release 2013.01.17

If the url doesn’t specify the protocol we can’t directly use it to download the webpage, we would need to build a new url.
Instead, we let the generic extractor add the protocol.

.gitignore

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

.travis.yml

@@ -3,9 +3,6 @@ python:
   - "2.6"
   - "2.7"
   - "3.3"
-before_install:
-  - sudo apt-get update -qq
-  - sudo apt-get install -qq rtmpdump
 script: nosetests test --verbose
 notifications:
   email:

README.md

@@ -30,13 +30,20 @@ which means you can modify it, redistribute it or use it however you like.
     --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
+    --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
+                               store some downloaded information permanently. By
                                default $XDG_CACHE_HOME/youtube-dl or ~/.cache
-                               /youtube-dl .
+                               /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
 
 ## Video Selection:
     --playlist-start NUMBER    playlist video to start at (default is 1)
@@ -51,12 +58,19 @@ which means you can modify it, redistribute it or use it however you like.
     --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
+    --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 present in the archive
-                               file. Record all downloaded videos in it.
+    --download-archive FILE    Download only videos not listed in the archive
+                               file. Record the IDs of all downloaded videos in
+                               it.
 
 ## Download Options:
     -r, --rate-limit LIMIT     maximum download rate in bytes per second (e.g.
@@ -79,13 +93,13 @@ which means you can modify it, redistribute it or use it however you like.
                                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
+                               (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
+                               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
@@ -96,6 +110,8 @@ which means you can modify it, redistribute it or use it however you like.
     --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
@@ -121,6 +137,7 @@ which means you can modify it, redistribute it or use it however you like.
     --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
@@ -128,21 +145,20 @@ which means you can modify it, redistribute it or use it however you like.
     --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
+    --dump-intermediate-pages  print downloaded pages to debug problems (very
                                verbose)
-    --write-pages              Write downloaded pages to files in the current
-                               directory
+    --write-pages              Write downloaded intermediary pages to files in
+                               the current directory to debug problems
 
 ## Video Format Options:
-    -f, --format FORMAT        video format code, specifiy the order of
+    -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, --list-formats         list all available formats
 
 ## Subtitle Options:
     --write-sub                write subtitle file
@@ -160,7 +176,7 @@ which means you can modify it, redistribute it or use it however you like.
     -u, --username USERNAME    account username
     -p, --password PASSWORD    account password
     -n, --netrc                use .netrc authentication data
-    --video-password PASSWORD  video password (vimeo only)
+    --video-password PASSWORD  video password (vimeo, smotri)
 
 ## Post-processing Options:
     -x, --extract-audio        convert video files to audio-only files (requires
@@ -178,11 +194,17 @@ which means you can modify it, redistribute it or use it however you like.
                                processed files are overwritten by default
     --embed-subs               embed subtitles in the video (only for mp4
                                videos)
-    --add-metadata             add metadata to the files
+    --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`.
+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`.
 
 # OUTPUT TEMPLATE
 
@@ -217,9 +239,12 @@ 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
 
@@ -272,14 +297,58 @@ This README file was originally written by Daniel Bolton (<https://github.com/db
 
 # BUGS
 
-Bugs and suggestions should be reported at: <https://github.com/rg3/youtube-dl/issues>
+Bugs and suggestions should be reported at: <https://github.com/rg3/youtube-dl/issues> . Unless you were prompted so or there is another pertinent reason (e.g. GitHub fails to accept the bug report), please do not send bug reports via personal email.
 
-Please include:
-
-* Your exact command line, like `youtube-dl -t "http://www.youtube.com/watch?v=uHlDtZ6Oc3s&feature=channel_video_title"`. A common mistake is not to escape the `&`. Putting URLs in quotes should solve this problem.
-* If possible re-run the command with `--verbose`, and include the full output, it is really helpful to us.
-* The output of `youtube-dl --version`
-* The output of `python --version`
-* The name and version of your Operating System ("Ubuntu 11.04 x64" or "Windows 7 x64" is usually enough).
+Please include the full output of the command when run with `--verbose`. The output (including the first lines) contain important debugging information. Issues without the full output are often not reproducible and therefore do not get solved in short order, if ever.
 
 For discussions, join us in the irc channel #youtube-dl on freenode.
+
+When you submit a request, please re-read it once to avoid a couple of mistakes (you can and should use this as a checklist):
+
+### Is the description of the issue itself sufficient?
+
+We often get issue reports that we cannot really decipher. While in most cases we eventually get the required information after asking back multiple times, this poses an unnecessary drain on our resources. Many contributors, including myself, are also not native speakers, so we may misread some parts.
+
+So please elaborate on what feature you are requesting, or what bug you want to be fixed. Make sure that it's obvious
+
+- What the problem is
+- How it could be fixed
+- How your proposed solution would look like
+
+If your report is shorter than two lines, it is almost certainly missing some of these, which makes it hard for us to respond to it. We're often too polite to close the issue outright, but the missing info makes misinterpretation likely. As a commiter myself, I often get frustrated by these issues, since the only possible way for me to move forward on them is to ask for clarification over and over.
+
+For bug reports, this means that your report should contain the *complete* output of youtube-dl when called with the -v flag. The error message you get for (most) bugs even says so, but you would not believe how many of our bug reports do not contain this information.
+
+Site support requests must contain an example URL. An example URL is a URL you might want to download, like http://www.youtube.com/watch?v=BaW_jenozKc . There should be an obvious video present. Except under very special circumstances, the main page of a video service (e.g. http://www.youtube.com/ ) is *not* an example URL.
+
+###  Are you using the latest version?
+
+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?
+
+Make sure that someone has not already opened the issue you're trying to open. Search at the top of the window or at https://github.com/rg3/youtube-dl/search?type=Issues . If there is an issue, feel free to write something along the lines of "This affects me as well, with version 2015.01.01. Here is some more information on the issue: ...". While some issues may be old, a new post into them often spurs rapid activity.
+
+###  Why are existing options not enough?
+
+Before requesting a new feature, please have a quick peek at [the list of supported options](https://github.com/rg3/youtube-dl/blob/master/README.md#synopsis). Many feature requests are for features that actually exist already! Please, absolutely do show off your work in the issue report and detail how the existing similar options do *not* solve your problem.
+
+###  Is there enough context in your bug report?
+
+People want to solve problems, and often think they do us a favor by breaking down their larger problems (e.g. wanting to skip already downloaded files) to a specific request (e.g. requesting us to look whether the file exists before downloading the info page). However, what often happens is that they break down the problem into two steps: One simple, and one impossible (or extremely complicated one).
+
+We are then presented with a very complicated request when the original problem could be solved far easier, e.g. by recording the downloaded video IDs in a separate file. To avoid this, you must include the greater context where it is non-obvious. In particular, every feature request that does not consist of adding support for a new site should contain a use case scenario that explains in what situation the missing feature would be useful.
+
+###  Does the issue involve one problem, and one problem only?
+
+Some of our users seem to think there is a limit of issues they can or should open. There is no limit of issues they can or should open. While it may seem appealing to be able to dump all your issues into one ticket, that means that someone who solves one of your issues cannot mark the issue as closed. Typically, reporting a bunch of issues leads to the ticket lingering since nobody wants to attack that behemoth, until someone mercifully splits the issue into multiple ones.
+
+In particular, every site support request issue should only pertain to services at one site (generally under a common domain, but always using the same backend technology). Do not request support for vimeo user videos, Whitehouse podcasts, and Google Plus pages in the same issue. Also, make sure that you don't post bug reports alongside feature requests. As a rule of thumb, a feature request does not include outputs of youtube-dl that are not immediately related to the feature at hand. Do not post reports of a network error alongside the request for a new video service.
+
+###  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.

devscripts/bash-completion.in

@@ -1,10 +1,21 @@
 __youtube_dl()
 {
-    local cur prev opts
+    local cur prev opts fileopts diropts keywords
     COMPREPLY=()
     cur="${COMP_WORDS[COMP_CWORD]}"
+    prev="${COMP_WORDS[COMP_CWORD-1]}"
     opts="{{flags}}"
-    keywords=":ytfavorites :ytrecommended :ytsubscriptions :ytwatchlater"
+    keywords=":ytfavorites :ytrecommended :ytsubscriptions :ytwatchlater :ythistory"
+    fileopts="-a|--batch-file|--download-archive|--cookies|--load-info"
+    diropts="--cache-dir"
+
+    if [[ ${prev} =~ ${fileopts} ]]; then
+        COMPREPLY=( $(compgen -f -- ${cur}) )
+        return 0
+    elif [[ ${prev} =~ ${diropts} ]]; then
+        COMPREPLY=( $(compgen -d -- ${cur}) )
+        return 0
+    fi
 
     if [[ ${cur} =~ : ]]; then
         COMPREPLY=( $(compgen -W "${keywords}" -- ${cur}) )

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

@@ -24,6 +24,8 @@ if [ -z "$1" ]; then echo "ERROR: specify version number like this: $0 1994.09.0
 version="$1"
 if [ ! -z "`git tag | grep "$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..."

setup.py

@@ -71,7 +71,10 @@ 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/parameters.json

@@ -39,5 +39,6 @@
     "writeinfojson": true, 
     "writesubtitles": false,
     "allsubtitles": false,
-    "listssubtitles": false
+    "listssubtitles": false,
+    "socket_timeout": 20
 }

test/test_YoutubeDL.py

@@ -7,6 +7,8 @@ import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from test.helper import FakeYDL
+from youtube_dl import YoutubeDL
+from youtube_dl.extractor import YoutubeIE
 
 
 class YDL(FakeYDL):
@@ -32,6 +34,8 @@ class TestFormatSelection(unittest.TestCase):
             {u'ext': u'mp4',  u'height': 460},
         ]
         info_dict = {u'formats': formats, u'extractor': u'test'}
+        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')
@@ -44,28 +48,46 @@ class TestFormatSelection(unittest.TestCase):
             {u'ext': u'mp4', u'height': 1080},
         ]
         info_dict[u'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')
 
-        # 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'mp4', u'height': 720},
             {u'ext': u'flv', u'height': 720},
         ]
         info_dict[u'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')
+
+        ydl = YDL()
+        ydl.params['prefer_free_formats'] = False
+        formats = [
+            {u'ext': u'flv', u'height': 720},
+            {u'ext': u'webm', u'height': 720},
+        ]
+        info_dict[u'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')
 
     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'},
+            {u'format_id': u'meh', u'url': u'http://example.com/meh', 'preference': 1},
+            {u'format_id': u'good', u'url': u'http://example.com/good', 'preference': 2},
+            {u'format_id': u'great', u'url': u'http://example.com/great', 'preference': 3},
+            {u'format_id': u'excellent', u'url': u'http://example.com/exc', 'preference': 4},
         ]
         info_dict = {
             u'formats': formats, u'extractor': u'test', 'id': 'testvid'}
@@ -77,12 +99,12 @@ class TestFormatSelection(unittest.TestCase):
 
         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')
 
         ydl = YDL({'format_limit': 'great', 'format': 'all'})
-        ydl.process_ie_result(info_dict)
+        ydl.process_ie_result(info_dict.copy())
         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')
@@ -90,44 +112,80 @@ class TestFormatSelection(unittest.TestCase):
 
         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')
 
     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'},
+            {u'format_id': u'35', u'ext': u'mp4', 'preference': 1},
+            {u'format_id': u'45', u'ext': u'webm', 'preference': 2},
+            {u'format_id': u'47', u'ext': u'webm', 'preference': 3},
+            {u'format_id': u'2', u'ext': u'flv', 'preference': 4},
         ]
         info_dict = {u'formats': formats, u'extractor': u'test'}
 
         ydl = YDL({'format': u'20/47'})
-        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'47')
 
         ydl = YDL({'format': u'20/71/worst'})
-        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'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')
 
         ydl = YDL({'format': u'webm/mp4'})
-        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'47')
 
         ydl = YDL({'format': u'3gp/40/mp4'})
-        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'35')
 
+    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
+            f2 = YoutubeIE._formats[f2id].copy()
+            f2['format_id'] = f2id
+
+            info_dict = {'formats': [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 = {'formats': [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 = {
             'extractor': 'Foo',
@@ -140,6 +198,20 @@ class TestFormatSelection(unittest.TestCase):
         self.assertEqual(test_dict['extractor'], 'Foo')
         self.assertEqual(test_dict['playlist'], 'funny videos')
 
+    def test_prepare_filename(self):
+        info = {
+            u'id': u'1234',
+            u'ext': u'mp4',
+            u'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')
+        # Replace missing fields with 'NA'
+        self.assertEqual(fname(u'%(uploader_date)s-%(id)s.%(ext)s'), u'NA-1234.mp4')
+
 
 if __name__ == '__main__':
     unittest.main()

test/test_all_urls.py

@@ -10,6 +10,7 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 from test.helper import get_testcases
 
 from youtube_dl.extractor import (
+    FacebookIE,
     gen_extractors,
     JustinTVIE,
     YoutubeIE,
@@ -87,12 +88,15 @@ class TestAllURLsMatching(unittest.TestCase):
         assertExtractId('http://www.youtube.com/watch?v=BaW_jenozKcsharePLED17F32AD9753930', 'BaW_jenozKc')
         assertExtractId('BaW_jenozKc', 'BaW_jenozKc')
 
+    def test_facebook_matching(self):
+        self.assertTrue(FacebookIE.suitable(u'https://www.facebook.com/Shiniknoh#!/photo.php?v=10153317450565268'))
+
     def test_no_duplicates(self):
         ies = gen_extractors()
         for tc in get_testcases():
             url = tc['url']
             for ie in ies:
-                if type(ie).__name__ in ['GenericIE', tc['name'] + 'IE']:
+                if type(ie).__name__ in ('GenericIE', tc['name'] + 'IE'):
                     self.assertTrue(ie.suitable(url), '%s should match URL %r' % (type(ie).__name__, url))
                 else:
                     self.assertFalse(ie.suitable(url), '%s should not match URL %r' % (type(ie).__name__, url))
@@ -106,6 +110,15 @@ class TestAllURLsMatching(unittest.TestCase):
         self.assertMatch(':colbertreport', ['ComedyCentralShows'])
         self.assertMatch(':cr', ['ComedyCentralShows'])
 
+    def test_vimeo_matching(self):
+        self.assertMatch('http://vimeo.com/channels/tributes', ['vimeo:channel'])
+        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'])
 
 if __name__ == '__main__':
     unittest.main()

test/test_download.py

@@ -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', {}))

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
@@ -12,16 +13,25 @@ from test.helper import FakeYDL
 
 
 from youtube_dl.extractor import (
+    AcademicEarthCourseIE,
     DailymotionPlaylistIE,
     DailymotionUserIE,
     VimeoChannelIE,
+    VimeoUserIE,
+    VimeoAlbumIE,
+    VimeoGroupsIE,
     UstreamChannelIE,
     SoundcloudSetIE,
     SoundcloudUserIE,
     LivestreamIE,
     NHLVideocenterIE,
     BambuserChannelIE,
-    BandcampAlbumIE
+    BandcampAlbumIE,
+    SmotriCommunityIE,
+    SmotriUserIE,
+    IviCompilationIE,
+    ImdbListIE,
+    KhanAcademyIE,
 )
 
 
@@ -35,7 +45,7 @@ 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):
@@ -43,7 +53,7 @@ class TestPlaylists(unittest.TestCase):
         ie = DailymotionUserIE(dl)
         result = ie.extract('http://www.dailymotion.com/user/generation-quoi/')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'Génération Quoi')
+        self.assertEqual(result['title'], 'Génération Quoi')
         self.assertTrue(len(result['entries']) >= 26)
 
     def test_vimeo_channel(self):
@@ -51,15 +61,39 @@ class TestPlaylists(unittest.TestCase):
         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):
+        dl = FakeYDL()
+        ie = VimeoUserIE(dl)
+        result = ie.extract('http://vimeo.com/nkistudio/videos')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['title'], 'Nki')
+        self.assertTrue(len(result['entries']) > 65)
+
+    def test_vimeo_album(self):
+        dl = FakeYDL()
+        ie = VimeoAlbumIE(dl)
+        result = ie.extract('http://vimeo.com/album/2632481')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['title'], 'Staff Favorites: November 2013')
+        self.assertTrue(len(result['entries']) > 12)
+
+    def test_vimeo_groups(self):
+        dl = FakeYDL()
+        ie = VimeoGroupsIE(dl)
+        result = ie.extract('http://vimeo.com/groups/rolexawards')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['title'], 'Rolex Awards for Enterprise')
+        self.assertTrue(len(result['entries']) > 72)
+
     def test_ustream_channel(self):
         dl = FakeYDL()
         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.assertEqual(result['id'], '5124905')
         self.assertTrue(len(result['entries']) >= 11)
 
     def test_soundcloud_set(self):
@@ -67,7 +101,7 @@ class TestPlaylists(unittest.TestCase):
         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):
@@ -75,7 +109,7 @@ 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_livestream_event(self):
@@ -83,7 +117,7 @@ class TestPlaylists(unittest.TestCase):
         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):
@@ -91,8 +125,8 @@ 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):
@@ -100,7 +134,7 @@ 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):
@@ -108,8 +142,74 @@ 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):
+        dl = FakeYDL()
+        ie = SmotriCommunityIE(dl)
+        result = ie.extract('http://smotri.com/community/video/kommuna')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'kommuna')
+        self.assertEqual(result['title'], 'КПРФ')
+        self.assertTrue(len(result['entries']) >= 4)
+        
+    def test_smotri_user(self):
+        dl = FakeYDL()
+        ie = SmotriUserIE(dl)
+        result = ie.extract('http://smotri.com/user/inspector')
+        self.assertIsPlaylist(result)
+        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('http://academicearth.org/courses/building-dynamic-websites/')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'building-dynamic-websites')
+        self.assertEqual(result['title'], '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)
+        
+    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'], 'dezhurnyi_angel')
+        self.assertEqual(result['title'], 'Дежурный ангел (2010 - 2012)')
+        self.assertTrue(len(result['entries']) >= 36)
+        
+    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'], 'dezhurnyi_angel/season2')
+        self.assertEqual(result['title'], 'Дежурный ангел (2010 - 2012) 2 сезон')
+        self.assertTrue(len(result['entries']) >= 20)
+        
+    def test_imdb_list(self):
+        dl = FakeYDL()
+        ie = ImdbListIE(dl)
+        result = ie.extract('http://www.imdb.com/list/sMjedvGDd8U')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'sMjedvGDd8U')
+        self.assertEqual(result['title'], 'Animated and Family Films')
+        self.assertTrue(len(result['entries']) >= 48)
+
+    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)
+
 
 if __name__ == '__main__':
     unittest.main()

test/test_subtitles.py

@@ -36,10 +36,6 @@ 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()
@@ -72,7 +68,7 @@ class TestYoutubeSubtitles(BaseTestSubtitles):
         self.DL.params['writesubtitles'] = True
         self.DL.params['subtitlesformat'] = 'vtt'
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '356cdc577fde0c6783b9b822e7206ff7')
+        self.assertEqual(md5(subtitles['en']), '3cb210999d3e021bd6c7f0ea751eab06')
 
     def test_youtube_list_subtitles(self):
         self.DL.expect_warning(u'Video doesn\'t have automatic captions')

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

@@ -13,19 +13,22 @@ import xml.etree.ElementTree
 
 #from youtube_dl.utils import htmlentity_transform
 from youtube_dl.utils import (
-    timeconvert,
-    sanitize_filename,
-    unescapeHTML,
-    orderedSet,
     DateRange,
-    unified_strdate,
+    encodeFilename,
     find_xpath_attr,
     get_meta_content,
-    xpath_with_ns,
-    smuggle_url,
-    unsmuggle_url,
+    orderedSet,
+    parse_duration,
+    sanitize_filename,
     shell_quote,
-    encodeFilename,
+    smuggle_url,
+    str_to_int,
+    timeconvert,
+    unescapeHTML,
+    unified_strdate,
+    unsmuggle_url,
+    url_basename,
+    xpath_with_ns,
 )
 
 if sys.version_info < (3, 0):
@@ -176,6 +179,26 @@ class TestUtil(unittest.TestCase):
         args = ['ffmpeg', '-i', encodeFilename(u'ñ€ß\'.mp4')]
         self.assertEqual(shell_quote(args), u"""ffmpeg -i 'ñ€ß'"'"'.mp4'""")
 
+    def test_str_to_int(self):
+        self.assertEqual(str_to_int('123,456'), 123456)
+        self.assertEqual(str_to_int('123.456'), 123456)
+
+    def test_url_basename(self):
+        self.assertEqual(url_basename(u'http://foo.de/'), u'')
+        self.assertEqual(url_basename(u'http://foo.de/bar/baz'), u'baz')
+        self.assertEqual(url_basename(u'http://foo.de/bar/baz?x=y'), u'baz')
+        self.assertEqual(url_basename(u'http://foo.de/bar/baz#x=y'), u'baz')
+        self.assertEqual(url_basename(u'http://foo.de/bar/baz/'), u'baz')
+        self.assertEqual(
+            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('x:y'), None)
 
 if __name__ == '__main__':
     unittest.main()

test/test_write_info_json.py

@@ -33,6 +33,7 @@ TEST_ID = 'BaW_jenozKc'
 INFO_JSON_FILE = TEST_ID + '.info.json'
 DESCRIPTION_FILE = TEST_ID + '.mp4.description'
 EXPECTED_DESCRIPTION = u'''test chars:  "'/\ä↭𝕐
+test URL: https://github.com/rg3/youtube-dl/issues/1892
 
 This is a test video for youtube-dl.
 

test/test_youtube_lists.py

@@ -15,6 +15,7 @@ from youtube_dl.extractor import (
     YoutubeIE,
     YoutubeChannelIE,
     YoutubeShowIE,
+    YoutubeTopListIE,
 )
 
 
@@ -107,5 +108,21 @@ class TestYoutubeLists(unittest.TestCase):
         result = ie.extract('http://www.youtube.com/show/airdisasters')
         self.assertTrue(len(result) >= 3)
 
+    def test_youtube_mix(self):
+        dl = FakeYDL()
+        ie = YoutubePlaylistIE(dl)
+        result = ie.extract('http://www.youtube.com/watch?v=lLJf9qJHR3E&list=RDrjFaenf1T-Y')
+        entries = result['entries']
+        self.assertTrue(len(entries) >= 20)
+        original_video = entries[0]
+        self.assertEqual(original_video['id'], 'rjFaenf1T-Y')
+
+    def test_youtube_toplist(self):
+        dl = FakeYDL()
+        ie = YoutubeTopListIE(dl)
+        result = ie.extract('yttoplist:music:Top Tracks')
+        entries = result['entries']
+        self.assertTrue(len(entries) >= 5)
+
 if __name__ == '__main__':
     unittest.main()

youtube_dl/FileDownloader.py

@@ -1,683 +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(self, percent, data_len_str, speed, eta):
-        """Report download progress."""
-        if self.params.get('noprogress', False):
-            return
-        clear_line = (u'\x1b[K' if sys.stderr.isatty() and os.name != 'nt' else u'')
-        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)
-        if self.params.get('progress_with_newline', False):
-            self.to_screen(u'[download] %s of %s at %s ETA %s' %
-                (percent_str, data_len_str, speed_str, eta_str))
-        else:
-            self.to_screen(u'\r%s[download] %s of %s at %s ETA %s' %
-                (clear_line, percent_str, data_len_str, speed_str, eta_str), skip_eol=True)
-        self.to_console_title(u'youtube-dl - %s of %s at %s ETA %s' %
-                (percent_str.strip(), data_len_str.strip(), speed_str.strip(), eta_str.strip()))
-
-    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 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:
-            clear_line = (u'\x1b[K' if sys.stderr.isatty() and os.name != 'nt' else u'')
-            self.to_screen(u'\r%s[download] 100%% of %s in %s' %
-                (clear_line, data_len_str, self.format_seconds(tot_time)))
-
-    def _download_with_rtmpdump(self, filename, url, player_url, page_url, play_path, tc_url, live):
-        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,
-                    })
-                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']
-        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))
-
-        # 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/__init__.py

@@ -36,18 +36,21 @@ __authors__  = (
     'Marcin Cieślak',
     'Anton Larionov',
     'Takuya Tsuchida',
+    'Sergey M.',
+    'Michael Orlitzky',
+    'Chris Gahan',
 )
 
 __license__ = 'Public Domain'
 
 import codecs
 import getpass
+import locale
 import optparse
 import os
 import random
 import re
 import shlex
-import subprocess
 import sys
 
 
@@ -55,12 +58,13 @@ from .utils import (
     compat_print,
     DateRange,
     decodeOption,
-    determine_ext,
+    get_term_width,
     DownloadError,
     get_cachedir,
     MaxDownloadsReached,
     preferredencoding,
     SameFileError,
+    setproctitle,
     std_headers,
     write_string,
 )
@@ -71,20 +75,21 @@ from .FileDownloader import (
 from .extractor import gen_extractors
 from .version import __version__
 from .YoutubeDL import YoutubeDL
-from .PostProcessor import (
+from .postprocessor import (
     FFmpegMetadataPP,
     FFmpegVideoConvertor,
     FFmpegExtractAudioPP,
     FFmpegEmbedSubtitlePP,
+    XAttrMetadataPP,
 )
 
 
 def parseOpts(overrideArguments=None):
-    def _readOptions(filename_bytes):
+    def _readOptions(filename_bytes, default=[]):
         try:
             optionf = open(filename_bytes)
         except IOError:
-            return [] # silently skip if file is not present
+            return default  # silently skip if file is not present
         try:
             res = []
             for l in optionf:
@@ -112,19 +117,6 @@ def parseOpts(overrideArguments=None):
     def _comma_separated_values_options_callback(option, opt_str, value, parser):
         setattr(parser.values, option.dest, value.split(','))
 
-    def _find_term_columns():
-        columns = os.environ.get('COLUMNS', None)
-        if columns:
-            return int(columns)
-
-        try:
-            sp = subprocess.Popen(['stty', 'size'], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
-            out,err = sp.communicate()
-            return int(out.split()[1])
-        except:
-            pass
-        return None
-
     def _hide_login_info(opts):
         opts = list(opts)
         for private_opt in ['-p', '--password', '-u', '--username', '--video-password']:
@@ -139,7 +131,7 @@ def parseOpts(overrideArguments=None):
     max_help_position = 80
 
     # No need to wrap help messages if we're on a wide console
-    columns = _find_term_columns()
+    columns = get_term_width()
     if columns: max_width = columns
 
     fmt = optparse.IndentedHelpFormatter(width=max_width, max_help_position=max_help_position)
@@ -190,20 +182,32 @@ def parseOpts(overrideArguments=None):
     general.add_option('--extractor-descriptions',
             action='store_true', dest='list_extractor_descriptions',
             help='Output descriptions of all supported extractors', default=False)
-    general.add_option('--proxy', dest='proxy', default=None, help='Use the specified HTTP/HTTPS proxy', metavar='URL')
+    general.add_option(
+        '--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(
         '--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=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')
 
 
-    selection.add_option('--playlist-start',
-            dest='playliststart', metavar='NUMBER', help='playlist video to start at (default is %default)', default=1)
-    selection.add_option('--playlist-end',
-            dest='playlistend', metavar='NUMBER', help='playlist video to end at (default is last)', default=-1)
+    selection.add_option(
+        '--playlist-start',
+        dest='playliststart', metavar='NUMBER', default=1, type=int,
+        help='playlist video to start at (default is %default)')
+    selection.add_option(
+        '--playlist-end',
+        dest='playlistend', metavar='NUMBER', default=None, type=int,
+        help='playlist video to end at (default is last)')
     selection.add_option('--match-title', dest='matchtitle', metavar='REGEX',help='download only matching titles (regex or caseless sub-string)')
     selection.add_option('--reject-title', dest='rejecttitle', metavar='REGEX',help='skip download for matching titles (regex or caseless sub-string)')
     selection.add_option('--max-downloads', metavar='NUMBER',
@@ -212,15 +216,27 @@ 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,
+        help="Do not download any videos with less than COUNT views",)
+    selection.add_option(
+        '--max-views', metavar='COUNT', dest='max_views',
+        default=None, type=int,
+        help="Do not download any videos with more than COUNT views",)
     selection.add_option('--no-playlist', action='store_true', dest='noplaylist', help='download only the currently playing video', default=False)
     selection.add_option('--age-limit', metavar='YEARS', dest='age_limit',
                          help='download only videos suitable for the given age',
                          default=None, type=int)
     selection.add_option('--download-archive', metavar='FILE',
                          dest='download_archive',
-                         help='Download only videos not present in the archive file. Record all downloaded videos in it.')
+                         help='Download only videos not listed in the archive file. Record the IDs of all downloaded videos in it.')
 
 
     authentication.add_option('-u', '--username',
@@ -230,12 +246,12 @@ 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, specifiy the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported')
+            help='video format code, specify the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported')
     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',
@@ -243,7 +259,7 @@ 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',
@@ -294,6 +310,9 @@ def parseOpts(overrideArguments=None):
     verbosity.add_option('--get-description',
             action='store_true', dest='getdescription',
             help='simulate, quiet but print video description', default=False)
+    verbosity.add_option('--get-duration',
+            action='store_true', dest='getduration',
+            help='simulate, quiet but print video length', default=False)
     verbosity.add_option('--get-filename',
             action='store_true', dest='getfilename',
             help='simulate, quiet but print output filename', default=False)
@@ -314,14 +333,16 @@ 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 pages to files in the current directory')
+            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=optparse.SUPPRESS_HELP)
 
     filesystem.add_option('-t', '--title',
             action='store_true', dest='usetitle', help='use title in file name (default)', default=False)
@@ -338,11 +359,11 @@ 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. '
                   '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\' .'))
@@ -354,6 +375,9 @@ def parseOpts(overrideArguments=None):
             help='Restrict filenames to only ASCII characters, and avoid "&" and spaces in filenames', default=False)
     filesystem.add_option('-a', '--batch-file',
             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)')
     filesystem.add_option('-w', '--no-overwrites',
             action='store_true', dest='nooverwrites', help='do not overwrite files', default=False)
     filesystem.add_option('-c', '--continue',
@@ -397,7 +421,13 @@ def parseOpts(overrideArguments=None):
     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('--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)
@@ -415,6 +445,8 @@ 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')
@@ -424,8 +456,31 @@ def parseOpts(overrideArguments=None):
             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')
-        systemConf = _readOptions('/etc/youtube-dl.conf')
-        userConf = _readOptions(userConfFile)
+        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:]
         argv = systemConf + userConf + commandLineConf
         opts, args = parser.parse_args(argv)
@@ -433,15 +488,20 @@ def parseOpts(overrideArguments=None):
             write_string(u'[debug] System config: ' + repr(_hide_login_info(systemConf)) + '\n')
             write_string(u'[debug] User config: ' + repr(_hide_login_info(userConf)) + '\n')
             write_string(u'[debug] Command-line args: ' + repr(_hide_login_info(commandLineConf)) + '\n')
+            write_string(u'[debug] Encodings: locale %r, fs %r, out %r, pref: %r\n' %
+                         (locale.getpreferredencoding(), sys.getfilesystemencoding(), sys.stdout.encoding, preferredencoding()))
 
     return parser, opts, args
 
+
 def _real_main(argv=None):
     # Compatibility fixes for Windows
     if sys.platform == 'win32':
         # https://github.com/rg3/youtube-dl/issues/820
         codecs.register(lambda name: codecs.lookup('utf-8') if name == 'cp65001' else None)
 
+    setproctitle(u'youtube-dl')
+
     parser, opts, args = parseOpts(argv)
 
     # Set user agent
@@ -474,6 +534,8 @@ def _real_main(argv=None):
             sys.exit(u'ERROR: batch file could not be read')
     all_urls = batchurls + 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()
 
@@ -481,7 +543,6 @@ def _real_main(argv=None):
         for ie in sorted(extractors, key=lambda ie: ie.IE_NAME.lower()):
             compat_print(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else ''))
             matchedUrls = [url for url in all_urls if ie.suitable(url)]
-            all_urls = [url for url in all_urls if url not in matchedUrls]
             for mu in matchedUrls:
                 compat_print(u'  ' + mu)
         sys.exit(0)
@@ -504,7 +565,7 @@ 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:
@@ -536,18 +597,10 @@ def _real_main(argv=None):
         if numeric_buffersize is None:
             parser.error(u'invalid buffer size specified')
         opts.buffersize = numeric_buffersize
-    try:
-        opts.playliststart = int(opts.playliststart)
-        if opts.playliststart <= 0:
-            raise ValueError(u'Playlist start must be positive')
-    except (TypeError, ValueError):
-        parser.error(u'invalid playlist start number specified')
-    try:
-        opts.playlistend = int(opts.playlistend)
-        if opts.playlistend != -1 and (opts.playlistend <= 0 or opts.playlistend < opts.playliststart):
-            raise ValueError(u'Playlist end must be greater than playlist start')
-    except (TypeError, ValueError):
-        parser.error(u'invalid playlist end number specified')
+    if opts.playliststart <= 0:
+        raise ValueError(u'Playlist start must be positive')
+    if opts.playlistend not in (-1, None) and opts.playlistend < opts.playliststart:
+        raise ValueError(u'Playlist end must be greater than playlist start')
     if opts.extractaudio:
         if opts.audioformat not in ['best', 'aac', 'mp3', 'm4a', 'opus', 'vorbis', 'wav']:
             parser.error(u'invalid audio format specified')
@@ -580,27 +633,31 @@ def _real_main(argv=None):
             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')
-    if '%(ext)s' not in outtmpl and opts.extractaudio:
+    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 "%%(ext)s" instead of %r' %
-                     determine_ext(outtmpl, u''))
+                     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,
         'username': opts.username,
         'password': opts.password,
         'videopassword': opts.videopassword,
-        'quiet': (opts.quiet or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.dumpjson),
+        'quiet': (opts.quiet or any_printing),
         'forceurl': opts.geturl,
         'forcetitle': opts.gettitle,
         'forceid': opts.getid,
         'forcethumbnail': opts.getthumbnail,
         'forcedescription': opts.getdescription,
+        'forceduration': opts.getduration,
         'forcefilename': opts.getfilename,
         'forceformat': opts.getformat,
         'forcejson': opts.dumpjson,
         'simulate': opts.simulate,
-        'skip_download': (opts.skip_download or opts.simulate or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.dumpjson),
+        'skip_download': (opts.skip_download or opts.simulate or any_printing),
         'format': opts.format,
         'format_limit': opts.format_limit,
         'listformats': opts.listformats,
@@ -644,13 +701,20 @@ def _real_main(argv=None):
         'keepvideo': opts.keepvideo,
         'min_filesize': opts.min_filesize,
         'max_filesize': opts.max_filesize,
+        'min_views': opts.min_views,
+        'max_views': opts.max_views,
         'daterange': date,
         '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,
+        'proxy': opts.proxy,
+        'socket_timeout': opts.socket_timeout,
+        'bidi_workaround': opts.bidi_workaround,
+        'debug_printtraffic': opts.debug_printtraffic,
+        'prefer_ffmpeg': opts.prefer_ffmpeg,
     }
 
     with YoutubeDL(ydl_opts) as ydl:
@@ -667,20 +731,25 @@ 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())
 
         # Update version
         if opts.update_self:
             update_self(ydl.to_screen, opts.verbose)
 
         # Maybe do nothing
-        if len(all_urls) < 1:
+        if (len(all_urls) < 1) and (opts.load_info_filename is None):
             if not opts.update_self:
                 parser.error(u'you must provide at least one URL')
             else:
                 sys.exit()
 
         try:
-            retcode = ydl.download(all_urls)
+            if opts.load_info_filename is not None:
+                retcode = ydl.download_with_info_file(opts.load_info_filename)
+            else:
+                retcode = ydl.download(all_urls)
         except MaxDownloadsReached:
             ydl.to_screen(u'--max-download limit reached, aborting.')
             retcode = 101

youtube_dl/aes.py

@@ -1,4 +1,4 @@
-__all__ = ['aes_encrypt', 'key_expansion', 'aes_ctr_decrypt', 'aes_decrypt_text']
+__all__ = ['aes_encrypt', 'key_expansion', 'aes_ctr_decrypt', 'aes_cbc_decrypt', 'aes_decrypt_text']
 
 import base64
 from math import ceil
@@ -32,6 +32,31 @@ def aes_ctr_decrypt(data, key, counter):
     
     return decrypted_data
 
+def aes_cbc_decrypt(data, key, iv):
+    """
+    Decrypt with aes in CBC mode
+    
+    @param {int[]} data        cipher
+    @param {int[]} key         16/24/32-Byte cipher key
+    @param {int[]} iv          16-Byte IV
+    @returns {int[]}           decrypted data
+    """
+    expanded_key = key_expansion(key)
+    block_count = int(ceil(float(len(data)) / BLOCK_SIZE_BYTES))
+    
+    decrypted_data=[]
+    previous_cipher_block = iv
+    for i in range(block_count):
+        block = data[i*BLOCK_SIZE_BYTES : (i+1)*BLOCK_SIZE_BYTES]
+        block += [0]*(BLOCK_SIZE_BYTES - len(block))
+        
+        decrypted_block = aes_decrypt(block, expanded_key)
+        decrypted_data += xor(decrypted_block, previous_cipher_block)
+        previous_cipher_block = block
+    decrypted_data = decrypted_data[:len(data)]
+    
+    return decrypted_data
+
 def key_expansion(data):
     """
     Generate key schedule
@@ -75,7 +100,7 @@ def aes_encrypt(data, expanded_key):
     @returns {int[]}             16-Byte cipher
     """
     rounds = len(expanded_key) // BLOCK_SIZE_BYTES - 1
-    
+
     data = xor(data, expanded_key[:BLOCK_SIZE_BYTES])
     for i in range(1, rounds+1):
         data = sub_bytes(data)
@@ -83,6 +108,26 @@ def aes_encrypt(data, expanded_key):
         if i != rounds:
             data = mix_columns(data)
         data = xor(data, expanded_key[i*BLOCK_SIZE_BYTES : (i+1)*BLOCK_SIZE_BYTES])
+
+    return data
+
+def aes_decrypt(data, expanded_key):
+    """
+    Decrypt one block with aes
+    
+    @param {int[]} data          16-Byte cipher
+    @param {int[]} expanded_key  176/208/240-Byte expanded key
+    @returns {int[]}             16-Byte state
+    """
+    rounds = len(expanded_key) // BLOCK_SIZE_BYTES - 1
+    
+    for i in range(rounds, 0, -1):
+        data = xor(data, expanded_key[i*BLOCK_SIZE_BYTES : (i+1)*BLOCK_SIZE_BYTES])
+        if i != rounds:
+            data = mix_columns_inv(data)
+        data = shift_rows_inv(data)
+        data = sub_bytes_inv(data)
+    data = xor(data, expanded_key[:BLOCK_SIZE_BYTES])
     
     return data
 
@@ -139,14 +184,69 @@ SBOX = (0x63, 0x7C, 0x77, 0x7B, 0xF2, 0x6B, 0x6F, 0xC5, 0x30, 0x01, 0x67, 0x2B,
         0x70, 0x3E, 0xB5, 0x66, 0x48, 0x03, 0xF6, 0x0E, 0x61, 0x35, 0x57, 0xB9, 0x86, 0xC1, 0x1D, 0x9E,
         0xE1, 0xF8, 0x98, 0x11, 0x69, 0xD9, 0x8E, 0x94, 0x9B, 0x1E, 0x87, 0xE9, 0xCE, 0x55, 0x28, 0xDF,
         0x8C, 0xA1, 0x89, 0x0D, 0xBF, 0xE6, 0x42, 0x68, 0x41, 0x99, 0x2D, 0x0F, 0xB0, 0x54, 0xBB, 0x16)
-MIX_COLUMN_MATRIX = ((2,3,1,1),
-                     (1,2,3,1),
-                     (1,1,2,3),
-                     (3,1,1,2))
+SBOX_INV = (0x52, 0x09, 0x6a, 0xd5, 0x30, 0x36, 0xa5, 0x38, 0xbf, 0x40, 0xa3, 0x9e, 0x81, 0xf3, 0xd7, 0xfb,
+            0x7c, 0xe3, 0x39, 0x82, 0x9b, 0x2f, 0xff, 0x87, 0x34, 0x8e, 0x43, 0x44, 0xc4, 0xde, 0xe9, 0xcb,
+            0x54, 0x7b, 0x94, 0x32, 0xa6, 0xc2, 0x23, 0x3d, 0xee, 0x4c, 0x95, 0x0b, 0x42, 0xfa, 0xc3, 0x4e,
+            0x08, 0x2e, 0xa1, 0x66, 0x28, 0xd9, 0x24, 0xb2, 0x76, 0x5b, 0xa2, 0x49, 0x6d, 0x8b, 0xd1, 0x25,
+            0x72, 0xf8, 0xf6, 0x64, 0x86, 0x68, 0x98, 0x16, 0xd4, 0xa4, 0x5c, 0xcc, 0x5d, 0x65, 0xb6, 0x92,
+            0x6c, 0x70, 0x48, 0x50, 0xfd, 0xed, 0xb9, 0xda, 0x5e, 0x15, 0x46, 0x57, 0xa7, 0x8d, 0x9d, 0x84,
+            0x90, 0xd8, 0xab, 0x00, 0x8c, 0xbc, 0xd3, 0x0a, 0xf7, 0xe4, 0x58, 0x05, 0xb8, 0xb3, 0x45, 0x06,
+            0xd0, 0x2c, 0x1e, 0x8f, 0xca, 0x3f, 0x0f, 0x02, 0xc1, 0xaf, 0xbd, 0x03, 0x01, 0x13, 0x8a, 0x6b,
+            0x3a, 0x91, 0x11, 0x41, 0x4f, 0x67, 0xdc, 0xea, 0x97, 0xf2, 0xcf, 0xce, 0xf0, 0xb4, 0xe6, 0x73,
+            0x96, 0xac, 0x74, 0x22, 0xe7, 0xad, 0x35, 0x85, 0xe2, 0xf9, 0x37, 0xe8, 0x1c, 0x75, 0xdf, 0x6e,
+            0x47, 0xf1, 0x1a, 0x71, 0x1d, 0x29, 0xc5, 0x89, 0x6f, 0xb7, 0x62, 0x0e, 0xaa, 0x18, 0xbe, 0x1b,
+            0xfc, 0x56, 0x3e, 0x4b, 0xc6, 0xd2, 0x79, 0x20, 0x9a, 0xdb, 0xc0, 0xfe, 0x78, 0xcd, 0x5a, 0xf4,
+            0x1f, 0xdd, 0xa8, 0x33, 0x88, 0x07, 0xc7, 0x31, 0xb1, 0x12, 0x10, 0x59, 0x27, 0x80, 0xec, 0x5f,
+            0x60, 0x51, 0x7f, 0xa9, 0x19, 0xb5, 0x4a, 0x0d, 0x2d, 0xe5, 0x7a, 0x9f, 0x93, 0xc9, 0x9c, 0xef,
+            0xa0, 0xe0, 0x3b, 0x4d, 0xae, 0x2a, 0xf5, 0xb0, 0xc8, 0xeb, 0xbb, 0x3c, 0x83, 0x53, 0x99, 0x61,
+            0x17, 0x2b, 0x04, 0x7e, 0xba, 0x77, 0xd6, 0x26, 0xe1, 0x69, 0x14, 0x63, 0x55, 0x21, 0x0c, 0x7d)
+MIX_COLUMN_MATRIX = ((0x2,0x3,0x1,0x1),
+                     (0x1,0x2,0x3,0x1),
+                     (0x1,0x1,0x2,0x3),
+                     (0x3,0x1,0x1,0x2))
+MIX_COLUMN_MATRIX_INV = ((0xE,0xB,0xD,0x9),
+                         (0x9,0xE,0xB,0xD),
+                         (0xD,0x9,0xE,0xB),
+                         (0xB,0xD,0x9,0xE))
+RIJNDAEL_EXP_TABLE = (0x01, 0x03, 0x05, 0x0F, 0x11, 0x33, 0x55, 0xFF, 0x1A, 0x2E, 0x72, 0x96, 0xA1, 0xF8, 0x13, 0x35,
+                      0x5F, 0xE1, 0x38, 0x48, 0xD8, 0x73, 0x95, 0xA4, 0xF7, 0x02, 0x06, 0x0A, 0x1E, 0x22, 0x66, 0xAA,
+                      0xE5, 0x34, 0x5C, 0xE4, 0x37, 0x59, 0xEB, 0x26, 0x6A, 0xBE, 0xD9, 0x70, 0x90, 0xAB, 0xE6, 0x31,
+                      0x53, 0xF5, 0x04, 0x0C, 0x14, 0x3C, 0x44, 0xCC, 0x4F, 0xD1, 0x68, 0xB8, 0xD3, 0x6E, 0xB2, 0xCD,
+                      0x4C, 0xD4, 0x67, 0xA9, 0xE0, 0x3B, 0x4D, 0xD7, 0x62, 0xA6, 0xF1, 0x08, 0x18, 0x28, 0x78, 0x88,
+                      0x83, 0x9E, 0xB9, 0xD0, 0x6B, 0xBD, 0xDC, 0x7F, 0x81, 0x98, 0xB3, 0xCE, 0x49, 0xDB, 0x76, 0x9A,
+                      0xB5, 0xC4, 0x57, 0xF9, 0x10, 0x30, 0x50, 0xF0, 0x0B, 0x1D, 0x27, 0x69, 0xBB, 0xD6, 0x61, 0xA3,
+                      0xFE, 0x19, 0x2B, 0x7D, 0x87, 0x92, 0xAD, 0xEC, 0x2F, 0x71, 0x93, 0xAE, 0xE9, 0x20, 0x60, 0xA0,
+                      0xFB, 0x16, 0x3A, 0x4E, 0xD2, 0x6D, 0xB7, 0xC2, 0x5D, 0xE7, 0x32, 0x56, 0xFA, 0x15, 0x3F, 0x41,
+                      0xC3, 0x5E, 0xE2, 0x3D, 0x47, 0xC9, 0x40, 0xC0, 0x5B, 0xED, 0x2C, 0x74, 0x9C, 0xBF, 0xDA, 0x75,
+                      0x9F, 0xBA, 0xD5, 0x64, 0xAC, 0xEF, 0x2A, 0x7E, 0x82, 0x9D, 0xBC, 0xDF, 0x7A, 0x8E, 0x89, 0x80,
+                      0x9B, 0xB6, 0xC1, 0x58, 0xE8, 0x23, 0x65, 0xAF, 0xEA, 0x25, 0x6F, 0xB1, 0xC8, 0x43, 0xC5, 0x54,
+                      0xFC, 0x1F, 0x21, 0x63, 0xA5, 0xF4, 0x07, 0x09, 0x1B, 0x2D, 0x77, 0x99, 0xB0, 0xCB, 0x46, 0xCA,
+                      0x45, 0xCF, 0x4A, 0xDE, 0x79, 0x8B, 0x86, 0x91, 0xA8, 0xE3, 0x3E, 0x42, 0xC6, 0x51, 0xF3, 0x0E,
+                      0x12, 0x36, 0x5A, 0xEE, 0x29, 0x7B, 0x8D, 0x8C, 0x8F, 0x8A, 0x85, 0x94, 0xA7, 0xF2, 0x0D, 0x17,
+                      0x39, 0x4B, 0xDD, 0x7C, 0x84, 0x97, 0xA2, 0xFD, 0x1C, 0x24, 0x6C, 0xB4, 0xC7, 0x52, 0xF6, 0x01)
+RIJNDAEL_LOG_TABLE = (0x00, 0x00, 0x19, 0x01, 0x32, 0x02, 0x1a, 0xc6, 0x4b, 0xc7, 0x1b, 0x68, 0x33, 0xee, 0xdf, 0x03,
+                      0x64, 0x04, 0xe0, 0x0e, 0x34, 0x8d, 0x81, 0xef, 0x4c, 0x71, 0x08, 0xc8, 0xf8, 0x69, 0x1c, 0xc1,
+                      0x7d, 0xc2, 0x1d, 0xb5, 0xf9, 0xb9, 0x27, 0x6a, 0x4d, 0xe4, 0xa6, 0x72, 0x9a, 0xc9, 0x09, 0x78,
+                      0x65, 0x2f, 0x8a, 0x05, 0x21, 0x0f, 0xe1, 0x24, 0x12, 0xf0, 0x82, 0x45, 0x35, 0x93, 0xda, 0x8e,
+                      0x96, 0x8f, 0xdb, 0xbd, 0x36, 0xd0, 0xce, 0x94, 0x13, 0x5c, 0xd2, 0xf1, 0x40, 0x46, 0x83, 0x38,
+                      0x66, 0xdd, 0xfd, 0x30, 0xbf, 0x06, 0x8b, 0x62, 0xb3, 0x25, 0xe2, 0x98, 0x22, 0x88, 0x91, 0x10,
+                      0x7e, 0x6e, 0x48, 0xc3, 0xa3, 0xb6, 0x1e, 0x42, 0x3a, 0x6b, 0x28, 0x54, 0xfa, 0x85, 0x3d, 0xba,
+                      0x2b, 0x79, 0x0a, 0x15, 0x9b, 0x9f, 0x5e, 0xca, 0x4e, 0xd4, 0xac, 0xe5, 0xf3, 0x73, 0xa7, 0x57,
+                      0xaf, 0x58, 0xa8, 0x50, 0xf4, 0xea, 0xd6, 0x74, 0x4f, 0xae, 0xe9, 0xd5, 0xe7, 0xe6, 0xad, 0xe8,
+                      0x2c, 0xd7, 0x75, 0x7a, 0xeb, 0x16, 0x0b, 0xf5, 0x59, 0xcb, 0x5f, 0xb0, 0x9c, 0xa9, 0x51, 0xa0,
+                      0x7f, 0x0c, 0xf6, 0x6f, 0x17, 0xc4, 0x49, 0xec, 0xd8, 0x43, 0x1f, 0x2d, 0xa4, 0x76, 0x7b, 0xb7,
+                      0xcc, 0xbb, 0x3e, 0x5a, 0xfb, 0x60, 0xb1, 0x86, 0x3b, 0x52, 0xa1, 0x6c, 0xaa, 0x55, 0x29, 0x9d,
+                      0x97, 0xb2, 0x87, 0x90, 0x61, 0xbe, 0xdc, 0xfc, 0xbc, 0x95, 0xcf, 0xcd, 0x37, 0x3f, 0x5b, 0xd1,
+                      0x53, 0x39, 0x84, 0x3c, 0x41, 0xa2, 0x6d, 0x47, 0x14, 0x2a, 0x9e, 0x5d, 0x56, 0xf2, 0xd3, 0xab,
+                      0x44, 0x11, 0x92, 0xd9, 0x23, 0x20, 0x2e, 0x89, 0xb4, 0x7c, 0xb8, 0x26, 0x77, 0x99, 0xe3, 0xa5,
+                      0x67, 0x4a, 0xed, 0xde, 0xc5, 0x31, 0xfe, 0x18, 0x0d, 0x63, 0x8c, 0x80, 0xc0, 0xf7, 0x70, 0x07)
 
 def sub_bytes(data):
     return [SBOX[x] for x in data]
 
+def sub_bytes_inv(data):
+    return [SBOX_INV[x] for x in data]
+
 def rotate(data):
     return data[1:] + [data[0]]
 
@@ -160,30 +260,31 @@ def key_schedule_core(data, rcon_iteration):
 def xor(data1, data2):
     return [x^y for x, y in zip(data1, data2)]
 
-def mix_column(data):
+def rijndael_mul(a, b):
+    if(a==0 or b==0):
+        return 0
+    return RIJNDAEL_EXP_TABLE[(RIJNDAEL_LOG_TABLE[a] + RIJNDAEL_LOG_TABLE[b]) % 0xFF]
+
+def mix_column(data, matrix):
     data_mixed = []
     for row in range(4):
         mixed = 0
         for column in range(4):
-            addend = data[column]
-            if MIX_COLUMN_MATRIX[row][column] in (2,3):
-                addend <<= 1
-                if addend > 0xff:
-                    addend &= 0xff
-                    addend ^= 0x1b
-                if MIX_COLUMN_MATRIX[row][column] == 3:
-                    addend ^= data[column]
-            mixed ^= addend & 0xff
+            # xor is (+) and (-)
+            mixed ^= rijndael_mul(data[column], matrix[row][column])
         data_mixed.append(mixed)
     return data_mixed
 
-def mix_columns(data):
+def mix_columns(data, matrix=MIX_COLUMN_MATRIX):
     data_mixed = []
     for i in range(4):
         column = data[i*4 : (i+1)*4]
-        data_mixed += mix_column(column)
+        data_mixed += mix_column(column, matrix)
     return data_mixed
 
+def mix_columns_inv(data):
+    return mix_columns(data, MIX_COLUMN_MATRIX_INV)
+
 def shift_rows(data):
     data_shifted = []
     for column in range(4):
@@ -191,6 +292,13 @@ def shift_rows(data):
             data_shifted.append( data[((column + row) & 0b11) * 4 + row] )
     return data_shifted
 
+def shift_rows_inv(data):
+    data_shifted = []
+    for column in range(4):
+        for row in range(4):
+            data_shifted.append( data[((column - row) & 0b11) * 4 + row] )
+    return data_shifted
+
 def inc(data):
     data = data[:] # copy
     for i in range(len(data)-1,-1,-1):

youtube_dl/downloader/__init__.py

@@ -0,0 +1,23 @@
+from .common import FileDownloader
+from .hls import HlsFD
+from .http import HttpFD
+from .mplayer import MplayerFD
+from .rtmp import RtmpFD
+
+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']
+
+    if url.startswith('rtmp'):
+        return RtmpFD
+    if determine_ext(url) == u'm3u8':
+        return HlsFD
+    if url.startswith('mms') or url.startswith('rtsp'):
+        return MplayerFD
+    else:
+        return HttpFD
+

youtube_dl/downloader/common.py

@@ -0,0 +1,317 @@
+import os
+import re
+import sys
+import time
+
+from ..utils import (
+    encodeFilename,
+    timeconvert,
+    format_bytes,
+)
+
+
+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' % 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/hls.py

@@ -0,0 +1,44 @@
+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', 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' % (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 +1,186 @@
+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):
+    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']
+        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:
+                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

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 +1,178 @@
+import os
+import re
+import subprocess
+import sys
+import time
+
+from .common import FileDownloader
+from ..utils import (
+    encodeFilename,
+    format_bytes,
+)
+
+
+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 = 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
+
+        url = info_dict['url']
+        player_url = info_dict.get('player_url', None)
+        page_url = info_dict.get('page_url', None)
+        play_path = info_dict.get('play_path', None)
+        tc_url = info_dict.get('tc_url', 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(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

youtube_dl/extractor/__init__.py

@@ -1,6 +1,8 @@
-from .appletrailers import AppleTrailersIE
+from .academicearth import AcademicEarthCourseIE
 from .addanime import AddAnimeIE
 from .anitube import AnitubeIE
+from .aparat import AparatIE
+from .appletrailers import AppleTrailersIE
 from .archiveorg import ArchiveOrgIE
 from .ard import ARDIE
 from .arte import (
@@ -8,10 +10,12 @@ from .arte import (
     ArteTVPlus7IE,
     ArteTVCreativeIE,
     ArteTVFutureIE,
+    ArteTVDDCIE,
 )
 from .auengine import AUEngineIE
 from .bambuser import BambuserIE, BambuserChannelIE
 from .bandcamp import BandcampIE, BandcampAlbumIE
+from .blinkx import BlinkxIE
 from .bliptv import BlipTVIE, BlipTVUserIE
 from .bloomberg import BloombergIE
 from .breakcom import BreakIE
@@ -19,13 +23,18 @@ from .brightcove import BrightcoveIE
 from .c56 import C56IE
 from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
+from .cbs import CBSIE
+from .channel9 import Channel9IE
 from .cinemassacre import CinemassacreIE
 from .clipfish import ClipfishIE
+from .clipsyndicate import ClipsyndicateIE
+from .cmt import CMTIE
 from .cnn import CNNIE
 from .collegehumor import CollegeHumorIE
 from .comedycentral import ComedyCentralIE, ComedyCentralShowsIE
 from .condenast import CondeNastIE
 from .criterion import CriterionIE
+from .crunchyroll import CrunchyrollIE
 from .cspan import CSpanIE
 from .d8 import D8IE
 from .dailymotion import (
@@ -55,8 +64,9 @@ from .flickr import FlickrIE
 from .francetv import (
     PluzzIE,
     FranceTvInfoIE,
-    France2IE,
-    GenerationQuoiIE
+    FranceTVIE,
+    GenerationQuoiIE,
+    CultureboxIE,
 )
 from .freesound import FreesoundIE
 from .funnyordie import FunnyOrDieIE
@@ -71,23 +81,40 @@ from .hotnewhiphop import HotNewHipHopIE
 from .howcast import HowcastIE
 from .hypem import HypemIE
 from .ign import IGNIE, OneUPIE
+from .imdb import (
+    ImdbIE,
+    ImdbListIE
+)
 from .ina import InaIE
 from .infoq import InfoQIE
 from .instagram import InstagramIE
 from .internetvideoarchive import InternetVideoArchiveIE
+from .ivi import (
+    IviIE,
+    IviCompilationIE
+)
 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 .liveleak import LiveLeakIE
 from .livestream import LivestreamIE, LivestreamOriginalIE
+from .lynda import (
+    LyndaIE,
+    LyndaCourseIE
+)
+from .macgamestore import MacGameStoreIE
+from .mdr import MDRIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
 from .mit import TechTVMITIE, MITIE
 from .mixcloud import MixcloudIE
+from .mpora import MporaIE
 from .mofosex import MofosexIE
 from .mtv import MTVIE
 from .muzu import MuzuTVIE
@@ -97,16 +124,23 @@ from .myvideo import MyVideoIE
 from .naver import NaverIE
 from .nba import NBAIE
 from .nbc import NBCNewsIE
+from .ndtv import NDTVIE
 from .newgrounds import NewgroundsIE
 from .nhl import NHLIE, NHLVideocenterIE
 from .niconico import NiconicoIE
+from .ninegag import NineGagIE
+from .novamov import NovamovIE
 from .nowvideo import NowVideoIE
 from .ooyala import OoyalaIE
 from .orf import ORFIE
 from .pbs import PBSIE
 from .photobucket import PhotobucketIE
+from .podomatic import PodomaticIE
+from .pornhd import PornHdIE
 from .pornhub import PornHubIE
 from .pornotube import PornotubeIE
+from .pyvideo import PyvideoIE
+from .radiofrance import RadioFranceIE
 from .rbmaradio import RBMARadioIE
 from .redtube import RedTubeIE
 from .ringtv import RingTVIE
@@ -118,6 +152,12 @@ from .rutube import RutubeIE
 from .sina import SinaIE
 from .slashdot import SlashdotIE
 from .slideshare import SlideshareIE
+from .smotri import (
+    SmotriIE,
+    SmotriCommunityIE,
+    SmotriUserIE,
+    SmotriBroadcastIE,
+)
 from .sohu import SohuIE
 from .soundcloud import SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE
 from .southparkstudios import (
@@ -136,6 +176,7 @@ from .teamcoco import TeamcocoIE
 from .techtalks import TechTalksIE
 from .ted import TEDIE
 from .tf1 import TF1IE
+from .theplatform import ThePlatformIE
 from .thisav import ThisAVIE
 from .toutv import TouTvIE
 from .traileraddict import TrailerAddictIE
@@ -156,7 +197,14 @@ from .viddler import ViddlerIE
 from .videodetective import VideoDetectiveIE
 from .videofyme import VideofyMeIE
 from .videopremium import VideoPremiumIE
-from .vimeo import VimeoIE, VimeoChannelIE
+from .vimeo import (
+    VimeoIE,
+    VimeoChannelIE,
+    VimeoUserIE,
+    VimeoAlbumIE,
+    VimeoGroupsIE,
+    VimeoReviewIE,
+)
 from .vine import VineIE
 from .viki import VikiIE
 from .vk import VKIE
@@ -164,12 +212,17 @@ from .wat import WatIE
 from .websurg import WeBSurgIE
 from .weibo import WeiboIE
 from .wimp import WimpIE
+from .wistia import WistiaIE
 from .worldstarhiphop import WorldStarHipHopIE
 from .xhamster import XHamsterIE
 from .xnxx import XNXXIE
 from .xvideos import XVideosIE
 from .xtube import XTubeIE
-from .yahoo import YahooIE, YahooSearchIE
+from .yahoo import (
+    YahooIE,
+    YahooNewsIE,
+    YahooSearchIE,
+)
 from .youjizz import YouJizzIE
 from .youku import YoukuIE
 from .youporn import YouPornIE
@@ -187,6 +240,7 @@ from .youtube import (
     YoutubeWatchLaterIE,
     YoutubeFavouritesIE,
     YoutubeHistoryIE,
+    YoutubeTopListIE,
 )
 from .zdf import ZDFIE
 

youtube_dl/extractor/academicearth.py

@@ -0,0 +1,32 @@
+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 = 'AcademicEarth:Course'
+
+    def _real_extract(self, url):
+        m = re.match(self._VALID_URL, url)
+        playlist_id = m.group('id')
+
+        webpage = self._download_webpage(url, playlist_id)
+        title = self._html_search_regex(
+            r'<h1 class="playlist-name">(.*?)</h1>', webpage, u'title')
+        description = self._html_search_regex(
+            r'<p class="excerpt">(.*?)</p>',
+            webpage, u'description', fatal=False)
+        urls = re.findall(
+            r'<h3 class="lecture-title"><a target="_blank" href="([^"]+)">',
+            webpage)
+        entries = [self.url_result(u) for u in urls]
+
+        return {
+            '_type': 'playlist',
+            'id': playlist_id,
+            'title': title,
+            'description': description,
+            'entries': entries,
+        }

youtube_dl/extractor/addanime.py

@@ -13,7 +13,7 @@ from ..utils import (
 
 class AddAnimeIE(InfoExtractor):
 
-    _VALID_URL = r'^http://(?:\w+\.)?add-anime\.net/watch_video.php\?(?:.*?)v=(?P<video_id>[\w_]+)(?:.*)'
+    _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',

youtube_dl/extractor/anitube.py

@@ -1,5 +1,4 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 
@@ -28,9 +27,8 @@ class AnitubeIE(InfoExtractor):
         key = self._html_search_regex(r'http://www\.anitube\.se/embed/([A-Za-z0-9_-]*)',
                                       webpage, u'key')
 
-        webpage_config = self._download_webpage('http://www.anitube.se/nuevo/econfig.php?key=%s' % key,
+        config_xml = self._download_xml('http://www.anitube.se/nuevo/econfig.php?key=%s' % key,
                                                 key)
-        config_xml = xml.etree.ElementTree.fromstring(webpage_config.encode('utf-8'))
 
         video_title = config_xml.find('title').text
 

youtube_dl/extractor/aparat.py

@@ -0,0 +1,56 @@
+#coding: utf-8
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    HEADRequest,
+)
+
+
+class AparatIE(InfoExtractor):
+    _VALID_URL = r'^https?://(?:www\.)?aparat\.com/(?:v/|video/video/embed/videohash/)(?P<id>[a-zA-Z0-9]+)'
+
+    _TEST = {
+        u'url': u'http://www.aparat.com/v/wP8On',
+        u'file': u'wP8On.mp4',
+        u'md5': u'6714e0af7e0d875c5a39c4dc4ab46ad1',
+        u'info_dict': {
+            u"title": u"تیم گلکسی 11 - زومیت",
+        },
+        #u'skip': u'Extremely unreliable',
+    }
+
+    def _real_extract(self, url):
+        m = re.match(self._VALID_URL, url)
+        video_id = m.group('id')
+
+        # Note: There is an easier-to-parse configuration at
+        # http://www.aparat.com/video/video/config/videohash/%video_id
+        # but the URL in there does not work
+        embed_url = (u'http://www.aparat.com/video/video/embed/videohash/' +
+                     video_id + u'/vt/frame')
+        webpage = self._download_webpage(embed_url, video_id)
+
+        video_urls = re.findall(r'fileList\[[0-9]+\]\s*=\s*"([^"]+)"', webpage)
+        for i, video_url in enumerate(video_urls):
+            req = HEADRequest(video_url)
+            res = self._request_webpage(
+                req, video_id, note=u'Testing video URL %d' % i, errnote=False)
+            if res:
+                break
+        else:
+            raise ExtractorError(u'No working video URLs found')
+
+        title = self._search_regex(r'\s+title:\s*"([^"]+)"', webpage, u'title')
+        thumbnail = self._search_regex(
+            r'\s+image:\s*"([^"]+)"', webpage, u'thumbnail', fatal=False)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'url': video_url,
+            'ext': 'mp4',
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/appletrailers.py

@@ -1,5 +1,6 @@
+from __future__ import unicode_literals
+
 import re
-import xml.etree.ElementTree
 import json
 
 from .common import InfoExtractor
@@ -10,48 +11,48 @@ from ..utils import (
 
 
 class AppleTrailersIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?trailers.apple.com/trailers/(?P<company>[^/]+)/(?P<movie>[^/]+)'
+    _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",
+                "file": "manofsteel-trailer4.mov",
+                "md5": "d97a8e575432dbcb81b7c3acb741f8a8",
+                "info_dict": {
+                    "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",
+                "file": "manofsteel-trailer3.mov",
+                "md5": "b8017b7131b721fb4e8d6f49e1df908c",
+                "info_dict": {
+                    "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",
+                "file": "manofsteel-trailer.mov",
+                "md5": "d0f1e1150989b9924679b441f3404d48",
+                "info_dict": {
+                    "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",
+                "file": "manofsteel-teaser.mov",
+                "md5": "5fe08795b943eb2e757fa95cb6def1cb",
+                "info_dict": {
+                    "duration": 93,
+                    "title": "Teaser",
+                    "upload_date": "20120721",
+                    "uploader_id": "wb",
                 },
             }
         ]
@@ -65,18 +66,18 @@ class AppleTrailersIE(InfoExtractor):
         uploader_id = mobj.group('company')
 
         playlist_url = compat_urlparse.urljoin(url, u'includes/playlists/itunes.inc')
-        playlist_snippet = self._download_webpage(playlist_url, movie)
-        playlist_cleaned = re.sub(r'(?s)<script[^<]*?>.*?</script>', u'', playlist_snippet)
-        playlist_cleaned = re.sub(r'<img ([^<]*?)>', r'<img \1/>', playlist_cleaned)
-        # The ' in the onClick attributes are not escaped, it couldn't be parsed
-        # with xml.etree.ElementTree.fromstring
-        # like: http://trailers.apple.com/trailers/wb/gravity/
-        def _clean_json(m):
-            return u'iTunes.playURL(%s);' % m.group(1).replace('\'', '&#39;')
-        playlist_cleaned = re.sub(self._JSON_RE, _clean_json, playlist_cleaned)
-        playlist_html = u'<html>' + playlist_cleaned + u'</html>'
+        def fix_html(s):
+            s = re.sub(r'(?s)<script[^<]*?>.*?</script>', u'', 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;')
+            s = re.sub(self._JSON_RE, _clean_json, s)
+            s = u'<html>' + s + u'</html>'
+            return s
+        doc = self._download_xml(playlist_url, movie, transform_source=fix_html)
 
-        doc = xml.etree.ElementTree.fromstring(playlist_html)
         playlist = []
         for li in doc.findall('./div/ul/li'):
             on_click = li.find('.//a').attrib['onClick']
@@ -111,9 +112,10 @@ class AppleTrailersIE(InfoExtractor):
                     'width': format['width'],
                     'height': int(format['height']),
                 })
-            formats = sorted(formats, key=lambda f: (f['height'], f['width']))
 
-            info = {
+            self._sort_formats(formats)
+
+            playlist.append({
                 '_type': 'video',
                 'id': video_id,
                 'title': title,
@@ -124,12 +126,7 @@ class AppleTrailersIE(InfoExtractor):
                 'upload_date': upload_date,
                 'uploader_id': uploader_id,
                 'user_agent': 'QuickTime compatible (youtube-dl)',
-            }
-            # TODO: Remove when #980 has been merged
-            info['url'] = formats[-1]['url']
-            info['ext'] = formats[-1]['ext']
-
-            playlist.append(info)
+            })
 
         return {
             '_type': 'playlist',

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,
 )
 
@@ -11,25 +12,24 @@ from ..utils import (
 class ArchiveOrgIE(InfoExtractor):
     IE_NAME = 'archive.org'
     IE_DESC = 'archive.org videos'
-    _VALID_URL = r'(?:https?://)?(?:www\.)?archive.org/details/(?P<id>[^?/]+)(?:[?].*)?$'
+    _VALID_URL = r'(?:https?://)?(?:www\.)?archive\.org/details/(?P<id>[^?/]+)(?:[?].*)?$'
     _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)
 
@@ -38,18 +38,18 @@ 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'])
 
-        info = {
+        self._sort_formats(formats)
+
+        return {
             '_type': 'video',
             'id': video_id,
             'title': title,
@@ -57,12 +57,5 @@ class ArchiveOrgIE(InfoExtractor):
             'description': description,
             'uploader': uploader,
             'upload_date': upload_date,
+            'thumbnail': data.get('misc', {}).get('image'),
         }
-        thumbnail = data.get('misc', {}).get('image')
-        if thumbnail:
-            info['thumbnail'] = thumbnail
-
-        # TODO: Remove when #980 has been merged
-        info.update(formats[-1])
-
-        return info

youtube_dl/extractor/arte.py

@@ -1,7 +1,8 @@
 # encoding: utf-8
+from __future__ import unicode_literals
+
 import re
 import json
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -11,6 +12,7 @@ from ..utils import (
     determine_ext,
     get_element_by_id,
     compat_str,
+    get_element_by_attribute,
 )
 
 # There are different sources of video in arte.tv, the extraction process 
@@ -18,11 +20,11 @@ from ..utils import (
 # 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>.+)'
+    _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'
+    IE_NAME = 'arte.tv'
 
     @classmethod
     def suitable(cls, url):
@@ -37,7 +39,7 @@ class ArteTvIE(InfoExtractor):
     #         r'src="(.*?/videothek_js.*?\.js)',
     #         0,
     #         [
-    #             (1, 'url', u'Invalid URL: %s' % url)
+    #             (1, 'url', 'Invalid URL: %s' % url)
     #         ]
     #     )
     #     http_host = url.split('/')[2]
@@ -49,12 +51,12 @@ class ArteTvIE(InfoExtractor):
     #             '(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)
+    #             (1, 'path',   'could not extract video path: %s' % url),
+    #             (2, 'player', 'could not extract video player: %s' % url),
+    #             (3, 'url',    'could not extract video url: %s' % url)
     #         ]
     #     )
-    #     video_url = u'%s/%s' % (info.get('url'), info.get('path'))
+    #     video_url = '%s/%s' % (info.get('url'), info.get('path'))
 
     def _real_extract(self, url):
         mobj = re.match(self._VIDEOS_URL, url)
@@ -78,8 +80,7 @@ class ArteTvIE(InfoExtractor):
         """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 = self._download_webpage(ref_xml_url, video_id, note=u'Downloading metadata')
-        ref_xml_doc = xml.etree.ElementTree.fromstring(ref_xml)
+        ref_xml_doc = self._download_xml(ref_xml_url, video_id, note=u'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')
@@ -108,10 +109,9 @@ class ArteTvIE(InfoExtractor):
     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_xml = self._download_webpage('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id,
-                                            video_id, u'Downloading information')
-        config_doc = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
+        video_id = self._search_regex(r'eventId=(\d+?)("|&)', webpage, 'event id')
+        config_doc = self._download_xml('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id,
+                                            video_id, 'Downloading information')
         event_doc = config_doc.find('event')
         url_node = event_doc.find('video').find('urlHd')
         if url_node is None:
@@ -126,7 +126,7 @@ class ArteTvIE(InfoExtractor):
 
 
 class ArteTVPlus7IE(InfoExtractor):
-    IE_NAME = u'arte.tv:+7'
+    IE_NAME = 'arte.tv:+7'
     _VALID_URL = r'https?://www\.arte.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
 
     @classmethod
@@ -145,7 +145,9 @@ class ArteTVPlus7IE(InfoExtractor):
 
     def _extract_from_webpage(self, webpage, video_id, lang):
         json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json 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)
@@ -207,7 +209,7 @@ 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 +218,7 @@ 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 +233,27 @@ 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',
+        'file': '050489-002.mp4',
+        'info_dict': {
+            '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',
+        'file': '050940-003.mp4',
+        'info_dict': {
+            'title': 'Les champignons au secours de la planète',
         },
     }
 
@@ -260,3 +262,21 @@ class ArteTVFutureIE(ArteTVPlus7IE):
         webpage = self._download_webpage(url, anchor_id)
         row = get_element_by_id(anchor_id, webpage)
         return self._extract_from_webpage(row, anchor_id, lang)
+
+
+class ArteTVDDCIE(ArteTVPlus7IE):
+    IE_NAME = 'arte.tv:ddc'
+    _VALID_URL = r'http?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)'
+
+    def _real_extract(self, url):
+        video_id, lang = self._extract_url_info(url)
+        if lang == 'folge':
+            lang = 'de'
+        elif lang == 'emission':
+            lang = 'fr'
+        webpage = self._download_webpage(url, video_id)
+        scriptElement = get_element_by_attribute('class', 'visu_video_block', webpage)
+        script_url = self._html_search_regex(r'src="(.*?)"', scriptElement, 'script url')
+        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)

youtube_dl/extractor/auengine.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -7,23 +9,24 @@ from ..utils import (
     ExtractorError,
 )
 
+
 class AUEngineIE(InfoExtractor):
     _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',
+        'file': 'lfvlytY6.mp4',
+        'md5': '48972bdbcf1a3a2f5533e62425b41d4f',
+        'info_dict': {
+            'title': '[Commie]The Legend of the Legendary Heroes - 03 - Replication Eye (Alpha Stigma)[F9410F5A]'
         }
     }
-    _VALID_URL = r'(?:http://)?(?:www\.)?auengine\.com/embed.php\?.*?file=([^&]+).*?'
+    _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)
         webpage = self._download_webpage(url, video_id)
         title = self._html_search_regex(r'<title>(?P<title>.+?)</title>',
-                webpage, u'title')
+                webpage, 'title')
         title = title.strip()
         links = re.findall(r'\s(?:file|url):\s*["\']([^\'"]+)["\']', webpage)
         links = map(compat_urllib_parse.unquote, links)
@@ -37,7 +40,7 @@ class AUEngineIE(InfoExtractor):
                 video_url = link
         if not video_url:
             raise ExtractorError(u'Could not find video URL')
-        ext = u'.' + determine_ext(video_url)
+        ext = '.' + determine_ext(video_url)
         if ext == title[-len(ext):]:
             title = title[:-len(ext)]
 

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,
         },
     }
 
@@ -53,8 +55,8 @@ class BambuserIE(InfoExtractor):
 
 
 class BambuserChannelIE(InfoExtractor):
-    IE_NAME = u'bambuser:channel'
-    _VALID_URL = r'http://bambuser.com/channel/(?P<user>.*?)(?:/|#|\?|$)'
+    IE_NAME = 'bambuser:channel'
+    _VALID_URL = r'https?://bambuser\.com/channel/(?P<user>.*?)(?:/|#|\?|$)'
     # 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>.*)'
     _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):
@@ -30,85 +32,98 @@ 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'
+    IE_NAME = 'Bandcamp:album'
     _VALID_URL = r'http://.*?\.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):
@@ -117,11 +132,11 @@ class BandcampAlbumIE(InfoExtractor):
         webpage = self._download_webpage(url, title)
         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',
             'title': title,

youtube_dl/extractor/blinkx.py

@@ -0,0 +1,95 @@
+from __future__ import unicode_literals
+
+import datetime
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    remove_start,
+)
+
+
+class BlinkxIE(InfoExtractor):
+    _VALID_URL = r'^(?:https?://(?:www\.)blinkx\.com/#?ce/|blinkx:)(?P<id>[^?]+)'
+    IE_NAME = 'blinkx'
+
+    _TEST = {
+        '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",
+            }],
+        },
+    }
+
+    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&' +
+                   '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'])
+        pload_date = dt.strftime('%Y%m%d')
+
+        duration = None
+        thumbnails = []
+        formats = []
+        for m in data['media']:
+            if m['type'] == 'jpg':
+                thumbnails.append({
+                    'url': m['link'],
+                    'width': int(m['w']),
+                    'height': int(m['h']),
+                })
+            elif m['type'] == 'original':
+                duration = m['d']
+            elif m['type'] == 'youtube':
+                yt_id = m['link']
+                self.to_screen(u'Youtube video detected: %s' % yt_id)
+                return self.url_result(yt_id, 'Youtube', video_id=yt_id)
+            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,
+                              tbr,
+                              m['w']))
+                formats.append({
+                    'format_id': format_id,
+                    'url': m['link'],
+                    'vcodec': vcodec,
+                    'acodec': acodec,
+                    'abr': int(m['abr']) // 1000,
+                    'vbr': int(m['vbr']) // 1000,
+                    'tbr': tbr,
+                    'width': int(m['w']),
+                    'height': int(m['h']),
+                })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': display_id,
+            'fullid': video_id,
+            'title': data['title'],
+            'formats': formats,
+            'uploader': data['channel_name'],
+            'upload_date': pload_date,
+            'description': data.get('description'),
+            'thumbnails': thumbnails,
+            'duration': duration,
+        }

youtube_dl/extractor/bliptv.py

@@ -1,16 +1,15 @@
+from __future__ import unicode_literals
+
 import datetime
 import json
-import os
 import re
 import socket
 
 from .common import InfoExtractor
 from ..utils import (
     compat_http_client,
-    compat_parse_qs,
     compat_str,
     compat_urllib_error,
-    compat_urllib_parse_urlparse,
     compat_urllib_request,
 
     ExtractorError,
@@ -22,43 +21,35 @@ class BlipTVIE(InfoExtractor):
     """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"
+        'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
+        'file': '5779306.mov',
+        'md5': 'c6934ad0b6acf2bd920720ec888eb812',
+        'info_dict': {
+            'upload_date': '20111205',
+            'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
+            'uploader': 'Comic Book Resources - CBR TV',
+            'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
         }
     }
 
     def report_direct_download(self, title):
         """Report information extraction."""
-        self.to_screen(u'%s: Direct download detected' % title)
+        self.to_screen('%s: Direct download detected' % title)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError('Invalid URL: %s' % url)
 
         # 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/'):
-            request = compat_urllib_request.Request(url)
-            response = compat_urllib_request.urlopen(request)
-            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)
-
+        embed_mobj = re.search(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="(\d+)', info_page,  'video_id')
+            return self.url_result('http://blip.tv/a/a-' + video_id, 'BlipTV')
 
         if '?' in url:
             cchar = '&'
@@ -68,69 +59,55 @@ class BlipTVIE(InfoExtractor):
         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,
+            'unable to download video info webpage')
+
         try:
-            urlh = compat_urllib_request.urlopen(request)
-            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)
-                info = {
-                    'id': title,
-                    'url': url,
-                    'uploader': None,
-                    'upload_date': None,
-                    'title': title,
-                    'ext': ext,
-                    'urlhandle': urlh
-                }
+            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'ERROR: unable to download video info webpage: %s' % compat_str(err))
-        if info is None: # Regular URL
-            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))
+            raise ExtractorError('Unable to read video info webpage: %s' % compat_str(err))
 
-            try:
-                json_data = json.loads(json_code)
-                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
 
-                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)
+            upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
+            formats = []
+            if 'additionalMedia' in data:
+                for f in sorted(data['additionalMedia'], key=lambda f: int(f['media_height'])):
+                    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']),
+                })
 
-                info = {
-                    '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))
+            self._sort_formats(formats)
 
-        return [info]
+            return {
+                'id': compat_str(data['item_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,
+            }
+        except (ValueError, KeyError) as err:
+            raise ExtractorError('Unable to parse video information: %s' % repr(err))
 
 
 class BlipTVUserIE(InfoExtractor):
@@ -138,19 +115,19 @@ class BlipTVUserIE(InfoExtractor):
 
     _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)
+            raise ExtractorError('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)
 
@@ -166,7 +143,7 @@ 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)
+                                          'Downloading video ids from page %d' % pagenum)
 
             # Extract video identifiers
             ids_in_page = []
@@ -188,6 +165,6 @@ 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)]

youtube_dl/extractor/bloomberg.py

@@ -1,10 +1,11 @@
 import re
 
 from .common import InfoExtractor
+from .ooyala import OoyalaIE
 
 
 class BloombergIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.bloomberg\.com/video/(?P<name>.+?).html'
+    _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',
@@ -23,5 +24,5 @@ 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')
+        ooyala_code = self._search_regex(r'<source src="http://player.ooyala.com/player/[^/]+/([^".]+)', webpage, u'ooyala url')
+        return OoyalaIE._build_url_result(ooyala_code)

youtube_dl/extractor/brightcove.py

@@ -1,4 +1,5 @@
 # encoding: utf-8
+from __future__ import unicode_literals
 
 import re
 import json
@@ -13,6 +14,7 @@ from ..utils import (
     compat_urllib_request,
 
     ExtractorError,
+    unsmuggle_url,
 )
 
 
@@ -24,35 +26,47 @@ class BrightcoveIE(InfoExtractor):
     _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'8eccab865181d29ec2958f32a6a754f5',
-            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/
+            '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',
             },
         },
     ]
@@ -68,13 +82,13 @@ 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_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'],
-                  }
+        assert 'BrightcoveExperience' in object_doc.attrib['class']
+        params = {
+            'playerID': find_xpath_attr(object_doc, './param', 'name', 'playerID').attrib['value'],
+        }
         def find_param(name):
             node = find_xpath_attr(object_doc, './param', 'name', name)
             if node is not None:
@@ -108,6 +122,8 @@ class BrightcoveIE(InfoExtractor):
             return None
 
     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
@@ -118,17 +134,22 @@ 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 = 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])
 
-    def _get_video_info(self, video_id, query_str, query):
+    def _get_video_info(self, video_id, query_str, query, referer=None):
         request_url = self._FEDERATED_URL_TEMPLATE % query_str
         req = compat_urllib_request.Request(request_url)
         linkBase = query.get('linkBaseURL')
         if linkBase is not None:
-            req.add_header('Referer', linkBase[0])
+            referer = linkBase[0]
+        if referer is not None:
+            req.add_header('Referer', referer)
         webpage = self._download_webpage(req, video_id)
 
         self.report_extraction(video_id)
@@ -140,11 +161,11 @@ class BrightcoveIE(InfoExtractor):
 
     def _get_playlist_info(self, player_key):
         playlist_info = self._download_webpage(self._PLAYLIST_URL_TEMPLATE % player_key,
-                                               player_key, u'Downloading playlist information')
+                                               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']]
 
@@ -173,5 +194,5 @@ class BrightcoveIE(InfoExtractor):
                 'url': video_info['FLVFullLengthURL'],
             })
         else:
-            raise ExtractorError(u'Unable to extract video url for %s' % info['id'])
+            raise ExtractorError('Unable to extract video url for %s' % info['id'])
         return info

youtube_dl/extractor/c56.py

@@ -1,21 +1,21 @@
 # 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',
+    IE_NAME = '56.com'
+    _TEST = {
+        'url': 'http://www.56.com/u39/v_OTM0NDA3MTY.html',
+        'file': '93440716.flv',
+        'md5': 'e59995ac63d0457783ea05f93f12a866',
+        'info_dict': {
+            'title': '网事知多少 第32期：车怒',
         },
     }
 
@@ -23,14 +23,18 @@ class C56IE(InfoExtractor):
         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')
+                                           text_id, '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']
+        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'],
-                'url': video_url,
-                'ext': determine_ext(video_url),
-                'thumbnail': info.get('bimg') or info.get('img'),
-                }
+        return {
+            'id': info['vid'],
+            'title': info['Subject'],
+            'formats': formats,
+            'thumbnail': info.get('bimg') or info.get('img'),
+        }

youtube_dl/extractor/canalplus.py

@@ -1,6 +1,5 @@
 # encoding: utf-8
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import unified_strdate
@@ -31,11 +30,10 @@ class CanalplusIE(InfoExtractor):
             webpage = self._download_webpage(url, mobj.group('path'))
             video_id = self._search_regex(r'videoId = "(\d+)";', webpage, u'video id')
         info_url = self._VIDEO_INFO_TEMPLATE % video_id
-        info_page = self._download_webpage(info_url,video_id, 
+        doc = self._download_xml(info_url,video_id, 
                                            u'Downloading video info')
 
         self.report_extraction(video_id)
-        doc = xml.etree.ElementTree.fromstring(info_page.encode('utf-8'))
         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')

youtube_dl/extractor/cbs.py

@@ -0,0 +1,30 @@
+import re
+
+from .common import InfoExtractor
+
+
+class CBSIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?cbs\.com/shows/[^/]+/video/(?P<id>[^/]+)/.*'
+
+    _TEST = {
+        u'url': u'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
+        u'file': u'4JUVEwq3wUT7.flv',
+        u'info_dict': {
+            u'title': u'Connect Chat feat. Garth Brooks',
+            u'description': u'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!',
+            u'duration': 1495,
+        },
+        u'params': {
+            # rtmp download
+            u'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)
+        real_id = self._search_regex(
+            r"video\.settings\.pid\s*=\s*'([^']+)';",
+            webpage, u'real video ID')
+        return self.url_result(u'theplatform:%s' % real_id)

youtube_dl/extractor/channel9.py

@@ -0,0 +1,271 @@
+# encoding: utf-8
+
+import re
+
+from .common import InfoExtractor
+from ..utils import ExtractorError
+
+class Channel9IE(InfoExtractor):
+    '''
+    Common extractor for channel9.msdn.com.
+
+    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.    
+    '''
+    IE_DESC = u'Channel 9'
+    IE_NAME = u'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' ],
+            },
+        },
+        {
+            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' ],
+            },
+        }
+    ]
+
+    _RSS_URL = 'http://channel9.msdn.com/%s/RSS'
+
+    # Sorted by quality
+    _known_formats = ['MP3', 'MP4', 'Mid Quality WMV', 'Mid Quality MP4', 'High Quality WMV', 'High Quality MP4']
+
+    def _restore_bytes(self, formatted_size):
+        if not formatted_size:
+            return 0
+        m = re.match(r'^(?P<size>\d+(?:\.\d+)?)\s+(?P<units>[a-zA-Z]+)', formatted_size)
+        if not m:
+            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())
+        except ValueError:
+            return 0
+        size = float(m.group('size'))
+        return int(size * (1024 ** exponent))
+
+    def _formats_from_html(self, html):
+        FORMAT_REGEX = r'''
+            (?x)
+            <a\s+href="(?P<url>[^"]+)">(?P<quality>[^<]+)</a>\s*
+            <span\s+class="usage">\((?P<note>[^\)]+)\)</span>\s*
+            (?:<div\s+class="popup\s+rounded">\s*
+            <h3>File\s+size</h3>\s*(?P<filesize>.*?)\s*
+            </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': u'%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')
+        if title is None:           
+            title = self._og_search_title(html)
+            TITLE_SUFFIX = u' (Channel 9)'
+            if title is not None and title.endswith(TITLE_SUFFIX):
+                title = title[:-len(TITLE_SUFFIX)]
+        return title
+
+    def _extract_description(self, html):
+        DESCRIPTION_REGEX = r'''(?sx)
+            <div\s+class="entry-content">\s*
+            <div\s+id="entry-body">\s*
+            (?P<description>.+?)\s*
+            </div>\s*
+            </div>
+        '''
+        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')
+
+    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)
+        return ((int(m.group('hours')) * 60 * 60) + (int(m.group('minutes')) * 60) + int(m.group('seconds'))) if m else None
+
+    def _extract_slides(self, html):
+        m = re.search(r'<a href="(?P<slidesurl>[^"]+)" class="slides">Slides</a>', html)
+        return m.group('slidesurl') if m is not None else None
+
+    def _extract_zip(self, html):
+        m = re.search(r'<a href="(?P<zipurl>[^"]+)" class="zip">Zip</a>', html)
+        return m.group('zipurl') if m is not None else None
+
+    def _extract_avg_rating(self, html):
+        m = re.search(r'<p class="avg-rating">Avg Rating: <span>(?P<avgrating>[^<]+)</span></p>', html)
+        return float(m.group('avgrating')) if m is not None else 0
+
+    def _extract_rating_count(self, html):
+        m = re.search(r'<div class="rating-count">\((?P<ratingcount>[^<]+)\)</div>', html)
+        return int(self._fix_count(m.group('ratingcount'))) if m is not None else 0
+
+    def _extract_view_count(self, html):
+        m = re.search(r'<li class="views">\s*<span class="count">(?P<viewcount>[^<]+)</span> Views\s*</li>', html)
+        return int(self._fix_count(m.group('viewcount'))) if m is not None else 0
+
+    def _extract_comment_count(self, html):
+        m = re.search(r'<li class="comments">\s*<a href="#comments">\s*<span class="count">(?P<commentcount>[^<]+)</span> Comments\s*</a>\s*</li>', html)
+        return int(self._fix_count(m.group('commentcount'))) if m is not None else 0
+
+    def _fix_count(self, count):
+        return int(str(count).replace(',', '')) if count is not None else None
+
+    def _extract_authors(self, html):
+        m = re.search(r'(?s)<li class="author">(.*?)</li>', html)
+        if m is None:
+            return None
+        return re.findall(r'<a href="/Niners/[^"]+">([^<]+)</a>', m.group(1))
+
+    def _extract_session_code(self, html):
+        m = re.search(r'<li class="code">\s*(?P<code>.+?)\s*</li>', html)
+        return m.group('code') if m is not None else None
+
+    def _extract_session_day(self, html):
+        m = re.search(r'<li class="day">\s*<a href="/Events/[^"]+">(?P<day>[^<]+)</a>\s*</li>', html)
+        return m.group('day') if m is not None else None
+
+    def _extract_session_room(self, html):
+        m = re.search(r'<li class="room">\s*(?P<room>.+?)\s*</li>', html)
+        return m.group('room') if m is not None else None
+
+    def _extract_session_speakers(self, html):
+        return re.findall(r'<a href="/Events/Speakers/[^"]+">([^<]+)</a>', html)
+
+    def _extract_content(self, html, content_path):
+        # Look for downloadable content        
+        formats = self._formats_from_html(html)
+        slides = self._extract_slides(html)
+        zip_ = self._extract_zip(html)
+
+        # 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)
+            return
+
+        # Extract meta
+        title = self._extract_title(html)
+        description = self._extract_description(html)
+        thumbnail = self._og_search_thumbnail(html)
+        duration = self._extract_duration(html)
+        avg_rating = self._extract_avg_rating(html)
+        rating_count = self._extract_rating_count(html)
+        view_count = self._extract_view_count(html)
+        comment_count = self._extract_comment_count(html)
+
+        common = {'_type': 'video',
+                  'id': content_path,
+                  'description': description,
+                  'thumbnail': thumbnail,
+                  'duration': duration,
+                  'avg_rating': avg_rating,
+                  'rating_count': rating_count,
+                  'view_count': view_count,
+                  'comment_count': comment_count,
+                }
+
+        result = []
+
+        if slides is not None:
+            d = common.copy()
+            d.update({ 'title': title + '-Slides', 'url': slides })
+            result.append(d)
+
+        if zip_ is not None:
+            d = common.copy()
+            d.update({ 'title': title + '-Zip', 'url': zip_ })
+            result.append(d)
+
+        if len(formats) > 0:
+            d = common.copy()
+            d.update({ 'title': title, 'formats': formats })
+            result.append(d)
+
+        return result
+
+    def _extract_entry_item(self, html, content_path):
+        contents = self._extract_content(html, content_path)
+        if contents is None:
+            return contents
+
+        authors = self._extract_authors(html)
+
+        for content in contents:
+            content['authors'] = authors
+
+        return contents
+
+    def _extract_session(self, html, content_path):
+        contents = self._extract_content(html, content_path)
+        if contents is None:
+            return contents
+
+        session_meta = {'session_code': self._extract_session_code(html),
+                        'session_day': self._extract_session_day(html),
+                        'session_room': self._extract_session_room(html),
+                        'session_speakers': self._extract_session_speakers(html),
+                        }
+
+        for content in contents:
+            content.update(session_meta)
+
+        return contents
+
+    def _extract_list(self, content_path):
+        rss = self._download_xml(self._RSS_URL % content_path, content_path, u'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
+        return self.playlist_result(entries, content_path, title_text)
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        content_path = mobj.group('contentpath')
+
+        webpage = self._download_webpage(url, content_path, u'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)
+
+        page_type = page_type_m.group('pagetype')
+        if page_type == 'List':         # List page, may contain list of 'item'-like objects
+            return self._extract_list(content_path)
+        elif page_type == 'Entry.Item': # Any 'item'-like page, may contain downloadable content
+            return self._extract_entry_item(webpage, content_path)
+        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)

youtube_dl/extractor/cinemassacre.py

@@ -12,21 +12,27 @@ class CinemassacreIE(InfoExtractor):
     _TESTS = [{
         u'url': u'http://cinemassacre.com/2012/11/10/avgn-the-movie-trailer/',
         u'file': u'19911.flv',
-        u'md5': u'f9bb7ede54d1229c9846e197b4737e06',
         u'info_dict': {
             u'upload_date': u'20121110',
             u'title': u'“Angry Video Game Nerd: The Movie” – Trailer',
             u'description': u'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'md5': u'9509ee44dcaa7c1068604817c19a9e50',
         u'info_dict': {
             u'upload_date': u'20131002',
             u'title': u'The Mummy’s Hand (1940)',
-        }
+        },
+        u'params': {
+            # rtmp download
+            u'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/clipfish.py

@@ -3,6 +3,7 @@ import time
 import xml.etree.ElementTree
 
 from .common import InfoExtractor
+from ..utils import ExtractorError
 
 
 class ClipfishIE(InfoExtractor):
@@ -10,13 +11,14 @@ class ClipfishIE(InfoExtractor):
 
     _VALID_URL = r'^https?://(?:www\.)?clipfish\.de/.*?/video/(?P<id>[0-9]+)/'
     _TEST = {
-        u'url': u'http://www.clipfish.de/special/supertalent/video/4028320/supertalent-2013-ivana-opacak-singt-nobodys-perfect/',
-        u'file': u'4028320.f4v',
-        u'md5': u'5e38bda8c329fbfb42be0386a3f5a382',
+        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'Supertalent 2013: Ivana Opacak singt Nobody\'s Perfect',
-            u'duration': 399,
-        }
+            u'title': u'FIFA 14 - E3 2013 Trailer',
+            u'duration': 82,
+        },
+        u'skip': 'Blocked in the US'
     }
 
     def _real_extract(self, url):
@@ -25,11 +27,14 @@ class ClipfishIE(InfoExtractor):
 
         info_url = ('http://www.clipfish.de/devxml/videoinfo/%s?ts=%d' %
                     (video_id, int(time.time())))
-        info_xml = self._download_webpage(
+        doc = self._download_xml(
             info_url, video_id, note=u'Downloading info page')
-        doc = xml.etree.ElementTree.fromstring(info_xml)
         title = doc.find('title').text
         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' %
+                                 xml_bytes)
         thumbnail = doc.find('imageurl').text
         duration_str = doc.find('duration').text
         m = re.match(

youtube_dl/extractor/clipsyndicate.py

@@ -0,0 +1,50 @@
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    find_xpath_attr,
+    fix_xml_all_ampersand,
+)
+
+
+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,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        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')
+        # it includes a required token
+        flvars = self._search_regex(r'flvars: "(.*?)"', js_player, u'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) 
+
+        track_doc = pdoc.find('trackList/track')
+        def find_param(name):
+            node = find_xpath_attr(track_doc, './/param', 'name', name)
+            if node is not None:
+                return node.attrib['value']
+
+        return {
+            'id': video_id,
+            'title': find_param('title'),
+            'url': track_doc.find('location').text,
+            'thumbnail': find_param('thumbnail'),
+            'duration': int(find_param('duration')),
+        }

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/cnn.py

@@ -1,8 +1,12 @@
+from __future__ import unicode_literals
+
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
-from ..utils import determine_ext
+from ..utils import (
+    int_or_none,
+    parse_duration,
+)
 
 
 class CNNIE(InfoExtractor):
@@ -10,12 +14,14 @@ 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',
         },
     },
     {
@@ -32,27 +38,62 @@ 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_xml = self._download_webpage(info_url, page_title)
-        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
+        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,
+        }

youtube_dl/extractor/collegehumor.py

@@ -1,82 +1,68 @@
+from __future__ import unicode_literals
+
+import json
 import re
 
 from .common import InfoExtractor
-from ..utils import (
-    compat_urllib_parse_urlparse,
-    determine_ext,
-
-    ExtractorError,
-)
 
 
 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',
+        'file': '6902724.mp4',
+        'md5': 'dcc0f5c1c8be98dc33889a191f4c26bd',
+        'info_dict': {
+            'title': 'Comic-Con Cosplay Catastrophe',
+            'description': 'Fans get creative this year at San Diego.  Too',
+            'age_limit': 13,
         },
     },
     {
-        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',
+        'file': '3505939.mp4',
+        'md5': '72fa701d8ef38664a4dbb9e2ab721816',
+        'info_dict': {
+            'title': 'Font Conference',
+            'description': 'This video wasn\'t long enough, so we made it double-spaced.',
+            'age_limit': 10,
         },
     }]
 
     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']
 
-        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[format_key].items():
+                formats.append({
+                    'format_id': format_key + '_' + qname,
+                    'url': qurl,
+                    'format': format_key,
+                    'preference': PREFS.get(qname),
+                })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': vdata['title'],
+            'description': vdata.get('description'),
+            'thumbnail': vdata.get('thumbnail'),
+            'formats': formats,
+            'age_limit': age_limit,
+        }

youtube_dl/extractor/comedycentral.py

@@ -1,8 +1,7 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
-from .mtv import MTVIE, _media_xml_tag
+from .mtv import MTVServicesInfoExtractor
 from ..utils import (
     compat_str,
     compat_urllib_parse,
@@ -12,8 +11,10 @@ from ..utils import (
 )
 
 
-class ComedyCentralIE(MTVIE):
-    _VALID_URL = r'http://www.comedycentral.com/(video-clips|episodes|cc-studios)/(?P<title>.*)'
+class ComedyCentralIE(MTVServicesInfoExtractor):
+    _VALID_URL = r'''(?x)https?://(?:www.)?comedycentral.com/
+        (video-clips|episodes|cc-studios|video-collections)
+        /(?P<title>.*)'''
     _FEED_URL = u'http://comedycentral.com/feeds/mrss/'
 
     _TEST = {
@@ -26,12 +27,6 @@ class ComedyCentralIE(MTVIE):
             u'description': u'After a certain point, breastfeeding becomes c**kblocking.',
         },
     }
-    # Overwrite MTVIE properties we don't want
-    _TESTS = []
-
-    def _get_thumbnail_url(self, uri, itemdoc):
-        search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
-        return itemdoc.find(search_path).attrib['url']
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -158,13 +153,12 @@ class ComedyCentralShowsIE(InfoExtractor):
 
         uri = mMovieParams[0][1]
         indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri})
-        indexXml = self._download_webpage(indexUrl, epTitle,
+        idoc = self._download_xml(indexUrl, epTitle,
                                           u'Downloading show index',
                                           u'unable to download episode index')
 
         results = []
 
-        idoc = xml.etree.ElementTree.fromstring(indexXml)
         itemEls = idoc.findall('.//item')
         for partNum,itemEl in enumerate(itemEls):
             mediaId = itemEl.findall('./guid')[0].text
@@ -175,10 +169,9 @@ class ComedyCentralShowsIE(InfoExtractor):
 
             configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' +
                         compat_urllib_parse.urlencode({'uri': mediaId}))
-            configXml = self._download_webpage(configUrl, epTitle,
+            cdoc = self._download_xml(configUrl, epTitle,
                                                u'Downloading configuration for %s' % shortMediaId)
 
-            cdoc = xml.etree.ElementTree.fromstring(configXml)
             turls = []
             for rendition in cdoc.findall('.//rendition'):
                 finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
@@ -200,7 +193,7 @@ class ComedyCentralShowsIE(InfoExtractor):
                 })
 
             effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1)
-            info = {
+            results.append({
                 'id': shortMediaId,
                 'formats': formats,
                 'uploader': showId,
@@ -208,11 +201,6 @@ class ComedyCentralShowsIE(InfoExtractor):
                 'title': effTitle,
                 'thumbnail': None,
                 'description': compat_str(officialTitle),
-            }
-
-            # TODO: Remove when #980 has been merged
-            info.update(info['formats'][-1])
-
-            results.append(info)
+            })
 
         return results

youtube_dl/extractor/common.py

@@ -1,4 +1,5 @@
 import base64
+import json
 import os
 import re
 import socket
@@ -9,6 +10,7 @@ import xml.etree.ElementTree
 from ..utils import (
     compat_http_client,
     compat_urllib_error,
+    compat_urllib_parse_urlparse,
     compat_str,
 
     clean_html,
@@ -18,6 +20,7 @@ from ..utils import (
     sanitize_filename,
     unescapeHTML,
 )
+_NO_DEFAULT = object()
 
 
 class InfoExtractor(object):
@@ -34,15 +37,54 @@ class InfoExtractor(object):
     The dictionaries must include the following fields:
 
     id:             Video identifier.
-    url:            Final video URL.
     title:          Video title, unescaped.
-    ext:            Video filename extension.
 
-    Instead of url and ext, formats can also specified.
+    Additionally, it must contain either a formats entry or a url one:
+
+    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
+                                 ("mp4 container with h264/opus").
+                                 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").
+                                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
+                    * vbr        Average video bitrate in KBit/s
+                    * vcodec     Name of the video codec in use
+                    * 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" or so.
+                    * preference Order number of this format. If this field is
+                                 present and not None, the formats get sorted
+                                 by this field.
+                                 -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).
 
     The following fields are optional:
 
-    format:         The video format, defaults to ext (used for --get-format)
     thumbnails:     A list of dictionaries (with the entries "resolution" and
                     "url") for the varying thumbnails
     thumbnail:      Full URL to a video thumbnail image.
@@ -51,32 +93,14 @@ class InfoExtractor(object):
     upload_date:    Video upload date (YYYYMMDD).
     uploader_id:    Nickname or id of the video uploader.
     location:       Physical location of the video.
-    player_url:     SWF Player URL (used for rtmpdump).
     subtitles:      The subtitle file contents as a dictionary in the format
                     {language: subtitles}.
+    duration:       Length of the video in seconds, as an integer.
     view_count:     How many users have watched the video on the platform.
-    urlhandle:      [internal] The urlHandle to be used to download the file,
-                    like returned by urllib.request.urlopen
+    like_count:     Number of positive ratings of the video
+    dislike_count:  Number of negative ratings of the video
+    comment_count:  Number of comments on the video
     age_limit:      Age restriction for the video, as an integer (years)
-    formats:        A list of dictionaries for each format available, it must
-                    be 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
-                                ("mp4 container with h264/opus").
-                                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")
-                    * 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
-                    * abr       Average audio bitrate in KBit/s
-                    * acodec    Name of the audio codec in use
-                    * vbr       Average video bitrate in KBit/s
-                    * vcodec    Name of the video codec in use
-                    * filesize  The number of bytes, if known in advance
     webpage_url:    The url to the video webpage, if given to youtube-dl it
                     should allow to get the same result again. (It will be set
                     by YoutubeDL if it's missing)
@@ -151,27 +175,40 @@ class InfoExtractor(object):
     def IE_NAME(self):
         return type(self).__name__[:-2]
 
-    def _request_webpage(self, url_or_request, video_id, note=None, errnote=None):
+    def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
         """ Returns the response handle """
         if note is None:
             self.report_download_webpage(video_id)
         elif note is not False:
-            self.to_screen(u'%s: %s' % (video_id, note))
+            if video_id is None:
+                self.to_screen(u'%s' % (note,))
+            else:
+                self.to_screen(u'%s: %s' % (video_id, note))
         try:
             return self._downloader.urlopen(url_or_request)
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
+            if errnote is False:
+                return False
             if errnote is None:
                 errnote = u'Unable to download webpage'
-            raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2], cause=err)
+            errmsg = u'%s: %s' % (errnote, compat_str(err))
+            if fatal:
+                raise ExtractorError(errmsg, sys.exc_info()[2], cause=err)
+            else:
+                self._downloader.report_warning(errmsg)
+                return False
 
-    def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None):
+    def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
         """ Returns a tuple (page content as string, URL handle) """
 
         # Strip hashes from the URL (#1038)
         if isinstance(url_or_request, (compat_str, str)):
             url_or_request = url_or_request.partition('#')[0]
 
-        urlh = self._request_webpage(url_or_request, video_id, note, errnote)
+        urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal)
+        if urlh is False:
+            assert not fatal
+            return False
         content_type = urlh.headers.get('Content-Type', '')
         webpage_bytes = urlh.read()
         m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
@@ -206,15 +243,38 @@ class InfoExtractor(object):
         content = webpage_bytes.decode(encoding, 'replace')
         return (content, urlh)
 
-    def _download_webpage(self, url_or_request, video_id, note=None, errnote=None):
+    def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
         """ Returns the data of the page as a string """
-        return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0]
+        res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal)
+        if res is False:
+            return res
+        else:
+            content, _ = res
+            return content
 
-    def _download_xml(self, url_or_request, video_id, note=u'Downloading XML', errnote=u'Unable to downloand XML'):
+    def _download_xml(self, url_or_request, video_id,
+                      note=u'Downloading XML', errnote=u'Unable to download XML',
+                      transform_source=None):
         """Return the xml as an xml.etree.ElementTree.Element"""
         xml_string = self._download_webpage(url_or_request, video_id, note, errnote)
+        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'):
+        json_string = self._download_webpage(url_or_request, video_id, note, errnote)
+        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))
@@ -236,7 +296,8 @@ class InfoExtractor(object):
         self.to_screen(u'Logging in')
 
     #Methods for following #608
-    def url_result(self, url, ie=None, video_id=None):
+    @staticmethod
+    def url_result(url, ie=None, video_id=None):
         """Returns a url that points to a page that should be processed"""
         #TODO: ie should be the class used for getting the info
         video_info = {'_type': 'url',
@@ -245,7 +306,8 @@ class InfoExtractor(object):
         if video_id is not None:
             video_info['id'] = video_id
         return video_info
-    def playlist_result(self, entries, playlist_id=None, playlist_title=None):
+    @staticmethod
+    def playlist_result(entries, playlist_id=None, playlist_title=None):
         """Returns a playlist"""
         video_info = {'_type': 'playlist',
                       'entries': entries}
@@ -255,7 +317,7 @@ class InfoExtractor(object):
             video_info['title'] = playlist_title
         return video_info
 
-    def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
+    def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0):
         """
         Perform a regex search on the given string, using a single or a list of
         patterns returning the first matching group.
@@ -269,7 +331,7 @@ class InfoExtractor(object):
                 mobj = re.search(p, string, flags)
                 if mobj: break
 
-        if sys.stderr.isatty() and os.name != 'nt':
+        if os.name != 'nt' and sys.stderr.isatty():
             _name = u'\033[0;34m%s\033[0m' % name
         else:
             _name = name
@@ -277,7 +339,7 @@ class InfoExtractor(object):
         if mobj:
             # return the first matching group
             return next(g for g in mobj.groups() if g is not None)
-        elif default is not None:
+        elif default is not _NO_DEFAULT:
             return default
         elif fatal:
             raise RegexNotFoundError(u'Unable to extract %s' % _name)
@@ -286,7 +348,7 @@ class InfoExtractor(object):
                 u'please report this issue on http://yt-dl.org/bug' % _name)
             return None
 
-    def _html_search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
+    def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0):
         """
         Like _search_regex, but strips HTML tags and unescapes entities.
         """
@@ -330,7 +392,7 @@ class InfoExtractor(object):
     @staticmethod
     def _og_regexes(prop):
         content_re = r'content=(?:"([^>]+?)"|\'(.+?)\')'
-        property_re = r'property=[\'"]og:%s[\'"]' % re.escape(prop)
+        property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop)
         template = r'<meta[^>]+?%s[^>]+?%s'
         return [
             template % (property_re, content_re),
@@ -363,7 +425,8 @@ class InfoExtractor(object):
         if display_name is None:
             display_name = name
         return self._html_search_regex(
-            r'''(?ix)<meta(?=[^>]+(?:name|property)=["\']%s["\'])
+            r'''(?ix)<meta
+                    (?=[^>]+(?:itemprop|name|property)=["\']%s["\'])
                     [^>]+content=["\']([^"\']+)["\']''' % re.escape(name),
             html, display_name, fatal=False)
 
@@ -394,6 +457,58 @@ class InfoExtractor(object):
         }
         return RATING_TABLE.get(rating.lower(), None)
 
+    def _sort_formats(self, formats):
+        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)
 
 
 class SearchInfoExtractor(InfoExtractor):

youtube_dl/extractor/crunchyroll.py

@@ -0,0 +1,171 @@
+# encoding: utf-8
+import re, base64, zlib
+from hashlib import sha1
+from math import pow, sqrt, floor
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    compat_urllib_parse,
+    compat_urllib_request,
+    bytes_to_intlist,
+    intlist_to_bytes,
+    unified_strdate,
+    clean_html,
+)
+from ..aes import (
+    aes_cbc_decrypt,
+    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',
+        },
+        u'params': {
+            # rtmp
+            u'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'),
+    }
+
+    def _decrypt_subtitles(self, data, iv, id):
+        data = bytes_to_intlist(data)
+        iv = bytes_to_intlist(iv)
+        id = int(id)
+
+        def obfuscate_key_aux(count, modulo, start):
+            output = list(start)
+            for _ in range(count):
+                output.append(output[-1] + output[-2])
+            # cut off start values
+            output = output[2:]
+            output = list(map(lambda x: x % modulo + 33, output))
+            return output
+
+        def obfuscate_key(key):
+            num1 = int(floor(pow(2, 25) * sqrt(6.9)))
+            num2 = (num1 ^ key) << 5
+            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())
+            # Extend 160 Bit hash to 256 Bit
+            return shaHash + [0] * 12
+        
+        key = obfuscate_key(id)
+        class Counter:
+            __value = iv
+            def next_value(self):
+                temp = self.__value
+                self.__value = inc(self.__value)
+                return temp
+        decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
+        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',')
+            text = clean_html(text)
+            text = text.replace(u'\\N', u'\n')
+            if not text:
+                continue
+            output += u'%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
+            i+=1
+        return output
+
+    def _real_extract(self,url):
+        mobj = re.match(self._VALID_URL, url)
+
+        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 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'')
+        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)
+        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)
+
+        playerdata_url = compat_urllib_parse.unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, u'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)
+
+        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/')
+            # 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')
+            formats.append({
+                u'url': video_url,
+                u'play_path':   video_play_path,
+                u'ext': 'flv',
+                u'format': video_format,
+                u'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)
+            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)
+            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,
+        }

youtube_dl/extractor/cspan.py

@@ -1,20 +1,25 @@
+from __future__ import unicode_literals
+
+import json
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_urllib_parse,
+    unescapeHTML,
 )
 
+
 class CSpanIE(InfoExtractor):
-    _VALID_URL = r'http://www.c-spanvideo.org/program/(.*)'
+    _VALID_URL = r'http://www\.c-spanvideo\.org/program/(.*)'
+    IE_DESC = 'C-SPAN'
     _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"
+        'url': 'http://www.c-spanvideo.org/program/HolderonV',
+        'file': '315139.mp4',
+        'md5': '8e44ce11f0f725527daccc453f553eb0',
+        'info_dict': {
+            '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'
     }
 
     def _real_extract(self, url):
@@ -22,30 +27,22 @@ class CSpanIE(InfoExtractor):
         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')
 
-        self.report_extraction(video_id)
+        title = self._html_search_regex(
+            r'<!-- title -->\n\s*<h1[^>]*>(.*?)</h1>', webpage, 'title')
+        description = self._og_search_description(webpage)
 
-        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_json = self._download_webpage(
+            info_url, video_id, 'Downloading video info')
+        data = json.loads(data_json)
 
-        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')
+        url = unescapeHTML(data['video']['files'][0]['path']['#text'])
 
-        return {'id': video_id,
-                'title': title,
-                'ext': 'flv',
-                'url': url,
-                'play_path': path,
-                'description': description,
-                'thumbnail': self._og_search_thumbnail(webpage),
-                }
+        return {
+            'id': video_id,
+            'title': title,
+            'url': url,
+            'description': description,
+            'thumbnail': self._og_search_thumbnail(webpage),
+        }

youtube_dl/extractor/dailymotion.py

@@ -11,6 +11,7 @@ from ..utils import (
     get_element_by_attribute,
     get_element_by_id,
     orderedSet,
+    str_to_int,
 
     ExtractorError,
 )
@@ -27,7 +28,7 @@ class DailymotionBaseInfoExtractor(InfoExtractor):
 class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
     """Information Extractor for Dailymotion"""
 
-    _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?video/([^/]+)'
+    _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)'
     IE_NAME = u'dailymotion'
 
     _FORMATS = [
@@ -80,7 +81,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
         # Extract id and simplified title from URL
         mobj = re.match(self._VALID_URL, url)
 
-        video_id = mobj.group(1).split('_')[0].split('?')[0]
+        video_id = mobj.group('id')
 
         url = 'http://www.dailymotion.com/video/%s' % video_id
 
@@ -100,10 +101,6 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
             self.to_screen(u'Vevo video detected: %s' % vevo_id)
             return self.url_result(u'vevo:%s' % vevo_id, ie='Vevo')
 
-        video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
-                                             # Looking for official user
-                                             r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
-                                            webpage, 'video uploader', fatal=False)
         age_limit = self._rta_search(webpage)
 
         video_upload_date = None
@@ -146,15 +143,21 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor):
             self._list_available_subtitles(video_id, webpage)
             return
 
+        view_count = self._search_regex(
+            r'video_views_count[^>]+>\s+([\d\.,]+)', webpage, u'view count', fatal=False)
+        if view_count is not None:
+            view_count = str_to_int(view_count)
+
         return {
             'id':       video_id,
             'formats': formats,
-            'uploader': video_uploader,
+            'uploader': info['owner_screenname'],
             'upload_date':  video_upload_date,
             'title':    self._og_search_title(webpage),
             'subtitles':    video_subtitles,
             'thumbnail': info['thumbnail_url'],
             'age_limit': age_limit,
+            'view_count': view_count,
         }
 
     def _get_available_subtitles(self, video_id, webpage):

youtube_dl/extractor/daum.py

@@ -1,6 +1,5 @@
 # encoding: utf-8
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -10,7 +9,7 @@ from ..utils import (
 
 
 class DaumIE(InfoExtractor):
-    _VALID_URL = r'https?://tvpot\.daum\.net/.*?clipid=(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/.*?clipid=(?P<id>\d+)'
     IE_NAME = u'daum.net'
 
     _TEST = {
@@ -29,17 +28,16 @@ class DaumIE(InfoExtractor):
         video_id = mobj.group(1)
         canonical_url = 'http://tvpot.daum.net/v/%s' % video_id
         webpage = self._download_webpage(canonical_url, video_id)
-        full_id = self._search_regex(r'<link rel="video_src" href=".+?vid=(.+?)"',
+        full_id = self._search_regex(
+            r'<iframe src="http://videofarm.daum.net/controller/video/viewer/Video.html\?.*?vid=(.+?)[&"]',
             webpage, u'full id')
         query = compat_urllib_parse.urlencode({'vid': full_id})
-        info_xml = self._download_webpage(
+        info = self._download_xml(
             'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id,
             u'Downloading video info')
-        urls_xml = self._download_webpage(
+        urls = self._download_xml(
             'http://videofarm.daum.net/controller/api/open/v1_2/MovieData.apixml?' + query,
             video_id, u'Downloading video formats info')
-        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
-        urls = xml.etree.ElementTree.fromstring(urls_xml.encode('utf-8'))
 
         self.to_screen(u'%s: Getting video urls' % video_id)
         formats = []
@@ -49,10 +47,9 @@ class DaumIE(InfoExtractor):
                 'vid': full_id,
                 'profile': profile,
             })
-            url_xml = self._download_webpage(
+            url_doc = self._download_xml(
                 'http://videofarm.daum.net/controller/api/open/v1_2/MovieLocation.apixml?' + format_query,
                 video_id, note=False)
-            url_doc = xml.etree.ElementTree.fromstring(url_xml.encode('utf-8'))
             format_url = url_doc.find('result/url').text
             formats.append({
                 'url': format_url,
@@ -60,7 +57,7 @@ class DaumIE(InfoExtractor):
                 'format_id': profile,
             })
 
-        info = {
+        return {
             'id': video_id,
             'title': info.find('TITLE').text,
             'formats': formats,
@@ -69,6 +66,3 @@ class DaumIE(InfoExtractor):
             'duration': int(info.find('DURATION').text),
             'upload_date': info.find('REGDTTM').text[:8],
         }
-        # TODO: Remove when #980 has been merged
-        info.update(formats[-1])
-        return info

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/dreisat.py

@@ -1,22 +1,20 @@
 # coding: utf-8
 
 import re
-import xml.etree.ElementTree
 
 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...", 
@@ -30,8 +28,7 @@ class DreiSatIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         details_url = 'http://www.3sat.de/mediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
-        details_xml = self._download_webpage(details_url, video_id, note=u'Downloading video details')
-        details_doc = xml.etree.ElementTree.fromstring(details_xml.encode('utf-8'))
+        details_doc = self._download_xml(details_url, video_id, note=u'Downloading video details')
 
         thumbnail_els = details_doc.findall('.//teaserimage')
         thumbnails = [{
@@ -54,20 +51,14 @@ 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)
 
-        info = {
+        return {
             '_type': 'video',
             'id': video_id,
             'title': video_title,
@@ -78,8 +69,3 @@ class DreiSatIE(InfoExtractor):
             'uploader': video_uploader,
             'upload_date': upload_date,
         }
-
-        # TODO: Remove when #980 has been merged
-        info.update(formats[-1])
-
-        return info

youtube_dl/extractor/ebaumsworld.py

@@ -1,5 +1,4 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import determine_ext
@@ -21,9 +20,8 @@ class EbaumsWorldIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
-        config_xml = self._download_webpage(
+        config = self._download_xml(
             'http://www.ebaumsworld.com/video/player/%s' % video_id, video_id)
-        config = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
         video_url = config.find('file').text
 
         return {

youtube_dl/extractor/eighttracks.py

@@ -10,7 +10,7 @@ from ..utils import (
 
 class EightTracksIE(InfoExtractor):
     IE_NAME = '8tracks'
-    _VALID_URL = r'https?://8tracks.com/(?P<user>[^/]+)/(?P<id>[^/#]+)(?:#.*)?$'
+    _VALID_URL = r'https?://8tracks\.com/(?P<user>[^/]+)/(?P<id>[^/#]+)(?:#.*)?$'
     _TEST = {
         u"name": u"EightTracks",
         u"url": u"http://8tracks.com/ytdl/youtube-dl-test-tracks-a",

youtube_dl/extractor/exfm.py

@@ -8,7 +8,7 @@ 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'
+    _SOUNDCLOUD_URL = r'(?:http://)?(?:www\.)?api\.soundcloud\.com/tracks/([^/]+)/stream'
     _TESTS = [
         {
             u'url': u'http://ex.fm/song/eh359',

youtube_dl/extractor/facebook.py

@@ -17,7 +17,7 @@ from ..utils import (
 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'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:[^#?]*#!/)?(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)'
     _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'
@@ -27,7 +27,7 @@ class FacebookIE(InfoExtractor):
         u'file': u'120708114770723.mp4',
         u'md5': u'48975a41ccc4b7a581abd68651c1a5a8',
         u'info_dict': {
-            u"duration": 279, 
+            u"duration": 279,
             u"title": u"PEOPLE ARE AWESOME 2013"
         }
     }

youtube_dl/extractor/faz.py

@@ -1,6 +1,5 @@
 # encoding: utf-8
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -10,7 +9,7 @@ from ..utils import (
 
 class FazIE(InfoExtractor):
     IE_NAME = u'faz.net'
-    _VALID_URL = r'https?://www\.faz\.net/multimedia/videos/.*?-(?P<id>\d+).html'
+    _VALID_URL = r'https?://www\.faz\.net/multimedia/videos/.*?-(?P<id>\d+)\.html'
 
     _TEST = {
         u'url': u'http://www.faz.net/multimedia/videos/stockholm-chemie-nobelpreis-fuer-drei-amerikanische-forscher-12610585.html',
@@ -28,9 +27,8 @@ class FazIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
         config_xml_url = self._search_regex(r'writeFLV\(\'(.+?)\',', webpage,
             u'config xml url')
-        config_xml = self._download_webpage(config_xml_url, video_id,
+        config = self._download_xml(config_xml_url, video_id,
             u'Downloading config xml')
-        config = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
 
         encodings = config.find('ENCODINGS')
         formats = []
@@ -46,13 +44,10 @@ class FazIE(InfoExtractor):
             })
 
         descr = self._html_search_regex(r'<p class="Content Copy">(.*?)</p>', webpage, u'description')
-        info = {
+        return {
             'id': video_id,
             'title': self._og_search_title(webpage),
             'formats': formats,
             'description': descr,
             'thumbnail': config.find('STILL/STILL_BIG').text,
         }
-        # TODO: Remove when #980 has been merged
-        info.update(formats[-1])
-        return info

youtube_dl/extractor/fktv.py

@@ -12,7 +12,7 @@ from ..utils import (
 
 class FKTVIE(InfoExtractor):
     IE_NAME = u'fernsehkritik.tv'
-    _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik.tv/folge-(?P<ep>[0-9]+)(?:/.*)?'
+    _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik\.tv/folge-(?P<ep>[0-9]+)(?:/.*)?'
 
     _TEST = {
         u'url': u'http://fernsehkritik.tv/folge-1',
@@ -52,7 +52,7 @@ class FKTVIE(InfoExtractor):
 
 class FKTVPosteckeIE(InfoExtractor):
     IE_NAME = u'fernsehkritik.tv:postecke'
-    _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik.tv/inline-video/postecke.php\?(.*&)?ep=(?P<ep>[0-9]+)(&|$)'
+    _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik\.tv/inline-video/postecke\.php\?(.*&)?ep=(?P<ep>[0-9]+)(&|$)'
     _TEST = {
         u'url': u'http://fernsehkritik.tv/inline-video/postecke.php?iframe=true&width=625&height=440&ep=120',
         u'file': u'0120.flv',

youtube_dl/extractor/francetv.py

@@ -1,6 +1,5 @@
 # encoding: utf-8
 import re
-import xml.etree.ElementTree
 import json
 
 from .common import InfoExtractor
@@ -11,11 +10,10 @@ from ..utils import (
 
 class FranceTVBaseInfoExtractor(InfoExtractor):
     def _extract_video(self, video_id):
-        xml_desc = self._download_webpage(
+        info = self._download_xml(
             'http://www.francetvinfo.fr/appftv/webservices/video/'
             'getInfosOeuvre.php?id-diffusion='
             + video_id, video_id, 'Downloading XML config')
-        info = xml.etree.ElementTree.fromstring(xml_desc.encode('utf-8'))
 
         manifest_url = info.find('videos/video/url').text
         video_url = manifest_url.replace('manifest.f4m', 'index_2_av.m3u8')
@@ -23,7 +21,7 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
         thumbnail_path = info.find('image').text
 
         return {'id': video_id,
-                'ext': 'mp4',
+                'ext': 'flv' if video_url.startswith('rtmp') else 'mp4',
                 'url': video_url,
                 'title': info.find('titre').text,
                 'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', thumbnail_path),
@@ -47,7 +45,7 @@ class PluzzIE(FranceTVBaseInfoExtractor):
 
 class FranceTvInfoIE(FranceTVBaseInfoExtractor):
     IE_NAME = u'francetvinfo.fr'
-    _VALID_URL = r'https?://www\.francetvinfo\.fr/replay.*/(?P<title>.+).html'
+    _VALID_URL = r'https?://www\.francetvinfo\.fr/replay.*/(?P<title>.+)\.html'
 
     _TEST = {
         u'url': u'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
@@ -68,35 +66,101 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor):
         return self._extract_video(video_id)
 
 
-class France2IE(FranceTVBaseInfoExtractor):
-    IE_NAME = u'france2.fr'
-    _VALID_URL = r'''(?x)https?://www\.france2\.fr/
+class FranceTVIE(FranceTVBaseInfoExtractor):
+    IE_NAME = u'francetv'
+    IE_DESC = u'France 2, 3, 4, 5 and Ô'
+    _VALID_URL = r'''(?x)https?://www\.france[2345o]\.fr/
         (?:
-            emissions/.*?/videos/(?P<id>\d+)
-        |   emission/(?P<key>[^/?]+)
+            emissions/.*?/(videos|emissions)/(?P<id>[^/?]+)
+        |   (emissions?|jt)/(?P<key>[^/?]+)
         )'''
 
-    _TEST = {
-        u'url': u'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104',
-        u'file': u'75540104.mp4',
-        u'info_dict': {
-            u'title': u'13h15, le samedi...',
-            u'description': u'md5:2e5b58ba7a2d3692b35c792be081a03d',
+    _TESTS = [
+        # france2
+        {
+            u'url': u'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104',
+            u'file': u'75540104.mp4',
+            u'info_dict': {
+                u'title': u'13h15, le samedi...',
+                u'description': u'md5:2e5b58ba7a2d3692b35c792be081a03d',
+            },
+            u'params': {
+                # m3u8 download
+                u'skip_download': True,
+            },
         },
-        u'params': {
-            u'skip_download': True,
+        # france3
+        {
+            u'url': u'http://www.france3.fr/emissions/pieces-a-conviction/diffusions/13-11-2013_145575',
+            u'info_dict': {
+                u'id': u'000702326_CAPP_PicesconvictionExtrait313022013_120220131722_Au',
+                u'ext': u'flv',
+                u'title': u'Le scandale du prix des médicaments',
+                u'description': u'md5:1384089fbee2f04fc6c9de025ee2e9ce',
+            },
+            u'params': {
+                # rtmp download
+                u'skip_download': True,
+            },
         },
-    }
+        # france4
+        {
+            u'url': u'http://www.france4.fr/emissions/hero-corp/videos/rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
+            u'info_dict': {
+                u'id': u'rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
+                u'ext': u'flv',
+                u'title': u'Hero Corp Making of - Extrait 1',
+                u'description': u'md5:c87d54871b1790679aec1197e73d650a',
+            },
+            u'params': {
+                # rtmp download
+                u'skip_download': True,
+            },
+        },
+        # france5
+        {
+            u'url': u'http://www.france5.fr/emissions/c-a-dire/videos/92837968',
+            u'info_dict': {
+                u'id': u'92837968',
+                u'ext': u'mp4',
+                u'title': u'C à dire ?!',
+                u'description': u'md5:fb1db1cbad784dcce7c7a7bd177c8e2f',
+            },
+            u'params': {
+                # m3u8 download
+                u'skip_download': True,
+            },
+        },
+        # franceo
+        {
+            u'url': u'http://www.franceo.fr/jt/info-afrique/04-12-2013',
+            u'info_dict': {
+                u'id': u'92327925',
+                u'ext': u'mp4',
+                u'title': u'Infô-Afrique',
+                u'description': u'md5:ebf346da789428841bee0fd2a935ea55',
+            },
+            u'params': {
+                # m3u8 download
+                u'skip_download': True,
+            },
+            u'skip': u'The id changes frequently',
+        },
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         if mobj.group('key'):
             webpage = self._download_webpage(url, mobj.group('key'))
-            video_id = self._html_search_regex(
-                r'''(?x)<div\s+class="video-player">\s*
+            id_res = [
+                (r'''(?x)<div\s+class="video-player">\s*
                     <a\s+href="http://videos.francetv.fr/video/([0-9]+)"\s+
-                    class="francetv-video-player">''',
-                webpage, u'video ID')
+                    class="francetv-video-player">'''),
+                (r'<a id="player_direct" href="http://info\.francetelevisions'
+                 '\.fr/\?id-video=([^"/&]+)'),
+                (r'<a class="video" id="ftv_player_(.+?)"'),
+            ]
+            video_id = self._html_search_regex(id_res, webpage, u'video ID')
         else:
             video_id = mobj.group('id')
         return self._extract_video(video_id)
@@ -127,3 +191,29 @@ class GenerationQuoiIE(InfoExtractor):
         info = json.loads(info_json)
         return self.url_result('http://www.dailymotion.com/video/%s' % info['id'],
             ie='Dailymotion')
+
+
+class CultureboxIE(FranceTVBaseInfoExtractor):
+    IE_NAME = u'culturebox.francetvinfo.fr'
+    _VALID_URL = r'https?://culturebox\.francetvinfo\.fr/(?P<name>.*?)(\?|$)'
+
+    _TEST = {
+        u'url': u'http://culturebox.francetvinfo.fr/einstein-on-the-beach-au-theatre-du-chatelet-146813',
+        u'info_dict': {
+            u'id': u'EV_6785',
+            u'ext': u'mp4',
+            u'title': u'Einstein on the beach au Théâtre du Châtelet',
+            u'description': u'md5:9ce2888b1efefc617b5e58b3f6200eeb',
+        },
+        u'params': {
+            # m3u8 download
+            u'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        name = mobj.group('name')
+        webpage = self._download_webpage(url, name)
+        video_id = self._search_regex(r'"http://videos\.francetv\.fr/video/(.*?)"', webpage, u'video id')
+        return self._extract_video(video_id)

youtube_dl/extractor/gamekings.py

@@ -4,7 +4,7 @@ from .common import InfoExtractor
 
 
 class GamekingsIE(InfoExtractor):
-    _VALID_URL = r'http?://www\.gamekings\.tv/videos/(?P<name>[0-9a-z\-]+)'
+    _VALID_URL = r'http://www\.gamekings\.tv/videos/(?P<name>[0-9a-z\-]+)'
     _TEST = {
         u"url": u"http://www.gamekings.tv/videos/phoenix-wright-ace-attorney-dual-destinies-review/",
         u'file': u'20130811.mp4',

youtube_dl/extractor/gamespot.py

@@ -47,13 +47,10 @@ class GameSpotIE(InfoExtractor):
                 'format_id': q,
             })
 
-        info = {
+        return {
             'id': data_video['guid'],
             'title': compat_urllib_parse.unquote(data_video['title']),
             'formats': formats,
             'description': get_meta_content('description', webpage),
             'thumbnail': self._og_search_thumbnail(webpage),
         }
-        # TODO: Remove when #980 has been merged
-        info.update(formats[-1])
-        return info

youtube_dl/extractor/gametrailers.py

@@ -1,13 +1,10 @@
 import re
 
-from .mtv import MTVIE, _media_xml_tag
+from .mtv import MTVServicesInfoExtractor
 
-class GametrailersIE(MTVIE):
-    """
-    Gametrailers use the same videos system as MTVIE, it just changes the feed
-    url, where the uri is and the method to get the thumbnails.
-    """
-    _VALID_URL = r'http://www.gametrailers.com/(?P<type>videos|reviews|full-episodes)/(?P<id>.*?)/(?P<title>.*)'
+
+class GametrailersIE(MTVServicesInfoExtractor):
+    _VALID_URL = r'http://www\.gametrailers\.com/(?P<type>videos|reviews|full-episodes)/(?P<id>.*?)/(?P<title>.*)'
     _TEST = {
         u'url': u'http://www.gametrailers.com/videos/zbvr8i/mirror-s-edge-2-e3-2013--debut-trailer',
         u'file': u'70e9a5d7-cf25-4a10-9104-6f3e7342ae0d.mp4',
@@ -17,15 +14,9 @@ class GametrailersIE(MTVIE):
             u'description': u'Faith is back!  Check out the World Premiere trailer for Mirror\'s Edge 2 straight from the EA Press Conference at E3 2013!',
         },
     }
-    # Overwrite MTVIE properties we don't want
-    _TESTS = []
 
     _FEED_URL = 'http://www.gametrailers.com/feeds/mrss'
 
-    def _get_thumbnail_url(self, uri, itemdoc):
-        search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
-        return itemdoc.find(search_path).attrib['url']
-
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')

youtube_dl/extractor/generic.py

@@ -1,9 +1,12 @@
 # encoding: utf-8
 
+from __future__ import unicode_literals
+
 import os
 import re
 
 from .common import InfoExtractor
+from .youtube import YoutubeIE
 from ..utils import (
     compat_urllib_error,
     compat_urllib_parse,
@@ -11,64 +14,89 @@ from ..utils import (
     compat_urlparse,
 
     ExtractorError,
+    HEADRequest,
     smuggle_url,
     unescapeHTML,
+    unified_strdate,
+    url_basename,
 )
 from .brightcove import BrightcoveIE
+from .ooyala import OoyalaIE
 
 
 class GenericIE(InfoExtractor):
-    IE_DESC = u'Generic downloader that works on some sites'
+    IE_DESC = 'Generic downloader that works on some sites'
     _VALID_URL = r'.*'
-    IE_NAME = u'generic'
+    IE_NAME = 'generic'
     _TESTS = [
         {
-            u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
-            u'file': u'13601338388002.mp4',
-            u'md5': u'6e15c93721d7ec9e9ca3fdbf07982cfd',
-            u'info_dict': {
-                u"uploader": u"www.hodiho.fr",
-                u"title": u"R\u00e9gis plante sa Jeep"
+            'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
+            'file': '13601338388002.mp4',
+            'md5': '6e15c93721d7ec9e9ca3fdbf07982cfd',
+            'info_dict': {
+                'uploader': 'www.hodiho.fr',
+                'title': 'R\u00e9gis plante sa Jeep',
             }
         },
         # embedded vimeo video
         {
-            u'add_ie': ['Vimeo'],
-            u'url': u'http://skillsmatter.com/podcast/home/move-semanticsperfect-forwarding-and-rvalue-references',
-            u'file': u'22444065.mp4',
-            u'md5': u'2903896e23df39722c33f015af0666e2',
-            u'info_dict': {
-                u'title': u'ACCU 2011: Move Semantics,Perfect Forwarding, and Rvalue references- Scott Meyers- 13/04/2011',
-                u"uploader_id": u"skillsmatter",
-                u"uploader": u"Skills Matter",
+            'add_ie': ['Vimeo'],
+            'url': 'http://skillsmatter.com/podcast/home/move-semanticsperfect-forwarding-and-rvalue-references',
+            'file': '22444065.mp4',
+            'md5': '2903896e23df39722c33f015af0666e2',
+            'info_dict': {
+                'title': 'ACCU 2011: Move Semantics,Perfect Forwarding, and Rvalue references- Scott Meyers- 13/04/2011',
+                'uploader_id': 'skillsmatter',
+                'uploader': 'Skills Matter',
             }
         },
         # bandcamp page with custom domain
         {
-            u'add_ie': ['Bandcamp'],
-            u'url': u'http://bronyrock.com/track/the-pony-mash',
-            u'file': u'3235767654.mp3',
-            u'info_dict': {
-                u'title': u'The Pony Mash',
-                u'uploader': u'M_Pallante',
+            'add_ie': ['Bandcamp'],
+            'url': 'http://bronyrock.com/track/the-pony-mash',
+            'file': '3235767654.mp3',
+            'info_dict': {
+                'title': 'The Pony Mash',
+                'uploader': 'M_Pallante',
             },
-            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',
         },
         # embedded brightcove video
         # it also tests brightcove videos that need to set the 'Referer' in the
         # http requests
         {
-            u'add_ie': ['Brightcove'],
-            u'url': u'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
-            u'info_dict': {
-                u'id': u'2765128793001',
-                u'ext': u'mp4',
-                u'title': u'Le cours de bourse : l’analyse technique',
-                u'description': u'md5:7e9ad046e968cb2d1114004aba466fd9',
-                u'uploader': u'BFM BUSINESS',
+            'add_ie': ['Brightcove'],
+            'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
+            'info_dict': {
+                'id': '2765128793001',
+                'ext': 'mp4',
+                'title': 'Le cours de bourse : l’analyse technique',
+                'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
+                'uploader': 'BFM BUSINESS',
             },
-            u'params': {
-                u'skip_download': True,
+            'params': {
+                'skip_download': True,
+            },
+        },
+        # Direct link to a video
+        {
+            'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
+            'file': 'trailer.mp4',
+            'md5': '67d406c2bcb6af27fa886f31aa934bbe',
+            'info_dict': {
+                'id': 'trailer',
+                'title': 'trailer',
+                'upload_date': '20100513',
+            }
+        },
+        # ooyala video
+        {
+            'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
+            'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
+            'info_dict': {
+                'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
+                'ext': 'mp4',
+                'title': '2cc213299525360.mov', #that's what we get
             },
         },
     ]
@@ -76,30 +104,27 @@ class GenericIE(InfoExtractor):
     def report_download_webpage(self, video_id):
         """Report webpage download."""
         if not self._downloader.params.get('test', False):
-            self._downloader.report_warning(u'Falling back on generic information extractor.')
+            self._downloader.report_warning('Falling back on generic information extractor.')
         super(GenericIE, self).report_download_webpage(video_id)
 
     def report_following_redirect(self, new_url):
         """Report information extraction."""
-        self._downloader.to_screen(u'[redirect] Following redirect to %s' % new_url)
+        self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
 
-    def _test_redirect(self, url):
+    def _send_head(self, url):
         """Check if it is a redirect, like url shorteners, in case return the new url."""
-        class HeadRequest(compat_urllib_request.Request):
-            def get_method(self):
-                return "HEAD"
 
         class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler):
             """
             Subclass the HTTPRedirectHandler to make it use our
-            HeadRequest also on the redirected URL
+            HEADRequest also on the redirected URL
             """
             def redirect_request(self, req, fp, code, msg, headers, newurl):
                 if code in (301, 302, 303, 307):
                     newurl = newurl.replace(' ', '%20')
                     newheaders = dict((k,v) for k,v in req.headers.items()
                                       if k.lower() not in ("content-length", "content-type"))
-                    return HeadRequest(newurl,
+                    return HEADRequest(newurl,
                                        headers=newheaders,
                                        origin_req_host=req.get_origin_req_host(),
                                        unverifiable=True)
@@ -128,38 +153,57 @@ class GenericIE(InfoExtractor):
                         compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]:
             opener.add_handler(handler())
 
-        response = opener.open(HeadRequest(url))
+        response = opener.open(HEADRequest(url))
         if response is None:
-            raise ExtractorError(u'Invalid URL protocol')
-        new_url = response.geturl()
-
-        if url == new_url:
-            return False
-
-        self.report_following_redirect(new_url)
-        return new_url
+            raise ExtractorError('Invalid URL protocol')
+        return response
 
     def _real_extract(self, url):
         parsed_url = compat_urlparse.urlparse(url)
         if not parsed_url.scheme:
             self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
             return self.url_result('http://' + url)
+        video_id = os.path.splitext(url.split('/')[-1])[0]
+
+        self.to_screen('%s: Requesting header' % video_id)
 
         try:
-            new_url = self._test_redirect(url)
-            if new_url:
-                return [self.url_result(new_url)]
+            response = self._send_head(url)
+
+            # Check for redirect
+            new_url = response.geturl()
+            if url != new_url:
+                self.report_following_redirect(new_url)
+                return self.url_result(new_url)
+
+            # Check for direct link to a video
+            content_type = response.headers.get('Content-Type', '')
+            m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
+            if m:
+                upload_date = response.headers.get('Last-Modified')
+                if upload_date:
+                    upload_date = unified_strdate(upload_date)
+                return {
+                    'id': video_id,
+                    'title': os.path.splitext(url_basename(url))[0],
+                    'formats': [{
+                        'format_id': m.group('format_id'),
+                        'url': url,
+                        'vcodec': 'none' if m.group('type') == 'audio' else None
+                    }],
+                    'upload_date': upload_date,
+                }
+
         except compat_urllib_error.HTTPError:
             # This may be a stupid server that doesn't like HEAD, our UA, or so
             pass
 
-        video_id = url.split('/')[-1]
         try:
             webpage = self._download_webpage(url, video_id)
         except ValueError:
             # since this is the last-resort InfoExtractor, if
             # this error is thrown, it'll be thrown here
-            raise ExtractorError(u'Failed to download URL: %s' % url)
+            raise ExtractorError('Failed to download URL: %s' % url)
 
         self.report_extraction(video_id)
 
@@ -169,32 +213,77 @@ class GenericIE(InfoExtractor):
         #   Site Name | Video Title
         #   Video Title - Tagline | Site Name
         # and so on and so forth; it's just not practical
-        video_title = self._html_search_regex(r'<title>(.*)</title>',
-            webpage, u'video title', default=u'video', flags=re.DOTALL)
+        video_title = self._html_search_regex(
+            r'(?s)<title>(.*?)</title>', webpage, 'video title',
+            default='video')
+
+        # video uploader is domain name
+        video_uploader = self._search_regex(
+            r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
 
         # Look for BrightCove:
         bc_url = BrightcoveIE._extract_brightcove_url(webpage)
         if bc_url is not None:
-            self.to_screen(u'Brightcove video detected.')
-            return self.url_result(bc_url, 'Brightcove')
+            self.to_screen('Brightcove video detected.')
+            surl = smuggle_url(bc_url, {'Referer': url})
+            return self.url_result(surl, 'Brightcove')
 
-        # Look for embedded Vimeo player
+        # Look for embedded (iframe) Vimeo player
         mobj = re.search(
-            r'<iframe[^>]+?src="(https?://player.vimeo.com/video/.+?)"', webpage)
+            r'<iframe[^>]+?src="((?:https?:)?//player.vimeo.com/video/.+?)"', webpage)
         if mobj:
             player_url = unescapeHTML(mobj.group(1))
             surl = smuggle_url(player_url, {'Referer': url})
             return self.url_result(surl, 'Vimeo')
 
+        # Look for embedded (swf embed) Vimeo player
+        mobj = re.search(
+            r'<embed[^>]+?src="(https?://(?:www\.)?vimeo.com/moogaloop.swf.+?)"', webpage)
+        if mobj:
+            return self.url_result(mobj.group(1), 'Vimeo')
+
         # Look for embedded YouTube player
-        matches = re.findall(
-            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?youtube.com/embed/.+?)\1', webpage)
+        matches = re.findall(r'''(?x)
+            (?:<iframe[^>]+?src=|embedSWF\(\s*)
+            (["\'])(?P<url>(?:https?:)?//(?:www\.)?youtube\.com/
+                (?:embed|v)/.+?)
+            \1''', webpage)
         if matches:
             urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube')
                      for tuppl in matches]
             return self.playlist_result(
                 urlrs, playlist_id=video_id, playlist_title=video_title)
 
+        # Look for embedded Dailymotion player
+        matches = re.findall(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
+        if matches:
+            urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Dailymotion')
+                     for tuppl in matches]
+            return self.playlist_result(
+                urlrs, playlist_id=video_id, playlist_title=video_title)
+
+        # Look for embedded Wistia player
+        match = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
+        if match:
+            return {
+                '_type': 'url_transparent',
+                'url': unescapeHTML(match.group('url')),
+                'ie_key': 'Wistia',
+                'uploader': video_uploader,
+                'title': video_title,
+                'id': video_id,
+            }
+
+        # Look for embedded blip.tv player
+        mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
+        if mobj:
+            return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
+        mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9]+)', webpage)
+        if mobj:
+            return self.url_result(mobj.group(1), 'BlipTV')
+
         # Look for Bandcamp pages with custom domain
         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
         if mobj is not None:
@@ -202,14 +291,44 @@ class GenericIE(InfoExtractor):
             # Don't set the extractor because it can be a track url or an album
             return self.url_result(burl)
 
+        # Look for embedded Vevo player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'))
+
+        # Look for Ooyala videos
+        mobj = re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=([^"&]+)', webpage)
+        if mobj is not None:
+            return OoyalaIE._build_url_result(mobj.group(1))
+
+        # Look for Aparat videos
+        mobj = re.search(r'<iframe src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group(1), 'Aparat')
+
+        # Look for MPORA videos
+        mobj = re.search(r'<iframe .*?src="(http://mpora\.com/videos/[^"]+)"', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group(1), 'Mpora')
+
+        # Look for embedded Novamov player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>http://(?:(?:embed|www)\.)?novamov\.com/embed\.php.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'Novamov')
+
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
+        if mobj is None:
+            # Look for gorilla-vid style embedding
+            mobj = re.search(r'(?s)jw_plugins.*?file:\s*["\'](.*?)["\']', webpage)
         if mobj is None:
             # Broaden the search a little bit
             mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
         if mobj is None:
             # Broaden the search a little bit: JWPlayer JS loader
-            mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http[^\'"]*)', webpage)
+            mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage)
         if mobj is None:
             # Try to find twitter cards info
             mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
@@ -224,28 +343,27 @@ class GenericIE(InfoExtractor):
             # HTML5 video
             mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
         if mobj is None:
-            raise ExtractorError(u'Unsupported URL: %s' % url)
+            raise ExtractorError('Unsupported URL: %s' % url)
 
         # It's possible that one of the regexes
         # matched, but returned an empty group:
         if mobj.group(1) is None:
-            raise ExtractorError(u'Did not find a valid video URL at %s' % url)
+            raise ExtractorError('Did not find a valid video URL at %s' % url)
 
         video_url = mobj.group(1)
         video_url = compat_urlparse.urljoin(url, video_url)
         video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
 
+        # Sometimes, jwplayer extraction will result in a YouTube URL
+        if YoutubeIE.suitable(video_url):
+            return self.url_result(video_url, 'Youtube')
+
         # here's a fun little line of code for you:
         video_id = os.path.splitext(video_id)[0]
 
-        # video uploader is domain name
-        video_uploader = self._search_regex(r'(?:https?://)?([^/]*)/.*',
-            url, u'video uploader')
-
         return {
-            'id':       video_id,
-            'url':      video_url,
+            'id': video_id,
+            'url': video_url,
             'uploader': video_uploader,
-            'upload_date':  None,
-            'title':    video_title,
+            'title': video_title,
         }

youtube_dl/extractor/hotnewhiphop.py

@@ -11,7 +11,7 @@ class HotNewHipHopIE(InfoExtractor):
         u'file': u'1435540.mp3',
         u'md5': u'2c2cd2f76ef11a9b3b581e8b232f3d96',
         u'info_dict': {
-            u"title": u"Freddie Gibbs - Lay It Down"
+            u"title": u'Freddie Gibbs "Lay It Down"'
         }
     }
 

youtube_dl/extractor/ign.py

@@ -44,7 +44,7 @@ class IGNIE(InfoExtractor):
                 {
                     u'file': u'638672ee848ae4ff108df2a296418ee2.mp4',
                     u'info_dict': {
-                        u'title': u'GTA 5\'s Twisted Beauty in Super Slow Motion',
+                        u'title': u'26 Twisted Moments from GTA 5 in Slow Motion',
                         u'description': u'The twisted beauty of GTA 5 in stunning slow motion.',
                     },
                 },
@@ -103,7 +103,7 @@ class IGNIE(InfoExtractor):
 class OneUPIE(IGNIE):
     """Extractor for 1up.com, it uses the ign videos system."""
 
-    _VALID_URL = r'https?://gamevideos.1up.com/(?P<type>video)/id/(?P<name_or_id>.+)'
+    _VALID_URL = r'https?://gamevideos\.1up\.com/(?P<type>video)/id/(?P<name_or_id>.+)'
     IE_NAME = '1up.com'
 
     _DESCRIPTION_RE = r'<div id="vid_summary">(.+?)</div>'

youtube_dl/extractor/imdb.py

@@ -0,0 +1,89 @@
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    get_element_by_attribute,
+)
+
+
+class ImdbIE(InfoExtractor):
+    IE_NAME = 'imdb'
+    IE_DESC = 'Internet Movie Database trailers'
+    _VALID_URL = r'http://(?:www|m)\.imdb\.com/video/imdb/vi(?P<id>\d+)'
+
+    _TEST = {
+        'url': 'http://www.imdb.com/video/imdb/vi2524815897',
+        'md5': '9f34fa777ade3a6e57a054fdbcb3a068',
+        'info_dict': {
+            'id': '2524815897',
+            'ext': 'mp4',
+            'title': 'Ice Age: Continental Drift Trailer (No. 2) - IMDb',
+            'description': 'md5:9061c2219254e5d14e03c25c98e96a81',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage('http://www.imdb.com/video/imdb/vi%s' % video_id, video_id)
+        descr = get_element_by_attribute('itemprop', 'description', webpage)
+        available_formats = re.findall(
+            r'case \'(?P<f_id>.*?)\' :$\s+url = \'(?P<path>.*?)\'', webpage,
+            flags=re.MULTILINE)
+        formats = []
+        for f_id, f_path in available_formats:
+            f_path = f_path.strip()
+            format_page = self._download_webpage(
+                compat_urlparse.urljoin(url, f_path),
+                'Downloading info for %s format' % f_id)
+            json_data = self._search_regex(
+                r'<script[^>]+class="imdb-player-data"[^>]*?>(.*?)</script>',
+                format_page, 'json data', flags=re.DOTALL)
+            info = json.loads(json_data)
+            format_info = info['videoPlayerObject']['video']
+            formats.append({
+                'format_id': f_id,
+                'url': format_info['url'],
+            })
+
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'formats': formats,
+            'description': descr,
+            'thumbnail': format_info['slate'],
+        }
+
+
+class ImdbListIE(InfoExtractor):
+    IE_NAME = 'imdb:list'
+    IE_DESC = 'Internet Movie Database lists'
+    _VALID_URL = r'http://www\.imdb\.com/list/(?P<id>[\da-zA-Z_-]{11})'
+    
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        list_id = mobj.group('id')
+        
+        # RSS XML is sometimes malformed
+        rss = self._download_webpage('http://rss.imdb.com/list/%s' % list_id, list_id, 'Downloading list RSS')
+        list_title = self._html_search_regex(r'<title>(.*?)</title>', rss, 'list title')
+        
+        # Export is independent of actual author_id, but returns 404 if no author_id is provided.
+        # However, passing dummy author_id seems to be enough.
+        csv = self._download_webpage('http://www.imdb.com/list/export?list_id=%s&author_id=ur00000000' % list_id,
+                                     list_id, 'Downloading list CSV')
+        
+        entries = []
+        for item in csv.split('\n')[1:]:
+            cols = item.split(',')
+            if len(cols) < 2:
+                continue
+            item_id = cols[1][1:-1]
+            if item_id.startswith('vi'):
+                entries.append(self.url_result('http://www.imdb.com/video/imdb/%s' % item_id, 'Imdb'))
+        
+        return self.playlist_result(entries, list_id, list_title)

youtube_dl/extractor/ina.py

@@ -7,7 +7,7 @@ class InaIE(InfoExtractor):
     """Information Extractor for Ina.fr"""
     _VALID_URL = r'(?:http://)?(?:www\.)?ina\.fr/video/(?P<id>I?[A-F0-9]+)/.*'
     _TEST = {
-        u'url': u'www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html',
+        u'url': u'http://www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html',
         u'file': u'I12055569.mp4',
         u'md5': u'a667021bf2b41f8dc6049479d9bb38a3',
         u'info_dict': {

youtube_dl/extractor/instagram.py

@@ -3,7 +3,7 @@ import re
 from .common import InfoExtractor
 
 class InstagramIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?instagram.com/p/(.*?)/'
+    _VALID_URL = r'(?:http://)?instagram\.com/p/(.*?)/'
     _TEST = {
         u'url': u'http://instagram.com/p/aye83DjauH/?foo=bar#abc',
         u'file': u'aye83DjauH.mp4',

youtube_dl/extractor/internetvideoarchive.py

@@ -1,12 +1,10 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
     compat_urlparse,
     compat_urllib_parse,
     xpath_with_ns,
-    determine_ext,
 )
 
 
@@ -43,9 +41,8 @@ class InternetVideoArchiveIE(InfoExtractor):
         video_id = query_dic['publishedid'][0]
         url = self._build_url(query)
 
-        flashconfiguration_xml = self._download_webpage(url, video_id,
+        flashconfiguration = self._download_xml(url, video_id,
             u'Downloading flash configuration')
-        flashconfiguration = xml.etree.ElementTree.fromstring(flashconfiguration_xml.encode('utf-8'))
         file_url = flashconfiguration.find('file').text
         file_url = file_url.replace('/playlist.aspx', '/mrssplaylist.aspx')
         # Replace some of the parameters in the query to get the best quality
@@ -53,9 +50,8 @@ class InternetVideoArchiveIE(InfoExtractor):
         file_url = re.sub(r'(?<=\?)(.+)$',
             lambda m: self._clean_query(m.group()),
             file_url)
-        info_xml = self._download_webpage(file_url, video_id,
+        info = self._download_xml(file_url, video_id,
             u'Downloading video info')
-        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
         item = info.find('channel/item')
 
         def _bp(p):
@@ -66,13 +62,17 @@ class InternetVideoArchiveIE(InfoExtractor):
         for content in item.findall(_bp('media:group/media:content')):
             attr = content.attrib
             f_url = attr['url']
+            width = int(attr['width'])
+            bitrate = int(attr['bitrate'])
+            format_id = '%d-%dk' % (width, bitrate)
             formats.append({
+                'format_id': format_id,
                 'url': f_url,
-                'ext': determine_ext(f_url),
-                'width': int(attr['width']),
-                'bitrate': int(attr['bitrate']),
+                'width': width,
+                'tbr': bitrate,
             })
-        formats = sorted(formats, key=lambda f: f['bitrate'])
+
+        self._sort_formats(formats)
 
         return {
             'id': video_id,

youtube_dl/extractor/ivi.py

@@ -0,0 +1,156 @@
+# encoding: utf-8
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_request,
+    ExtractorError,
+)
+
+
+class IviIE(InfoExtractor):
+    IE_DESC = u'ivi.ru'
+    IE_NAME = u'ivi'
+    _VALID_URL = r'^https?://(?:www\.)?ivi\.ru/watch(?:/(?P<compilationid>[^/]+))?/(?P<videoid>\d+)'
+
+    _TESTS = [
+        # Single movie
+        {
+            u'url': u'http://www.ivi.ru/watch/53141',
+            u'file': u'53141.mp4',
+            u'md5': u'6ff5be2254e796ed346251d117196cf4',
+            u'info_dict': {
+                u'title': u'Иван Васильевич меняет профессию',
+                u'description': u'md5:14d8eda24e9d93d29b5857012c6d6346',
+                u'duration': 5498,
+                u'thumbnail': u'http://thumbs.ivi.ru/f20.vcp.digitalaccess.ru/contents/d/1/c3c885163a082c29bceeb7b5a267a6.jpg',
+            },
+            u'skip': u'Only works from Russia',
+        },
+        # Serial's serie
+        {
+            u'url': u'http://www.ivi.ru/watch/dezhurnyi_angel/74791',
+            u'file': u'74791.mp4',
+            u'md5': u'3e6cc9a848c1d2ebcc6476444967baa9',
+            u'info_dict': {
+                u'title': u'Дежурный ангел - 1 серия',
+                u'duration': 2490,
+                u'thumbnail': u'http://thumbs.ivi.ru/f7.vcp.digitalaccess.ru/contents/8/e/bc2f6c2b6e5d291152fdd32c059141.jpg',
+            },
+            u'skip': u'Only works from Russia',
+         }
+    ]
+    
+    # Sorted by quality
+    _known_formats = ['MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi', 'MP4-SHQ']
+
+    # Sorted by size
+    _known_thumbnails = ['Thumb-120x90', 'Thumb-160', 'Thumb-640x480']
+
+    def _extract_description(self, html):
+        m = re.search(r'<meta name="description" content="(?P<description>[^"]+)"/>', html)
+        return m.group('description') if m is not None else None
+
+    def _extract_comment_count(self, html):
+        m = re.search(u'(?s)<a href="#" id="view-comments" class="action-button dim gradient">\s*Комментарии:\s*(?P<commentcount>\d+)\s*</a>', html)
+        return int(m.group('commentcount')) if m is not None else 0
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('videoid')
+
+        api_url = 'http://api.digitalaccess.ru/api/json/'
+
+        data = {u'method': u'da.content.get',
+                u'params': [video_id, {u'site': u's183',
+                                       u'referrer': u'http://www.ivi.ru/watch/%s' % video_id,
+                                       u'contentid': video_id
+                                    }
+                            ]
+                }
+
+        request = compat_urllib_request.Request(api_url, json.dumps(data))
+
+        video_json_page = self._download_webpage(request, video_id, u'Downloading video JSON')
+        video_json = json.loads(video_json_page)
+
+        if u'error' in video_json:
+            error = video_json[u'error']
+            if error[u'origin'] == u'NoRedisValidData':
+                raise ExtractorError(u'Video %s does not exist' % video_id, expected=True)
+            raise ExtractorError(u'Unable to download video %s: %s' % (video_id, error[u'message']), expected=True)
+
+        result = video_json[u'result']
+
+        formats = [{
+            'url': x[u'url'],
+            'format_id': x[u'content_format'],
+            'preference': self._known_formats.index(x[u'content_format']),
+        } for x in result[u'files'] if x[u'content_format'] in self._known_formats]
+
+        self._sort_formats(formats)
+
+        if not formats:
+            raise ExtractorError(u'No media links available for %s' % video_id)
+
+        duration = result[u'duration']
+        compilation = result[u'compilation']
+        title = result[u'title']
+
+        title = '%s - %s' % (compilation, title) if compilation is not None else title  
+
+        previews = result[u'preview']
+        previews.sort(key=lambda fmt: self._known_thumbnails.index(fmt['content_format']))
+        thumbnail = previews[-1][u'url'] if len(previews) > 0 else None
+
+        video_page = self._download_webpage(url, video_id, u'Downloading video page')
+        description = self._extract_description(video_page)
+        comment_count = self._extract_comment_count(video_page)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'description': description,
+            'duration': duration,
+            'comment_count': comment_count,
+            'formats': formats,
+        }
+
+
+class IviCompilationIE(InfoExtractor):
+    IE_DESC = u'ivi.ru compilations'
+    IE_NAME = u'ivi:compilation'
+    _VALID_URL = r'^https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
+
+    def _extract_entries(self, html, compilation_id):
+        return [self.url_result('http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), 'Ivi')
+                for serie in re.findall(r'<strong><a href="/watch/%s/(\d+)">(?:[^<]+)</a></strong>' % compilation_id, html)]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        compilation_id = mobj.group('compilationid')
+        season_id = mobj.group('seasonid')
+
+        if season_id is not None: # Season link
+            season_page = self._download_webpage(url, compilation_id, u'Downloading season %s web page' % season_id)
+            playlist_id = '%s/season%s' % (compilation_id, season_id)
+            playlist_title = self._html_search_meta(u'title', season_page, u'title')
+            entries = self._extract_entries(season_page, compilation_id)
+        else: # Compilation link            
+            compilation_page = self._download_webpage(url, compilation_id, u'Downloading compilation web page')
+            playlist_id = compilation_id
+            playlist_title = self._html_search_meta(u'title', compilation_page, u'title')
+            seasons = re.findall(r'<a href="/watch/%s/season(\d+)">[^<]+</a>' % compilation_id, compilation_page)
+            if len(seasons) == 0: # No seasons in this compilation
+                entries = self._extract_entries(compilation_page, compilation_id)
+            else:
+                entries = []
+                for season_id in seasons:
+                    season_page = self._download_webpage('http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
+                                                         compilation_id, u'Downloading season %s web page' % season_id)
+                    entries.extend(self._extract_entries(season_page, compilation_id))
+
+        return self.playlist_result(entries, playlist_id, playlist_title)

youtube_dl/extractor/jeuxvideo.py

@@ -2,7 +2,6 @@
 
 import json
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 
@@ -32,12 +31,9 @@ class JeuxVideoIE(InfoExtractor):
             r'http://www\.jeuxvideo\.com/config/\w+/\d+/(.*?)/\d+_player\.xml',
             xml_link, u'video ID')
 
-        xml_config = self._download_webpage(
+        config = self._download_xml(
             xml_link, title, u'Downloading XML config')
-        config = xml.etree.ElementTree.fromstring(xml_config.encode('utf-8'))
-        info_json = self._search_regex(
-            r'(?sm)<format\.json>(.*?)</format\.json>',
-            xml_config, u'JSON information')
+        info_json = config.find('format.json').text
         info = json.loads(info_json)['versions'][0]
         
         video_url = 'http://video720.jeuxvideo.com/' + info['file']

youtube_dl/extractor/jpopsukitv.py

@@ -0,0 +1,73 @@
+# coding=utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    unified_strdate,
+)
+
+
+class JpopsukiIE(InfoExtractor):
+    IE_NAME = 'jpopsuki.tv'
+    _VALID_URL = r'https?://(?:www\.)?jpopsuki\.tv/video/(.*?)/(?P<id>\S+)'
+
+    _TEST = {
+        'url': 'http://www.jpopsuki.tv/video/ayumi-hamasaki---evolution/00be659d23b0b40508169cdee4545771',
+        'md5': '88018c0c1a9b1387940e90ec9e7e198e',
+        'file': '00be659d23b0b40508169cdee4545771.mp4',
+        'info_dict': {
+            'id': '00be659d23b0b40508169cdee4545771',
+            'title': 'ayumi hamasaki - evolution',
+            'description': 'Release date: 2001.01.31\r\n浜崎あゆみ - evolution',
+            'thumbnail': 'http://www.jpopsuki.tv/cache/89722c74d2a2ebe58bcac65321c115b2.jpg',
+            'uploader': 'plama_chan',
+            'uploader_id': '404',
+            'upload_date': '20121101'
+        }
+    }
+
+    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_url = 'http://www.jpopsuki.tv' + self._html_search_regex(
+            r'<source src="(.*?)" type', webpage, 'video url')
+
+        video_title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage)
+        thumbnail = self._og_search_thumbnail(webpage)
+        uploader = self._html_search_regex(
+            r'<li>from: <a href="/user/view/user/(.*?)/uid/',
+            webpage, 'video uploader', fatal=False)
+        uploader_id = self._html_search_regex(
+            r'<li>from: <a href="/user/view/user/\S*?/uid/(\d*)',
+            webpage, 'video uploader_id', fatal=False)
+        upload_date = self._html_search_regex(
+            r'<li>uploaded: (.*?)</li>', webpage, 'video upload_date',
+            fatal=False)
+        if upload_date is not None:
+            upload_date = unified_strdate(upload_date)
+        view_count_str = self._html_search_regex(
+            r'<li>Hits: ([0-9]+?)</li>', webpage, 'video view_count',
+            fatal=False)
+        comment_count_str = self._html_search_regex(
+            r'<h2>([0-9]+?) comments</h2>', webpage, 'video comment_count',
+            fatal=False)
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': video_title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'upload_date': upload_date,
+            'view_count': int_or_none(view_count_str),
+            'comment_count': int_or_none(comment_count_str),
+        }

youtube_dl/extractor/jukebox.py

@@ -8,7 +8,7 @@ from ..utils import (
 )
 
 class JukeboxIE(InfoExtractor):
-    _VALID_URL = r'^http://www\.jukebox?\..+?\/.+[,](?P<video_id>[a-z0-9\-]+).html'
+    _VALID_URL = r'^http://www\.jukebox?\..+?\/.+[,](?P<video_id>[a-z0-9\-]+)\.html'
     _IFRAME = r'<iframe .*src="(?P<iframe>[^"]*)".*>'
     _VIDEO_URL = r'"config":{"file":"(?P<video_url>http:[^"]+[.](?P<video_ext>[^.?]+)[?]mdtk=[0-9]+)"'
     _TITLE = r'<h1 class="inline">(?P<title>[^<]+)</h1>.*<span id="infos_article_artist">(?P<artist>[^<]+)</span>'

youtube_dl/extractor/justintv.py

@@ -1,7 +1,6 @@
 import json
 import os
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -94,10 +93,9 @@ class JustinTVIE(InfoExtractor):
             archive_id = m.group(1)
 
             api = api_base + '/broadcast/by_chapter/%s.xml' % chapter_id
-            chapter_info_xml = self._download_webpage(api, chapter_id,
+            doc = self._download_xml(api, chapter_id,
                                              note=u'Downloading chapter information',
                                              errnote=u'Chapter information download failed')
-            doc = xml.etree.ElementTree.fromstring(chapter_info_xml)
             for a in doc.findall('.//archive'):
                 if archive_id == a.find('./id').text:
                     break

youtube_dl/extractor/khanacademy.py

@@ -0,0 +1,71 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    unified_strdate,
+)
+
+
+class KhanAcademyIE(InfoExtractor):
+    _VALID_URL = r'^https?://(?:www\.)?khanacademy\.org/(?P<key>[^/]+)/(?:[^/]+/){,2}(?P<id>[^?#/]+)(?:$|[?#])'
+    IE_NAME = 'KhanAcademy'
+
+    _TEST = {
+        'url': 'http://www.khanacademy.org/video/one-time-pad',
+        'file': 'one-time-pad.mp4',
+        'md5': '7021db7f2d47d4fff89b13177cb1e8f4',
+        'info_dict': {
+            'title': 'The one-time pad',
+            'description': 'The perfect cipher',
+            'duration': 176,
+            'uploader': 'Brit Cruise',
+            'upload_date': '20120411',
+        }
+    }
+
+    def _real_extract(self, url):
+        m = re.match(self._VALID_URL, url)
+        video_id = m.group('id')
+
+        if m.group('key') == 'video':
+            data = self._download_json(
+                'http://api.khanacademy.org/api/v1/videos/' + video_id,
+                video_id, 'Downloading video info')
+
+            upload_date = unified_strdate(data['date_added'])
+            uploader = ', '.join(data['author_names'])
+            return {
+                '_type': 'url_transparent',
+                'url': data['url'],
+                'id': video_id,
+                'title': data['title'],
+                'thumbnail': data['image_url'],
+                'duration': data['duration'],
+                'description': data['description'],
+                'uploader': uploader,
+                'upload_date': upload_date,
+            }
+        else:
+            # topic
+            data = self._download_json(
+                'http://api.khanacademy.org/api/v1/topic/' + video_id,
+                video_id, 'Downloading topic info')
+
+            entries = [
+                {
+                    '_type': 'url',
+                    'url': c['url'],
+                    'id': c['id'],
+                    'title': c['title'],
+                }
+                for c in data['children'] if c['kind'] in ('Video', 'Topic')]
+
+            return {
+                '_type': 'playlist',
+                'id': video_id,
+                'title': data['title'],
+                'description': data['description'],
+                'entries': entries,
+            }

youtube_dl/extractor/liveleak.py

@@ -8,7 +8,7 @@ from ..utils import (
 
 class LiveLeakIE(InfoExtractor):
 
-    _VALID_URL = r'^(?:http?://)?(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<video_id>[\w_]+)(?:.*)'
+    _VALID_URL = r'^(?:http://)?(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<video_id>[\w_]+)(?:.*)'
     IE_NAME = u'liveleak'
     _TEST = {
         u'url': u'http://www.liveleak.com/view?i=757_1364311680',

youtube_dl/extractor/livestream.py

@@ -1,6 +1,5 @@
 import re
 import json
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -12,7 +11,7 @@ from ..utils import (
 
 class LivestreamIE(InfoExtractor):
     IE_NAME = u'livestream'
-    _VALID_URL = r'http://new.livestream.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>\d+))?/?$'
+    _VALID_URL = r'http://new\.livestream\.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>\d+))?/?$'
     _TEST = {
         u'url': u'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370',
         u'file': u'4719370.mp4',
@@ -80,8 +79,7 @@ class LivestreamOriginalIE(InfoExtractor):
         user = mobj.group('user')
         api_url = 'http://x{0}x.api.channel.livestream.com/2.0/clipdetails?extendedInfo=true&id={1}'.format(user, video_id)
 
-        api_response = self._download_webpage(api_url, video_id)
-        info = xml.etree.ElementTree.fromstring(api_response.encode('utf-8'))
+        info = self._download_xml(api_url, video_id)
         item = info.find('channel').find('item')
         ns = {'media': 'http://search.yahoo.com/mrss'}
         thumbnail_url = item.find(xpath_with_ns('media:thumbnail', ns)).attrib['url']

youtube_dl/extractor/lynda.py

@@ -0,0 +1,201 @@
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .subtitles import SubtitlesInfoExtractor
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    compat_urllib_request,
+    ExtractorError
+)
+
+
+class LyndaIE(SubtitlesInfoExtractor):
+    IE_NAME = 'lynda'
+    IE_DESC = 'lynda.com videos'
+    _VALID_URL = r'https?://www\.lynda\.com/[^/]+/[^/]+/\d+/(\d+)-\d\.html'
+    _LOGIN_URL = 'https://www.lynda.com/login/login.aspx'
+    _NETRC_MACHINE = 'lynda'
+
+    _SUCCESSFUL_LOGIN_REGEX = r'<a href="https://www.lynda.com/home/userAccount/ChangeContactInfo.aspx" data-qa="eyebrow_account_menu">My account'
+    _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
+
+    ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
+
+    _TEST = {
+        'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
+        'file': '114408.mp4',
+        'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
+        'info_dict': {
+            'title': 'Using the exercise files',
+            'duration': 68
+        }
+    }
+
+    def _real_initialize(self):
+        self._login()
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group(1)
+
+        page = self._download_webpage('http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id,
+                                      video_id, 'Downloading video JSON')
+        video_json = json.loads(page)
+
+        if 'Status' in video_json:
+            raise ExtractorError('lynda returned error: %s' % video_json['Message'], expected=True)
+
+        if video_json['HasAccess'] is False:
+            raise ExtractorError('Video %s is only available for members. ' % video_id + self.ACCOUNT_CREDENTIALS_HINT, expected=True)
+
+        video_id = video_json['ID']
+        duration = video_json['DurationInSeconds']
+        title = video_json['Title']
+
+        formats = [{'url': fmt['Url'],
+                    'ext': fmt['Extension'],
+                    'width': fmt['Width'],
+                    'height': fmt['Height'],
+                    'filesize': fmt['FileSize'],
+                    'format_id': str(fmt['Resolution'])
+                    } for fmt in video_json['Formats']]
+
+        self._sort_formats(formats)
+
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(video_id, page)
+            return
+
+        subtitles = self._fix_subtitles(self.extract_subtitles(video_id, page))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'duration': duration,
+            'subtitles': subtitles,
+            'formats': formats
+        }
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None:
+            return
+
+        login_form = {
+            'username': username,
+            'password': password,
+            'remember': 'false',
+            'stayPut': 'false'
+        }        
+        request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
+        login_page = self._download_webpage(request, None, note='Logging in as %s' % username)
+
+        # Not (yet) logged in
+        m = re.search(r'loginResultJson = \'(?P<json>[^\']+)\';', login_page)
+        if m is not None:
+            response = m.group('json')
+            response_json = json.loads(response)            
+            state = response_json['state']
+
+            if state == 'notlogged':
+                raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
+
+            # This is when we get popup:
+            # > You're already logged in to lynda.com on two devices.
+            # > If you log in here, we'll log you out of another device.
+            # So, we need to confirm this.
+            if state == 'conflicted':
+                confirm_form = {
+                    'username': '',
+                    'password': '',
+                    'resolve': 'true',
+                    'remember': 'false',
+                    'stayPut': 'false',
+                }
+                request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form))
+                login_page = self._download_webpage(request, None, note='Confirming log in and log out from another device')
+
+        if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None:
+            raise ExtractorError('Unable to log in')
+
+    def _fix_subtitles(self, subtitles):
+        if subtitles is None:
+            return subtitles  # subtitles not requested
+
+        fixed_subtitles = {}
+        for k, v in subtitles.items():
+            subs = json.loads(v)
+            if len(subs) == 0:
+                continue
+            srt = ''
+            for pos in range(0, len(subs) - 1):
+                seq_current = subs[pos]
+                m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
+                if m_current is None:
+                    continue
+                seq_next = subs[pos + 1]
+                m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
+                if m_next is None:
+                    continue
+                appear_time = m_current.group('timecode')
+                disappear_time = m_next.group('timecode')
+                text = seq_current['Caption']
+                srt += '%s\r\n%s --> %s\r\n%s' % (str(pos), appear_time, disappear_time, text)
+            if srt:
+                fixed_subtitles[k] = srt
+        return fixed_subtitles
+
+    def _get_available_subtitles(self, video_id, webpage):
+        url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
+        sub = self._download_webpage(url, None, note=False)
+        sub_json = json.loads(sub)
+        return {'en': url} if len(sub_json) > 0 else {}
+
+
+class LyndaCourseIE(InfoExtractor):
+    IE_NAME = 'lynda:course'
+    IE_DESC = 'lynda.com online courses'
+
+    # Course link equals to welcome/introduction video link of same course
+    # We will recognize it as course link
+    _VALID_URL = r'https?://(?:www|m)\.lynda\.com/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        course_path = mobj.group('coursepath')
+        course_id = mobj.group('courseid')
+        
+        page = self._download_webpage('http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
+                                      course_id, 'Downloading course JSON')
+        course_json = json.loads(page)
+
+        if 'Status' in course_json and course_json['Status'] == 'NotFound':
+            raise ExtractorError('Course %s does not exist' % course_id, expected=True)
+
+        unaccessible_videos = 0
+        videos = []
+        (username, _) = self._get_login_info()
+
+        for chapter in course_json['Chapters']:
+            for video in chapter['Videos']:
+                if username is None and video['HasAccess'] is False:
+                    unaccessible_videos += 1
+                    continue
+                videos.append(video['ID'])
+
+        if unaccessible_videos > 0:
+            self._downloader.report_warning('%s videos are only available for members and will not be downloaded. '
+                                            % unaccessible_videos + LyndaIE.ACCOUNT_CREDENTIALS_HINT)
+
+        entries = [
+            self.url_result('http://www.lynda.com/%s/%s-4.html' %
+                            (course_path, video_id),
+                            'Lynda')
+            for video_id in videos]
+
+        course_title = course_json['Title']
+
+        return self.playlist_result(entries, course_id, course_title)

youtube_dl/extractor/macgamestore.py

@@ -0,0 +1,43 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import ExtractorError
+
+
+class MacGameStoreIE(InfoExtractor):
+    IE_NAME = 'macgamestore'
+    IE_DESC = 'MacGameStore trailers'
+    _VALID_URL = r'https?://www\.macgamestore\.com/mediaviewer\.php\?trailer=(?P<id>\d+)'
+
+    _TEST = {
+        'url': 'http://www.macgamestore.com/mediaviewer.php?trailer=2450',
+        'file': '2450.m4v',
+        'md5': '8649b8ea684b6666b4c5be736ecddc61',
+        'info_dict': {
+            'title': 'Crow',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id, 'Downloading trailer page')
+
+        if re.search(r'>Missing Media<', webpage) is not None:
+            raise ExtractorError('Trailer %s does not exist' % video_id, expected=True)
+
+        video_title = self._html_search_regex(
+            r'<title>MacGameStore: (.*?) Trailer</title>', webpage, 'title')
+
+        video_url = self._html_search_regex(
+            r'(?s)<div\s+id="video-player".*?href="([^"]+)"\s*>',
+            webpage, 'video URL')
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': video_title
+        }

youtube_dl/extractor/mdr.py

@@ -0,0 +1,64 @@
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+)
+
+
+class MDRIE(InfoExtractor):
+    _VALID_URL = r'^(?P<domain>(?:https?://)?(?:www\.)?mdr\.de)/mediathek/(?:.*)/(?P<type>video|audio)(?P<video_id>[^/_]+)_.*'
+    
+    # No tests, MDR regularily deletes its videos
+
+    def _real_extract(self, url):
+        m = re.match(self._VALID_URL, url)
+        video_id = m.group('video_id')
+        domain = m.group('domain')
+
+        # determine title and media streams from webpage
+        html = self._download_webpage(url, video_id)
+
+        title = self._html_search_regex(r'<h2>(.*?)</h2>', html, u'title')
+        xmlurl = self._search_regex(
+            r'(/mediathek/(?:.+)/(?:video|audio)[0-9]+-avCustom.xml)', html, u'XML URL')
+
+        doc = self._download_xml(domain + xmlurl, video_id)
+        formats = []
+        for a in doc.findall('./assets/asset'):
+            url_el = a.find('.//progressiveDownloadUrl')
+            if url_el is None:
+                continue
+            abr = int(a.find('bitrateAudio').text) // 1000
+            media_type = a.find('mediaType').text
+            format = {
+                'abr': abr,
+                'filesize': int(a.find('fileSize').text),
+                'url': url_el.text,
+            }
+
+            vbr_el = a.find('bitrateVideo')
+            if vbr_el is None:
+                format.update({
+                    'vcodec': 'none',
+                    'format_id': u'%s-%d' % (media_type, abr),
+                })
+            else:
+                vbr = int(vbr_el.text) // 1000
+                format.update({
+                    'vbr': vbr,
+                    'width': int(a.find('frameWidth').text),
+                    'height': int(a.find('frameHeight').text),
+                    'format_id': u'%s-%d' % (media_type, vbr),
+                })
+            formats.append(format)
+        if not formats:
+            raise ExtractorError(u'Could not find any valid formats')
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+        }

youtube_dl/extractor/metacafe.py

@@ -1,14 +1,10 @@
 import re
-import socket
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_http_client,
     compat_parse_qs,
-    compat_urllib_error,
     compat_urllib_parse,
     compat_urllib_request,
-    compat_str,
     determine_ext,
     ExtractorError,
 )
@@ -69,6 +65,21 @@ class MetacafeIE(InfoExtractor):
             u'age_limit': 18,
         },
     },
+    # cbs video
+    {
+        u'url': u'http://www.metacafe.com/watch/cb-0rOxMBabDXN6/samsung_galaxy_note_2_samsungs_next_generation_phablet/',
+        u'info_dict': {
+            u'id': u'0rOxMBabDXN6',
+            u'ext': u'flv',
+            u'title': u'Samsung Galaxy Note 2: Samsung\'s next-generation phablet',
+            u'description': u'md5:54d49fac53d26d5a0aaeccd061ada09d',
+            u'duration': 129,
+        },
+        u'params': {
+            # rtmp download
+            u'skip_download': True,
+        },
+    },
     ]
 
 
@@ -78,12 +89,8 @@ class MetacafeIE(InfoExtractor):
 
     def _real_initialize(self):
         # Retrieve disclaimer
-        request = compat_urllib_request.Request(self._DISCLAIMER)
-        try:
-            self.report_disclaimer()
-            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 disclaimer: %s' % compat_str(err))
+        self.report_disclaimer()
+        self._download_webpage(self._DISCLAIMER, None, False, u'Unable to retrieve disclaimer')
 
         # Confirm age
         disclaimer_form = {
@@ -92,11 +99,8 @@ class MetacafeIE(InfoExtractor):
             }
         request = compat_urllib_request.Request(self._FILTER_POST, compat_urllib_parse.urlencode(disclaimer_form))
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-        try:
-            self.report_age_confirmation()
-            compat_urllib_request.urlopen(request).read()
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
+        self.report_age_confirmation()
+        self._download_webpage(request, None, False, u'Unable to confirm age')
 
     def _real_extract(self, url):
         # Extract id and simplified title from URL
@@ -106,10 +110,16 @@ class MetacafeIE(InfoExtractor):
 
         video_id = mobj.group(1)
 
-        # Check if video comes from YouTube
-        mobj2 = re.match(r'^yt-(.*)$', video_id)
-        if mobj2 is not None:
-            return [self.url_result('http://www.youtube.com/watch?v=%s' % mobj2.group(1), 'Youtube')]
+        # the video may come from an external site
+        m_external = re.match('^(\w{2})-(.*)$', video_id)
+        if m_external is not None:
+            prefix, ext_id = m_external.groups()
+            # Check if video comes from YouTube
+            if prefix == 'yt':
+                return self.url_result('http://www.youtube.com/watch?v=%s' % ext_id, 'Youtube')
+            # CBS videos use theplatform.com
+            if prefix == 'cb':
+                return self.url_result('theplatform:%s' % ext_id, 'ThePlatform')
 
         # Retrieve video webpage to extract further information
         req = compat_urllib_request.Request('http://www.metacafe.com/watch/%s/' % video_id)

youtube_dl/extractor/metacritic.py

@@ -1,20 +1,23 @@
+from __future__ import unicode_literals
+
 import re
-import xml.etree.ElementTree
-import operator
 
 from .common import InfoExtractor
+from ..utils import (
+    fix_xml_all_ampersand,
+)
 
 
 class MetacriticIE(InfoExtractor):
     _VALID_URL = r'https?://www\.metacritic\.com/.+?/trailers/(?P<id>\d+)'
 
     _TEST = {
-        u'url': u'http://www.metacritic.com/game/playstation-4/infamous-second-son/trailers/3698222',
-        u'file': u'3698222.mp4',
-        u'info_dict': {
-            u'title': u'inFamous: Second Son - inSide Sucker Punch: Smoke & Mirrors',
-            u'description': u'Take a peak behind-the-scenes to see how Sucker Punch brings smoke into the universe of inFAMOUS Second Son on the PS4.',
-            u'duration': 221,
+        'url': 'http://www.metacritic.com/game/playstation-4/infamous-second-son/trailers/3698222',
+        'file': '3698222.mp4',
+        'info_dict': {
+            'title': 'inFamous: Second Son - inSide Sucker Punch: Smoke & Mirrors',
+            'description': 'Take a peak behind-the-scenes to see how Sucker Punch brings smoke into the universe of inFAMOUS Second Son on the PS4.',
+            'duration': 221,
         },
     }
 
@@ -23,9 +26,8 @@ class MetacriticIE(InfoExtractor):
         video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
         # The xml is not well formatted, there are raw '&'
-        info_xml = self._download_webpage('http://www.metacritic.com/video_data?video=' + video_id,
-            video_id, u'Downloading info xml').replace('&', '&amp;')
-        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
+        info = self._download_xml('http://www.metacritic.com/video_data?video=' + video_id,
+            video_id, 'Downloading info xml', transform_source=fix_xml_all_ampersand)
 
         clip = next(c for c in info.findall('playList/clip') if c.find('id').text == video_id)
         formats = []
@@ -36,20 +38,17 @@ class MetacriticIE(InfoExtractor):
                 'url': video_url,
                 'ext': 'mp4',
                 'format_id': rate_str,
-                'rate': int(rate_str),
+                'tbr': int(rate_str),
             })
-        formats.sort(key=operator.itemgetter('rate'))
+        self._sort_formats(formats)
 
         description = self._html_search_regex(r'<b>Description:</b>(.*?)</p>',
-            webpage, u'description', flags=re.DOTALL)
+            webpage, 'description', flags=re.DOTALL)
 
-        info = {
+        return {
             'id': video_id,
             'title': clip.find('title').text,
             'formats': formats,
             'description': description,
             'duration': int(clip.find('duration').text),
         }
-        # TODO: Remove when #980 has been merged
-        info.update(formats[-1])
-        return info

youtube_dl/extractor/mit.py

@@ -33,8 +33,18 @@ class TechTVMITIE(InfoExtractor):
             raw_page, u'base url')
         formats_json = self._search_regex(r'bitrates: (\[.+?\])', raw_page,
             u'video formats')
-        formats = json.loads(formats_json)
-        formats = sorted(formats, key=lambda f: f['bitrate'])
+        formats_mit = json.loads(formats_json)
+        formats = [
+            {
+                'format_id': f['label'],
+                'url': base_url + f['url'].partition(':')[2],
+                'ext': f['url'].partition(':')[0],
+                'format': f['label'],
+                'width': f['width'],
+                'vbr': f['bitrate'],
+            }
+            for f in formats_mit
+        ]
 
         title = get_element_by_id('edit-title', clean_page)
         description = clean_html(get_element_by_id('edit-description', clean_page))
@@ -43,8 +53,7 @@ class TechTVMITIE(InfoExtractor):
 
         return {'id': video_id,
                 'title': title,
-                'url': base_url + formats[-1]['url'].replace('mp4:', ''),
-                'ext': 'mp4',
+                'formats': formats,
                 'description': description,
                 'thumbnail': thumbnail,
                 }

youtube_dl/extractor/mixcloud.py

@@ -1,13 +1,10 @@
 import json
 import re
-import socket
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_http_client,
-    compat_urllib_error,
-    compat_urllib_request,
     unified_strdate,
+    ExtractorError,
 )
 
 
@@ -31,13 +28,18 @@ class MixcloudIE(InfoExtractor):
         """Returns 1st active url from list"""
         for url in url_list:
             try:
-                compat_urllib_request.urlopen(url)
+                # We only want to know if the request succeed
+                # don't download the whole file
+                self._request_webpage(url, None, False)
                 return url
-            except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error):
+            except ExtractorError:
                 url = None
 
         return None
 
+    def _get_url(self, template_url):
+        return self.check_urls(template_url % i for i in range(30))
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
 
@@ -51,15 +53,20 @@ class MixcloudIE(InfoExtractor):
         info = json.loads(json_data)
 
         preview_url = self._search_regex(r'data-preview-url="(.+?)"', webpage, u'preview url')
-        song_url = preview_url.replace('/previews/', '/cloudcasts/originals/')
+        song_url = preview_url.replace('/previews/', '/c/originals/')
         template_url = re.sub(r'(stream\d*)', 'stream%d', song_url)
-        final_song_url = self.check_urls(template_url % i for i in range(30))
+        final_song_url = self._get_url(template_url)
+        if final_song_url is None:
+            self.to_screen('Trying with m4a extension')
+            template_url = template_url.replace('.mp3', '.m4a').replace('originals/', 'm4a/64/')
+            final_song_url = self._get_url(template_url)
+        if final_song_url is None:
+            raise ExtractorError(u'Unable to extract track url')
 
         return {
             'id': track_id,
             'title': info['name'],
             'url': final_song_url,
-            'ext': 'mp3',
             'description': info.get('description'),
             'thumbnail': info['pictures'].get('extra_large'),
             'uploader': info['user']['name'],

youtube_dl/extractor/mpora.py

@@ -0,0 +1,66 @@
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+)
+
+
+class MporaIE(InfoExtractor):
+    _VALID_URL = r'^https?://(www\.)?mpora\.(?:com|de)/videos/(?P<id>[^?#/]+)'
+    IE_NAME = 'MPORA'
+
+    _TEST = {
+        'url': 'http://mpora.de/videos/AAdo8okx4wiz/embed?locale=de',
+        'file': 'AAdo8okx4wiz.mp4',
+        'md5': 'a7a228473eedd3be741397cf452932eb',
+        'info_dict': {
+            'title': 'Katy Curd -  Winter in the Forest',
+            'duration': 416,
+            'uploader': 'petenewman',
+        },
+    }
+
+    def _real_extract(self, url):
+        m = re.match(self._VALID_URL, url)
+        video_id = m.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        data_json = self._search_regex(
+            r"new FM\.Player\('[^']+',\s*(\{.*?)\);\n", webpage, 'json')
+
+        data = json.loads(data_json)
+
+        uploader = data['info_overlay']['name']
+        duration = data['video']['duration'] // 1000
+        thumbnail = data['video']['encodings']['sd']['poster']
+        title = data['info_overlay']['title']
+
+        formats = []
+        for encoding_id, edata in data['video']['encodings'].items():
+            for src in edata['sources']:
+                width_str = self._search_regex(
+                    r'_([0-9]+)\.[a-zA-Z0-9]+$', src['src'],
+                    False, default=None)
+                vcodec = src['type'].partition('/')[2]
+                
+                formats.append({
+                    'format_id': encoding_id + '-' + vcodec,
+                    'url': src['src'],
+                    'vcodec': vcodec,
+                    'width': int_or_none(width_str),
+                })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'uploader': uploader,
+            'duration': duration,
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/mtv.py

@@ -10,35 +10,8 @@ from ..utils import (
 def _media_xml_tag(tag):
     return '{http://search.yahoo.com/mrss/}%s' % tag
 
-class MTVIE(InfoExtractor):
-    _VALID_URL = r'^https?://(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$'
-
-    _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/'
-
-    _TESTS = [
-        {
-            u'url': u'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
-            u'file': u'853555.mp4',
-            u'md5': u'850f3f143316b1e71fa56a4edfd6e0f8',
-            u'info_dict': {
-                u'title': u'Taylor Swift - "Ours (VH1 Storytellers)"',
-                u'description': u'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
-            },
-        },
-        {
-            u'add_ie': ['Vevo'],
-            u'url': u'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
-            u'file': u'USCJY1331283.mp4',
-            u'md5': u'73b4e7fcadd88929292fe52c3ced8caf',
-            u'info_dict': {
-                u'title': u'Everything Has Changed',
-                u'upload_date': u'20130606',
-                u'uploader': u'Taylor Swift',
-            },
-            u'skip': u'VEVO is only available in some countries',
-        },
-    ]
 
+class MTVServicesInfoExtractor(InfoExtractor):
     @staticmethod
     def _id_from_uri(uri):
         return uri.split(':')[-1]
@@ -53,7 +26,12 @@ class MTVIE(InfoExtractor):
         return base + m.group('finalid')
 
     def _get_thumbnail_url(self, uri, itemdoc):
-        return 'http://mtv.mtvnimages.com/uri/' + uri
+        search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
+        thumb_node = itemdoc.find(search_path)
+        if thumb_node is None:
+            return None
+        else:
+            return thumb_node.attrib['url']
 
     def _extract_video_formats(self, metadataXml):
         if '/error_country_block.swf' in metadataXml:
@@ -93,7 +71,7 @@ class MTVIE(InfoExtractor):
         else:
             description = None
 
-        info = {
+        return {
             'title': itemdoc.find('title').text,
             'formats': self._extract_video_formats(mediagen_page),
             'id': video_id,
@@ -101,32 +79,67 @@ class MTVIE(InfoExtractor):
             'description': description,
         }
 
-        # TODO: Remove when #980 has been merged
-        info.update(info['formats'][-1])
-
-        return info
-
     def _get_videos_info(self, uri):
         video_id = self._id_from_uri(uri)
         data = compat_urllib_parse.urlencode({'uri': uri})
-        infoXml = self._download_webpage(self._FEED_URL +'?' + data, video_id,
-                                         u'Downloading info')
-        idoc = xml.etree.ElementTree.fromstring(infoXml.encode('utf-8'))
+
+        def fix_ampersand(s):
+            """ Fix unencoded ampersand in XML """
+            return s.replace(u'& ', '&amp; ')
+        idoc = self._download_xml(
+            self._FEED_URL + '?' + data, video_id,
+            u'Downloading info', transform_source=fix_ampersand)
         return [self._get_video_info(item) for item in idoc.findall('.//item')]
 
+
+class MTVIE(MTVServicesInfoExtractor):
+    _VALID_URL = r'''(?x)^https?://
+        (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$|
+           m\.mtv\.com/videos/video\.rbml\?.*?id=(?P<mgid>[^&]+))'''
+
+    _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/'
+
+    _TESTS = [
+        {
+            u'url': u'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
+            u'file': u'853555.mp4',
+            u'md5': u'850f3f143316b1e71fa56a4edfd6e0f8',
+            u'info_dict': {
+                u'title': u'Taylor Swift - "Ours (VH1 Storytellers)"',
+                u'description': u'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
+            },
+        },
+        {
+            u'add_ie': ['Vevo'],
+            u'url': u'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
+            u'file': u'USCJY1331283.mp4',
+            u'md5': u'73b4e7fcadd88929292fe52c3ced8caf',
+            u'info_dict': {
+                u'title': u'Everything Has Changed',
+                u'upload_date': u'20130606',
+                u'uploader': u'Taylor Swift',
+            },
+            u'skip': u'VEVO is only available in some countries',
+        },
+    ]
+
+    def _get_thumbnail_url(self, uri, itemdoc):
+        return 'http://mtv.mtvnimages.com/uri/' + uri
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('videoid')
-
-        webpage = self._download_webpage(url, video_id)
-
-        # Some videos come from Vevo.com
-        m_vevo = re.search(r'isVevoVideo = true;.*?vevoVideoId = "(.*?)";',
-                           webpage, re.DOTALL)
-        if m_vevo:
-            vevo_id = m_vevo.group(1);
-            self.to_screen(u'Vevo video detected: %s' % vevo_id)
-            return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
-
-        uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, u'uri')
+        uri = mobj.groupdict().get('mgid')
+        if uri is None:
+            webpage = self._download_webpage(url, video_id)
+    
+            # Some videos come from Vevo.com
+            m_vevo = re.search(r'isVevoVideo = true;.*?vevoVideoId = "(.*?)";',
+                               webpage, re.DOTALL)
+            if m_vevo:
+                vevo_id = m_vevo.group(1);
+                self.to_screen(u'Vevo video detected: %s' % vevo_id)
+                return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
+    
+            uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, u'uri')
         return self._get_videos_info(uri)

youtube_dl/extractor/muzu.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class MuzuTVIE(InfoExtractor):
-    _VALID_URL = r'https?://www.muzu.tv/(.+?)/(.+?)/(?P<id>\d+)'
+    _VALID_URL = r'https?://www\.muzu\.tv/(.+?)/(.+?)/(?P<id>\d+)'
     IE_NAME = u'muzu.tv'
 
     _TEST = {

youtube_dl/extractor/myspass.py

@@ -1,5 +1,4 @@
 import os.path
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -10,7 +9,7 @@ from ..utils import (
 
 
 class MySpassIE(InfoExtractor):
-    _VALID_URL = r'http://www.myspass.de/.*'
+    _VALID_URL = r'http://www\.myspass\.de/.*'
     _TEST = {
         u'url': u'http://www.myspass.de/myspass/shows/tvshows/absolute-mehrheit/Absolute-Mehrheit-vom-17022013-Die-Highlights-Teil-2--/11741/',
         u'file': u'11741.mp4',
@@ -33,8 +32,7 @@ class MySpassIE(InfoExtractor):
 
         # get metadata
         metadata_url = META_DATA_URL_TEMPLATE % video_id
-        metadata_text = self._download_webpage(metadata_url, video_id)
-        metadata = xml.etree.ElementTree.fromstring(metadata_text.encode('utf-8'))
+        metadata = self._download_xml(metadata_url, video_id)
 
         # extract values from metadata
         url_flv_el = metadata.find('url_flv')

youtube_dl/extractor/myvideo.py

@@ -143,8 +143,10 @@ class MyVideoIE(InfoExtractor):
         if mobj:
             video_url = compat_urllib_parse.unquote(mobj.group(1))
             if 'myvideo2flash' in video_url:
-                self._downloader.report_warning(u'forcing RTMPT ...')
-                video_url = video_url.replace('rtmpe://', 'rtmpt://')
+                self.report_warning(
+                    u'Rewriting URL to use unencrypted rtmp:// ...',
+                    video_id)
+                video_url = video_url.replace('rtmpe://', 'rtmp://')
 
         if not video_url:
             # extract non rtmp videos

youtube_dl/extractor/naver.py

@@ -1,6 +1,5 @@
 # encoding: utf-8
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -10,7 +9,7 @@ from ..utils import (
 
 
 class NaverIE(InfoExtractor):
-    _VALID_URL = r'https?://tvcast\.naver\.com/v/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:m\.)?tvcast\.naver\.com/v/(?P<id>\d+)'
 
     _TEST = {
         u'url': u'http://tvcast.naver.com/v/81652',
@@ -38,14 +37,12 @@ class NaverIE(InfoExtractor):
             'protocol': 'p2p',
             'inKey': key,
         })
-        info_xml = self._download_webpage(
+        info = self._download_xml(
             'http://serviceapi.rmcnmv.naver.com/flash/videoInfo.nhn?' + query,
             video_id, u'Downloading video info')
-        urls_xml = self._download_webpage(
+        urls = self._download_xml(
             'http://serviceapi.rmcnmv.naver.com/flash/playableEncodingOption.nhn?' + query_urls,
             video_id, u'Downloading video formats info')
-        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
-        urls = xml.etree.ElementTree.fromstring(urls_xml.encode('utf-8'))
 
         formats = []
         for format_el in urls.findall('EncodingOptions/EncodingOption'):
@@ -59,7 +56,7 @@ class NaverIE(InfoExtractor):
                 'height': int(format_el.find('height').text),
             })
 
-        info = {
+        return {
             'id': video_id,
             'title': info.find('Subject').text,
             'formats': formats,
@@ -68,6 +65,3 @@ class NaverIE(InfoExtractor):
             'upload_date': info.find('WriteDate').text.replace('.', ''),
             'view_count': int(info.find('PlayCount').text),
         }
-        # TODO: Remove when #980 has been merged
-        info.update(formats[-1])
-        return info