
+
+        video_info = config['videoInfo']
+        sources = config['sources']
+        quality_order = qualities(['sd', 'hq'])
+
+        formats = [{
+            'format_id': src['streamQuality'],
+            'url': src['src'],
+            'quality': quality_order(src['streamQuality']),
+        } for src in sources]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': video_info['title'],
+            'formats': formats,
+            'description': clean_html(video_info.get('description')),
+            'thumbnail': config.get('poster'),
+        }

youtube_dl/extractor/cmt.py

@@ -0,0 +1,19 @@
+from .mtv import MTVIE
+
+class CMTIE(MTVIE):
+    IE_NAME = u'cmt.com'
+    _VALID_URL = r'https?://www\.cmt\.com/videos/.+?/(?P<videoid>[^/]+)\.jhtml'
+    _FEED_URL = 'http://www.cmt.com/sitewide/apps/player/embed/rss/'
+
+    _TESTS = [
+        {
+            u'url': u'http://www.cmt.com/videos/garth-brooks/989124/the-call-featuring-trisha-yearwood.jhtml#artist=30061',
+            u'md5': u'e6b7ef3c4c45bbfae88061799bbba6c2',
+            u'info_dict': {
+                u'id': u'989124',
+                u'ext': u'mp4',
+                u'title': u'Garth Brooks - "The Call (featuring Trisha Yearwood)"',
+                u'description': u'Blame It All On My Roots',
+            },
+        },
+    ]

youtube_dl/extractor/cnet.py

@@ -0,0 +1,75 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+)
+
+
+class CNETIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?cnet\.com/videos/(?P<id>[^/]+)/'
+    _TEST = {
+        'url': 'http://www.cnet.com/videos/hands-on-with-microsofts-windows-8-1-update/',
+        'md5': '041233212a0d06b179c87cbcca1577b8',
+        'info_dict': {
+            'id': '56f4ea68-bd21-4852-b08c-4de5b8354c60',
+            'ext': 'mp4',
+            'title': 'Hands-on with Microsoft Windows 8.1 Update',
+            'description': 'The new update to the Windows 8 OS brings improved performance for mouse and keyboard users.',
+            'thumbnail': 're:^http://.*/flmswindows8.jpg$',
+            'uploader_id': 'sarah.mitroff@cbsinteractive.com',
+            'uploader': 'Sarah Mitroff',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, display_id)
+        data_json = self._html_search_regex(
+            r"<div class=\"cnetVideoPlayer\"\s+.*?data-cnet-video-options='([^']+)'",
+            webpage, 'data json')
+        data = json.loads(data_json)
+        vdata = data['video']
+        if not vdata:
+            vdata = data['videos'][0]
+        if not vdata:
+            raise ExtractorError('Cannot find video data')
+
+        video_id = vdata['id']
+        title = vdata['headline']
+        description = vdata.get('dek')
+        thumbnail = vdata.get('image', {}).get('path')
+        author = vdata.get('author')
+        if author:
+            uploader = '%s %s' % (author['firstName'], author['lastName'])
+            uploader_id = author.get('email')
+        else:
+            uploader = None
+            uploader_id = None
+
+        formats = [{
+            'format_id': '%s-%s-%s' % (
+                f['type'], f['format'],
+                int_or_none(f.get('bitrate'), 1000, default='')),
+            'url': f['uri'],
+            'tbr': int_or_none(f.get('bitrate'), 1000),
+        } for f in vdata['files']['data']]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'formats': formats,
+            'description': description,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/cnn.py

@@ -1,7 +1,13 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
-from ..utils import determine_ext
+from ..utils import (
+    int_or_none,
+    parse_duration,
+    url_basename,
+)
 
 
 class CNNIE(InfoExtractor):
@@ -9,21 +15,24 @@ class CNNIE(InfoExtractor):
         (?P<path>.+?/(?P<title>[^/]+?)(?:\.cnn|(?=&)))'''
 
     _TESTS = [{
-        u'url': u'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn',
-        u'file': u'sports_2013_06_09_nadal-1-on-1.cnn.mp4',
-        u'md5': u'3e6121ea48df7e2259fe73a0628605c4',
-        u'info_dict': {
-            u'title': u'Nadal wins 8th French Open title',
-            u'description': u'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.',
+        'url': 'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn',
+        'file': 'sports_2013_06_09_nadal-1-on-1.cnn.mp4',
+        'md5': '3e6121ea48df7e2259fe73a0628605c4',
+        'info_dict': {
+            'title': 'Nadal wins 8th French Open title',
+            'description': 'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.',
+            'duration': 135,
+            'upload_date': '20130609',
         },
     },
     {
-        u"url": u"http://edition.cnn.com/video/?/video/us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fcnn_topstories+%28RSS%3A+Top+Stories%29",
-        u"file": u"us_2013_08_21_sot-student-gives-epic-speech.georgia-institute-of-technology.mp4",
-        u"md5": u"b5cc60c60a3477d185af8f19a2a26f4e",
-        u"info_dict": {
-            u"title": "Student's epic speech stuns new freshmen",
-            u"description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\""
+        "url": "http://edition.cnn.com/video/?/video/us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fcnn_topstories+%28RSS%3A+Top+Stories%29",
+        "file": "us_2013_08_21_sot-student-gives-epic-speech.georgia-institute-of-technology.mp4",
+        "md5": "b5cc60c60a3477d185af8f19a2a26f4e",
+        "info_dict": {
+            "title": "Student's epic speech stuns new freshmen",
+            "description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\"",
+            "upload_date": "20130821",
         }
     }]
 
@@ -31,26 +40,87 @@ class CNNIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         path = mobj.group('path')
         page_title = mobj.group('title')
-        info_url = u'http://cnn.com/video/data/3.0/%s/index.xml' % path
+        info_url = 'http://cnn.com/video/data/3.0/%s/index.xml' % path
         info = self._download_xml(info_url, page_title)
 
         formats = []
+        rex = re.compile(r'''(?x)
+            (?P<width>[0-9]+)x(?P<height>[0-9]+)
+            (?:_(?P<bitrate>[0-9]+)k)?
+        ''')
         for f in info.findall('files/file'):
-            mf = re.match(r'(\d+)x(\d+)(?:_(.*)k)?',f.attrib['bitrate'])
-            if mf is not None:
-                formats.append((int(mf.group(1)), int(mf.group(2)), int(mf.group(3) or 0), f.text))
-        formats = sorted(formats)
-        (_,_,_, video_path) = formats[-1]
-        video_url = 'http://ht.cdn.turner.com/cnn/big%s' % video_path
+            video_url = 'http://ht.cdn.turner.com/cnn/big%s' % (f.text.strip())
+            fdct = {
+                'format_id': f.attrib['bitrate'],
+                'url': video_url,
+            }
+
+            mf = rex.match(f.attrib['bitrate'])
+            if mf:
+                fdct['width'] = int(mf.group('width'))
+                fdct['height'] = int(mf.group('height'))
+                fdct['tbr'] = int_or_none(mf.group('bitrate'))
+            else:
+                mf = rex.search(f.text)
+                if mf:
+                    fdct['width'] = int(mf.group('width'))
+                    fdct['height'] = int(mf.group('height'))
+                    fdct['tbr'] = int_or_none(mf.group('bitrate'))
+                else:
+                    mi = re.match(r'ios_(audio|[0-9]+)$', f.attrib['bitrate'])
+                    if mi:
+                        if mi.group(1) == 'audio':
+                            fdct['vcodec'] = 'none'
+                            fdct['ext'] = 'm4a'
+                        else:
+                            fdct['tbr'] = int(mi.group(1))
+
+            formats.append(fdct)
+
+        self._sort_formats(formats)
 
         thumbnails = sorted([((int(t.attrib['height']),int(t.attrib['width'])), t.text) for t in info.findall('images/image')])
         thumbs_dict = [{'resolution': res, 'url': t_url} for (res, t_url) in thumbnails]
 
-        return {'id': info.attrib['id'],
-                'title': info.find('headline').text,
-                'url': video_url,
-                'ext': determine_ext(video_url),
-                'thumbnail': thumbnails[-1][1],
-                'thumbnails': thumbs_dict,
-                'description': info.find('description').text,
-                }
+        metas_el = info.find('metas')
+        upload_date = (
+            metas_el.attrib.get('version') if metas_el is not None else None)
+
+        duration_el = info.find('length')
+        duration = parse_duration(duration_el.text)
+
+        return {
+            'id': info.attrib['id'],
+            'title': info.find('headline').text,
+            'formats': formats,
+            'thumbnail': thumbnails[-1][1],
+            'thumbnails': thumbs_dict,
+            'description': info.find('description').text,
+            'duration': duration,
+            'upload_date': upload_date,
+        }
+
+
+class CNNBlogsIE(InfoExtractor):
+    _VALID_URL = r'https?://[^\.]+\.blogs\.cnn\.com/.+'
+    _TEST = {
+        'url': 'http://reliablesources.blogs.cnn.com/2014/02/09/criminalizing-journalism/',
+        'md5': '3e56f97b0b6ffb4b79f4ea0749551084',
+        'info_dict': {
+            'id': 'bestoftv/2014/02/09/criminalizing-journalism.cnn',
+            'ext': 'mp4',
+            'title': 'Criminalizing journalism?',
+            'description': 'Glenn Greenwald responds to comments made this week on Capitol Hill that journalists could be criminal accessories.',
+            'upload_date': '20140209',
+        },
+        'add_ie': ['CNN'],
+    }
+
+    def _real_extract(self, url):
+        webpage = self._download_webpage(url, url_basename(url))
+        cnn_url = self._html_search_regex(r'data-url="(.+?)"', webpage, 'cnn url')
+        return {
+            '_type': 'url',
+            'url': cnn_url,
+            'ie_key': CNNIE.ie_key(),
+        }

youtube_dl/extractor/collegehumor.py

@@ -1,82 +1,102 @@
+from __future__ import unicode_literals
+
+import json
 import re
 
 from .common import InfoExtractor
-from ..utils import (
-    compat_urllib_parse_urlparse,
-    determine_ext,
-
-    ExtractorError,
-)
+from ..utils import int_or_none
 
 
 class CollegeHumorIE(InfoExtractor):
     _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/(video|embed|e)/(?P<videoid>[0-9]+)/?(?P<shorttitle>.*)$'
 
     _TESTS = [{
-        u'url': u'http://www.collegehumor.com/video/6902724/comic-con-cosplay-catastrophe',
-        u'file': u'6902724.mp4',
-        u'md5': u'1264c12ad95dca142a9f0bf7968105a0',
-        u'info_dict': {
-            u'title': u'Comic-Con Cosplay Catastrophe',
-            u'description': u'Fans get creative this year at San Diego.  Too creative.  And yes, that\'s really Joss Whedon.',
+        'url': 'http://www.collegehumor.com/video/6902724/comic-con-cosplay-catastrophe',
+        'md5': 'dcc0f5c1c8be98dc33889a191f4c26bd',
+        'info_dict': {
+            'id': '6902724',
+            'ext': 'mp4',
+            'title': 'Comic-Con Cosplay Catastrophe',
+            'description': "Fans get creative this year at San Diego.  Too creative.  And yes, that's really Joss Whedon.",
+            'age_limit': 13,
+            'duration': 187,
         },
     },
     {
-        u'url': u'http://www.collegehumor.com/video/3505939/font-conference',
-        u'file': u'3505939.mp4',
-        u'md5': u'c51ca16b82bb456a4397987791a835f5',
-        u'info_dict': {
-            u'title': u'Font Conference',
-            u'description': u'This video wasn\'t long enough, so we made it double-spaced.',
+        'url': 'http://www.collegehumor.com/video/3505939/font-conference',
+        'md5': '72fa701d8ef38664a4dbb9e2ab721816',
+        'info_dict': {
+            'id': '3505939',
+            'ext': 'mp4',
+            'title': 'Font Conference',
+            'description': "This video wasn't long enough, so we made it double-spaced.",
+            'age_limit': 10,
+            'duration': 179,
         },
-    }]
+    },
+    # embedded youtube video
+    {
+        'url': 'http://www.collegehumor.com/embed/6950306',
+        'info_dict': {
+            'id': 'Z-bao9fg6Yc',
+            'ext': 'mp4',
+            'title': 'Young Americans Think President John F. Kennedy Died THIS MORNING IN A CAR ACCIDENT!!!',
+            'uploader': 'Mark Dice',
+            'uploader_id': 'MarkDice',
+            'description': 'md5:62c3dab9351fac7bb44b53b69511d87f',
+            'upload_date': '20140127',
+        },
+        'params': {
+            'skip_download': True,
+        },
+        'add_ie': ['Youtube'],
+    },
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
         video_id = mobj.group('videoid')
 
-        info = {
-            'id': video_id,
-            'uploader': None,
-            'upload_date': None,
-        }
+        jsonUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id + '.json'
+        data = json.loads(self._download_webpage(
+            jsonUrl, video_id, 'Downloading info JSON'))
+        vdata = data['video']
+        if vdata.get('youtubeId') is not None:
+            return {
+                '_type': 'url',
+                'url': vdata['youtubeId'],
+                'ie_key': 'Youtube',
+            }
 
-        self.report_extraction(video_id)
-        xmlUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id
-        mdoc = self._download_xml(xmlUrl, video_id,
-                                         u'Downloading info XML',
-                                         u'Unable to download video info XML')
-
-        try:
-            videoNode = mdoc.findall('./video')[0]
-            youtubeIdNode = videoNode.find('./youtubeID')
-            if youtubeIdNode is not None:
-                return self.url_result(youtubeIdNode.text, 'Youtube')
-            info['description'] = videoNode.findall('./description')[0].text
-            info['title'] = videoNode.findall('./caption')[0].text
-            info['thumbnail'] = videoNode.findall('./thumbnail')[0].text
-            next_url = videoNode.findall('./file')[0].text
-        except IndexError:
-            raise ExtractorError(u'Invalid metadata XML file')
-
-        if next_url.endswith(u'manifest.f4m'):
-            manifest_url = next_url + '?hdcore=2.10.3'
-            adoc = self._download_xml(manifest_url, video_id,
-                                         u'Downloading XML manifest',
-                                         u'Unable to download video info XML')
-
-            try:
-                video_id = adoc.findall('./{http://ns.adobe.com/f4m/1.0}id')[0].text
-            except IndexError:
-                raise ExtractorError(u'Invalid manifest file')
-            url_pr = compat_urllib_parse_urlparse(info['thumbnail'])
-            info['url'] = url_pr.scheme + '://' + url_pr.netloc + video_id[:-2].replace('.csmil','').replace(',','')
-            info['ext'] = 'mp4'
+        AGE_LIMITS = {'nc17': 18, 'r': 18, 'pg13': 13, 'pg': 10, 'g': 0}
+        rating = vdata.get('rating')
+        if rating:
+            age_limit = AGE_LIMITS.get(rating.lower())
         else:
-            # Old-style direct links
-            info['url'] = next_url
-            info['ext'] = determine_ext(info['url'])
+            age_limit = None  # None = No idea
 
-        return info
+        PREFS = {'high_quality': 2, 'low_quality': 0}
+        formats = []
+        for format_key in ('mp4', 'webm'):
+            for qname, qurl in vdata.get(format_key, {}).items():
+                formats.append({
+                    'format_id': format_key + '_' + qname,
+                    'url': qurl,
+                    'format': format_key,
+                    'preference': PREFS.get(qname),
+                })
+        self._sort_formats(formats)
+
+        duration = int_or_none(vdata.get('duration'), 1000)
+        like_count = int_or_none(vdata.get('likes'))
+
+        return {
+            'id': video_id,
+            'title': vdata['title'],
+            'description': vdata.get('description'),
+            'thumbnail': vdata.get('thumbnail'),
+            'formats': formats,
+            'age_limit': age_limit,
+            'duration': duration,
+            'like_count': like_count,
+        }

youtube_dl/extractor/comedycentral.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -5,62 +7,59 @@ from .mtv import MTVServicesInfoExtractor
 from ..utils import (
     compat_str,
     compat_urllib_parse,
-
     ExtractorError,
+    float_or_none,
     unified_strdate,
 )
 
 
 class ComedyCentralIE(MTVServicesInfoExtractor):
-    _VALID_URL = r'https?://(?:www.)?comedycentral.com/(video-clips|episodes|cc-studios)/(?P<title>.*)'
-    _FEED_URL = u'http://comedycentral.com/feeds/mrss/'
+    _VALID_URL = r'''(?x)https?://(?:www\.)?(comedycentral|cc)\.com/
+        (video-clips|episodes|cc-studios|video-collections)
+        /(?P<title>.*)'''
+    _FEED_URL = 'http://comedycentral.com/feeds/mrss/'
 
     _TEST = {
-        u'url': u'http://www.comedycentral.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
-        u'md5': u'4167875aae411f903b751a21f357f1ee',
-        u'info_dict': {
-            u'id': u'cef0cbb3-e776-4bc9-b62e-8016deccb354',
-            u'ext': u'mp4',
-            u'title': u'Uncensored - Greg Fitzsimmons - Too Good of a Mother',
-            u'description': u'After a certain point, breastfeeding becomes c**kblocking.',
+        'url': 'http://www.comedycentral.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
+        'md5': 'c4f48e9eda1b16dd10add0744344b6d8',
+        'info_dict': {
+            'id': 'cef0cbb3-e776-4bc9-b62e-8016deccb354',
+            'ext': 'mp4',
+            'title': 'CC:Stand-Up|Greg Fitzsimmons: Life on Stage|Uncensored - Too Good of a Mother',
+            'description': 'After a certain point, breastfeeding becomes c**kblocking.',
         },
     }
 
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        title = mobj.group('title')
-        webpage = self._download_webpage(url, title)
-        mgid = self._search_regex(r'data-mgid="(?P<mgid>mgid:.*?)"',
-                                  webpage, u'mgid')
-        return self._get_videos_info(mgid)
-
 
 class ComedyCentralShowsIE(InfoExtractor):
-    IE_DESC = u'The Daily Show / Colbert Report'
+    IE_DESC = 'The Daily Show / The Colbert Report'
     # urls can be abbreviations like :thedailyshow or :colbert
     # urls for episodes like:
     # or urls for clips like: http://www.thedailyshow.com/watch/mon-december-10-2012/any-given-gun-day
     #                     or: http://www.colbertnation.com/the-colbert-report-videos/421667/november-29-2012/moon-shattering-news
     #                     or: http://www.colbertnation.com/the-colbert-report-collections/422008/festival-of-lights/79524
-    _VALID_URL = r"""^(:(?P<shortname>tds|thedailyshow|cr|colbert|colbertnation|colbertreport)
-                      |(https?://)?(www\.)?
-                          (?P<showname>thedailyshow|colbertnation)\.com/
-                         (full-episodes/(?P<episode>.*)|
+    _VALID_URL = r'''(?x)^(:(?P<shortname>tds|thedailyshow|cr|colbert|colbertnation|colbertreport)
+                      |https?://(:www\.)?
+                          (?P<showname>thedailyshow|thecolbertreport)\.(?:cc\.)?com/
+                         ((?:full-)?episodes/(?:[0-9a-z]{6}/)?(?P<episode>.*)|
                           (?P<clip>
-                              (the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
-                              |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)))|
+                              (?:(?:guests/[^/]+|videos|video-playlists|special-editions)/[^/]+/(?P<videotitle>[^/?#]+))
+                              |(the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
+                              |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*))
+                          )|
                           (?P<interview>
-                              extended-interviews/(?P<interID>[0-9]+)/playlist_tds_extended_(?P<interview_title>.*?)/.*?)))
-                     $"""
+                              extended-interviews/(?P<interID>[0-9a-z]+)/(?:playlist_tds_extended_)?(?P<interview_title>.*?)(/.*?)?)))
+                     (?:[?#].*|$)'''
     _TEST = {
-        u'url': u'http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart',
-        u'file': u'422212.mp4',
-        u'md5': u'4e2f5cb088a83cd8cdb7756132f9739d',
-        u'info_dict': {
-            u"upload_date": u"20121214", 
-            u"description": u"Kristen Stewart", 
-            u"uploader": u"thedailyshow", 
-            u"title": u"thedailyshow-kristen-stewart part 1"
+        'url': 'http://thedailyshow.cc.com/watch/thu-december-13-2012/kristen-stewart',
+        'md5': '4e2f5cb088a83cd8cdb7756132f9739d',
+        'info_dict': {
+            'id': 'ab9ab3e7-5a98-4dbe-8b21-551dc0523d55',
+            'ext': 'mp4',
+            'upload_date': '20121213',
+            'description': 'Kristen Stewart learns to let loose in "On the Road."',
+            'uploader': 'thedailyshow',
+            'title': 'thedailyshow kristen-stewart part 1',
         }
     }
 
@@ -83,34 +82,31 @@ class ComedyCentralShowsIE(InfoExtractor):
         '400': (384, 216),
     }
 
-    @classmethod
-    def suitable(cls, url):
-        """Receives a URL and returns True if suitable for this IE."""
-        return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
-
     @staticmethod
     def _transform_rtmp_url(rtmp_video_url):
-        m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url)
+        m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\.comedystor/.*)$', rtmp_video_url)
         if not m:
-            raise ExtractorError(u'Cannot transform RTMP url')
+            raise ExtractorError('Cannot transform RTMP url')
         base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
         return base + m.group('finalid')
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
         if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError('Invalid URL: %s' % url)
 
         if mobj.group('shortname'):
             if mobj.group('shortname') in ('tds', 'thedailyshow'):
-                url = u'http://www.thedailyshow.com/full-episodes/'
+                url = 'http://thedailyshow.cc.com/full-episodes/'
             else:
-                url = u'http://www.colbertnation.com/full-episodes/'
+                url = 'http://thecolbertreport.cc.com/full-episodes/'
             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
             assert mobj is not None
 
         if mobj.group('clip'):
-            if mobj.group('showname') == 'thedailyshow':
+            if mobj.group('videotitle'):
+                epTitle = mobj.group('videotitle')
+            elif mobj.group('showname') == 'thedailyshow':
                 epTitle = mobj.group('tdstitle')
             else:
                 epTitle = mobj.group('cntitle')
@@ -124,81 +120,96 @@ class ComedyCentralShowsIE(InfoExtractor):
                 epTitle = mobj.group('showname')
             else:
                 epTitle = mobj.group('episode')
+        show_name = mobj.group('showname')
 
-        self.report_extraction(epTitle)
-        webpage,htmlHandle = self._download_webpage_handle(url, epTitle)
+        webpage, htmlHandle = self._download_webpage_handle(url, epTitle)
         if dlNewest:
             url = htmlHandle.geturl()
             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
             if mobj is None:
-                raise ExtractorError(u'Invalid redirected URL: ' + url)
+                raise ExtractorError('Invalid redirected URL: ' + url)
             if mobj.group('episode') == '':
-                raise ExtractorError(u'Redirected URL is still not specific: ' + url)
-            epTitle = mobj.group('episode')
+                raise ExtractorError('Redirected URL is still not specific: ' + url)
+            epTitle = mobj.group('episode').rpartition('/')[-1]
 
         mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
-
         if len(mMovieParams) == 0:
             # The Colbert Report embeds the information in a without
             # a URL prefix; so extract the alternate reference
             # and then add the URL prefix manually.
 
-            altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video).*?:.*?)"', webpage)
+            altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video|playlist).*?:.*?)"', webpage)
             if len(altMovieParams) == 0:
-                raise ExtractorError(u'unable to find Flash URL in webpage ' + url)
+                raise ExtractorError('unable to find Flash URL in webpage ' + url)
             else:
                 mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
 
         uri = mMovieParams[0][1]
-        indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri})
-        idoc = self._download_xml(indexUrl, epTitle,
-                                          u'Downloading show index',
-                                          u'unable to download episode index')
+        # Correct cc.com in uri
+        uri = re.sub(r'(episode:[^.]+)(\.cc)?\.com', r'\1.cc.com', uri)
 
-        results = []
+        index_url = 'http://%s.cc.com/feeds/mrss?%s' % (show_name, compat_urllib_parse.urlencode({'uri': uri}))
+        idoc = self._download_xml(
+            index_url, epTitle,
+            'Downloading show index', 'Unable to download episode index')
 
-        itemEls = idoc.findall('.//item')
-        for partNum,itemEl in enumerate(itemEls):
-            mediaId = itemEl.findall('./guid')[0].text
-            shortMediaId = mediaId.split(':')[-1]
-            showId = mediaId.split(':')[-2].replace('.com', '')
-            officialTitle = itemEl.findall('./title')[0].text
-            officialDate = unified_strdate(itemEl.findall('./pubDate')[0].text)
+        title = idoc.find('./channel/title').text
+        description = idoc.find('./channel/description').text
 
-            configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' +
-                        compat_urllib_parse.urlencode({'uri': mediaId}))
-            cdoc = self._download_xml(configUrl, epTitle,
-                                               u'Downloading configuration for %s' % shortMediaId)
+        entries = []
+        item_els = idoc.findall('.//item')
+        for part_num, itemEl in enumerate(item_els):
+            upload_date = unified_strdate(itemEl.findall('./pubDate')[0].text)
+            thumbnail = itemEl.find('.//{http://search.yahoo.com/mrss/}thumbnail').attrib.get('url')
+
+            content = itemEl.find('.//{http://search.yahoo.com/mrss/}content')
+            duration = float_or_none(content.attrib.get('duration'))
+            mediagen_url = content.attrib['url']
+            guid = itemEl.find('./guid').text.rpartition(':')[-1]
+
+            cdoc = self._download_xml(
+                mediagen_url, epTitle,
+                'Downloading configuration for segment %d / %d' % (part_num + 1, len(item_els)))
 
             turls = []
             for rendition in cdoc.findall('.//rendition'):
                 finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
                 turls.append(finfo)
 
-            if len(turls) == 0:
-                self._downloader.report_error(u'unable to download ' + mediaId + ': No videos found')
-                continue
-
             formats = []
             for format, rtmp_video_url in turls:
                 w, h = self._video_dimensions.get(format, (None, None))
                 formats.append({
+                    'format_id': 'vhttp-%s' % format,
                     'url': self._transform_rtmp_url(rtmp_video_url),
                     'ext': self._video_extensions.get(format, 'mp4'),
-                    'format_id': format,
                     'height': h,
                     'width': w,
                 })
+                formats.append({
+                    'format_id': 'rtmp-%s' % format,
+                    'url': rtmp_video_url,
+                    'ext': self._video_extensions.get(format, 'mp4'),
+                    'height': h,
+                    'width': w,
+                })
+                self._sort_formats(formats)
 
-            effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1)
-            results.append({
-                'id': shortMediaId,
+            virtual_id = show_name + ' ' + epTitle + ' part ' + compat_str(part_num + 1)
+            entries.append({
+                'id': guid,
+                'title': virtual_id,
                 'formats': formats,
-                'uploader': showId,
-                'upload_date': officialDate,
-                'title': effTitle,
-                'thumbnail': None,
-                'description': compat_str(officialTitle),
+                'uploader': show_name,
+                'upload_date': upload_date,
+                'duration': duration,
+                'thumbnail': thumbnail,
+                'description': description,
             })
 
-        return results
+        return {
+            '_type': 'playlist',
+            'entries': entries,
+            'title': show_name + ' ' + title,
+            'description': description,
+        }

youtube_dl/extractor/common.py

@@ -1,4 +1,6 @@
 import base64
+import hashlib
+import json
 import os
 import re
 import socket
@@ -9,6 +11,7 @@ import xml.etree.ElementTree
 from ..utils import (
     compat_http_client,
     compat_urllib_error,
+    compat_urllib_parse_urlparse,
     compat_str,
 
     clean_html,
@@ -37,10 +40,12 @@ class InfoExtractor(object):
     id:             Video identifier.
     title:          Video title, unescaped.
 
-    Additionally, it must contain either a formats entry or url and ext:
+    Additionally, it must contain either a formats entry or a url one:
 
-    formats:        A list of dictionaries for each format available, it must
-                    be ordered from worst to best quality. Potential fields:
+    formats:        A list of dictionaries for each format available, ordered
+                    from worst to best quality.
+
+                    Potential fields:
                     * url        Mandatory. The URL of the video file
                     * ext        Will be calculated from url if missing
                     * format     A human-readable description of the format
@@ -48,32 +53,53 @@ class InfoExtractor(object):
                                  Calculated from the format_id, width, height.
                                  and format_note fields if missing.
                     * format_id  A short description of the format
-                                 ("mp4_h264_opus" or "19")
+                                 ("mp4_h264_opus" or "19").
+                                Technically optional, but strongly recommended.
                     * format_note Additional info about the format
                                  ("3D" or "DASH video")
                     * width      Width of the video, if known
                     * height     Height of the video, if known
+                    * resolution Textual description of width and height
+                    * tbr        Average bitrate of audio and video in KBit/s
                     * abr        Average audio bitrate in KBit/s
                     * acodec     Name of the audio codec in use
+                    * asr        Audio sampling rate in Hertz
                     * vbr        Average video bitrate in KBit/s
                     * vcodec     Name of the video codec in use
+                    * container  Name of the container format
                     * filesize   The number of bytes, if known in advance
                     * player_url SWF Player URL (used for rtmpdump).
+                    * protocol   The protocol that will be used for the actual
+                                 download, lower-case.
+                                 "http", "https", "rtsp", "rtmp", "m3u8" or so.
+                    * preference Order number of this format. If this field is
+                                 present and not None, the formats get sorted
+                                 by this field, regardless of all other values.
+                                 -1 for default (order by other properties),
+                                 -2 or smaller for less than default.
+                    * quality    Order number of the video quality of this
+                                 format, irrespective of the file format.
+                                 -1 for default (order by other properties),
+                                 -2 or smaller for less than default.
     url:            Final video URL.
     ext:            Video filename extension.
     format:         The video format, defaults to ext (used for --get-format)
     player_url:     SWF Player URL (used for rtmpdump).
-    urlhandle:      [internal] The urlHandle to be used to download the file,
-                    like returned by urllib.request.urlopen
 
     The following fields are optional:
 
+    display_id      An alternative identifier for the video, not necessarily
+                    unique, but available before title. Typically, id is
+                    something like "4234987", title "Dancing naked mole rats",
+                    and display_id "dancing-naked-mole-rats"
     thumbnails:     A list of dictionaries (with the entries "resolution" and
                     "url") for the varying thumbnails
     thumbnail:      Full URL to a video thumbnail image.
     description:    One-line video description.
     uploader:       Full name of the video uploader.
+    timestamp:      UNIX timestamp of the moment the video became available.
     upload_date:    Video upload date (YYYYMMDD).
+                    If not explicitly set, calculated from timestamp.
     uploader_id:    Nickname or id of the video uploader.
     location:       Physical location of the video.
     subtitles:      The subtitle file contents as a dictionary in the format
@@ -94,9 +120,6 @@ class InfoExtractor(object):
     _real_extract() methods and define a _VALID_URL regexp.
     Probably, they should also be added to the list of extractors.
 
-    _real_extract() must return a *list* of information dictionaries as
-    described above.
-
     Finally, the _WORKING attribute should be set to False for broken IEs
     in order to warn the users and skip the tests.
     """
@@ -202,6 +225,8 @@ class InfoExtractor(object):
                           webpage_bytes[:1024])
             if m:
                 encoding = m.group(1).decode('ascii')
+            elif webpage_bytes.startswith(b'\xff\xfe'):
+                encoding = 'utf-16'
             else:
                 encoding = 'utf-8'
         if self._downloader.params.get('dump_intermediate_pages', False):
@@ -217,13 +242,30 @@ class InfoExtractor(object):
                 url = url_or_request.get_full_url()
             except AttributeError:
                 url = url_or_request
+            if len(url) > 200:
+                h = u'___' + hashlib.md5(url.encode('utf-8')).hexdigest()
+                url = url[:200 - len(h)] + h
             raw_filename = ('%s_%s.dump' % (video_id, url))
             filename = sanitize_filename(raw_filename, restricted=True)
             self.to_screen(u'Saving request to ' + filename)
             with open(filename, 'wb') as outf:
                 outf.write(webpage_bytes)
 
-        content = webpage_bytes.decode(encoding, 'replace')
+        try:
+            content = webpage_bytes.decode(encoding, 'replace')
+        except LookupError:
+            content = webpage_bytes.decode('utf-8', 'replace')
+
+        if (u'<title>Access to this site is blocked</title>' in content and
+                u'Websense' in content[:512]):
+            msg = u'Access to this webpage has been blocked by Websense filtering software in your network.'
+            blocked_iframe = self._html_search_regex(
+                r'<iframe src="([^"]+)"', content,
+                u'Websense information URL', default=None)
+            if blocked_iframe:
+                msg += u' Visit %s for more details' % blocked_iframe
+            raise ExtractorError(msg, expected=True)
+
         return (content, urlh)
 
     def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
@@ -237,13 +279,33 @@ class InfoExtractor(object):
 
     def _download_xml(self, url_or_request, video_id,
                       note=u'Downloading XML', errnote=u'Unable to download XML',
-                      transform_source=None):
+                      transform_source=None, fatal=True):
         """Return the xml as an xml.etree.ElementTree.Element"""
-        xml_string = self._download_webpage(url_or_request, video_id, note, errnote)
+        xml_string = self._download_webpage(
+            url_or_request, video_id, note, errnote, fatal=fatal)
+        if xml_string is False:
+            return xml_string
         if transform_source:
             xml_string = transform_source(xml_string)
         return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8'))
 
+    def _download_json(self, url_or_request, video_id,
+                       note=u'Downloading JSON metadata',
+                       errnote=u'Unable to download JSON metadata',
+                       transform_source=None):
+        json_string = self._download_webpage(url_or_request, video_id, note, errnote)
+        if transform_source:
+            json_string = transform_source(json_string)
+        try:
+            return json.loads(json_string)
+        except ValueError as ve:
+            raise ExtractorError('Failed to download JSON', cause=ve)
+
+    def report_warning(self, msg, video_id=None):
+        idstr = u'' if video_id is None else u'%s: ' % video_id
+        self._downloader.report_warning(
+            u'[%s] %s%s' % (self.IE_NAME, idstr, msg))
+
     def to_screen(self, msg):
         """Print msg to screen, prefixing it with '[ie_name]'"""
         self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg))
@@ -360,8 +422,8 @@ class InfoExtractor(object):
     # Helper functions for extracting OpenGraph info
     @staticmethod
     def _og_regexes(prop):
-        content_re = r'content=(?:"([^>]+?)"|\'(.+?)\')'
-        property_re = r'property=[\'"]og:%s[\'"]' % re.escape(prop)
+        content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')'
+        property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop)
         template = r'<meta[^>]+?%s[^>]+?%s'
         return [
             template % (property_re, content_re),
@@ -390,14 +452,14 @@ class InfoExtractor(object):
         if secure: regexes = self._og_regexes('video:secure_url') + regexes
         return self._html_search_regex(regexes, html, name, **kargs)
 
-    def _html_search_meta(self, name, html, display_name=None):
+    def _html_search_meta(self, name, html, display_name=None, fatal=False):
         if display_name is None:
             display_name = name
         return self._html_search_regex(
             r'''(?ix)<meta
                     (?=[^>]+(?:itemprop|name|property)=["\']%s["\'])
                     [^>]+content=["\']([^"\']+)["\']''' % re.escape(name),
-            html, display_name, fatal=False)
+            html, display_name, fatal=fatal)
 
     def _dc_search_uploader(self, html):
         return self._html_search_meta('dc.creator', html, 'uploader')
@@ -426,6 +488,72 @@ class InfoExtractor(object):
         }
         return RATING_TABLE.get(rating.lower(), None)
 
+    def _twitter_search_player(self, html):
+        return self._html_search_meta('twitter:player', html,
+            'twitter card player')
+
+    def _sort_formats(self, formats):
+        if not formats:
+            raise ExtractorError(u'No video formats found')
+
+        def _formats_key(f):
+            # TODO remove the following workaround
+            from ..utils import determine_ext
+            if not f.get('ext') and 'url' in f:
+                f['ext'] = determine_ext(f['url'])
+
+            preference = f.get('preference')
+            if preference is None:
+                proto = f.get('protocol')
+                if proto is None:
+                    proto = compat_urllib_parse_urlparse(f.get('url', '')).scheme
+
+                preference = 0 if proto in ['http', 'https'] else -0.1
+                if f.get('ext') in ['f4f', 'f4m']:  # Not yet supported
+                    preference -= 0.5
+
+            if f.get('vcodec') == 'none':  # audio only
+                if self._downloader.params.get('prefer_free_formats'):
+                    ORDER = [u'aac', u'mp3', u'm4a', u'webm', u'ogg', u'opus']
+                else:
+                    ORDER = [u'webm', u'opus', u'ogg', u'mp3', u'aac', u'm4a']
+                ext_preference = 0
+                try:
+                    audio_ext_preference = ORDER.index(f['ext'])
+                except ValueError:
+                    audio_ext_preference = -1
+            else:
+                if self._downloader.params.get('prefer_free_formats'):
+                    ORDER = [u'flv', u'mp4', u'webm']
+                else:
+                    ORDER = [u'webm', u'flv', u'mp4']
+                try:
+                    ext_preference = ORDER.index(f['ext'])
+                except ValueError:
+                    ext_preference = -1
+                audio_ext_preference = 0
+
+            return (
+                preference,
+                f.get('quality') if f.get('quality') is not None else -1,
+                f.get('height') if f.get('height') is not None else -1,
+                f.get('width') if f.get('width') is not None else -1,
+                ext_preference,
+                f.get('tbr') if f.get('tbr') is not None else -1,
+                f.get('vbr') if f.get('vbr') is not None else -1,
+                f.get('abr') if f.get('abr') is not None else -1,
+                audio_ext_preference,
+                f.get('filesize') if f.get('filesize') is not None else -1,
+                f.get('format_id'),
+            )
+        formats.sort(key=_formats_key)
+
+    def http_scheme(self):
+        """ Either "https:" or "https:", depending on the user's preferences """
+        return (
+            'http:'
+            if self._downloader.params.get('prefer_insecure', False)
+            else 'https:')
 
 
 class SearchInfoExtractor(InfoExtractor):
@@ -470,3 +598,4 @@ class SearchInfoExtractor(InfoExtractor):
     @property
     def SEARCH_KEY(self):
         return self._SEARCH_KEY
+

youtube_dl/extractor/condenast.py

@@ -1,4 +1,5 @@
 # coding: utf-8
+from __future__ import unicode_literals
 
 import re
 import json
@@ -20,30 +21,33 @@ class CondeNastIE(InfoExtractor):
 
     # The keys are the supported sites and the values are the name to be shown
     # to the user and in the extractor description.
-    _SITES = {'wired': u'WIRED',
-              'gq': u'GQ',
-              'vogue': u'Vogue',
-              'glamour': u'Glamour',
-              'wmagazine': u'W Magazine',
-              'vanityfair': u'Vanity Fair',
-              }
+    _SITES = {
+        'wired': 'WIRED',
+        'gq': 'GQ',
+        'vogue': 'Vogue',
+        'glamour': 'Glamour',
+        'wmagazine': 'W Magazine',
+        'vanityfair': 'Vanity Fair',
+        'cnevids': 'Condé Nast',
+    }
 
-    _VALID_URL = r'http://(video|www).(?P<site>%s).com/(?P<type>watch|series|video)/(?P<id>.+)' % '|'.join(_SITES.keys())
-    IE_DESC = u'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
+    _VALID_URL = r'http://(video|www|player)\.(?P<site>%s)\.com/(?P<type>watch|series|video|embed)/(?P<id>[^/?#]+)' % '|'.join(_SITES.keys())
+    IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
 
     _TEST = {
-        u'url': u'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
-        u'file': u'5171b343c2b4c00dd0c1ccb3.mp4',
-        u'md5': u'1921f713ed48aabd715691f774c451f7',
-        u'info_dict': {
-            u'title': u'3D Printed Speakers Lit With LED',
-            u'description': u'Check out these beautiful 3D printed LED speakers.  You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
+        'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
+        'md5': '1921f713ed48aabd715691f774c451f7',
+        'info_dict': {
+            'id': '5171b343c2b4c00dd0c1ccb3',
+            'ext': 'mp4',
+            'title': '3D Printed Speakers Lit With LED',
+            'description': 'Check out these beautiful 3D printed LED speakers.  You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
         }
     }
 
     def _extract_series(self, url, webpage):
         title = self._html_search_regex(r'<div class="cne-series-info">.*?<h1>(.+?)</h1>',
-                                        webpage, u'series title', flags=re.DOTALL)
+                                        webpage, 'series title', flags=re.DOTALL)
         url_object = compat_urllib_parse_urlparse(url)
         base_url = '%s://%s' % (url_object.scheme, url_object.netloc)
         m_paths = re.finditer(r'<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]',
@@ -53,54 +57,60 @@ class CondeNastIE(InfoExtractor):
         entries = [self.url_result(build_url(path), 'CondeNast') for path in paths]
         return self.playlist_result(entries, playlist_title=title)
 
-    def _extract_video(self, webpage):
-        description = self._html_search_regex([r'<div class="cne-video-description">(.+?)</div>',
-                                               r'<div class="video-post-content">(.+?)</div>',
-                                               ],
-                                              webpage, u'description',
-                                              fatal=False, flags=re.DOTALL)
+    def _extract_video(self, webpage, url_type):
+        if url_type != 'embed':
+            description = self._html_search_regex(
+                [
+                    r'<div class="cne-video-description">(.+?)</div>',
+                    r'<div class="video-post-content">(.+?)</div>',
+                ],
+                webpage, 'description', fatal=False, flags=re.DOTALL)
+        else:
+            description = None
         params = self._search_regex(r'var params = {(.+?)}[;,]', webpage,
-                                    u'player params', flags=re.DOTALL)
-        video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, u'video id')
-        player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, u'player id')
-        target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, u'target')
+                                    'player params', flags=re.DOTALL)
+        video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id')
+        player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id')
+        target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target')
         data = compat_urllib_parse.urlencode({'videoId': video_id,
                                               'playerId': player_id,
                                               'target': target,
                                               })
         base_info_url = self._search_regex(r'url = [\'"](.+?)[\'"][,;]',
-                                           webpage, u'base info url',
+                                           webpage, 'base info url',
                                            default='http://player.cnevids.com/player/loader.js?')
         info_url = base_info_url + data
         info_page = self._download_webpage(info_url, video_id,
-                                           u'Downloading video info')
-        video_info = self._search_regex(r'var video = ({.+?});', info_page, u'video info')
+                                           'Downloading video info')
+        video_info = self._search_regex(r'var video = ({.+?});', info_page, 'video info')
         video_info = json.loads(video_info)
 
-        def _formats_sort_key(f):
-            type_ord = 1 if f['type'] == 'video/mp4' else 0
-            quality_ord = 1 if f['quality'] == 'high' else 0
-            return (quality_ord, type_ord)
-        best_format = sorted(video_info['sources'][0], key=_formats_sort_key)[-1]
+        formats = [{
+            'format_id': '%s-%s' % (fdata['type'].split('/')[-1], fdata['quality']),
+            'url': fdata['src'],
+            'ext': fdata['type'].split('/')[-1],
+            'quality': 1 if fdata['quality'] == 'high' else 0,
+        } for fdata in video_info['sources'][0]]
+        self._sort_formats(formats)
 
-        return {'id': video_id,
-                'url': best_format['src'],
-                'ext': best_format['type'].split('/')[-1],
-                'title': video_info['title'],
-                'thumbnail': video_info['poster_frame'],
-                'description': description,
-                }
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': video_info['title'],
+            'thumbnail': video_info['poster_frame'],
+            'description': description,
+        }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         site = mobj.group('site')
         url_type = mobj.group('type')
-        id = mobj.group('id')
+        item_id = mobj.group('id')
 
-        self.to_screen(u'Extracting from %s with the Condé Nast extractor' % self._SITES[site])
-        webpage = self._download_webpage(url, id)
+        self.to_screen('Extracting from %s with the Condé Nast extractor' % self._SITES[site])
+        webpage = self._download_webpage(url, item_id)
 
         if url_type == 'series':
             return self._extract_series(url, webpage)
         else:
-            return self._extract_video(webpage)
+            return self._extract_video(webpage, url_type)

youtube_dl/extractor/crunchyroll.py

@@ -1,5 +1,11 @@
 # encoding: utf-8
-import re, base64, zlib
+from __future__ import unicode_literals
+
+import re
+import json
+import base64
+import zlib
+
 from hashlib import sha1
 from math import pow, sqrt, floor
 from .common import InfoExtractor
@@ -17,30 +23,32 @@ from ..aes import (
     inc,
 )
 
+
 class CrunchyrollIE(InfoExtractor):
-    _VALID_URL = r'(?:https?://)?(?:www\.)?(?P<url>crunchyroll\.com/[^/]*/[^/?&]*?(?P<video_id>[0-9]+))(?:[/?&]|$)'
-    _TESTS = [{
-        u'url': u'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
-        u'file': u'645513.flv',
-        #u'md5': u'b1639fd6ddfaa43788c85f6d1dddd412',
-        u'info_dict': {
-            u'title': u'Wanna be the Strongest in the World Episode 1 – An Idol-Wrestler is Born!',
-            u'description': u'md5:2d17137920c64f2f49981a7797d275ef',
-            u'thumbnail': u'http://img1.ak.crunchyroll.com/i/spire1-tmb/20c6b5e10f1a47b10516877d3c039cae1380951166_full.jpg',
-            u'uploader': u'Yomiuri Telecasting Corporation (YTV)',
-            u'upload_date': u'20131013',
+    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)'
+    _TEST = {
+        'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
+        #'md5': 'b1639fd6ddfaa43788c85f6d1dddd412',
+        'info_dict': {
+            'id': '645513',
+            'ext': 'flv',
+            'title': 'Wanna be the Strongest in the World Episode 1 – An Idol-Wrestler is Born!',
+            'description': 'md5:2d17137920c64f2f49981a7797d275ef',
+            'thumbnail': 'http://img1.ak.crunchyroll.com/i/spire1-tmb/20c6b5e10f1a47b10516877d3c039cae1380951166_full.jpg',
+            'uploader': 'Yomiuri Telecasting Corporation (YTV)',
+            'upload_date': '20131013',
         },
-        u'params': {
+        'params': {
             # rtmp
-            u'skip_download': True,
+            'skip_download': True,
         },
-    }]
+    }
 
     _FORMAT_IDS = {
-        u'360': (u'60', u'106'),
-        u'480': (u'61', u'106'),
-        u'720': (u'62', u'106'),
-        u'1080': (u'80', u'108'),
+        '360': ('60', '106'),
+        '480': ('61', '106'),
+        '720': ('62', '106'),
+        '1080': ('80', '108'),
     }
 
     def _decrypt_subtitles(self, data, iv, id):
@@ -63,10 +71,10 @@ class CrunchyrollIE(InfoExtractor):
             num3 = key ^ num1
             num4 = num3 ^ (num3 >> 3) ^ num2
             prefix = intlist_to_bytes(obfuscate_key_aux(20, 97, (1, 2)))
-            shaHash = bytes_to_intlist(sha1(prefix + str(num4).encode(u'ascii')).digest())
+            shaHash = bytes_to_intlist(sha1(prefix + str(num4).encode('ascii')).digest())
             # Extend 160 Bit hash to 256 Bit
             return shaHash + [0] * 12
-        
+
         key = obfuscate_key(id)
         class Counter:
             __value = iv
@@ -78,94 +86,103 @@ class CrunchyrollIE(InfoExtractor):
         return zlib.decompress(decrypted_data)
 
     def _convert_subtitles_to_srt(self, subtitles):
-        i=1
-        output = u''
-        for start, end, text in re.findall(r'<event [^>]*?start="([^"]+)" [^>]*?end="([^"]+)" [^>]*?text="([^"]+)"[^>]*?>', subtitles):
-            start = start.replace(u'.', u',')
-            end = end.replace(u'.', u',')
+        output = ''
+        for i, (start, end, text) in enumerate(re.findall(r'<event [^>]*?start="([^"]+)" [^>]*?end="([^"]+)" [^>]*?text="([^"]+)"[^>]*?>', subtitles), 1):
+            start = start.replace('.', ',')
+            end = end.replace('.', ',')
             text = clean_html(text)
-            text = text.replace(u'\\N', u'\n')
+            text = text.replace('\\N', '\n')
             if not text:
                 continue
-            output += u'%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
-            i+=1
+            output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
         return output
 
     def _real_extract(self,url):
         mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('video_id')
 
-        webpage_url = u'http://www.' + mobj.group('url')
-        video_id = mobj.group(u'video_id')
-        webpage = self._download_webpage(webpage_url, video_id)
-        note_m = self._html_search_regex(r'<div class="showmedia-trailer-notice">(.+?)</div>', webpage, u'trailer-notice', default=u'')
+        if mobj.group('prefix') == 'm':
+            mobile_webpage = self._download_webpage(url, video_id, 'Downloading mobile webpage')
+            webpage_url = self._search_regex(r'<link rel="canonical" href="([^"]+)" />', mobile_webpage, 'webpage_url')
+        else:
+            webpage_url = 'http://www.' + mobj.group('url')
+
+        webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage')
+        note_m = self._html_search_regex(r'<div class="showmedia-trailer-notice">(.+?)</div>', webpage, 'trailer-notice', default='')
         if note_m:
             raise ExtractorError(note_m)
 
-        video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, u'video_title', flags=re.DOTALL)
-        video_title = re.sub(r' {2,}', u' ', video_title)
-        video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, u'video_description', default=u'')
+        mobj = re.search(r'Page\.messaging_box_controller\.addItems\(\[(?P<msg>{.+?})\]\)', webpage)
+        if mobj:
+            msg = json.loads(mobj.group('msg'))
+            if msg.get('type') == 'error':
+                raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True)
+
+        video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL)
+        video_title = re.sub(r' {2,}', ' ', video_title)
+        video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='')
         if not video_description:
             video_description = None
-        video_upload_date = self._html_search_regex(r'<div>Availability for free users:(.+?)</div>', webpage, u'video_upload_date', fatal=False, flags=re.DOTALL)
+        video_upload_date = self._html_search_regex(r'<div>Availability for free users:(.+?)</div>', webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)
         if video_upload_date:
             video_upload_date = unified_strdate(video_upload_date)
-        video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, u'video_uploader', fatal=False, flags=re.DOTALL)
+        video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, 'video_uploader', fatal=False, flags=re.DOTALL)
 
-        playerdata_url = compat_urllib_parse.unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, u'playerdata_url'))
+        playerdata_url = compat_urllib_parse.unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
         playerdata_req = compat_urllib_request.Request(playerdata_url)
-        playerdata_req.data = compat_urllib_parse.urlencode({u'current_page': webpage_url})
-        playerdata_req.add_header(u'Content-Type', u'application/x-www-form-urlencoded')
-        playerdata = self._download_webpage(playerdata_req, video_id, note=u'Downloading media info')
-        
-        stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, u'stream_id')
-        video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, u'thumbnail', fatal=False)
+        playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url})
+        playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info')
+
+        stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, 'stream_id')
+        video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
 
         formats = []
         for fmt in re.findall(r'\?p([0-9]{3,4})=1', webpage):
             stream_quality, stream_format = self._FORMAT_IDS[fmt]
-            video_format = fmt+u'p'
-            streamdata_req = compat_urllib_request.Request(u'http://www.crunchyroll.com/xml/')
+            video_format = fmt+'p'
+            streamdata_req = compat_urllib_request.Request('http://www.crunchyroll.com/xml/')
             # urlencode doesn't work!
-            streamdata_req.data = u'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality='+stream_quality+u'&media%5Fid='+stream_id+u'&video%5Fformat='+stream_format
-            streamdata_req.add_header(u'Content-Type', u'application/x-www-form-urlencoded')
-            streamdata_req.add_header(u'Content-Length', str(len(streamdata_req.data)))
-            streamdata = self._download_webpage(streamdata_req, video_id, note=u'Downloading media info for '+video_format)
-            video_url = self._search_regex(r'<host>([^<]+)', streamdata, u'video_url')
-            video_play_path = self._search_regex(r'<file>([^<]+)', streamdata, u'video_play_path')
+            streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality='+stream_quality+'&media%5Fid='+stream_id+'&video%5Fformat='+stream_format
+            streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
+            streamdata_req.add_header('Content-Length', str(len(streamdata_req.data)))
+            streamdata = self._download_webpage(streamdata_req, video_id, note='Downloading media info for '+video_format)
+            video_url = self._search_regex(r'<host>([^<]+)', streamdata, 'video_url')
+            video_play_path = self._search_regex(r'<file>([^<]+)', streamdata, 'video_play_path')
             formats.append({
-                u'url': video_url,
-                u'play_path':   video_play_path,
-                u'ext': 'flv',
-                u'format': video_format,
-                u'format_id': video_format,
+                'url': video_url,
+                'play_path':   video_play_path,
+                'ext': 'flv',
+                'format': video_format,
+                'format_id': video_format,
             })
 
         subtitles = {}
         for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
-            sub_page = self._download_webpage(u'http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id='+sub_id,\
-                                              video_id, note=u'Downloading subtitles for '+sub_name)
-            id = self._search_regex(r'id=\'([0-9]+)', sub_page, u'subtitle_id', fatal=False)
-            iv = self._search_regex(r'<iv>([^<]+)', sub_page, u'subtitle_iv', fatal=False)
-            data = self._search_regex(r'<data>([^<]+)', sub_page, u'subtitle_data', fatal=False)
+            sub_page = self._download_webpage('http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id='+sub_id,\
+                                              video_id, note='Downloading subtitles for '+sub_name)
+            id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False)
+            iv = self._search_regex(r'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
+            data = self._search_regex(r'<data>([^<]+)', sub_page, 'subtitle_data', fatal=False)
             if not id or not iv or not data:
                 continue
             id = int(id)
             iv = base64.b64decode(iv)
             data = base64.b64decode(data)
 
-            subtitle = self._decrypt_subtitles(data, iv, id).decode(u'utf-8')
-            lang_code = self._search_regex(r'lang_code=\'([^\']+)', subtitle, u'subtitle_lang_code', fatal=False)
+            subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
+            lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
             if not lang_code:
                 continue
             subtitles[lang_code] = self._convert_subtitles_to_srt(subtitle)
 
         return {
-            u'id':          video_id,
-            u'title':       video_title,
-            u'description': video_description,
-            u'thumbnail':   video_thumbnail,
-            u'uploader':    video_uploader,
-            u'upload_date': video_upload_date,
-            u'subtitles':   subtitles,
-            u'formats':     formats,
+            'id':          video_id,
+            'title':       video_title,
+            'description': video_description,
+            'thumbnail':   video_thumbnail,
+            'uploader':    video_uploader,
+            'upload_date': video_upload_date,
+            'subtitles':   subtitles,
+            'formats':     formats,
         }

youtube_dl/extractor/cspan.py

@@ -1,51 +1,83 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_urllib_parse,
+    int_or_none,
+    unescapeHTML,
+    find_xpath_attr,
 )
 
+
 class CSpanIE(InfoExtractor):
-    _VALID_URL = r'http://www\.c-spanvideo\.org/program/(.*)'
-    _TEST = {
-        u'url': u'http://www.c-spanvideo.org/program/HolderonV',
-        u'file': u'315139.flv',
-        u'md5': u'74a623266956f69e4df0068ab6c80fe4',
-        u'info_dict': {
-            u"title": u"Attorney General Eric Holder on Voting Rights Act Decision"
+    _VALID_URL = r'http://(?:www\.)?c-span\.org/video/\?(?P<id>[0-9a-f]+)'
+    IE_DESC = 'C-SPAN'
+    _TESTS = [{
+        'url': 'http://www.c-span.org/video/?313572-1/HolderonV',
+        'md5': '8e44ce11f0f725527daccc453f553eb0',
+        'info_dict': {
+            'id': '315139',
+            'ext': 'mp4',
+            'title': 'Attorney General Eric Holder on Voting Rights Act Decision',
+            'description': 'Attorney General Eric Holder spoke to reporters following the Supreme Court decision in Shelby County v. Holder in which the court ruled that the preclearance provisions of the Voting Rights Act could not be enforced until Congress established new guidelines for review.',
         },
-        u'skip': u'Requires rtmpdump'
-    }
+        'skip': 'Regularly fails on travis, for unknown reasons',
+    }, {
+        'url': 'http://www.c-span.org/video/?c4486943/cspan-international-health-care-models',
+        # For whatever reason, the served video alternates between
+        # two different ones
+        #'md5': 'dbb0f047376d457f2ab8b3929cbb2d0c',
+        'info_dict': {
+            'id': '340723',
+            'ext': 'mp4',
+            'title': 'International Health Care Models',
+            'description': 'md5:7a985a2d595dba00af3d9c9f0783c967',
+        }
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        prog_name = mobj.group(1)
-        webpage = self._download_webpage(url, prog_name)
-        video_id = self._search_regex(r'programid=(.*?)&', webpage, 'video id')
-        data = compat_urllib_parse.urlencode({'programid': video_id,
-                                              'dynamic':'1'})
-        info_url = 'http://www.c-spanvideo.org/common/services/flashXml.php?' + data
-        video_info = self._download_webpage(info_url, video_id, u'Downloading video info')
+        page_id = mobj.group('id')
+        webpage = self._download_webpage(url, page_id)
+        video_id = self._search_regex(r'progid=\'?([0-9]+)\'?>', webpage, 'video id')
 
-        self.report_extraction(video_id)
+        description = self._html_search_regex(
+            [
+                # The full description
+                r'<div class=\'expandable\'>(.*?)<a href=\'#\'',
+                # If the description is small enough the other div is not
+                # present, otherwise this is a stripped version
+                r'<p class=\'initial\'>(.*?)</p>'
+            ],
+            webpage, 'description', flags=re.DOTALL)
 
-        title = self._html_search_regex(r'<string name="title">(.*?)</string>',
-                                        video_info, 'title')
-        description = self._html_search_regex(r'<meta (?:property="og:|name=")description" content="(.*?)"',
-                                              webpage, 'description',
-                                              flags=re.MULTILINE|re.DOTALL)
+        info_url = 'http://c-spanvideo.org/videoLibrary/assets/player/ajax-player.php?os=android&html5=program&id=' + video_id
+        data = self._download_json(info_url, video_id)
 
-        url = self._search_regex(r'<string name="URL">(.*?)</string>',
-                                 video_info, 'video url')
-        url = url.replace('$(protocol)', 'rtmp').replace('$(port)', '443')
-        path = self._search_regex(r'<string name="path">(.*?)</string>',
-                            video_info, 'rtmp play path')
+        doc = self._download_xml(
+            'http://www.c-span.org/common/services/flashXml.php?programid=' + video_id,
+            video_id)
 
-        return {'id': video_id,
-                'title': title,
-                'ext': 'flv',
-                'url': url,
-                'play_path': path,
-                'description': description,
-                'thumbnail': self._og_search_thumbnail(webpage),
-                }
+        title = find_xpath_attr(doc, './/string', 'name', 'title').text
+        thumbnail = find_xpath_attr(doc, './/string', 'name', 'poster').text
+
+        files = data['video']['files']
+
+        entries = [{
+            'id': '%s_%d' % (video_id, partnum + 1),
+            'title': (
+                title if len(files) == 1 else
+                '%s part %d' % (title, partnum + 1)),
+            'url': unescapeHTML(f['path']['#text']),
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': int_or_none(f.get('length', {}).get('#text')),
+        } for partnum, f in enumerate(files)]
+
+        return {
+            '_type': 'playlist',
+            'entries': entries,
+            'title': title,
+            'id': video_id,
+        }

youtube_dl/extractor/d8.py

@@ -1,22 +1,25 @@
 # encoding: utf-8
+from __future__ import unicode_literals
+
 from .canalplus import CanalplusIE
 
 
 class D8IE(CanalplusIE):
     _VALID_URL = r'https?://www\.d8\.tv/.*?/(?P<path>.*)'
     _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/d8/%s'
-    IE_NAME = u'd8.tv'
+    IE_NAME = 'd8.tv'
 
     _TEST = {
-        u'url': u'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html',
-        u'file': u'966289.flv',
-        u'info_dict': {
-            u'title': u'Campagne intime - Documentaire exceptionnel',
-            u'description': u'md5:d2643b799fb190846ae09c61e59a859f',
-            u'upload_date': u'20131108',
+        'url': 'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html',
+        'file': '966289.flv',
+        'info_dict': {
+            'title': 'Campagne intime - Documentaire exceptionnel',
+            'description': 'md5:d2643b799fb190846ae09c61e59a859f',
+            'upload_date': '20131108',
         },
-        u'params': {
+        'params': {
             # rtmp
-            u'skip_download': True,
+            'skip_download': True,
         },
+        'skip': 'videos get deleted after a while',
     }

youtube_dl/extractor/dailymotion.py

@@ -8,12 +8,11 @@ from .subtitles import SubtitlesInfoExtractor
 from ..utils import (
     compat_urllib_request,
     compat_str,
-    get_element_by_attribute,
-    get_element_by_id,
     orderedSet,
     str_to_int,
-
+    int_or_none,
     ExtractorError,
+    unescapeHTML,
 )
 
 class DailymotionBaseInfoExtractor(InfoExtractor):
@@ -124,7 +123,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
             if video_url is not None:
                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
                 if m_size is not None:
-                    width, height = m_size.group(1), m_size.group(2)
+                    width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
                 else:
                     width, height = None, None
                 formats.append({
@@ -179,7 +178,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
     IE_NAME = u'dailymotion:playlist'
     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
-    _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/playlist/.+?".*?>.*?</a>.*?</div>'
+    _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 
     def _extract_entries(self, id):
@@ -189,10 +188,9 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
             webpage = self._download_webpage(request,
                                              id, u'Downloading page %s' % pagenum)
 
-            playlist_el = get_element_by_attribute(u'class', u'row video_list', webpage)
-            video_ids.extend(re.findall(r'data-id="(.+?)"', playlist_el))
+            video_ids.extend(re.findall(r'data-xid="(.+?)"', webpage))
 
-            if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
+            if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
                 break
         return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
                    for video_id in orderedSet(video_ids)]
@@ -202,26 +200,26 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
         playlist_id = mobj.group('id')
         webpage = self._download_webpage(url, playlist_id)
 
-        return {'_type': 'playlist',
-                'id': playlist_id,
-                'title': get_element_by_id(u'playlist_name', webpage),
-                'entries': self._extract_entries(playlist_id),
-                }
+        return {
+            '_type': 'playlist',
+            'id': playlist_id,
+            'title': self._og_search_title(webpage),
+            'entries': self._extract_entries(playlist_id),
+        }
 
 
 class DailymotionUserIE(DailymotionPlaylistIE):
     IE_NAME = u'dailymotion:user'
-    _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/user/(?P<user>[^/]+)'
-    _MORE_PAGES_INDICATOR = r'<div class="next">.*?<a.*?href="/user/.+?".*?>.*?</a>.*?</div>'
+    _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/user/(?P<user>[^/]+)'
     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         user = mobj.group('user')
         webpage = self._download_webpage(url, user)
-        full_user = self._html_search_regex(
-            r'<a class="label" href="/%s".*?>(.*?)</' % re.escape(user),
-            webpage, u'user', flags=re.DOTALL)
+        full_user = unescapeHTML(self._html_search_regex(
+            r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
+            webpage, u'user', flags=re.DOTALL))
 
         return {
             '_type': 'playlist',

youtube_dl/extractor/daum.py

@@ -1,25 +1,28 @@
 # encoding: utf-8
+
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
-    determine_ext,
 )
 
 
 class DaumIE(InfoExtractor):
     _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/.*?clipid=(?P<id>\d+)'
-    IE_NAME = u'daum.net'
+    IE_NAME = 'daum.net'
 
     _TEST = {
-        u'url': u'http://tvpot.daum.net/clip/ClipView.do?clipid=52554690',
-        u'file': u'52554690.mp4',
-        u'info_dict': {
-            u'title': u'DOTA 2GETHER 시즌2 6회 - 2부',
-            u'description': u'DOTA 2GETHER 시즌2 6회 - 2부',
-            u'upload_date': u'20130831',
-            u'duration': 3868,
+        'url': 'http://tvpot.daum.net/clip/ClipView.do?clipid=52554690',
+        'info_dict': {
+            'id': '52554690',
+            'ext': 'mp4',
+            'title': 'DOTA 2GETHER 시즌2 6회 - 2부',
+            'description': 'DOTA 2GETHER 시즌2 6회 - 2부',
+            'upload_date': '20130831',
+            'duration': 3868,
         },
     }
 
@@ -30,14 +33,14 @@ class DaumIE(InfoExtractor):
         webpage = self._download_webpage(canonical_url, video_id)
         full_id = self._search_regex(
             r'<iframe src="http://videofarm.daum.net/controller/video/viewer/Video.html\?.*?vid=(.+?)[&"]',
-            webpage, u'full id')
+            webpage, 'full id')
         query = compat_urllib_parse.urlencode({'vid': full_id})
         info = self._download_xml(
             'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id,
-            u'Downloading video info')
+            'Downloading video info')
         urls = self._download_xml(
             'http://videofarm.daum.net/controller/api/open/v1_2/MovieData.apixml?' + query,
-            video_id, u'Downloading video formats info')
+            video_id, 'Downloading video formats info')
 
         self.to_screen(u'%s: Getting video urls' % video_id)
         formats = []
@@ -53,7 +56,6 @@ class DaumIE(InfoExtractor):
             format_url = url_doc.find('result/url').text
             formats.append({
                 'url': format_url,
-                'ext': determine_ext(format_url),
                 'format_id': profile,
             })
 

youtube_dl/extractor/defense.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -5,15 +7,14 @@ from .common import InfoExtractor
 
 
 class DefenseGouvFrIE(InfoExtractor):
-    _IE_NAME = 'defense.gouv.fr'
+    IE_NAME = 'defense.gouv.fr'
     _VALID_URL = (r'http://.*?\.defense\.gouv\.fr/layout/set/'
         r'ligthboxvideo/base-de-medias/webtv/(.*)')
 
     _TEST = {
-        u'url': (u'http://www.defense.gouv.fr/layout/set/ligthboxvideo/'
-        u'base-de-medias/webtv/attaque-chimique-syrienne-du-21-aout-2013-1'),
-        u'file': u'11213.mp4',
-        u'md5': u'75bba6124da7e63d2d60b5244ec9430c',
+        'url': 'http://www.defense.gouv.fr/layout/set/ligthboxvideo/base-de-medias/webtv/attaque-chimique-syrienne-du-21-aout-2013-1',
+        'file': '11213.mp4',
+        'md5': '75bba6124da7e63d2d60b5244ec9430c',
         "info_dict": {
             "title": "attaque-chimique-syrienne-du-21-aout-2013-1"
         }

youtube_dl/extractor/depositfiles.py

@@ -1,60 +0,0 @@
-import re
-import os
-import socket
-
-from .common import InfoExtractor
-from ..utils import (
-    compat_http_client,
-    compat_str,
-    compat_urllib_error,
-    compat_urllib_parse,
-    compat_urllib_request,
-
-    ExtractorError,
-)
-
-
-class DepositFilesIE(InfoExtractor):
-    """Information extractor for depositfiles.com"""
-
-    _VALID_URL = r'(?:http://)?(?:\w+\.)?depositfiles\.com/(?:../(?#locale))?files/(.+)'
-
-    def _real_extract(self, url):
-        file_id = url.split('/')[-1]
-        # Rebuild url in english locale
-        url = 'http://depositfiles.com/en/files/' + file_id
-
-        # Retrieve file webpage with 'Free download' button pressed
-        free_download_indication = {'gateway_result' : '1'}
-        request = compat_urllib_request.Request(url, compat_urllib_parse.urlencode(free_download_indication))
-        try:
-            self.report_download_webpage(file_id)
-            webpage = compat_urllib_request.urlopen(request).read()
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            raise ExtractorError(u'Unable to retrieve file webpage: %s' % compat_str(err))
-
-        # Search for the real file URL
-        mobj = re.search(r'<form action="(http://fileshare.+?)"', webpage)
-        if (mobj is None) or (mobj.group(1) is None):
-            # Try to figure out reason of the error.
-            mobj = re.search(r'<strong>(Attention.*?)</strong>', webpage, re.DOTALL)
-            if (mobj is not None) and (mobj.group(1) is not None):
-                restriction_message = re.sub('\s+', ' ', mobj.group(1)).strip()
-                raise ExtractorError(u'%s' % restriction_message)
-            else:
-                raise ExtractorError(u'Unable to extract download URL from: %s' % url)
-
-        file_url = mobj.group(1)
-        file_extension = os.path.splitext(file_url)[1][1:]
-
-        # Search for file title
-        file_title = self._search_regex(r'<b title="(.*?)">', webpage, u'title')
-
-        return [{
-            'id':       file_id.decode('utf-8'),
-            'url':      file_url.decode('utf-8'),
-            'uploader': None,
-            'upload_date':  None,
-            'title':    file_title,
-            'ext':      file_extension.decode('utf-8'),
-        }]

youtube_dl/extractor/discovery.py

@@ -0,0 +1,47 @@
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+
+
+class DiscoveryIE(InfoExtractor):
+    _VALID_URL = r'http://dsc\.discovery\.com\/[a-zA-Z0-9\-]*/[a-zA-Z0-9\-]*/videos/(?P<id>[a-zA-Z0-9\-]*)(.htm)?'
+    _TEST = {
+        'url': 'http://dsc.discovery.com/tv-shows/mythbusters/videos/mission-impossible-outtakes.htm',
+        'md5': 'e12614f9ee303a6ccef415cb0793eba2',
+        'info_dict': {
+            'id': '614784',
+            'ext': 'mp4',
+            'title': 'MythBusters: Mission Impossible Outtakes',
+            'description': ('Watch Jamie Hyneman and Adam Savage practice being'
+                ' each other -- to the point of confusing Jamie\'s dog -- and '
+                'don\'t miss Adam moon-walking as Jamie ... behind Jamie\'s'
+                ' back.'),
+            'duration': 156,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+
+        video_list_json = self._search_regex(r'var videoListJSON = ({.*?});',
+            webpage, 'video list', flags=re.DOTALL)
+        video_list = json.loads(video_list_json)
+        info = video_list['clips'][0]
+        formats = []
+        for f in info['mp4']:
+            formats.append(
+                {'url': f['src'], 'ext': 'mp4', 'tbr': int(f['bitrate'][:-1])})
+
+        return {
+            'id': info['contentId'],
+            'title': video_list['name'],
+            'formats': formats,
+            'description': info['videoCaption'],
+            'thumbnail': info.get('videoStillURL') or info.get('thumbnailURL'),
+            'duration': info['duration'],
+        }

youtube_dl/extractor/divxstage.py

@@ -0,0 +1,27 @@
+from __future__ import unicode_literals
+
+from .novamov import NovaMovIE
+
+
+class DivxStageIE(NovaMovIE):
+    IE_NAME = 'divxstage'
+    IE_DESC = 'DivxStage'
+
+    _VALID_URL = NovaMovIE._VALID_URL_TEMPLATE % {'host': 'divxstage\.(?:eu|net|ch|co|at|ag)'}
+
+    _HOST = 'www.divxstage.eu'
+
+    _FILE_DELETED_REGEX = r'>This file no longer exists on our servers.<'
+    _TITLE_REGEX = r'<div class="video_det">\s*<strong>([^<]+)</strong>'
+    _DESCRIPTION_REGEX = r'<div class="video_det">\s*<strong>[^<]+</strong>\s*<p>([^<]+)</p>'
+
+    _TEST = {
+        'url': 'http://www.divxstage.eu/video/57f238e2e5e01',
+        'md5': '63969f6eb26533a1968c4d325be63e72',
+        'info_dict': {
+            'id': '57f238e2e5e01',
+            'ext': 'flv',
+            'title': 'youtubedl test video',
+            'description': 'This is a test video for youtubedl.',
+        }
+    }

youtube_dl/extractor/dotsub.py

@@ -1,41 +1,42 @@
+from __future__ import unicode_literals
+
 import re
-import json
 import time
 
 from .common import InfoExtractor
 
 
 class DotsubIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?dotsub\.com/view/([^/]+)'
+    _VALID_URL = r'http://(?:www\.)?dotsub\.com/view/(?P<id>[^/]+)'
     _TEST = {
-        u'url': u'http://dotsub.com/view/aed3b8b2-1889-4df5-ae63-ad85f5572f27',
-        u'file': u'aed3b8b2-1889-4df5-ae63-ad85f5572f27.flv',
-        u'md5': u'0914d4d69605090f623b7ac329fea66e',
-        u'info_dict': {
-            u"title": u"Pyramids of Waste (2010), AKA The Lightbulb Conspiracy - Planned obsolescence documentary",
-            u"uploader": u"4v4l0n42",
-            u'description': u'Pyramids of Waste (2010) also known as "The lightbulb conspiracy" is a documentary about how our economic system based on consumerism  and planned obsolescence is breaking our planet down.\r\n\r\nSolutions to this can be found at:\r\nhttp://robotswillstealyourjob.com\r\nhttp://www.federicopistono.org\r\n\r\nhttp://opensourceecology.org\r\nhttp://thezeitgeistmovement.com',
-            u'thumbnail': u'http://dotsub.com/media/aed3b8b2-1889-4df5-ae63-ad85f5572f27/p',
-            u'upload_date': u'20101213',
+        'url': 'http://dotsub.com/view/aed3b8b2-1889-4df5-ae63-ad85f5572f27',
+        'md5': '0914d4d69605090f623b7ac329fea66e',
+        'info_dict': {
+            'id': 'aed3b8b2-1889-4df5-ae63-ad85f5572f27',
+            'ext': 'flv',
+            'title': 'Pyramids of Waste (2010), AKA The Lightbulb Conspiracy - Planned obsolescence documentary',
+            'uploader': '4v4l0n42',
+            'description': 'Pyramids of Waste (2010) also known as "The lightbulb conspiracy" is a documentary about how our economic system based on consumerism  and planned obsolescence is breaking our planet down.\r\n\r\nSolutions to this can be found at:\r\nhttp://robotswillstealyourjob.com\r\nhttp://www.federicopistono.org\r\n\r\nhttp://opensourceecology.org\r\nhttp://thezeitgeistmovement.com',
+            'thumbnail': 'http://dotsub.com/media/aed3b8b2-1889-4df5-ae63-ad85f5572f27/p',
+            'upload_date': '20101213',
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1)
-        info_url = "https://dotsub.com/api/media/%s/metadata" %(video_id)
-        webpage = self._download_webpage(info_url, video_id)
-        info = json.loads(webpage)
+        video_id = mobj.group('id')
+        info_url = "https://dotsub.com/api/media/%s/metadata" % video_id
+        info = self._download_json(info_url, video_id)
         date = time.gmtime(info['dateCreated']/1000) # The timestamp is in miliseconds
 
-        return [{
-            'id':          video_id,
-            'url':         info['mediaURI'],
-            'ext':         'flv',
-            'title':       info['title'],
-            'thumbnail':   info['screenshotURI'],
+        return {
+            'id': video_id,
+            'url': info['mediaURI'],
+            'ext': 'flv',
+            'title': info['title'],
+            'thumbnail': info['screenshotURI'],
             'description': info['description'],
-            'uploader':    info['user'],
-            'view_count':  info['numberOfViews'],
-            'upload_date': u'%04i%02i%02i' % (date.tm_year, date.tm_mon, date.tm_mday),
-        }]
+            'uploader': info['user'],
+            'view_count': info['numberOfViews'],
+            'upload_date': '%04i%02i%02i' % (date.tm_year, date.tm_mon, date.tm_mday),
+        }

youtube_dl/extractor/dreisat.py

@@ -4,18 +4,17 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
     unified_strdate,
 )
 
 
 class DreiSatIE(InfoExtractor):
     IE_NAME = '3sat'
-    _VALID_URL = r'(?:http://)?(?:www\.)?3sat\.de/mediathek/index\.php\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$'
+    _VALID_URL = r'(?:http://)?(?:www\.)?3sat\.de/mediathek/(?:index\.php)?\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$'
     _TEST = {
         u"url": u"http://www.3sat.de/mediathek/index.php?obj=36983",
-        u'file': u'36983.webm',
-        u'md5': u'57c97d0469d71cf874f6815aa2b7c944',
+        u'file': u'36983.mp4',
+        u'md5': u'9dcfe344732808dbfcc901537973c922',
         u'info_dict': {
             u"title": u"Kaffeeland Schweiz",
             u"description": u"Über 80 Kaffeeröstereien liefern in der Schweiz das Getränk, in das das Land so vernarrt ist: Mehr als 1000 Tassen trinkt ein Schweizer pro Jahr. SCHWEIZWEIT nimmt die Kaffeekultur unter die...", 
@@ -52,18 +51,12 @@ class DreiSatIE(InfoExtractor):
             'width': int(fe.find('./width').text),
             'height': int(fe.find('./height').text),
             'url': fe.find('./url').text,
-            'ext': determine_ext(fe.find('./url').text),
             'filesize': int(fe.find('./filesize').text),
             'video_bitrate': int(fe.find('./videoBitrate').text),
-            '3sat_qualityname': fe.find('./quality').text,
         } for fe in format_els
             if not fe.find('./url').text.startswith('http://www.metafilegenerator.de/')]
 
-        def _sortkey(format):
-            qidx = ['low', 'med', 'high', 'veryhigh'].index(format['3sat_qualityname'])
-            prefer_http = 1 if 'rtmp' in format['url'] else 0
-            return (qidx, prefer_http, format['video_bitrate'])
-        formats.sort(key=_sortkey)
+        self._sort_formats(formats)
 
         return {
             '_type': 'video',

youtube_dl/extractor/dropbox.py

@@ -0,0 +1,32 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import os.path
+import re
+
+from .common import InfoExtractor
+
+
+class DropboxIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?dropbox[.]com/s/(?P<id>[a-zA-Z0-9]{15})/(?P<title>[^?#]*)'
+    _TEST = {
+        'url': 'https://www.dropbox.com/s/0qr9sai2veej4f8/THE_DOCTOR_GAMES.mp4',
+        'md5': '8ae17c51172fb7f93bdd6a214cc8c896',
+        'info_dict': {
+            'id': '0qr9sai2veej4f8',
+            'ext': 'mp4',
+            'title': 'THE_DOCTOR_GAMES'
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        title = os.path.splitext(mobj.group('title'))[0]
+        video_url = url + '?dl=1'
+
+        return {
+            'id': video_id,
+            'title': title,
+            'url': video_url,
+        }

youtube_dl/extractor/ehow.py

@@ -1,23 +1,25 @@
+from __future__ import unicode_literals
+
 import re
 
 from ..utils import (
     compat_urllib_parse,
-    determine_ext
 )
 from .common import InfoExtractor
 
 
 class EHowIE(InfoExtractor):
-    IE_NAME = u'eHow'
-    _VALID_URL = r'(?:https?://)?(?:www\.)?ehow\.com/[^/_?]*_(?P<id>[0-9]+)'
+    IE_NAME = 'eHow'
+    _VALID_URL = r'https?://(?:www\.)?ehow\.com/[^/_?]*_(?P<id>[0-9]+)'
     _TEST = {
-        u'url': u'http://www.ehow.com/video_12245069_hardwood-flooring-basics.html',
-        u'file': u'12245069.flv',
-        u'md5': u'9809b4e3f115ae2088440bcb4efbf371',
-        u'info_dict': {
-            u"title": u"Hardwood Flooring Basics",
-            u"description": u"Hardwood flooring may be time consuming, but its ultimately a pretty straightforward concept. Learn about hardwood flooring basics with help from a hardware flooring business owner in this free video...",
-   			u"uploader": u"Erick Nathan"
+        'url': 'http://www.ehow.com/video_12245069_hardwood-flooring-basics.html',
+        'md5': '9809b4e3f115ae2088440bcb4efbf371',
+        'info_dict': {
+            'id': '12245069',
+            'ext': 'flv',
+            'title': 'Hardwood Flooring Basics',
+            'description': 'Hardwood flooring may be time consuming, but its ultimately a pretty straightforward concept. Learn about hardwood flooring basics with help from a hardware flooring business owner in this free video...',
+            'uploader': 'Erick Nathan',
         }
     }
 
@@ -26,21 +28,16 @@ class EHowIE(InfoExtractor):
         video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
         video_url = self._search_regex(r'(?:file|source)=(http[^\'"&]*)',
-            webpage, u'video URL')
-        final_url = compat_urllib_parse.unquote(video_url)        
-        uploader = self._search_regex(r'<meta name="uploader" content="(.+?)" />',
-            webpage, u'uploader')
+            webpage, 'video URL')
+        final_url = compat_urllib_parse.unquote(video_url)
+        uploader = self._html_search_meta('uploader', webpage)
         title = self._og_search_title(webpage).replace(' | eHow', '')
-        ext = determine_ext(final_url)
 
         return {
-            '_type':       'video',
-            'id':          video_id,
-            'url':         final_url,
-            'ext':         ext,
-            'title':       title,
-            'thumbnail':   self._og_search_thumbnail(webpage),
+            'id': video_id,
+            'url': final_url,
+            'title': title,
+            'thumbnail': self._og_search_thumbnail(webpage),
             'description': self._og_search_description(webpage),
-            'uploader':    uploader,
+            'uploader': uploader,
         }
-

youtube_dl/extractor/elpais.py

@@ -0,0 +1,58 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import unified_strdate
+
+
+class ElPaisIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:[^.]+\.)?elpais\.com/.*/(?P<id>[^/#?]+)\.html(?:$|[?#])'
+    IE_DESC = 'El País'
+
+    _TEST = {
+        'url': 'http://blogs.elpais.com/la-voz-de-inaki/2014/02/tiempo-nuevo-recetas-viejas.html',
+        'md5': '98406f301f19562170ec071b83433d55',
+        'info_dict': {
+            'id': 'tiempo-nuevo-recetas-viejas',
+            'ext': 'mp4',
+            'title': 'Tiempo nuevo, recetas viejas',
+            'description': 'De lunes a viernes, a partir de las ocho de la mañana, Iñaki Gabilondo nos cuenta su visión de la actualidad nacional e internacional.',
+            'upload_date': '20140206',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        prefix = self._html_search_regex(
+            r'var url_cache = "([^"]+)";', webpage, 'URL prefix')
+        video_suffix = self._search_regex(
+            r"URLMediaFile = url_cache \+ '([^']+)'", webpage, 'video URL')
+        video_url = prefix + video_suffix
+        thumbnail_suffix = self._search_regex(
+            r"URLMediaStill = url_cache \+ '([^']+)'", webpage, 'thumbnail URL',
+            fatal=False)
+        thumbnail = (
+            None if thumbnail_suffix is None
+            else prefix + thumbnail_suffix)
+        title = self._html_search_regex(
+            '<h2 class="entry-header entry-title.*?>(.*?)</h2>',
+            webpage, 'title')
+        date_str = self._search_regex(
+            r'<p class="date-header date-int updated"\s+title="([^"]+)">',
+            webpage, 'upload date', fatal=False)
+        upload_date = (None if date_str is None else unified_strdate(date_str))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'description': self._og_search_description(webpage),
+            'thumbnail': thumbnail,
+            'upload_date': upload_date,
+        }

youtube_dl/extractor/engadget.py

@@ -0,0 +1,43 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from .fivemin import FiveMinIE
+from ..utils import (
+    url_basename,
+)
+
+
+class EngadgetIE(InfoExtractor):
+    _VALID_URL = r'''(?x)https?://www.engadget.com/
+        (?:video/5min/(?P<id>\d+)|
+            [\d/]+/.*?)
+        '''
+
+    _TEST = {
+        'url': 'http://www.engadget.com/video/5min/518153925/',
+        'md5': 'c6820d4828a5064447a4d9fc73f312c9',
+        'info_dict': {
+            'id': '518153925',
+            'ext': 'mp4',
+            'title': 'Samsung Galaxy Tab Pro 8.4 Review',
+        },
+        'add_ie': ['FiveMin'],
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        if video_id is not None:
+            return FiveMinIE._build_result(video_id)
+        else:
+            title = url_basename(url)
+            webpage = self._download_webpage(url, title)
+            ids = re.findall(r'<iframe[^>]+?playList=(\d+)', webpage)
+            return {
+                '_type': 'playlist',
+                'title': title,
+                'entries': [FiveMinIE._build_result(id) for id in ids]
+            }

youtube_dl/extractor/escapist.py

@@ -1,9 +1,9 @@
-import json
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_str,
     compat_urllib_parse,
 
     ExtractorError,
@@ -11,70 +11,68 @@ from ..utils import (
 
 
 class EscapistIE(InfoExtractor):
-    _VALID_URL = r'^https?://?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<episode>[^/?]+)[/?]?.*$'
+    _VALID_URL = r'^https?://?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<id>[0-9]+)-'
     _TEST = {
-        u'url': u'http://www.escapistmagazine.com/videos/view/the-escapist-presents/6618-Breaking-Down-Baldurs-Gate',
-        u'file': u'6618-Breaking-Down-Baldurs-Gate.mp4',
-        u'md5': u'ab3a706c681efca53f0a35f1415cf0d1',
-        u'info_dict': {
-            u"description": u"Baldur's Gate: Original, Modded or Enhanced Edition? I'll break down what you can expect from the new Baldur's Gate: Enhanced Edition.", 
-            u"uploader": u"the-escapist-presents", 
-            u"title": u"Breaking Down Baldur's Gate"
+        'url': 'http://www.escapistmagazine.com/videos/view/the-escapist-presents/6618-Breaking-Down-Baldurs-Gate',
+        'md5': 'ab3a706c681efca53f0a35f1415cf0d1',
+        'info_dict': {
+            'id': '6618',
+            'ext': 'mp4',
+            'description': "Baldur's Gate: Original, Modded or Enhanced Edition? I'll break down what you can expect from the new Baldur's Gate: Enhanced Edition.",
+            'uploader': 'the-escapist-presents',
+            'title': "Breaking Down Baldur's Gate",
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         showName = mobj.group('showname')
-        videoId = mobj.group('episode')
+        video_id = mobj.group('id')
 
-        self.report_extraction(videoId)
-        webpage = self._download_webpage(url, videoId)
+        self.report_extraction(video_id)
+        webpage = self._download_webpage(url, video_id)
 
         videoDesc = self._html_search_regex(
             r'<meta name="description" content="([^"]*)"',
-            webpage, u'description', fatal=False)
+            webpage, 'description', fatal=False)
 
         playerUrl = self._og_search_video_url(webpage, name=u'player URL')
 
         title = self._html_search_regex(
             r'<meta name="title" content="([^"]*)"',
-            webpage, u'title').split(' : ')[-1]
+            webpage, 'title').split(' : ')[-1]
 
-        configUrl = self._search_regex('config=(.*)$', playerUrl, u'config URL')
+        configUrl = self._search_regex('config=(.*)$', playerUrl, 'config URL')
         configUrl = compat_urllib_parse.unquote(configUrl)
 
         formats = []
 
-        def _add_format(name, cfgurl):
-            configJSON = self._download_webpage(
-                cfgurl, videoId,
-                u'Downloading ' + name + ' configuration',
-                u'Unable to download ' + name + ' configuration')
+        def _add_format(name, cfgurl, quality):
+            config = self._download_json(
+                cfgurl, video_id,
+                'Downloading ' + name + ' configuration',
+                'Unable to download ' + name + ' configuration',
+                transform_source=lambda s: s.replace("'", '"'))
 
-            # Technically, it's JavaScript, not JSON
-            configJSON = configJSON.replace("'", '"')
-
-            try:
-                config = json.loads(configJSON)
-            except (ValueError,) as err:
-                raise ExtractorError(u'Invalid JSON in configuration file: ' + compat_str(err))
             playlist = config['playlist']
             formats.append({
                 'url': playlist[1]['url'],
                 'format_id': name,
+                'quality': quality,
             })
 
-        _add_format(u'normal', configUrl)
+        _add_format('normal', configUrl, quality=0)
         hq_url = (configUrl +
                   ('&hq=1' if '?' in configUrl else configUrl + '?hq=1'))
         try:
-            _add_format(u'hq', hq_url)
+            _add_format('hq', hq_url, quality=1)
         except ExtractorError:
             pass  # That's fine, we'll just use normal quality
 
+        self._sort_formats(formats)
+
         return {
-            'id': videoId,
+            'id': video_id,
             'formats': formats,
             'uploader': showName,
             'title': title,

youtube_dl/extractor/everyonesmixtape.py

@@ -0,0 +1,69 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_request,
+    ExtractorError,
+)
+
+
+class EveryonesMixtapeIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?everyonesmixtape\.com/#/mix/(?P<id>[0-9a-zA-Z]+)(?:/(?P<songnr>[0-9]))?$'
+
+    _TEST = {
+        'url': 'http://everyonesmixtape.com/#/mix/m7m0jJAbMQi/5',
+        'file': '5bfseWNmlds.mp4',
+        "info_dict": {
+            "title": "Passion Pit - \"Sleepyhead\" (Official Music Video)",
+            "uploader": "FKR.TV",
+            "uploader_id": "frenchkissrecords",
+            "description": "Music video for \"Sleepyhead\" from Passion Pit's debut EP Chunk Of Change.\nBuy on iTunes: https://itunes.apple.com/us/album/chunk-of-change-ep/id300087641\n\nDirected by The Wilderness.\n\nhttp://www.passionpitmusic.com\nhttp://www.frenchkissrecords.com",
+            "upload_date": "20081015"
+        },
+        'params': {
+            'skip_download': True,  # This is simply YouTube
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('id')
+
+        pllist_url = 'http://everyonesmixtape.com/mixtape.php?a=getMixes&u=-1&linked=%s&explore=' % playlist_id
+        pllist_req = compat_urllib_request.Request(pllist_url)
+        pllist_req.add_header('X-Requested-With', 'XMLHttpRequest')
+
+        playlist_list = self._download_json(
+            pllist_req, playlist_id, note='Downloading playlist metadata')
+        try:
+            playlist_no = next(playlist['id']
+                               for playlist in playlist_list
+                               if playlist['code'] == playlist_id)
+        except StopIteration:
+            raise ExtractorError('Playlist id not found')
+
+        pl_url = 'http://everyonesmixtape.com/mixtape.php?a=getMix&id=%s&userId=null&code=' % playlist_no
+        pl_req = compat_urllib_request.Request(pl_url)
+        pl_req.add_header('X-Requested-With', 'XMLHttpRequest')
+        playlist = self._download_json(
+            pl_req, playlist_id, note='Downloading playlist info')
+
+        entries = [{
+            '_type': 'url',
+            'url': t['url'],
+            'title': t['title'],
+        } for t in playlist['tracks']]
+
+        if mobj.group('songnr'):
+            songnr = int(mobj.group('songnr')) - 1
+            return entries[songnr]
+
+        playlist_title = playlist['mixData']['name']
+        return {
+            '_type': 'playlist',
+            'id': playlist_id,
+            'title': playlist_title,
+            'entries': entries,
+        }

youtube_dl/extractor/exfm.py

@@ -1,56 +1,58 @@
+from __future__ import unicode_literals
+
 import re
-import json
 
 from .common import InfoExtractor
 
 
 class ExfmIE(InfoExtractor):
-    IE_NAME = u'exfm'
-    IE_DESC = u'ex.fm'
-    _VALID_URL = r'(?:http://)?(?:www\.)?ex\.fm/song/([^/]+)'
-    _SOUNDCLOUD_URL = r'(?:http://)?(?:www\.)?api\.soundcloud\.com/tracks/([^/]+)/stream'
+    IE_NAME = 'exfm'
+    IE_DESC = 'ex.fm'
+    _VALID_URL = r'http://(?:www\.)?ex\.fm/song/(?P<id>[^/]+)'
+    _SOUNDCLOUD_URL = r'http://(?:www\.)?api\.soundcloud\.com/tracks/([^/]+)/stream'
     _TESTS = [
         {
-            u'url': u'http://ex.fm/song/eh359',
-            u'file': u'44216187.mp3',
-            u'md5': u'e45513df5631e6d760970b14cc0c11e7',
-            u'info_dict': {
-                u"title": u"Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive",
-                u"uploader": u"deadjournalist",
-                u'upload_date': u'20120424',
-                u'description': u'Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive',
+            'url': 'http://ex.fm/song/eh359',
+            'md5': 'e45513df5631e6d760970b14cc0c11e7',
+            'info_dict': {
+                'id': '44216187',
+                'ext': 'mp3',
+                'title': 'Test House "Love Is Not Enough" (Extended Mix) DeadJournalist Exclusive',
+                'uploader': 'deadjournalist',
+                'upload_date': '20120424',
+                'description': 'Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive',
             },
-            u'note': u'Soundcloud song',
-            u'skip': u'The site is down too often',
+            'note': 'Soundcloud song',
+            'skip': 'The site is down too often',
         },
         {
-            u'url': u'http://ex.fm/song/wddt8',
-            u'file': u'wddt8.mp3',
-            u'md5': u'966bd70741ac5b8570d8e45bfaed3643',
-            u'info_dict': {
-                u'title': u'Safe and Sound',
-                u'uploader': u'Capital Cities',
+            'url': 'http://ex.fm/song/wddt8',
+            'md5': '966bd70741ac5b8570d8e45bfaed3643',
+            'info_dict': {
+                'id': 'wddt8',
+                'ext': 'mp3',
+                'title': 'Safe and Sound',
+                'uploader': 'Capital Cities',
             },
-            u'skip': u'The site is down too often',
+            'skip': 'The site is down too often',
         },
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        song_id = mobj.group(1)
-        info_url = "http://ex.fm/api/v3/song/%s" %(song_id)
-        webpage = self._download_webpage(info_url, song_id)
-        info = json.loads(webpage)
-        song_url = info['song']['url']
+        song_id = mobj.group('id')
+        info_url = "http://ex.fm/api/v3/song/%s" % song_id
+        info = self._download_json(info_url, song_id)['song']
+        song_url = info['url']
         if re.match(self._SOUNDCLOUD_URL, song_url) is not None:
             self.to_screen('Soundcloud song detected')
-            return self.url_result(song_url.replace('/stream',''), 'Soundcloud')
-        return [{
-            'id':          song_id,
-            'url':         song_url,
-            'ext':         'mp3',
-            'title':       info['song']['title'],
-            'thumbnail':   info['song']['image']['large'],
-            'uploader':    info['song']['artist'],
-            'view_count':  info['song']['loved_count'],
-        }]
+            return self.url_result(song_url.replace('/stream', ''), 'Soundcloud')
+        return {
+            'id': song_id,
+            'url': song_url,
+            'ext': 'mp3',
+            'title': info['title'],
+            'thumbnail': info['image']['large'],
+            'uploader': info['artist'],
+            'view_count': info['loved_count'],
+        }

youtube_dl/extractor/extremetube.py

@@ -1,4 +1,5 @@
-import os
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -8,18 +9,23 @@ from ..utils import (
     compat_urllib_parse,
 )
 
+
 class ExtremeTubeIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>extremetube\.com/video/.+?(?P<videoid>[0-9]+))(?:[/?&]|$)'
-    _TEST = {
-        u'url': u'http://www.extremetube.com/video/music-video-14-british-euro-brit-european-cumshots-swallow-652431',
-        u'file': u'652431.mp4',
-        u'md5': u'1fb9228f5e3332ec8c057d6ac36f33e0',
-        u'info_dict': {
-            u"title": u"Music Video 14 british euro brit european cumshots swallow",
-            u"uploader": u"unknown",
-            u"age_limit": 18,
+    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>extremetube\.com/.*?video/.+?(?P<videoid>[0-9]+))(?:[/?&]|$)'
+    _TESTS = [{
+        'url': 'http://www.extremetube.com/video/music-video-14-british-euro-brit-european-cumshots-swallow-652431',
+        'md5': '1fb9228f5e3332ec8c057d6ac36f33e0',
+        'info_dict': {
+            'id': '652431',
+            'ext': 'mp4',
+            'title': 'Music Video 14 british euro brit european cumshots swallow',
+            'uploader': 'unknown',
+            'age_limit': 18,
         }
-    }
+    }, {
+        'url': 'http://www.extremetube.com/gay/video/abcde-1234',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -30,11 +36,14 @@ class ExtremeTubeIE(InfoExtractor):
         req.add_header('Cookie', 'age_verified=1')
         webpage = self._download_webpage(req, video_id)
 
-        video_title = self._html_search_regex(r'<h1 [^>]*?title="([^"]+)"[^>]*>\1<', webpage, u'title')
-        uploader = self._html_search_regex(r'>Posted by:(?=<)(?:\s|<[^>]*>)*(.+?)\|', webpage, u'uploader', fatal=False)
-        video_url = compat_urllib_parse.unquote(self._html_search_regex(r'video_url=(.+?)&amp;', webpage, u'video_url'))
+        video_title = self._html_search_regex(
+            r'<h1 [^>]*?title="([^"]+)"[^>]*>\1<', webpage, 'title')
+        uploader = self._html_search_regex(
+            r'>Posted by:(?=<)(?:\s|<[^>]*>)*(.+?)\|', webpage, 'uploader',
+            fatal=False)
+        video_url = compat_urllib_parse.unquote(self._html_search_regex(
+            r'video_url=(.+?)&amp;', webpage, 'video_url'))
         path = compat_urllib_parse_urlparse(video_url).path
-        extension = os.path.splitext(path)[1][1:]
         format = path.split('/')[5].split('_')[:2]
         format = "-".join(format)
 
@@ -43,7 +52,6 @@ class ExtremeTubeIE(InfoExtractor):
             'title': video_title,
             'uploader': uploader,
             'url': video_url,
-            'ext': extension,
             'format': format,
             'format_id': format,
             'age_limit': 18,

youtube_dl/extractor/facebook.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import json
 import re
 import socket
@@ -9,33 +11,34 @@ from ..utils import (
     compat_urllib_error,
     compat_urllib_parse,
     compat_urllib_request,
+    urlencode_postdata,
 
     ExtractorError,
 )
 
 
 class FacebookIE(InfoExtractor):
-    """Information Extractor for Facebook"""
-
-    _VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:[^#?]*#!/)?(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)'
+    _VALID_URL = r'''(?x)
+        https?://(?:\w+\.)?facebook\.com/
+        (?:[^#?]*\#!/)?
+        (?:video/video\.php|photo\.php|video/embed)\?(?:.*?)
+        (?:v|video_id)=(?P<id>[0-9]+)
+        (?:.*)'''
     _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
     _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
     _NETRC_MACHINE = 'facebook'
-    IE_NAME = u'facebook'
+    IE_NAME = 'facebook'
     _TEST = {
-        u'url': u'https://www.facebook.com/photo.php?v=120708114770723',
-        u'file': u'120708114770723.mp4',
-        u'md5': u'48975a41ccc4b7a581abd68651c1a5a8',
-        u'info_dict': {
-            u"duration": 279,
-            u"title": u"PEOPLE ARE AWESOME 2013"
+        'url': 'https://www.facebook.com/photo.php?v=120708114770723',
+        'md5': '48975a41ccc4b7a581abd68651c1a5a8',
+        'info_dict': {
+            'id': '120708114770723',
+            'ext': 'mp4',
+            'duration': 279,
+            'title': 'PEOPLE ARE AWESOME 2013',
         }
     }
 
-    def report_login(self):
-        """Report attempt to log in."""
-        self.to_screen(u'Logging in')
-
     def _login(self):
         (useremail, password) = self._get_login_info()
         if useremail is None:
@@ -43,11 +46,13 @@ class FacebookIE(InfoExtractor):
 
         login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
         login_page_req.add_header('Cookie', 'locale=en_US')
-        self.report_login()
-        login_page = self._download_webpage(login_page_req, None, note=False,
-            errnote=u'Unable to download login page')
-        lsd = self._search_regex(r'"lsd":"(\w*?)"', login_page, u'lsd')
-        lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, u'lgnrnd')
+        login_page = self._download_webpage(login_page_req, None,
+            note='Downloading login page',
+            errnote='Unable to download login page')
+        lsd = self._search_regex(
+            r'<input type="hidden" name="lsd" value="([^"]*)"',
+            login_page, 'lsd')
+        lgnrnd = self._search_regex(r'name="lgnrnd" value="([^"]*?)"', login_page, 'lgnrnd')
 
         login_form = {
             'email': useremail,
@@ -60,27 +65,28 @@ class FacebookIE(InfoExtractor):
             'timezone': '-60',
             'trynum': '1',
             }
-        request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
+        request = compat_urllib_request.Request(self._LOGIN_URL, urlencode_postdata(login_form))
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         try:
-            login_results = compat_urllib_request.urlopen(request).read()
+            login_results = self._download_webpage(request, None,
+                note='Logging in', errnote='unable to fetch login page')
             if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
-                self._downloader.report_warning(u'unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
+                self._downloader.report_warning('unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
                 return
 
             check_form = {
-                'fb_dtsg': self._search_regex(r'"fb_dtsg":"(.*?)"', login_results, u'fb_dtsg'),
-                'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, u'nh'),
+                'fb_dtsg': self._search_regex(r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg'),
+                'h': self._search_regex(r'name="h" value="(\w*?)"', login_results, 'h'),
                 'name_action_selected': 'dont_save',
-                'submit[Continue]': self._search_regex(r'<input value="(.*?)" name="submit\[Continue\]"', login_results, u'continue'),
             }
-            check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, compat_urllib_parse.urlencode(check_form))
+            check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
             check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
-            check_response = compat_urllib_request.urlopen(check_req).read()
+            check_response = self._download_webpage(check_req, None,
+                note='Confirming login')
             if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
-                self._downloader.report_warning(u'Unable to confirm login, you have to login in your brower and authorize the login.')
+                self._downloader.report_warning('Unable to confirm login, you have to login in your brower and authorize the login.')
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
+            self._downloader.report_warning('unable to log in: %s' % compat_str(err))
             return
 
     def _real_initialize(self):
@@ -88,9 +94,7 @@ class FacebookIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
-        video_id = mobj.group('ID')
+        video_id = mobj.group('id')
 
         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
         webpage = self._download_webpage(url, video_id)
@@ -102,10 +106,10 @@ class FacebookIE(InfoExtractor):
             m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage)
             if m_msg is not None:
                 raise ExtractorError(
-                    u'The video is not available, Facebook said: "%s"' % m_msg.group(1),
+                    'The video is not available, Facebook said: "%s"' % m_msg.group(1),
                     expected=True)
             else:
-                raise ExtractorError(u'Cannot parse data')
+                raise ExtractorError('Cannot parse data')
         data = dict(json.loads(m.group(1)))
         params_raw = compat_urllib_parse.unquote(data['params'])
         params = json.loads(params_raw)
@@ -114,19 +118,15 @@ class FacebookIE(InfoExtractor):
         if not video_url:
             video_url = video_data['sd_src']
         if not video_url:
-            raise ExtractorError(u'Cannot find video URL')
-        video_duration = int(video_data['video_duration'])
-        thumbnail = video_data['thumbnail_src']
+            raise ExtractorError('Cannot find video URL')
 
         video_title = self._html_search_regex(
-            r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, u'title')
+            r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, 'title')
 
-        info = {
+        return {
             'id': video_id,
             'title': video_title,
             'url': video_url,
-            'ext': 'mp4',
-            'duration': video_duration,
-            'thumbnail': thumbnail,
+            'duration': int(video_data['video_duration']),
+            'thumbnail': video_data['thumbnail_src'],
         }
-        return [info]