release 2016.08.07

- extract f4m and dash formats
- improve format sorting and listing
- improve extraction of articles with `otherSettings.playlist`

.github/ISSUE_TEMPLATE.md

@@ -6,8 +6,8 @@
 
 ---
 
-### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.04.24*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
-- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.04.24**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.08.07*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
+- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.08.07**
 
 ### Before submitting an *issue* make sure you have:
 - [ ] At least skimmed through [README](https://github.com/rg3/youtube-dl/blob/master/README.md) and **most notably** [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections
@@ -35,7 +35,7 @@ $ youtube-dl -v <your command line>
 [debug] User config: []
 [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
 [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
-[debug] youtube-dl version 2016.04.24
+[debug] youtube-dl version 2016.08.07
 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2
 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4
 [debug] Proxy map: {}

.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,22 @@
+## Please follow the guide below
+
+- You will be asked some questions, please read them **carefully** and answer honestly
+- Put an `x` into all the boxes [ ] relevant to your *pull request* (like that [x])
+- Use *Preview* tab to see how your *pull request* will actually look like
+
+---
+
+### Before submitting a *pull request* make sure you have:
+- [ ] At least skimmed through [adding new extractor tutorial](https://github.com/rg3/youtube-dl#adding-support-for-a-new-site) and [youtube-dl coding conventions](https://github.com/rg3/youtube-dl#youtube-dl-coding-conventions) sections
+- [ ] [Searched](https://github.com/rg3/youtube-dl/search?q=is%3Apr&type=Issues) the bugtracker for similar pull requests
+
+### What is the purpose of your *pull request*?
+- [ ] Bug fix
+- [ ] New extractor
+- [ ] New feature
+
+---
+
+### Description of your *pull request* and other information
+
+Explanation of your *pull request* in arbitrary form goes here. Please make sure the description explains the purpose and effect of your *pull request* and is worded well enough to be understood. Provide as much context and examples as possible.

.gitignore

@@ -28,10 +28,16 @@ updates_key.pem
 *.mp4
 *.m4a
 *.m4v
+*.mp3
 *.part
 *.swp
 test/testdata
+test/local_parameters.json
 .tox
 youtube-dl.zsh
+
+# IntelliJ related files
 .idea
-.idea/*
+*.iml
+
+tmp/

.travis.yml

@@ -11,7 +11,6 @@ script: nosetests test --verbose
 notifications:
   email:
     - filippo.valsorda@gmail.com
-    - phihag@phihag.de
     - yasoob.khld@gmail.com
 #  irc:
 #    channels:

AUTHORS

@@ -168,3 +168,14 @@ José Joaquín Atria
 Viťas Strádal
 Kagami Hiiragi
 Philip Huppert
+blahgeek
+Kevin Deldycke
+inondle
+Tomáš Čech
+Déstin Reed
+Roman Tsiupa
+Artur Krysiak
+Jakub Adam Wieczorek
+Aleksandar Topuzović
+Nehal Patel
+Rob van Bekkum

CONTRIBUTING.md

@@ -97,9 +97,17 @@ If you want to add support for a new site, first of all **make sure** this site
 After you have ensured this site is distributing it's content legally, you can follow this quick list (assuming your service is called `yourextractor`):
 
 1. [Fork this repository](https://github.com/rg3/youtube-dl/fork)
-2. Check out the source code with `git clone git@github.com:YOUR_GITHUB_USERNAME/youtube-dl.git`
-3. Start a new git branch with `cd youtube-dl; git checkout -b yourextractor`
+2. Check out the source code with:
+
+        git clone git@github.com:YOUR_GITHUB_USERNAME/youtube-dl.git
+
+3. Start a new git branch with
+
+        cd youtube-dl
+        git checkout -b yourextractor
+
 4. Start with this simple template and save it to `youtube_dl/extractor/yourextractor.py`:
+
     ```python
     # coding: utf-8
     from __future__ import unicode_literals
@@ -142,17 +150,149 @@ After you have ensured this site is distributing it's content legally, you can f
     ```
 5. Add an import in [`youtube_dl/extractor/extractors.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/extractors.py).
 6. Run `python test/test_download.py TestDownload.test_YourExtractor`. This *should fail* at first, but you can continually re-run it until you're done. If you decide to add more than one test, then rename ``_TEST`` to ``_TESTS`` and make it into a list of dictionaries. The tests will then be named `TestDownload.test_YourExtractor`, `TestDownload.test_YourExtractor_1`, `TestDownload.test_YourExtractor_2`, etc.
-7. Have a look at [`youtube_dl/extractor/common.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py) for possible helper methods and a [detailed description of what your extractor should and may return](https://github.com/rg3/youtube-dl/blob/58525c94d547be1c8167d16c298bdd75506db328/youtube_dl/extractor/common.py#L68-L226). Add tests and code for as many as you want.
-8. Keep in mind that the only mandatory fields in info dict for successful extraction process are `id`, `title` and either `url` or `formats`, i.e. these are the critical data the extraction does not make any sense without. This means that [any field](https://github.com/rg3/youtube-dl/blob/58525c94d547be1c8167d16c298bdd75506db328/youtube_dl/extractor/common.py#L138-L226) apart from aforementioned mandatory ones should be treated **as optional** and extraction should be **tolerate** to situations when sources for these fields can potentially be unavailable (even if they always available at the moment) and **future-proof** in order not to break the extraction of general purpose mandatory fields. For example, if you have some intermediate dict `meta` that is a source of metadata and it has a key `summary` that you want to extract and put into resulting info dict as `description`, you should be ready that this key may be missing from the `meta` dict, i.e. you should extract it as `meta.get('summary')` and not `meta['summary']`. Similarly, you should pass `fatal=False` when extracting data from a webpage with `_search_regex/_html_search_regex`.
-9. Check the code with [flake8](https://pypi.python.org/pypi/flake8).
-10. When the tests pass, [add](http://git-scm.com/docs/git-add) the new files and [commit](http://git-scm.com/docs/git-commit) them and [push](http://git-scm.com/docs/git-push) the result, like this:
+7. Have a look at [`youtube_dl/extractor/common.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py) for possible helper methods and a [detailed description of what your extractor should and may return](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L74-L252). Add tests and code for as many as you want.
+8. Make sure your code follows [youtube-dl coding conventions](#youtube-dl-coding-conventions) and check the code with [flake8](https://pypi.python.org/pypi/flake8). Also make sure your code works under all [Python](http://www.python.org/) versions claimed supported by youtube-dl, namely 2.6, 2.7, and 3.2+.
+9. When the tests pass, [add](http://git-scm.com/docs/git-add) the new files and [commit](http://git-scm.com/docs/git-commit) them and [push](http://git-scm.com/docs/git-push) the result, like this:
 
         $ git add youtube_dl/extractor/extractors.py
         $ git add youtube_dl/extractor/yourextractor.py
         $ git commit -m '[yourextractor] Add new extractor'
         $ git push origin yourextractor
 
-11. Finally, [create a pull request](https://help.github.com/articles/creating-a-pull-request). We'll then review and merge it.
+10. Finally, [create a pull request](https://help.github.com/articles/creating-a-pull-request). We'll then review and merge it.
 
 In any case, thank you very much for your contributions!
 
+## youtube-dl coding conventions
+
+This section introduces a guide lines for writing idiomatic, robust and future-proof extractor code.
+
+Extractors are very fragile by nature since they depend on the layout of the source data provided by 3rd party media hoster out of your control and this layout tend to change. As an extractor implementer your task is not only to write code that will extract media links and metadata correctly but also to minimize code dependency on source's layout changes and even to make the code foresee potential future changes and be ready for that. This is important because it will allow extractor not to break on minor layout changes thus keeping old youtube-dl versions working. Even though this breakage issue is easily fixed by emitting a new version of youtube-dl with fix incorporated all the previous version become broken in all repositories and distros' packages that may not be so prompt in fetching the update from us. Needless to say some may never receive an update at all that is possible for non rolling release distros.
+
+### Mandatory and optional metafields
+
+For extraction to work youtube-dl relies on metadata your extractor extracts and provides to youtube-dl expressed by [information dictionary](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L75-L257) or simply *info dict*. Only the following meta fields in *info dict* are considered mandatory for successful extraction process by youtube-dl:
+
+ - `id` (media identifier)
+ - `title` (media title)
+ - `url` (media download URL) or `formats`
+
+In fact only the last option is technically mandatory (i.e. if you can't figure out the download location of the media the extraction does not make any sense). But by convention youtube-dl also treats `id` and `title` to be mandatory. Thus aforementioned metafields are the critical data the extraction does not make any sense without and if any of them fail to be extracted then extractor is considered completely broken.
+
+[Any field](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L149-L257) apart from the aforementioned ones are considered **optional**. That means that extraction should be **tolerate** to situations when sources for these fields can potentially be unavailable (even if they are always available at the moment) and **future-proof** in order not to break the extraction of general purpose mandatory fields.
+
+#### Example
+
+Say you have some source dictionary `meta` that you've fetched as JSON with HTTP request and it has a key `summary`:
+
+```python
+meta = self._download_json(url, video_id)
+```
+    
+Assume at this point `meta`'s layout is:
+
+```python
+{
+    ...
+    "summary": "some fancy summary text",
+    ...
+}
+```
+
+Assume you want to extract `summary` and put into resulting info dict as `description`. Since `description` is optional metafield you should be ready that this key may be missing from the `meta` dict, so that you should extract it like:
+
+```python
+description = meta.get('summary')  # correct
+```
+
+and not like:
+
+```python
+description = meta['summary']  # incorrect
+```
+
+The latter will break extraction process with `KeyError` if `summary` disappears from `meta` at some time later but with former approach extraction will just go ahead with `description` set to `None` that is perfectly fine (remember `None` is equivalent for absence of data). 
+
+Similarly, you should pass `fatal=False` when extracting optional data from a webpage with `_search_regex`, `_html_search_regex` or similar methods, for instance:
+
+```python
+description = self._search_regex(
+    r'<span[^>]+id="title"[^>]*>([^<]+)<',
+    webpage, 'description', fatal=False)
+```
+
+With `fatal` set to `False` if `_search_regex` fails to extract `description` it will emit a warning and continue extraction.
+
+You can also pass `default=<some fallback value>`, for example:
+
+```python
+description = self._search_regex(
+    r'<span[^>]+id="title"[^>]*>([^<]+)<',
+    webpage, 'description', default=None)
+```
+
+On failure this code will silently continue the extraction with `description` set to `None`. That is useful for metafields that are known to may or may not be present.
+ 
+### Provide fallbacks
+
+When extracting metadata try to provide several scenarios for that. For example if `title` is present in several places/sources try extracting from at least some of them. This would make it more future-proof in case some of the sources became unavailable.
+
+#### Example
+
+Say `meta` from previous example has a `title` and you are about to extract it. Since `title` is mandatory meta field you should end up with something like:
+
+```python
+title = meta['title']
+```
+
+If `title` disappeares from `meta` in future due to some changes on hoster's side the extraction would fail since `title` is mandatory. That's expected.
+
+Assume that you have some another source you can extract `title` from, for example `og:title` HTML meta of a `webpage`. In this case you can provide a fallback scenario:
+
+```python
+title = meta.get('title') or self._og_search_title(webpage)
+```
+
+This code will try to extract from `meta` first and if it fails it will try extracting `og:title` from a `webpage`.
+
+### Make regular expressions flexible
+
+When using regular expressions try to write them fuzzy and flexible.
+ 
+#### Example
+
+Say you need to extract `title` from the following HTML code:
+
+```html
+<span style="position: absolute; left: 910px; width: 90px; float: right; z-index: 9999;" class="title">some fancy title</span>
+```
+
+The code for that task should look similar to:
+
+```python
+title = self._search_regex(
+    r'<span[^>]+class="title"[^>]*>([^<]+)', webpage, 'title')
+```
+
+Or even better:
+
+```python
+title = self._search_regex(
+    r'<span[^>]+class=(["\'])title\1[^>]*>(?P<title>[^<]+)',
+    webpage, 'title', group='title')
+```
+
+Note how you tolerate potential changes in `style` attribute's value or switch from using double quotes to single for `class` attribute: 
+
+The code definitely should not look like:
+
+```python
+title = self._search_regex(
+    r'<span style="position: absolute; left: 910px; width: 90px; float: right; z-index: 9999;" class="title">(.*?)</span>',
+    webpage, 'title', group='title')
+```
+
+### Use safe conversion functions
+
+Wrap all extracted numeric data into safe functions from `utils`: `int_or_none`, `float_or_none`. Use them for string to number conversions as well.
+

ChangeLog

@@ -0,0 +1,333 @@
+version 2016.08.07
+
+Core
++ Add support for TV Parental Guidelines ratings in parse_age_limit
++ Add decode_png (#9706)
++ Add support for partOfTVSeries in JSON-LD
+* Lower master M3U8 manifest preference for better format sorting
+
+Extractors
++ [discoverygo] Add extractor (#10245)
+* [flipagram] Make JSON-LD extraction non fatal
+* [generic] Make JSON-LD extraction non fatal
++ [bbc] Add support for morph embeds (#10239)
+* [tnaflixnetworkbase] Improve title extraction
+* [tnaflix] Fix metadata extraction (#10249)
+* [fox] Fix theplatform release URL query
+* [openload] Fix extraction (#9706)
+* [bbc] Skip duplicate manifest URLs
+* [bbc] Improve format code
++ [bbc] Add support for DASH and F4M
+* [bbc] Improve format sorting and listing
+* [bbc] Improve playlist extraction
++ [pokemon] Add extractor (#10093)
++ [condenast] Add fallback scenario for video info extraction
+
+
+version 2016.08.06
+
+Core
+* Add support for JSON-LD root list entries (#10203)
+* Improve unified_timestamp
+* Lower preference of RTSP formats in generic sorting
++ Add support for multiple properties in _og_search_property
+* Improve password hiding from verbose output
+
+Extractors
++ [adultswim] Add support for trailers (#10235)
+* [archiveorg] Improve extraction (#10219)
++ [jwplatform] Add support for playlists
++ [jwplatform] Add support for relative URLs
+* [jwplatform] Improve audio detection
++ [tvplay] Capture and output native error message
++ [tvplay] Extract series metadata
++ [tvplay] Add support for subtitles (#10194)
+* [tvp] Improve extraction (#7799)
+* [cbslocal] Fix timestamp parsing (#10213)
++ [naver] Add support for subtitles (#8096)
+* [naver] Improve extraction
+* [condenast] Improve extraction
+* [engadget] Relax URL regular expression
+* [5min] Fix extraction
++ [nationalgeographic] Add support for Episode Guide
++ [kaltura] Add support for subtitles
+* [kaltura] Optimize network requests
++ [vodplatform] Add extractor for vod-platform.net
+- [gamekings] Remove extractor
+* [limelight] Extract HTTP formats
+* [ntvru] Fix extraction
++ [comedycentral] Re-add :tds and :thedailyshow shortnames
+
+
+version 2016.08.01
+
+Fixed/improved extractors
+- [yandexmusic:track] Adapt to changes in track location JSON (#10193)
+- [bloomberg] Support another form of player (#10187)
+- [limelight] Skip DRM protected videos
+- [safari] Relax regular expressions for URL matching (#10202)
+- [cwtv] Add support for cwtvpr.com (#10196)
+
+
+version 2016.07.30
+
+Fixed/improved extractors
+- [twitch:clips] Sort formats
+- [tv2] Use m3u8_native
+- [tv2:article] Fix video detection (#10188)
+- rtve (#10076)
+- [dailymotion:playlist] Optimize download archive processing (#10180)
+
+
+version 2016.07.28
+
+Fixed/improved extractors
+- shared (#10170)
+- soundcloud (#10179)
+- twitch (#9767)
+
+
+version 2016.07.26.2
+
+Fixed/improved extractors
+- smotri
+- camdemy
+- mtv
+- comedycentral
+- cmt
+- cbc
+- mgtv
+- orf
+
+
+version 2016.07.24
+
+New extractors
+- arkena (#8682)
+- lcp (#8682)
+
+Fixed/improved extractors
+- facebook (#10151)
+- dailymail
+- telegraaf
+- dcn
+- onet
+- tvp
+
+Miscellaneous
+- Support $Time$ in DASH manifests
+
+
+version 2016.07.22
+
+New extractors
+- odatv (#9285)
+
+Fixed/improved extractors
+- bbc
+- youjizz (#10131)
+- youtube (#10140)
+- pornhub (#10138)
+- eporner (#10139)
+
+
+version 2016.07.17
+
+New extractors
+- nintendo (#9986)
+- streamable (#9122)
+
+Fixed/improved extractors
+- ard (#10095)
+- mtv
+- comedycentral (#10101)
+- viki (#10098)
+- spike (#10106)
+
+Miscellaneous
+- Improved twitter player detection (#10090)
+
+
+version 2016.07.16
+
+New extractors
+- ninenow (#5181)
+
+Fixed/improved extractors
+- rtve (#10076)
+- brightcove
+- 3qsdn
+- syfy (#9087, #3820, #2388)
+- youtube (#10083)
+
+Miscellaneous
+- Fix subtitle embedding for video-only and audio-only files (#10081)
+
+
+version 2016.07.13
+
+New extractors
+- rudo
+
+Fixed/improved extractors
+- biobiochiletv
+- tvplay
+- dbtv
+- brightcove
+- tmz
+- youtube (#10059)
+- shahid (#10062)
+- vk
+- ellentv (#10067)
+
+
+version 2016.07.11
+
+New Extractors
+- roosterteeth (#9864)
+
+Fixed/improved extractors
+- miomio (#9605)
+- vuclip
+- youtube
+- vidzi (#10058)
+
+
+version 2016.07.09.2
+
+Fixed/improved extractors
+- vimeo (#1638)
+- facebook (#10048)
+- lynda (#10047)
+- animeondemand
+
+Fixed/improved features
+- Embedding subtitles no longer throws an error with problematic inputs (#9063)
+
+
+version 2016.07.09.1
+
+Fixed/improved extractors
+- youtube
+- ard
+- srmediatek (#9373)
+
+
+version 2016.07.09
+
+New extractors
+- Flipagram (#9898)
+
+Fixed/improved extractors
+- telecinco
+- toutv
+- radiocanada
+- tweakers (#9516)
+- lynda
+- nick (#7542)
+- polskieradio (#10028)
+- le
+- facebook (#9851)
+- mgtv
+- animeondemand (#10031)
+
+Fixed/improved features
+- `--postprocessor-args` and `--downloader-args` now accepts non-ASCII inputs
+  on non-Windows systems
+
+
+version 2016.07.07
+
+New extractors
+- kamcord (#10001)
+
+Fixed/improved extractors
+- spiegel (#10018)
+- metacafe (#8539, #3253)
+- onet (#9950)
+- francetv (#9955)
+- brightcove (#9965)
+- daum (#9972)
+
+
+version 2016.07.06
+
+Fixed/improved extractors
+- youtube (#10007, #10009)
+- xuite
+- stitcher
+- spiegel
+- slideshare
+- sandia
+- rtvnh
+- prosiebensat1
+- onionstudios
+
+
+version 2016.07.05
+
+Fixed/improved extractors
+- brightcove
+- yahoo (#9995)
+- pornhub (#9997)
+- iqiyi
+- kaltura (#5557)
+- la7
+- Changed features
+- Rename --cn-verfication-proxy to --geo-verification-proxy
+Miscellaneous
+- Add script for displaying downloads statistics
+
+
+version 2016.07.03.1
+
+Fixed/improved extractors
+- theplatform
+- aenetworks
+- nationalgeographic
+- hrti (#9482)
+- facebook (#5701)
+- buzzfeed (#5701)
+- rai (#8617, #9157, #9232, #8552, #8551)
+- nationalgeographic (#9991)
+- iqiyi
+
+
+version 2016.07.03
+
+New extractors
+- hrti (#9482)
+
+Fixed/improved extractors
+- vk (#9981)
+- facebook (#9938)
+- xtube (#9953, #9961)
+
+
+version 2016.07.02
+
+New extractors
+- fusion (#9958)
+
+Fixed/improved extractors
+- twitch (#9975)
+- vine (#9970)
+- periscope (#9967)
+- pornhub (#8696)
+
+
+version 2016.07.01
+
+New extractors
+- 9c9media
+- ctvnews (#2156)
+- ctv (#4077)
+
+Fixed/Improved extractors
+- rds
+- meta (#8789)
+- pornhub (#9964)
+- sixplay (#2183)
+
+New features
+- Accept quoted strings across multiple lines (#9940)

Makefile

@@ -1,7 +1,7 @@
 all: youtube-dl README.md CONTRIBUTING.md README.txt youtube-dl.1 youtube-dl.bash-completion youtube-dl.zsh youtube-dl.fish supportedsites
 
 clean:
-	rm -rf youtube-dl.1.temp.md youtube-dl.1 youtube-dl.bash-completion README.txt MANIFEST build/ dist/ .coverage cover/ youtube-dl.tar.gz youtube-dl.zsh youtube-dl.fish youtube_dl/extractor/lazy_extractors.py *.dump *.part *.info.json *.mp4 *.flv *.mp3 *.avi CONTRIBUTING.md.tmp ISSUE_TEMPLATE.md.tmp youtube-dl youtube-dl.exe
+	rm -rf youtube-dl.1.temp.md youtube-dl.1 youtube-dl.bash-completion README.txt MANIFEST build/ dist/ .coverage cover/ youtube-dl.tar.gz youtube-dl.zsh youtube-dl.fish youtube_dl/extractor/lazy_extractors.py *.dump *.part *.info.json *.mp4 *.m4a *.flv *.mp3 *.avi *.mkv *.webm *.jpg *.png CONTRIBUTING.md.tmp ISSUE_TEMPLATE.md.tmp youtube-dl youtube-dl.exe
 	find . -name "*.pyc" -delete
 	find . -name "*.class" -delete
 
@@ -37,7 +37,7 @@ test:
 ot: offlinetest
 
 offlinetest: codetest
-	$(PYTHON) -m nose --verbose test --exclude test_download.py --exclude test_age_restriction.py --exclude test_subtitles.py --exclude test_write_annotations.py --exclude test_youtube_lists.py --exclude test_iqiyi_sdk_interpreter.py
+	$(PYTHON) -m nose --verbose test --exclude test_download.py --exclude test_age_restriction.py --exclude test_subtitles.py --exclude test_write_annotations.py --exclude test_youtube_lists.py --exclude test_iqiyi_sdk_interpreter.py --exclude test_socks.py
 
 tar: youtube-dl.tar.gz
 
@@ -69,7 +69,7 @@ README.txt: README.md
 	pandoc -f markdown -t plain README.md -o README.txt
 
 youtube-dl.1: README.md
-	$(PYTHON) devscripts/prepare_manpage.py >youtube-dl.1.temp.md
+	$(PYTHON) devscripts/prepare_manpage.py youtube-dl.1.temp.md
 	pandoc -s -f markdown -t man youtube-dl.1.temp.md -o youtube-dl.1
 	rm -f youtube-dl.1.temp.md
 
@@ -94,7 +94,7 @@ _EXTRACTOR_FILES != find youtube_dl/extractor -iname '*.py' -and -not -iname 'la
 youtube_dl/extractor/lazy_extractors.py: devscripts/make_lazy_extractors.py devscripts/lazy_load_template.py $(_EXTRACTOR_FILES)
 	$(PYTHON) devscripts/make_lazy_extractors.py $@
 
-youtube-dl.tar.gz: youtube-dl README.md README.txt youtube-dl.1 youtube-dl.bash-completion youtube-dl.zsh youtube-dl.fish
+youtube-dl.tar.gz: youtube-dl README.md README.txt youtube-dl.1 youtube-dl.bash-completion youtube-dl.zsh youtube-dl.fish ChangeLog
 	@tar -czf youtube-dl.tar.gz --transform "s|^|youtube-dl/|" --owner 0 --group 0 \
 		--exclude '*.DS_Store' \
 		--exclude '*.kate-swp' \
@@ -107,7 +107,7 @@ youtube-dl.tar.gz: youtube-dl README.md README.txt youtube-dl.1 youtube-dl.bash-
 		--exclude 'docs/_build' \
 		-- \
 		bin devscripts test youtube_dl docs \
-		LICENSE README.md README.txt \
+		ChangeLog LICENSE README.md README.txt \
 		Makefile MANIFEST.in youtube-dl.1 youtube-dl.bash-completion \
 		youtube-dl.zsh youtube-dl.fish setup.py \
 		youtube-dl

README.md

@@ -17,7 +17,7 @@ youtube-dl - download videos from youtube.com or other video platforms
 
 To install it right away for all UNIX users (Linux, OS X, etc.), type:
 
-    sudo curl https://yt-dl.org/latest/youtube-dl -o /usr/local/bin/youtube-dl
+    sudo curl -L https://yt-dl.org/downloads/latest/youtube-dl -o /usr/local/bin/youtube-dl
     sudo chmod a+rx /usr/local/bin/youtube-dl
 
 If you do not have curl, you can alternatively use a recent wget:
@@ -25,20 +25,26 @@ If you do not have curl, you can alternatively use a recent wget:
     sudo wget https://yt-dl.org/downloads/latest/youtube-dl -O /usr/local/bin/youtube-dl
     sudo chmod a+rx /usr/local/bin/youtube-dl
 
-Windows users can [download a .exe file](https://yt-dl.org/latest/youtube-dl.exe) and place it in their home directory or any other location on their [PATH](http://en.wikipedia.org/wiki/PATH_%28variable%29).
-
-OS X users can install **youtube-dl** with [Homebrew](http://brew.sh/).
-
-    brew install youtube-dl
+Windows users can [download an .exe file](https://yt-dl.org/latest/youtube-dl.exe) and place it in any location on their [PATH](http://en.wikipedia.org/wiki/PATH_%28variable%29) except for `%SYSTEMROOT%\System32` (e.g. **do not** put in `C:\Windows\System32`).
 
 You can also use pip:
 
-    sudo pip install youtube-dl
+    sudo pip install --upgrade youtube-dl
+    
+This command will update youtube-dl if you have already installed it. See the [pypi page](https://pypi.python.org/pypi/youtube_dl) for more information.
+
+OS X users can install youtube-dl with [Homebrew](http://brew.sh/):
+
+    brew install youtube-dl
+
+Or with [MacPorts](https://www.macports.org/):
+
+    sudo port install youtube-dl
 
 Alternatively, refer to the [developer instructions](#developer-instructions) for how to check out and work with the git repository. For further options, including PGP signatures, see the [youtube-dl Download Page](https://rg3.github.io/youtube-dl/download.html).
 
 # DESCRIPTION
-**youtube-dl** is a small command-line program to download videos from
+**youtube-dl** is a command-line program to download videos from
 YouTube.com and a few more sites. It requires the Python interpreter, version
 2.6, 2.7, or 3.2+, and it is not platform specific. It should work on
 your Unix box, on Windows or on Mac OS X. It is released to the public domain,
@@ -73,8 +79,8 @@ which means you can modify it, redistribute it or use it however you like.
                                      repairs broken URLs, but emits an error if
                                      this is not possible instead of searching.
     --ignore-config                  Do not read configuration files. When given
-                                     in the global configuration file /etc
-                                     /youtube-dl.conf: Do not read the user
+                                     in the global configuration file
+                                     /etc/youtube-dl.conf: Do not read the user
                                      configuration in ~/.config/youtube-
                                      dl/config (%APPDATA%/youtube-dl/config.txt
                                      on Windows)
@@ -85,9 +91,11 @@ which means you can modify it, redistribute it or use it however you like.
     --no-color                       Do not emit color codes in output
 
 ## Network Options:
-    --proxy URL                      Use the specified HTTP/HTTPS proxy. Pass in
-                                     an empty string (--proxy "") for direct
-                                     connection
+    --proxy URL                      Use the specified HTTP/HTTPS/SOCKS proxy.
+                                     To enable experimental SOCKS proxy, specify
+                                     a proper scheme. For example
+                                     socks5://127.0.0.1:1080/. Pass in an empty
+                                     string (--proxy "") for direct connection
     --socket-timeout SECONDS         Time to wait before giving up, in seconds
     --source-address IP              Client-side IP address to bind to
                                      (experimental)
@@ -95,9 +103,9 @@ which means you can modify it, redistribute it or use it however you like.
                                      (experimental)
     -6, --force-ipv6                 Make all connections via IPv6
                                      (experimental)
-    --cn-verification-proxy URL      Use this proxy to verify the IP address for
-                                     some Chinese sites. The default proxy
-                                     specified by --proxy (or none, if the
+    --geo-verification-proxy URL     Use this proxy to verify the IP address for
+                                     some geo-restricted sites. The default
+                                     proxy specified by --proxy (or none, if the
                                      options is not present) is used for the
                                      actual downloading. (experimental)
 
@@ -160,7 +168,7 @@ which means you can modify it, redistribute it or use it however you like.
                                      (experimental)
 
 ## Download Options:
-    -r, --rate-limit LIMIT           Maximum download rate in bytes per second
+    -r, --limit-rate RATE            Maximum download rate in bytes per second
                                      (e.g. 50K or 4.2M)
     -R, --retries RETRIES            Number of retries (default is 10), or
                                      "infinite".
@@ -247,18 +255,19 @@ which means you can modify it, redistribute it or use it however you like.
     --write-info-json                Write video metadata to a .info.json file
     --write-annotations              Write video annotations to a
                                      .annotations.xml file
-    --load-info FILE                 JSON file containing the video information
+    --load-info-json FILE            JSON file containing the video information
                                      (created with the "--write-info-json"
                                      option)
     --cookies FILE                   File to read cookies from and dump cookie
                                      jar in
     --cache-dir DIR                  Location in the filesystem where youtube-dl
                                      can store some downloaded information
-                                     permanently. By default $XDG_CACHE_HOME
-                                     /youtube-dl or ~/.cache/youtube-dl . At the
-                                     moment, only YouTube player files (for
-                                     videos with obfuscated signatures) are
-                                     cached, but that may change.
+                                     permanently. By default
+                                     $XDG_CACHE_HOME/youtube-dl or
+                                     ~/.cache/youtube-dl . At the moment, only
+                                     YouTube player files (for videos with
+                                     obfuscated signatures) are cached, but that
+                                     may change.
     --no-cache-dir                   Disable filesystem caching
     --rm-cache-dir                   Delete all filesystem cache files
 
@@ -415,7 +424,7 @@ which means you can modify it, redistribute it or use it however you like.
 
 # CONFIGURATION
 
-You can configure youtube-dl by placing any supported command line option to a configuration file. On Linux, the system wide configuration file is located at `/etc/youtube-dl.conf` and the user wide configuration file at `~/.config/youtube-dl/config`. On Windows, the user wide configuration file locations are `%APPDATA%\youtube-dl\config.txt` or `C:\Users\<user name>\youtube-dl.conf`.
+You can configure youtube-dl by placing any supported command line option to a configuration file. On Linux and OS X, the system wide configuration file is located at `/etc/youtube-dl.conf` and the user wide configuration file at `~/.config/youtube-dl/config`. On Windows, the user wide configuration file locations are `%APPDATA%\youtube-dl\config.txt` or `C:\Users\<user name>\youtube-dl.conf`. Note that by default configuration file may not exist so you may need to create it yourself.
 
 For example, with the following configuration file youtube-dl will always extract the audio, not copy the mtime, use a proxy and save all videos under `Movies` directory in your home directory:
 ```
@@ -423,6 +432,7 @@ For example, with the following configuration file youtube-dl will always extrac
 --no-mtime
 --proxy 127.0.0.1:3128
 -o ~/Movies/%(title)s.%(ext)s
+# Lines starting with # are comments
 ```
 
 Note that options in configuration file are just the same options aka switches used in regular command line calls thus there **must be no whitespace** after `-` or `--`, e.g. `-o` or `--proxy` but not `- o` or `-- proxy`.
@@ -431,7 +441,7 @@ You can use `--ignore-config` if you want to disable the configuration file for
 
 ### Authentication with `.netrc` file
 
-You may also want to configure automatic credentials storage for extractors that support authentication (by providing login and password with `--username` and `--password`) in order not to pass credentials as command line arguments on every youtube-dl execution and prevent tracking plain text passwords in the shell command history. You can achieve this using a [`.netrc` file](http://stackoverflow.com/tags/.netrc/info) on per extractor basis. For that you will need to create a`.netrc` file in your `$HOME` and restrict permissions to read/write by you only:
+You may also want to configure automatic credentials storage for extractors that support authentication (by providing login and password with `--username` and `--password`) in order not to pass credentials as command line arguments on every youtube-dl execution and prevent tracking plain text passwords in the shell command history. You can achieve this using a [`.netrc` file](http://stackoverflow.com/tags/.netrc/info) on per extractor basis. For that you will need to create a `.netrc` file in your `$HOME` and restrict permissions to read/write by you only:
 ```
 touch $HOME/.netrc
 chmod a-rwx,u+rw $HOME/.netrc
@@ -465,7 +475,7 @@ The basic usage is not to set any template arguments when downloading a single f
  - `display_id`: An alternative identifier for the video
  - `uploader`: Full name of the video uploader
  - `license`: License name the video is licensed under
- - `creator`: The main artist who created the video
+ - `creator`: The creator of the video
  - `release_date`: The date (YYYYMMDD) when the video was released
  - `timestamp`: UNIX timestamp of the moment the video became available
  - `upload_date`: Video upload date (YYYYMMDD)
@@ -502,6 +512,9 @@ The basic usage is not to set any template arguments when downloading a single f
  - `autonumber`: Five-digit number that will be increased with each download, starting at zero
  - `playlist`: Name or id of the playlist that contains the video
  - `playlist_index`: Index of the video in the playlist padded with leading zeros according to the total length of the playlist
+ - `playlist_id`: Playlist identifier
+ - `playlist_title`: Playlist title
+
 
 Available for the video that belongs to some logical chapter or section:
  - `chapter`: Name or title of the chapter the video belongs to
@@ -541,6 +554,10 @@ The current default template is `%(title)s-%(id)s.%(ext)s`.
 
 In some cases, you don't want special characters such as 中, spaces, or &, such as when transferring the downloaded filename to a Windows system or the filename through an 8bit-unsafe channel. In these cases, add the `--restrict-filenames` flag to get a shorter title:
 
+#### Output template and Windows batch files
+
+If you are using output template inside a Windows batch file then you must escape plain percent characters (`%`) by doubling, so that `-o "%(title)s-%(id)s.%(ext)s"` should become `-o "%%(title)s-%%(id)s.%%(ext)s"`. However you should not touch `%`'s that are not plain characters, e.g. environment variables for expansion should stay intact: `-o "C:\%HOMEPATH%\Desktop\%%(title)s.%%(ext)s"`.
+
 #### Output template examples
 
 Note on Windows you may need to use double quotes instead of single.
@@ -691,12 +708,20 @@ hash -r
 
 Again, from then on you'll be able to update with `sudo youtube-dl -U`.
 
+### youtube-dl is extremely slow to start on Windows
+
+Add a file exclusion for `youtube-dl.exe` in Windows Defender settings.
+
 ### I'm getting an error `Unable to extract OpenGraph title` on YouTube playlists
 
 YouTube changed their playlist format in March 2014 and later on, so you'll need at least youtube-dl 2014.07.25 to download all YouTube videos.
 
 If you have installed youtube-dl with a package manager, pip, setup.py or a tarball, please use that to update. Note that Ubuntu packages do not seem to get updated anymore. Since we are not affiliated with Ubuntu, there is little we can do. Feel free to [report bugs](https://bugs.launchpad.net/ubuntu/+source/youtube-dl/+filebug) to the [Ubuntu packaging guys](mailto:ubuntu-motu@lists.ubuntu.com?subject=outdated%20version%20of%20youtube-dl) - all they have to do is update the package to a somewhat recent version. See above for a way to update.
 
+### I'm getting an error when trying to use output template: `error: using output template conflicts with using title, video ID or auto number`
+
+Make sure you are not using `-o` with any of these options `-t`, `--title`, `--id`, `-A` or `--auto-number` set in command line or in a configuration file. Remove the latter if any.
+
 ### Do I always have to pass `-citw`?
 
 By default, youtube-dl intends to have the best options (incidentally, if you have a convincing case that these should be different, [please file an issue where you explain that](https://yt-dl.org/bug)). Therefore, it is unnecessary and sometimes harmful to copy long option strings from webpages. In particular, the only option out of `-citw` that is regularly useful is `-i`.
@@ -717,7 +742,7 @@ Videos or video formats streamed via RTMP protocol can only be downloaded when [
 
 ### I have downloaded a video but how can I play it?
 
-Once the video is fully downloaded, use any video player, such as [vlc](http://www.videolan.org) or [mplayer](http://www.mplayerhq.hu/).
+Once the video is fully downloaded, use any video player, such as [mpv](https://mpv.io/), [vlc](http://www.videolan.org) or [mplayer](http://www.mplayerhq.hu/).
 
 ### I extracted a video URL with `-g`, but it does not play on another machine / in my webbrowser.
 
@@ -774,9 +799,9 @@ means you're using an outdated version of Python. Please update to Python 2.6 or
 
 Since June 2012 ([#342](https://github.com/rg3/youtube-dl/issues/342)) youtube-dl is packed as an executable zipfile, simply unzip it (might need renaming to `youtube-dl.zip` first on some systems) or clone the git repository, as laid out above. If you modify the code, you can run it by executing the `__main__.py` file. To recompile the executable, run `make youtube-dl`.
 
-### The exe throws a *Runtime error from Visual C++*
+### The exe throws an error due to missing `MSVCR100.dll`
 
-To run the exe you need to install first the [Microsoft Visual C++ 2008 Redistributable Package](http://www.microsoft.com/en-us/download/details.aspx?id=29).
+To run the exe you need to install first the [Microsoft Visual C++ 2010 Redistributable Package (x86)](https://www.microsoft.com/en-US/download/details.aspx?id=5555).
 
 ### On Windows, how should I set up ffmpeg and youtube-dl? Where should I put the exe files?
 
@@ -831,6 +856,12 @@ It is *not* possible to detect whether a URL is supported or not. That's because
 
 If you want to find out whether a given URL is supported, simply call youtube-dl with it. If you get no videos back, chances are the URL is either not referring to a video or unsupported. You can find out which by examining the output (if you run youtube-dl on the console) or catching an `UnsupportedError` exception if you run it from a Python program.
 
+# Why do I need to go through that much red tape when filing bugs?
+
+Before we had the issue template, despite our extensive [bug reporting instructions](#bugs), about 80% of the issue reports we got were useless, for instance because people used ancient versions hundreds of releases old, because of simple syntactic errors (not in youtube-dl but in general shell usage), because the problem was alrady reported multiple times before, because people did not actually read an error message, even if it said "please install ffmpeg", because people did not mention the URL they were trying to download and many more simple, easy-to-avoid problems, many of whom were totally unrelated to youtube-dl.
+
+youtube-dl is an open-source project manned by too few volunteers, so we'd rather spend time fixing bugs where we are certain none of those simple problems apply, and where we can be reasonably confident to be able to reproduce the issue without asking the reporter repeatedly. As such, the output of `youtube-dl -v YOUR_URL_HERE` is really all that's required to file an issue. The issue template also guides you through some basic steps you can do, such as checking that your version of youtube-dl is current.
+
 # DEVELOPER INSTRUCTIONS
 
 Most users do not need to build youtube-dl and can [download the builds](http://rg3.github.io/youtube-dl/download.html) or get them from their distribution.
@@ -860,9 +891,17 @@ If you want to add support for a new site, first of all **make sure** this site
 After you have ensured this site is distributing it's content legally, you can follow this quick list (assuming your service is called `yourextractor`):
 
 1. [Fork this repository](https://github.com/rg3/youtube-dl/fork)
-2. Check out the source code with `git clone git@github.com:YOUR_GITHUB_USERNAME/youtube-dl.git`
-3. Start a new git branch with `cd youtube-dl; git checkout -b yourextractor`
+2. Check out the source code with:
+
+        git clone git@github.com:YOUR_GITHUB_USERNAME/youtube-dl.git
+
+3. Start a new git branch with
+
+        cd youtube-dl
+        git checkout -b yourextractor
+
 4. Start with this simple template and save it to `youtube_dl/extractor/yourextractor.py`:
+
     ```python
     # coding: utf-8
     from __future__ import unicode_literals
@@ -905,20 +944,152 @@ After you have ensured this site is distributing it's content legally, you can f
     ```
 5. Add an import in [`youtube_dl/extractor/extractors.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/extractors.py).
 6. Run `python test/test_download.py TestDownload.test_YourExtractor`. This *should fail* at first, but you can continually re-run it until you're done. If you decide to add more than one test, then rename ``_TEST`` to ``_TESTS`` and make it into a list of dictionaries. The tests will then be named `TestDownload.test_YourExtractor`, `TestDownload.test_YourExtractor_1`, `TestDownload.test_YourExtractor_2`, etc.
-7. Have a look at [`youtube_dl/extractor/common.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py) for possible helper methods and a [detailed description of what your extractor should and may return](https://github.com/rg3/youtube-dl/blob/58525c94d547be1c8167d16c298bdd75506db328/youtube_dl/extractor/common.py#L68-L226). Add tests and code for as many as you want.
-8. Keep in mind that the only mandatory fields in info dict for successful extraction process are `id`, `title` and either `url` or `formats`, i.e. these are the critical data the extraction does not make any sense without. This means that [any field](https://github.com/rg3/youtube-dl/blob/58525c94d547be1c8167d16c298bdd75506db328/youtube_dl/extractor/common.py#L138-L226) apart from aforementioned mandatory ones should be treated **as optional** and extraction should be **tolerate** to situations when sources for these fields can potentially be unavailable (even if they always available at the moment) and **future-proof** in order not to break the extraction of general purpose mandatory fields. For example, if you have some intermediate dict `meta` that is a source of metadata and it has a key `summary` that you want to extract and put into resulting info dict as `description`, you should be ready that this key may be missing from the `meta` dict, i.e. you should extract it as `meta.get('summary')` and not `meta['summary']`. Similarly, you should pass `fatal=False` when extracting data from a webpage with `_search_regex/_html_search_regex`.
-9. Check the code with [flake8](https://pypi.python.org/pypi/flake8).
-10. When the tests pass, [add](http://git-scm.com/docs/git-add) the new files and [commit](http://git-scm.com/docs/git-commit) them and [push](http://git-scm.com/docs/git-push) the result, like this:
+7. Have a look at [`youtube_dl/extractor/common.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py) for possible helper methods and a [detailed description of what your extractor should and may return](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L74-L252). Add tests and code for as many as you want.
+8. Make sure your code follows [youtube-dl coding conventions](#youtube-dl-coding-conventions) and check the code with [flake8](https://pypi.python.org/pypi/flake8). Also make sure your code works under all [Python](http://www.python.org/) versions claimed supported by youtube-dl, namely 2.6, 2.7, and 3.2+.
+9. When the tests pass, [add](http://git-scm.com/docs/git-add) the new files and [commit](http://git-scm.com/docs/git-commit) them and [push](http://git-scm.com/docs/git-push) the result, like this:
 
         $ git add youtube_dl/extractor/extractors.py
         $ git add youtube_dl/extractor/yourextractor.py
         $ git commit -m '[yourextractor] Add new extractor'
         $ git push origin yourextractor
 
-11. Finally, [create a pull request](https://help.github.com/articles/creating-a-pull-request). We'll then review and merge it.
+10. Finally, [create a pull request](https://help.github.com/articles/creating-a-pull-request). We'll then review and merge it.
 
 In any case, thank you very much for your contributions!
 
+## youtube-dl coding conventions
+
+This section introduces a guide lines for writing idiomatic, robust and future-proof extractor code.
+
+Extractors are very fragile by nature since they depend on the layout of the source data provided by 3rd party media hoster out of your control and this layout tend to change. As an extractor implementer your task is not only to write code that will extract media links and metadata correctly but also to minimize code dependency on source's layout changes and even to make the code foresee potential future changes and be ready for that. This is important because it will allow extractor not to break on minor layout changes thus keeping old youtube-dl versions working. Even though this breakage issue is easily fixed by emitting a new version of youtube-dl with fix incorporated all the previous version become broken in all repositories and distros' packages that may not be so prompt in fetching the update from us. Needless to say some may never receive an update at all that is possible for non rolling release distros.
+
+### Mandatory and optional metafields
+
+For extraction to work youtube-dl relies on metadata your extractor extracts and provides to youtube-dl expressed by [information dictionary](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L75-L257) or simply *info dict*. Only the following meta fields in *info dict* are considered mandatory for successful extraction process by youtube-dl:
+
+ - `id` (media identifier)
+ - `title` (media title)
+ - `url` (media download URL) or `formats`
+
+In fact only the last option is technically mandatory (i.e. if you can't figure out the download location of the media the extraction does not make any sense). But by convention youtube-dl also treats `id` and `title` to be mandatory. Thus aforementioned metafields are the critical data the extraction does not make any sense without and if any of them fail to be extracted then extractor is considered completely broken.
+
+[Any field](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L149-L257) apart from the aforementioned ones are considered **optional**. That means that extraction should be **tolerate** to situations when sources for these fields can potentially be unavailable (even if they are always available at the moment) and **future-proof** in order not to break the extraction of general purpose mandatory fields.
+
+#### Example
+
+Say you have some source dictionary `meta` that you've fetched as JSON with HTTP request and it has a key `summary`:
+
+```python
+meta = self._download_json(url, video_id)
+```
+    
+Assume at this point `meta`'s layout is:
+
+```python
+{
+    ...
+    "summary": "some fancy summary text",
+    ...
+}
+```
+
+Assume you want to extract `summary` and put into resulting info dict as `description`. Since `description` is optional metafield you should be ready that this key may be missing from the `meta` dict, so that you should extract it like:
+
+```python
+description = meta.get('summary')  # correct
+```
+
+and not like:
+
+```python
+description = meta['summary']  # incorrect
+```
+
+The latter will break extraction process with `KeyError` if `summary` disappears from `meta` at some time later but with former approach extraction will just go ahead with `description` set to `None` that is perfectly fine (remember `None` is equivalent for absence of data). 
+
+Similarly, you should pass `fatal=False` when extracting optional data from a webpage with `_search_regex`, `_html_search_regex` or similar methods, for instance:
+
+```python
+description = self._search_regex(
+    r'<span[^>]+id="title"[^>]*>([^<]+)<',
+    webpage, 'description', fatal=False)
+```
+
+With `fatal` set to `False` if `_search_regex` fails to extract `description` it will emit a warning and continue extraction.
+
+You can also pass `default=<some fallback value>`, for example:
+
+```python
+description = self._search_regex(
+    r'<span[^>]+id="title"[^>]*>([^<]+)<',
+    webpage, 'description', default=None)
+```
+
+On failure this code will silently continue the extraction with `description` set to `None`. That is useful for metafields that are known to may or may not be present.
+ 
+### Provide fallbacks
+
+When extracting metadata try to provide several scenarios for that. For example if `title` is present in several places/sources try extracting from at least some of them. This would make it more future-proof in case some of the sources became unavailable.
+
+#### Example
+
+Say `meta` from previous example has a `title` and you are about to extract it. Since `title` is mandatory meta field you should end up with something like:
+
+```python
+title = meta['title']
+```
+
+If `title` disappeares from `meta` in future due to some changes on hoster's side the extraction would fail since `title` is mandatory. That's expected.
+
+Assume that you have some another source you can extract `title` from, for example `og:title` HTML meta of a `webpage`. In this case you can provide a fallback scenario:
+
+```python
+title = meta.get('title') or self._og_search_title(webpage)
+```
+
+This code will try to extract from `meta` first and if it fails it will try extracting `og:title` from a `webpage`.
+
+### Make regular expressions flexible
+
+When using regular expressions try to write them fuzzy and flexible.
+ 
+#### Example
+
+Say you need to extract `title` from the following HTML code:
+
+```html
+<span style="position: absolute; left: 910px; width: 90px; float: right; z-index: 9999;" class="title">some fancy title</span>
+```
+
+The code for that task should look similar to:
+
+```python
+title = self._search_regex(
+    r'<span[^>]+class="title"[^>]*>([^<]+)', webpage, 'title')
+```
+
+Or even better:
+
+```python
+title = self._search_regex(
+    r'<span[^>]+class=(["\'])title\1[^>]*>(?P<title>[^<]+)',
+    webpage, 'title', group='title')
+```
+
+Note how you tolerate potential changes in `style` attribute's value or switch from using double quotes to single for `class` attribute: 
+
+The code definitely should not look like:
+
+```python
+title = self._search_regex(
+    r'<span style="position: absolute; left: 910px; width: 90px; float: right; z-index: 9999;" class="title">(.*?)</span>',
+    webpage, 'title', group='title')
+```
+
+### Use safe conversion functions
+
+Wrap all extracted numeric data into safe functions from `utils`: `int_or_none`, `float_or_none`. Use them for string to number conversions as well.
+
 # EMBEDDING YOUTUBE-DL
 
 youtube-dl makes the best effort to be a good command-line program, and thus should be callable from any programming language. If you encounter any problems parsing its output, feel free to [create a report](https://github.com/rg3/youtube-dl/issues/new).
@@ -934,7 +1105,7 @@ with youtube_dl.YoutubeDL(ydl_opts) as ydl:
     ydl.download(['http://www.youtube.com/watch?v=BaW_jenozKc'])
 ```
 
-Most likely, you'll want to use various options. For a list of what can be done, have a look at [`youtube_dl/YoutubeDL.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/YoutubeDL.py#L121-L269). For a start, if you want to intercept youtube-dl's output, set a `logger` object.
+Most likely, you'll want to use various options. For a list of options available, have a look at [`youtube_dl/YoutubeDL.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/YoutubeDL.py#L128-L278). For a start, if you want to intercept youtube-dl's output, set a `logger` object.
 
 Here's a more complete example of a program that outputs only errors (and a short message after the download is finished), and downloads/converts the video to an mp3 file:
 

devscripts/buildserver.py

@@ -1,17 +1,38 @@
 #!/usr/bin/python3
 
-from http.server import HTTPServer, BaseHTTPRequestHandler
-from socketserver import ThreadingMixIn
 import argparse
 import ctypes
 import functools
+import shutil
+import subprocess
 import sys
+import tempfile
 import threading
 import traceback
 import os.path
 
+sys.path.insert(0, os.path.dirname(os.path.dirname((os.path.abspath(__file__)))))
+from youtube_dl.compat import (
+    compat_input,
+    compat_http_server,
+    compat_str,
+    compat_urlparse,
+)
 
-class BuildHTTPServer(ThreadingMixIn, HTTPServer):
+# These are not used outside of buildserver.py thus not in compat.py
+
+try:
+    import winreg as compat_winreg
+except ImportError:  # Python 2
+    import _winreg as compat_winreg
+
+try:
+    import socketserver as compat_socketserver
+except ImportError:  # Python 2
+    import SocketServer as compat_socketserver
+
+
+class BuildHTTPServer(compat_socketserver.ThreadingMixIn, compat_http_server.HTTPServer):
     allow_reuse_address = True
 
 
@@ -191,7 +212,7 @@ def main(args=None):
                         action='store_const', dest='action', const='service',
                         help='Run as a Windows service')
     parser.add_argument('-b', '--bind', metavar='<host:port>',
-                        action='store', default='localhost:8142',
+                        action='store', default='0.0.0.0:8142',
                         help='Bind to host:port (default %default)')
     options = parser.parse_args(args=args)
 
@@ -216,7 +237,7 @@ def main(args=None):
     srv = BuildHTTPServer((host, port), BuildHTTPRequestHandler)
     thr = threading.Thread(target=srv.serve_forever)
     thr.start()
-    input('Press ENTER to shut down')
+    compat_input('Press ENTER to shut down')
     srv.shutdown()
     thr.join()
 
@@ -231,8 +252,6 @@ def rmtree(path):
             os.remove(fname)
     os.rmdir(path)
 
-#==============================================================================
-
 
 class BuildError(Exception):
     def __init__(self, output, code=500):
@@ -249,15 +268,25 @@ class HTTPError(BuildError):
 
 class PythonBuilder(object):
     def __init__(self, **kwargs):
-        pythonVersion = kwargs.pop('python', '2.7')
-        try:
-            key = _winreg.OpenKey(_winreg.HKEY_LOCAL_MACHINE, r'SOFTWARE\Python\PythonCore\%s\InstallPath' % pythonVersion)
+        python_version = kwargs.pop('python', '3.4')
+        python_path = None
+        for node in ('Wow6432Node\\', ''):
             try:
-                self.pythonPath, _ = _winreg.QueryValueEx(key, '')
-            finally:
-                _winreg.CloseKey(key)
-        except Exception:
-            raise BuildError('No such Python version: %s' % pythonVersion)
+                key = compat_winreg.OpenKey(
+                    compat_winreg.HKEY_LOCAL_MACHINE,
+                    r'SOFTWARE\%sPython\PythonCore\%s\InstallPath' % (node, python_version))
+                try:
+                    python_path, _ = compat_winreg.QueryValueEx(key, '')
+                finally:
+                    compat_winreg.CloseKey(key)
+                break
+            except Exception:
+                pass
+
+        if not python_path:
+            raise BuildError('No such Python version: %s' % python_version)
+
+        self.pythonPath = python_path
 
         super(PythonBuilder, self).__init__(**kwargs)
 
@@ -305,8 +334,10 @@ class YoutubeDLBuilder(object):
 
     def build(self):
         try:
-            subprocess.check_output([os.path.join(self.pythonPath, 'python.exe'), 'setup.py', 'py2exe'],
-                                    cwd=self.buildPath)
+            proc = subprocess.Popen([os.path.join(self.pythonPath, 'python.exe'), 'setup.py', 'py2exe'], stdin=subprocess.PIPE, cwd=self.buildPath)
+            proc.wait()
+            #subprocess.check_output([os.path.join(self.pythonPath, 'python.exe'), 'setup.py', 'py2exe'],
+            #                        cwd=self.buildPath)
         except subprocess.CalledProcessError as e:
             raise BuildError(e.output)
 
@@ -369,12 +400,12 @@ class Builder(PythonBuilder, GITBuilder, YoutubeDLBuilder, DownloadBuilder, Clea
     pass
 
 
-class BuildHTTPRequestHandler(BaseHTTPRequestHandler):
+class BuildHTTPRequestHandler(compat_http_server.BaseHTTPRequestHandler):
     actionDict = {'build': Builder, 'download': Builder}  # They're the same, no more caching.
 
     def do_GET(self):
-        path = urlparse.urlparse(self.path)
-        paramDict = dict([(key, value[0]) for key, value in urlparse.parse_qs(path.query).items()])
+        path = compat_urlparse.urlparse(self.path)
+        paramDict = dict([(key, value[0]) for key, value in compat_urlparse.parse_qs(path.query).items()])
         action, _, path = path.path.strip('/').partition('/')
         if path:
             path = path.split('/')
@@ -388,7 +419,7 @@ class BuildHTTPRequestHandler(BaseHTTPRequestHandler):
                         builder.close()
                 except BuildError as e:
                     self.send_response(e.code)
-                    msg = unicode(e).encode('UTF-8')
+                    msg = compat_str(e).encode('UTF-8')
                     self.send_header('Content-Type', 'text/plain; charset=UTF-8')
                     self.send_header('Content-Length', len(msg))
                     self.end_headers()
@@ -400,7 +431,5 @@ class BuildHTTPRequestHandler(BaseHTTPRequestHandler):
         else:
             self.send_response(500, 'Malformed URL')
 
-#==============================================================================
-
 if __name__ == '__main__':
     main()

devscripts/create-github-release.py

@@ -0,0 +1,111 @@
+#!/usr/bin/env python
+from __future__ import unicode_literals
+
+import base64
+import json
+import mimetypes
+import netrc
+import optparse
+import os
+import sys
+
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl.compat import (
+    compat_basestring,
+    compat_input,
+    compat_getpass,
+    compat_print,
+    compat_urllib_request,
+)
+from youtube_dl.utils import (
+    make_HTTPS_handler,
+    sanitized_Request,
+)
+
+
+class GitHubReleaser(object):
+    _API_URL = 'https://api.github.com/repos/rg3/youtube-dl/releases'
+    _UPLOADS_URL = 'https://uploads.github.com/repos/rg3/youtube-dl/releases/%s/assets?name=%s'
+    _NETRC_MACHINE = 'github.com'
+
+    def __init__(self, debuglevel=0):
+        self._init_github_account()
+        https_handler = make_HTTPS_handler({}, debuglevel=debuglevel)
+        self._opener = compat_urllib_request.build_opener(https_handler)
+
+    def _init_github_account(self):
+        try:
+            info = netrc.netrc().authenticators(self._NETRC_MACHINE)
+            if info is not None:
+                self._username = info[0]
+                self._password = info[2]
+                compat_print('Using GitHub credentials found in .netrc...')
+                return
+            else:
+                compat_print('No GitHub credentials found in .netrc')
+        except (IOError, netrc.NetrcParseError):
+            compat_print('Unable to parse .netrc')
+        self._username = compat_input(
+            'Type your GitHub username or email address and press [Return]: ')
+        self._password = compat_getpass(
+            'Type your GitHub password and press [Return]: ')
+
+    def _call(self, req):
+        if isinstance(req, compat_basestring):
+            req = sanitized_Request(req)
+        # Authorizing manually since GitHub does not response with 401 with
+        # WWW-Authenticate header set (see
+        # https://developer.github.com/v3/#basic-authentication)
+        b64 = base64.b64encode(
+            ('%s:%s' % (self._username, self._password)).encode('utf-8')).decode('ascii')
+        req.add_header('Authorization', 'Basic %s' % b64)
+        response = self._opener.open(req).read().decode('utf-8')
+        return json.loads(response)
+
+    def list_releases(self):
+        return self._call(self._API_URL)
+
+    def create_release(self, tag_name, name=None, body='', draft=False, prerelease=False):
+        data = {
+            'tag_name': tag_name,
+            'target_commitish': 'master',
+            'name': name,
+            'body': body,
+            'draft': draft,
+            'prerelease': prerelease,
+        }
+        req = sanitized_Request(self._API_URL, json.dumps(data).encode('utf-8'))
+        return self._call(req)
+
+    def create_asset(self, release_id, asset):
+        asset_name = os.path.basename(asset)
+        url = self._UPLOADS_URL % (release_id, asset_name)
+        # Our files are small enough to be loaded directly into memory.
+        data = open(asset, 'rb').read()
+        req = sanitized_Request(url, data)
+        mime_type, _ = mimetypes.guess_type(asset_name)
+        req.add_header('Content-Type', mime_type or 'application/octet-stream')
+        return self._call(req)
+
+
+def main():
+    parser = optparse.OptionParser(usage='%prog VERSION BUILDPATH')
+    options, args = parser.parse_args()
+    if len(args) != 2:
+        parser.error('Expected a version and a build directory')
+
+    version, build_path = args
+
+    releaser = GitHubReleaser()
+
+    new_release = releaser.create_release(version, name='youtube-dl %s' % version)
+    release_id = new_release['id']
+
+    for asset in os.listdir(build_path):
+        compat_print('Uploading %s...' % asset)
+        releaser.create_asset(release_id, os.path.join(build_path, asset))
+
+
+if __name__ == '__main__':
+    main()

devscripts/gh-pages/generate-download.py

@@ -15,13 +15,9 @@ data = urllib.request.urlopen(URL).read()
 with open('download.html.in', 'r', encoding='utf-8') as tmplf:
     template = tmplf.read()
 
-md5sum = hashlib.md5(data).hexdigest()
-sha1sum = hashlib.sha1(data).hexdigest()
 sha256sum = hashlib.sha256(data).hexdigest()
 template = template.replace('@PROGRAM_VERSION@', version)
 template = template.replace('@PROGRAM_URL@', URL)
-template = template.replace('@PROGRAM_MD5SUM@', md5sum)
-template = template.replace('@PROGRAM_SHA1SUM@', sha1sum)
 template = template.replace('@PROGRAM_SHA256SUM@', sha256sum)
 template = template.replace('@EXE_URL@', versions_info['versions'][version]['exe'][0])
 template = template.replace('@EXE_SHA256SUM@', versions_info['versions'][version]['exe'][1])

devscripts/make_lazy_extractors.py

@@ -14,15 +14,17 @@ if os.path.exists(lazy_extractors_filename):
     os.remove(lazy_extractors_filename)
 
 from youtube_dl.extractor import _ALL_CLASSES
-from youtube_dl.extractor.common import InfoExtractor
+from youtube_dl.extractor.common import InfoExtractor, SearchInfoExtractor
 
 with open('devscripts/lazy_load_template.py', 'rt') as f:
     module_template = f.read()
 
-module_contents = [module_template + '\n' + getsource(InfoExtractor.suitable)]
+module_contents = [
+    module_template + '\n' + getsource(InfoExtractor.suitable) + '\n',
+    'class LazyLoadSearchExtractor(LazyLoadExtractor):\n    pass\n']
 
 ie_template = '''
-class {name}(LazyLoadExtractor):
+class {name}({bases}):
     _VALID_URL = {valid_url!r}
     _module = '{module}'
 '''
@@ -34,10 +36,20 @@ make_valid_template = '''
 '''
 
 
+def get_base_name(base):
+    if base is InfoExtractor:
+        return 'LazyLoadExtractor'
+    elif base is SearchInfoExtractor:
+        return 'LazyLoadSearchExtractor'
+    else:
+        return base.__name__
+
+
 def build_lazy_ie(ie, name):
     valid_url = getattr(ie, '_VALID_URL', None)
     s = ie_template.format(
         name=name,
+        bases=', '.join(map(get_base_name, ie.__bases__)),
         valid_url=valid_url,
         module=ie.__module__)
     if ie.suitable.__func__ is not InfoExtractor.suitable.__func__:
@@ -47,12 +59,35 @@ def build_lazy_ie(ie, name):
         s += make_valid_template.format(valid_url=ie._make_valid_url())
     return s
 
+# find the correct sorting and add the required base classes so that sublcasses
+# can be correctly created
+classes = _ALL_CLASSES[:-1]
+ordered_cls = []
+while classes:
+    for c in classes[:]:
+        bases = set(c.__bases__) - set((object, InfoExtractor, SearchInfoExtractor))
+        stop = False
+        for b in bases:
+            if b not in classes and b not in ordered_cls:
+                if b.__name__ == 'GenericIE':
+                    exit()
+                classes.insert(0, b)
+                stop = True
+        if stop:
+            break
+        if all(b in ordered_cls for b in bases):
+            ordered_cls.append(c)
+            classes.remove(c)
+            break
+ordered_cls.append(_ALL_CLASSES[-1])
+
 names = []
-for ie in list(sorted(_ALL_CLASSES[:-1], key=lambda cls: cls.ie_key())) + _ALL_CLASSES[-1:]:
-    name = ie.ie_key() + 'IE'
+for ie in ordered_cls:
+    name = ie.__name__
     src = build_lazy_ie(ie, name)
     module_contents.append(src)
-    names.append(name)
+    if ie in _ALL_CLASSES:
+        names.append(name)
 
 module_contents.append(
     '_ALL_CLASSES = [{0}]'.format(', '.join(names)))

devscripts/prepare_manpage.py

@@ -1,13 +1,46 @@
 from __future__ import unicode_literals
 
 import io
+import optparse
 import os.path
-import sys
 import re
 
 ROOT_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 README_FILE = os.path.join(ROOT_DIR, 'README.md')
 
+PREFIX = '''%YOUTUBE-DL(1)
+
+# NAME
+
+youtube\-dl \- download videos from youtube.com or other video platforms
+
+# SYNOPSIS
+
+**youtube-dl** \[OPTIONS\] URL [URL...]
+
+'''
+
+
+def main():
+    parser = optparse.OptionParser(usage='%prog OUTFILE.md')
+    options, args = parser.parse_args()
+    if len(args) != 1:
+        parser.error('Expected an output filename')
+
+    outfile, = args
+
+    with io.open(README_FILE, encoding='utf-8') as f:
+        readme = f.read()
+
+    readme = re.sub(r'(?s)^.*?(?=# DESCRIPTION)', '', readme)
+    readme = re.sub(r'\s+youtube-dl \[OPTIONS\] URL \[URL\.\.\.\]', '', readme)
+    readme = PREFIX + readme
+
+    readme = filter_options(readme)
+
+    with io.open(outfile, 'w', encoding='utf-8') as outf:
+        outf.write(readme)
+
 
 def filter_options(readme):
     ret = ''
@@ -37,27 +70,5 @@ def filter_options(readme):
 
     return ret
 
-with io.open(README_FILE, encoding='utf-8') as f:
-    readme = f.read()
-
-PREFIX = '''%YOUTUBE-DL(1)
-
-# NAME
-
-youtube\-dl \- download videos from youtube.com or other video platforms
-
-# SYNOPSIS
-
-**youtube-dl** \[OPTIONS\] URL [URL...]
-
-'''
-readme = re.sub(r'(?s)^.*?(?=# DESCRIPTION)', '', readme)
-readme = re.sub(r'\s+youtube-dl \[OPTIONS\] URL \[URL\.\.\.\]', '', readme)
-readme = PREFIX + readme
-
-readme = filter_options(readme)
-
-if sys.version_info < (3, 0):
-    print(readme.encode('utf-8'))
-else:
-    print(readme)
+if __name__ == '__main__':
+    main()

devscripts/release.sh

@@ -6,7 +6,7 @@
 # * the git config user.signingkey is properly set
 
 # You will need
-# pip install coverage nose rsa
+# pip install coverage nose rsa wheel
 
 # TODO
 # release notes
@@ -15,10 +15,33 @@
 set -e
 
 skip_tests=true
-if [ "$1" = '--run-tests' ]; then
-    skip_tests=false
-    shift
-fi
+gpg_sign_commits=""
+buildserver='localhost:8142'
+
+while true
+do
+case "$1" in
+    --run-tests)
+        skip_tests=false
+        shift
+    ;;
+    --gpg-sign-commits|-S)
+        gpg_sign_commits="-S"
+        shift
+    ;;
+    --buildserver)
+        buildserver="$2"
+        shift 2
+    ;;
+    --*)
+        echo "ERROR: unknown option $1"
+        exit 1
+    ;;
+    *)
+        break
+    ;;
+esac
+done
 
 if [ -z "$1" ]; then echo "ERROR: specify version number like this: $0 1994.09.06"; exit 1; fi
 version="$1"
@@ -33,6 +56,9 @@ if [ ! -z "`git status --porcelain | grep -v CHANGELOG`" ]; then echo 'ERROR: th
 useless_files=$(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
+if ! type pandoc >/dev/null 2>/dev/null; then echo 'ERROR: pandoc is missing'; exit 1; fi
+if ! python3 -c 'import rsa' 2>/dev/null; then echo 'ERROR: python3-rsa is missing'; exit 1; fi
+if ! python3 -c 'import wheel' 2>/dev/null; then echo 'ERROR: wheel is missing'; exit 1; fi
 
 /bin/echo -e "\n### First of all, testing..."
 make clean
@@ -45,10 +71,13 @@ fi
 /bin/echo -e "\n### Changing version in version.py..."
 sed -i "s/__version__ = '.*'/__version__ = '$version'/" youtube_dl/version.py
 
+/bin/echo -e "\n### Changing version in ChangeLog..."
+sed -i "s/<unreleased>/$version/" ChangeLog
+
 /bin/echo -e "\n### Committing documentation, templates and youtube_dl/version.py..."
 make README.md CONTRIBUTING.md .github/ISSUE_TEMPLATE.md supportedsites
-git add README.md CONTRIBUTING.md .github/ISSUE_TEMPLATE.md docs/supportedsites.md youtube_dl/version.py
-git commit -m "release $version"
+git add README.md CONTRIBUTING.md .github/ISSUE_TEMPLATE.md docs/supportedsites.md youtube_dl/version.py ChangeLog
+git commit $gpg_sign_commits -m "release $version"
 
 /bin/echo -e "\n### Now tagging, signing and pushing..."
 git tag -s -m "Release $version" "$version"
@@ -64,7 +93,7 @@ git push origin "$version"
 REV=$(git rev-parse HEAD)
 make youtube-dl youtube-dl.tar.gz
 read -p "VM running? (y/n) " -n 1
-wget "http://localhost:8142/build/rg3/youtube-dl/youtube-dl.exe?rev=$REV" -O youtube-dl.exe
+wget "http://$buildserver/build/rg3/youtube-dl/youtube-dl.exe?rev=$REV" -O youtube-dl.exe
 mkdir -p "build/$version"
 mv youtube-dl youtube-dl.exe "build/$version"
 mv youtube-dl.tar.gz "build/$version/youtube-dl-$version.tar.gz"
@@ -74,15 +103,16 @@ RELEASE_FILES="youtube-dl youtube-dl.exe youtube-dl-$version.tar.gz"
 (cd build/$version/ && sha256sum $RELEASE_FILES > SHA2-256SUMS)
 (cd build/$version/ && sha512sum $RELEASE_FILES > SHA2-512SUMS)
 
-/bin/echo -e "\n### Signing and uploading the new binaries to yt-dl.org ..."
+/bin/echo -e "\n### Signing and uploading the new binaries to GitHub..."
 for f in $RELEASE_FILES; do gpg --passphrase-repeat 5 --detach-sig "build/$version/$f"; done
-scp -r "build/$version" ytdl@yt-dl.org:html/tmp/
-ssh ytdl@yt-dl.org "mv html/tmp/$version html/downloads/"
+
+ROOT=$(pwd)
+python devscripts/create-github-release.py $version "$ROOT/build/$version"
+
 ssh ytdl@yt-dl.org "sh html/update_latest.sh $version"
 
 /bin/echo -e "\n### Now switching to gh-pages..."
 git clone --branch gh-pages --single-branch . build/gh-pages
-ROOT=$(pwd)
 (
     set -e
     ORIGIN_URL=$(git config --get remote.origin.url)
@@ -94,7 +124,7 @@ ROOT=$(pwd)
     "$ROOT/devscripts/gh-pages/update-copyright.py"
     "$ROOT/devscripts/gh-pages/update-sites.py"
     git add *.html *.html.in update
-    git commit -m "release $version"
+    git commit $gpg_sign_commits -m "release $version"
     git push "$ROOT" gh-pages
     git push "$ORIGIN_URL" gh-pages
 )

devscripts/show-downloads-statistics.py

@@ -0,0 +1,47 @@
+#!/usr/bin/env python
+from __future__ import unicode_literals
+
+import itertools
+import json
+import os
+import re
+import sys
+
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl.compat import (
+    compat_print,
+    compat_urllib_request,
+)
+from youtube_dl.utils import format_bytes
+
+
+def format_size(bytes):
+    return '%s (%d bytes)' % (format_bytes(bytes), bytes)
+
+
+total_bytes = 0
+
+for page in itertools.count(1):
+    releases = json.loads(compat_urllib_request.urlopen(
+        'https://api.github.com/repos/rg3/youtube-dl/releases?page=%s' % page
+    ).read().decode('utf-8'))
+
+    if not releases:
+        break
+
+    for release in releases:
+        compat_print(release['name'])
+        for asset in release['assets']:
+            asset_name = asset['name']
+            total_bytes += asset['download_count'] * asset['size']
+            if all(not re.match(p, asset_name) for p in (
+                    r'^youtube-dl$',
+                    r'^youtube-dl-\d{4}\.\d{2}\.\d{2}(?:\.\d+)?\.tar\.gz$',
+                    r'^youtube-dl\.exe$')):
+                continue
+            compat_print(
+                ' %s size: %s downloads: %d'
+                % (asset_name, format_size(asset['size']), asset['download_count']))
+
+compat_print('total downloads traffic: %s' % format_size(total_bytes))

docs/supportedsites.md

@@ -6,6 +6,7 @@
  - **22tracks:genre**
  - **22tracks:track**
  - **24video**
+ - **3qsdn**: 3Q SDN
  - **3sat**
  - **4tube**
  - **56.com**
@@ -13,8 +14,11 @@
  - **8tracks**
  - **91porn**
  - **9gag**
+ - **9now.com.au**
  - **abc.net.au**
  - **Abc7News**
+ - **abcnews**
+ - **abcnews:video**
  - **AcademicEarth:Course**
  - **acast**
  - **acast:channel**
@@ -25,6 +29,7 @@
  - **AdobeTVVideo**
  - **AdultSwim**
  - **aenetworks**: A+E Networks: A&E, Lifetime, History.com, FYI Network
+ - **AfreecaTV**: afreecatv.com
  - **Aftonbladet**
  - **AirMozilla**
  - **AlJazeera**
@@ -40,8 +45,8 @@
  - **appletrailers:section**
  - **archive.org**: archive.org videos
  - **ARD**
- - **ARD:mediathek**: Saarländischer Rundfunk
  - **ARD:mediathek**
+ - **Arkena**
  - **arte.tv**
  - **arte.tv:+7**
  - **arte.tv:cinema**
@@ -52,6 +57,7 @@
  - **arte.tv:future**
  - **arte.tv:info**
  - **arte.tv:magazine**
+ - **arte.tv:playlist**
  - **AtresPlayer**
  - **ATTTechChannel**
  - **AudiMedia**
@@ -69,6 +75,8 @@
  - **bbc**: BBC
  - **bbc.co.uk**: BBC iPlayer
  - **bbc.co.uk:article**: BBC articles
+ - **bbc.co.uk:iplayer:playlist**
+ - **bbc.co.uk:playlist**
  - **BeatportPro**
  - **Beeg**
  - **BehindKink**
@@ -77,6 +85,7 @@
  - **Bild**: Bild.de
  - **BiliBili**
  - **BioBioChileTV**
+ - **BIQLE**
  - **BleacherReport**
  - **BleacherReportCMS**
  - **blinkx**
@@ -98,10 +107,13 @@
  - **canalc2.tv**
  - **Canalplus**: canalplus.fr, piwiplus.fr and d8.tv
  - **Canvas**
+ - **CarambaTV**
+ - **CarambaTVPage**
  - **CBC**
  - **CBCPlayer**
  - **CBS**
  - **CBSInteractive**
+ - **CBSLocal**
  - **CBSNews**: CBS News
  - **CBSNewsLiveVideo**: CBS News Live Videos
  - **CBSSports**
@@ -113,11 +125,11 @@
  - **chirbit**
  - **chirbit:profile**
  - **Cinchcast**
- - **Cinemassacre**
  - **Clipfish**
  - **cliphunter**
  - **ClipRs**
  - **Clipsyndicate**
+ - **CloserToTruth**
  - **cloudtime**: CloudTime
  - **Cloudy**
  - **Clubic**
@@ -127,12 +139,13 @@
  - **CNN**
  - **CNNArticle**
  - **CNNBlogs**
- - **CollegeHumor**
  - **CollegeRama**
  - **ComCarCoff**
  - **ComedyCentral**
- - **ComedyCentralShows**: The Daily Show / The Colbert Report
+ - **ComedyCentralShortname**
+ - **ComedyCentralTV**
  - **CondeNast**: Condé Nast media group: Allure, Architectural Digest, Ars Technica, Bon Appétit, Brides, Condé Nast, Condé Nast Traveler, Details, Epicurious, GQ, Glamour, Golf Digest, SELF, Teen Vogue, The New Yorker, Vanity Fair, Vogue, W Magazine, WIRED
+ - **Coub**
  - **Cracked**
  - **Crackle**
  - **Criterion**
@@ -142,9 +155,12 @@
  - **CSNNE**
  - **CSpan**: C-SPAN
  - **CtsNews**: 華視新聞
+ - **CTV**
+ - **CTVNews**
  - **culturebox.francetvinfo.fr**
  - **CultureUnplugged**
  - **CWTV**
+ - **DailyMail**
  - **dailymotion**
  - **dailymotion:playlist**
  - **dailymotion:user**
@@ -166,6 +182,7 @@
  - **DigitallySpeaking**
  - **Digiteka**
  - **Discovery**
+ - **DiscoveryGo**
  - **Dotsub**
  - **DouyuTV**: 斗鱼
  - **DPlay**
@@ -201,6 +218,7 @@
  - **exfm**: ex.fm
  - **ExpoTV**
  - **ExtremeTube**
+ - **EyedoTV**
  - **facebook**
  - **faz.net**
  - **fc2**
@@ -210,8 +228,10 @@
  - **Firstpost**
  - **FiveTV**
  - **Flickr**
+ - **Flipagram**
  - **Folketinget**: Folketinget (ft.dk; Danish parliament)
  - **FootyRoom**
+ - **Formula1**
  - **FOX**
  - **Foxgay**
  - **FoxNews**: Fox News and Fox Business Video
@@ -227,14 +247,13 @@
  - **FreeVideo**
  - **Funimation**
  - **FunnyOrDie**
+ - **Fusion**
  - **GameInformer**
- - **Gamekings**
  - **GameOne**
  - **gameone:playlist**
  - **Gamersyde**
  - **GameSpot**
  - **GameStar**
- - **Gametrailers**
  - **Gazeta**
  - **GDCVault**
  - **generic**: Generic downloader that works on some sites
@@ -245,6 +264,7 @@
  - **Globo**
  - **GloboArticle**
  - **GodTube**
+ - **GodTV**
  - **GoldenMoustache**
  - **Golem**
  - **GoogleDrive**
@@ -259,6 +279,7 @@
  - **Helsinki**: helsinki.fi
  - **HentaiStigma**
  - **HistoricFilms**
+ - **history:topic**: History.com Topic
  - **hitbox**
  - **hitbox:live**
  - **HornBunny**
@@ -266,6 +287,8 @@
  - **HotStar**
  - **Howcast**
  - **HowStuffWorks**
+ - **HRTi**
+ - **HRTiPlaylist**
  - **HuffPost**: Huffington Post
  - **Hypem**
  - **Iconosquare**
@@ -293,6 +316,7 @@
  - **jpopsuki.tv**
  - **JWPlatform**
  - **Kaltura**
+ - **Kamcord**
  - **KanalPlay**: Kanal 5/9/11 Play
  - **Kankan**
  - **Karaoketv**
@@ -312,23 +336,29 @@
  - **kuwo:mv**: 酷我音乐 - MV
  - **kuwo:singer**: 酷我音乐 - 歌手
  - **kuwo:song**: 酷我音乐
- - **la7.tv**
+ - **la7.it**
  - **Laola1Tv**
+ - **Lcp**
+ - **LcpPlay**
  - **Le**: 乐视网
+ - **Learnr**
  - **Lecture2Go**
  - **Lemonde**
  - **LePlaylist**
  - **LetvCloud**: 乐视云
  - **Libsyn**
+ - **life**: Life.ru
  - **life:embed**
- - **lifenews**: LIFE | NEWS
  - **limelight**
  - **limelight:channel**
  - **limelight:channel_list**
+ - **LiTV**
  - **LiveLeak**
  - **livestream**
  - **livestream:original**
  - **LnkGo**
+ - **loc**: Library of Congress
+ - **LocalNews8**
  - **LoveHomePorn**
  - **lrt.lt**
  - **lynda**: lynda.com videos
@@ -338,10 +368,10 @@
  - **mailru**: Видео@Mail.Ru
  - **MakersChannel**
  - **MakerTV**
- - **Malemotion**
  - **MatchTV**
  - **MDR**: MDR.DE and KiKA
  - **media.ccc.de**
+ - **META**
  - **metacafe**
  - **Metacritic**
  - **Mgoon**
@@ -368,15 +398,16 @@
  - **MovieFap**
  - **Moviezine**
  - **MPORA**
- - **MSNBC**
+ - **MSN**
  - **MTV**
  - **mtv.de**
- - **mtviggy.com**
  - **mtvservices:embedded**
  - **MuenchenTV**: münchen.tv
  - **MusicPlayOn**
- - **muzu.tv**
+ - **mva**: Microsoft Virtual Academy videos
+ - **mva:course**: Microsoft Virtual Academy courses
  - **Mwave**
+ - **MwaveMeetGreet**
  - **MySpace**
  - **MySpace:album**
  - **MySpass**
@@ -385,7 +416,8 @@
  - **MyVidster**
  - **n-tv.de**
  - **natgeo**
- - **natgeo:channel**
+ - **natgeo:episodeguide**
+ - **natgeo:video**
  - **Naver**
  - **NBA**
  - **NBC**
@@ -409,7 +441,6 @@
  - **Newstube**
  - **NextMedia**: 蘋果日報
  - **NextMediaActionNews**: 蘋果日報 - 動新聞
- - **nextmovie.com**
  - **nfb**: National Film Board of Canada
  - **nfl.com**
  - **nhl.com**
@@ -417,8 +448,11 @@
  - **nhl.com:videocenter**
  - **nhl.com:videocenter:category**: NHL videocenter category
  - **nick.com**
+ - **nick.de**
  - **niconico**: ニコニコ動画
  - **NiconicoPlaylist**
+ - **NineCNineMedia**
+ - **Nintendo**
  - **njoy**: N-JOY
  - **njoy:embed**
  - **Noco**
@@ -446,9 +480,12 @@
  - **NYTimes**
  - **NYTimesArticle**
  - **ocw.mit.edu**
+ - **OdaTV**
  - **Odnoklassniki**
  - **OktoberfestTV**
  - **on.aol.com**
+ - **onet.tv**
+ - **onet.tv:channel**
  - **OnionStudios**
  - **Ooyala**
  - **OoyalaExternal**
@@ -464,7 +501,8 @@
  - **pbs**: Public Broadcasting Service (PBS) and member stations: PBS: Public Broadcasting Service, APT - Alabama Public Television (WBIQ), GPB/Georgia Public Broadcasting (WGTV), Mississippi Public Broadcasting (WMPN), Nashville Public Television (WNPT), WFSU-TV (WFSU), WSRE (WSRE), WTCI (WTCI), WPBA/Channel 30 (WPBA), Alaska Public Media (KAKM), Arizona PBS (KAET), KNME-TV/Channel 5 (KNME), Vegas PBS (KLVX), AETN/ARKANSAS ETV NETWORK (KETS), KET (WKLE), WKNO/Channel 10 (WKNO), LPB/LOUISIANA PUBLIC BROADCASTING (WLPB), OETA (KETA), Ozarks Public Television (KOZK), WSIU Public Broadcasting (WSIU), KEET TV (KEET), KIXE/Channel 9 (KIXE), KPBS San Diego (KPBS), KQED (KQED), KVIE Public Television (KVIE), PBS SoCal/KOCE (KOCE), ValleyPBS (KVPT), CONNECTICUT PUBLIC TELEVISION (WEDH), KNPB Channel 5 (KNPB), SOPTV (KSYS), Rocky Mountain PBS (KRMA), KENW-TV3 (KENW), KUED Channel 7 (KUED), Wyoming PBS (KCWC), Colorado Public Television / KBDI 12 (KBDI), KBYU-TV (KBYU), Thirteen/WNET New York (WNET), WGBH/Channel 2 (WGBH), WGBY (WGBY), NJTV Public Media NJ (WNJT), WLIW21 (WLIW), mpt/Maryland Public Television (WMPB), WETA Television and Radio (WETA), WHYY (WHYY), PBS 39 (WLVT), WVPT - Your Source for PBS and More! (WVPT), Howard University Television (WHUT), WEDU PBS (WEDU), WGCU Public Media (WGCU), WPBT2 (WPBT), WUCF TV (WUCF), WUFT/Channel 5 (WUFT), WXEL/Channel 42 (WXEL), WLRN/Channel 17 (WLRN), WUSF Public Broadcasting (WUSF), ETV (WRLK), UNC-TV (WUNC), PBS Hawaii - Oceanic Cable Channel 10 (KHET), Idaho Public Television (KAID), KSPS (KSPS), OPB (KOPB), KWSU/Channel 10 & KTNW/Channel 31 (KWSU), WILL-TV (WILL), Network Knowledge - WSEC/Springfield (WSEC), WTTW11 (WTTW), Iowa Public Television/IPTV (KDIN), Nine Network (KETC), PBS39 Fort Wayne (WFWA), WFYI Indianapolis (WFYI), Milwaukee Public Television (WMVS), WNIN (WNIN), WNIT Public Television (WNIT), WPT (WPNE), WVUT/Channel 22 (WVUT), WEIU/Channel 51 (WEIU), WQPT-TV (WQPT), WYCC PBS Chicago (WYCC), WIPB-TV (WIPB), WTIU (WTIU), CET  (WCET), ThinkTVNetwork (WPTD), WBGU-TV (WBGU), WGVU TV (WGVU), NET1 (KUON), Pioneer Public Television (KWCM), SDPB Television (KUSD), TPT (KTCA), KSMQ (KSMQ), KPTS/Channel 8 (KPTS), KTWU/Channel 11 (KTWU), East Tennessee PBS (WSJK), WCTE-TV (WCTE), WLJT, Channel 11 (WLJT), WOSU TV (WOSU), WOUB/WOUC (WOUB), WVPB (WVPB), WKYU-PBS (WKYU), KERA 13 (KERA), MPBN (WCBB), Mountain Lake PBS (WCFE), NHPTV (WENH), Vermont PBS (WETK), witf (WITF), WQED Multimedia (WQED), WMHT Educational Telecommunications (WMHT), Q-TV (WDCQ), WTVS Detroit Public TV (WTVS), CMU Public Television (WCMU), WKAR-TV (WKAR), WNMU-TV Public TV 13 (WNMU), WDSE - WRPT (WDSE), WGTE TV (WGTE), Lakeland Public Television (KAWE), KMOS-TV - Channels 6.1, 6.2 and 6.3 (KMOS), MontanaPBS (KUSM), KRWG/Channel 22 (KRWG), KACV (KACV), KCOS/Channel 13 (KCOS), WCNY/Channel 24 (WCNY), WNED (WNED), WPBS (WPBS), WSKG Public TV (WSKG), WXXI (WXXI), WPSU (WPSU), WVIA Public Media Studios (WVIA), WTVI (WTVI), Western Reserve PBS (WNEO), WVIZ/PBS ideastream (WVIZ), KCTS 9 (KCTS), Basin PBS (KPBT), KUHT / Channel 8 (KUHT), KLRN (KLRN), KLRU (KLRU), WTJX Channel 12 (WTJX), WCVE PBS (WCVE), KBTC Public Television (KBTC)
  - **pcmag**
  - **People**
- - **Periscope**: Periscope
+ - **periscope**: Periscope
+ - **periscope:user**: Periscope user videos
  - **PhilharmonieDeParis**: Philharmonie de Paris
  - **phoenix.de**
  - **Photobucket**
@@ -481,8 +519,10 @@
  - **plus.google**: Google Plus
  - **pluzz.francetv.fr**
  - **podomatic**
+ - **Pokemon**
+ - **PolskieRadio**
  - **PornHd**
- - **PornHub**
+ - **PornHub**: PornHub and Thumbzilla
  - **PornHubPlaylist**
  - **PornHubUserVideos**
  - **Pornotube**
@@ -500,8 +540,11 @@
  - **qqmusic:singer**: QQ音乐 - 歌手
  - **qqmusic:toplist**: QQ音乐 - 排行榜
  - **R7**
+ - **R7Article**
  - **radio.de**
  - **radiobremen**
+ - **radiocanada**
+ - **RadioCanadaAudioVideo**
  - **radiofrance**
  - **RadioJavan**
  - **Rai**
@@ -511,10 +554,14 @@
  - **RedTube**
  - **RegioTV**
  - **Restudy**
+ - **Reuters**
  - **ReverbNation**
- - **Revision3**
+ - **revision**
+ - **revision3:embed**
  - **RICE**
  - **RingTV**
+ - **RockstarGames**
+ - **RoosterTeeth**
  - **RottenTomatoes**
  - **Roxwel**
  - **RTBF**
@@ -527,7 +574,9 @@
  - **rtve.es:alacarta**: RTVE a la carta
  - **rtve.es:infantil**: RTVE infantil
  - **rtve.es:live**: RTVE.es live streams
+ - **rtve.es:television**
  - **RTVNH**
+ - **Rudo**
  - **RUHD**
  - **RulePorn**
  - **rutube**: Rutube videos
@@ -551,24 +600,25 @@
  - **ScreencastOMatic**
  - **ScreenJunkies**
  - **ScreenwaveMedia**
+ - **Seeker**
  - **SenateISVP**
+ - **SendtoNews**
  - **ServingSys**
  - **Sexu**
- - **SexyKarma**: Sexy Karma and Watch Indian Porn
  - **Shahid**
  - **Shared**: shared.sx and vivo.sx
  - **ShareSix**
  - **Sina**
+ - **SixPlay**
+ - **skynewsarabia:article**
  - **skynewsarabia:video**
- - **skynewsarabia:video**
+ - **SkySports**
  - **Slideshare**
  - **Slutload**
  - **smotri**: Smotri.com
  - **smotri:broadcast**: Smotri.com broadcasts
  - **smotri:community**: Smotri.com community videos
  - **smotri:user**: Smotri.com user videos
- - **SnagFilms**
- - **SnagFilmsEmbed**
  - **Snotr**
  - **Sohu**
  - **soundcloud**
@@ -594,12 +644,14 @@
  - **SportBoxEmbed**
  - **SportDeutschland**
  - **Sportschau**
+ - **sr:mediathek**: Saarländischer Rundfunk
  - **SRGSSR**
  - **SRGSSRPlay**: srf.ch, rts.ch, rsi.ch, rtr.ch and swissinfo.ch play sites
  - **SSA**
  - **stanfordoc**: Stanford Open ClassRoom
  - **Steam**
  - **Stitcher**
+ - **Streamable**
  - **streamcloud.eu**
  - **StreamCZ**
  - **StreetVoice**
@@ -610,6 +662,7 @@
  - **Syfy**
  - **SztvHu**
  - **Tagesschau**
+ - **tagesschau:player**
  - **Tapely**
  - **Tass**
  - **TDSLifeway**
@@ -627,6 +680,7 @@
  - **Telegraaf**
  - **TeleMB**
  - **TeleTask**
+ - **Telewebion**
  - **TF1**
  - **TheIntercept**
  - **ThePlatform**
@@ -645,6 +699,7 @@
  - **TNAFlix**
  - **TNAFlixNetworkEmbed**
  - **toggle**
+ - **Tosh**: Tosh.0
  - **tou.tv**
  - **Toypics**: Toypics user profile
  - **ToypicsUser**: Toypics user profile
@@ -673,12 +728,13 @@
  - **TVCArticle**
  - **tvigle**: Интернет-телевидение Tvigle.ru
  - **tvland.com**
- - **tvp.pl**
- - **tvp.pl:Series**
+ - **tvp**: Telewizja Polska
+ - **tvp:embed**: Telewizja Polska
+ - **tvp:series**
  - **TVPlay**: TV3Play and related services
  - **Tweakers**
- - **twitch:bookmarks**
  - **twitch:chapter**
+ - **twitch:clips**
  - **twitch:past_broadcasts**
  - **twitch:profile**
  - **twitch:stream**
@@ -692,10 +748,12 @@
  - **UDNEmbed**: 聯合影音
  - **Unistra**
  - **Urort**: NRK P3 Urørt
+ - **URPlay**
  - **USAToday**
  - **ustream**
  - **ustream:channel**
- - **Ustudio**
+ - **ustudio**
+ - **ustudio:embed**
  - **Varzesh3**
  - **Vbox7**
  - **VeeHD**
@@ -703,10 +761,12 @@
  - **Vessel**
  - **Vesti**: Вести.Ru
  - **Vevo**
+ - **VevoPlaylist**
  - **VGTV**: VGTV, BTTV, FTV, Aftenposten and Aftonbladet
  - **vh1.com**
  - **Vice**
  - **ViceShow**
+ - **Vidbit**
  - **Viddler**
  - **video.google:search**: Google Video search
  - **video.mit.edu**
@@ -719,12 +779,15 @@
  - **VideoPremium**
  - **VideoTt**: video.tt - Your True Tube (Currently broken)
  - **videoweed**: VideoWeed
+ - **Vidio**
  - **vidme**
  - **vidme:user**
  - **vidme:user:likes**
  - **Vidzi**
  - **vier**
  - **vier:videos**
+ - **ViewLift**
+ - **ViewLiftEmbed**
  - **Viewster**
  - **Viidea**
  - **viki**
@@ -743,8 +806,10 @@
  - **vine:user**
  - **vk**: VK
  - **vk:uservideos**: VK - User's Videos
+ - **vk:wallpost**
  - **vlive**
  - **Vodlocker**
+ - **VODPlatform**
  - **VoiceRepublic**
  - **VoxMedia**
  - **Vporn**
@@ -752,16 +817,15 @@
  - **VRT**
  - **vube**: Vube.com
  - **VuClip**
- - **vulture.com**
  - **Walla**
- - **WashingtonPost**
+ - **washingtonpost**
+ - **washingtonpost:article**
  - **wat.tv**
+ - **WatchIndianPorn**: Watch Indian Porn
  - **WDR**
  - **wdr:mobile**
- - **WDRMaus**: Sendung mit der Maus
  - **WebOfStories**
  - **WebOfStoriesPlaylist**
- - **Weibo**
  - **WeiqiTV**: WQTV
  - **wholecloud**: WholeCloud
  - **Wimp**
@@ -769,12 +833,17 @@
  - **WNL**
  - **WorldStarHipHop**
  - **wrzuta.pl**
+ - **wrzuta.pl:playlist**
  - **WSJ**: Wall Street Journal
  - **XBef**
  - **XboxClips**
- - **XFileShare**: XFileShare based sites: GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net, filehoot.com and vidto.me
+ - **XFileShare**: XFileShare based sites: DaClips, FileHoot, GorillaVid, MovPod, PowerWatch, Rapidvideo.ws, TheVideoBee, Vidto, Streamin.To, XVIDSTAGE
  - **XHamster**
  - **XHamsterEmbed**
+ - **xiami:album**: 虾米音乐 - 专辑
+ - **xiami:artist**: 虾米音乐 - 歌手
+ - **xiami:collection**: 虾米音乐 - 精选集
+ - **xiami:song**: 虾米音乐
  - **XMinus**
  - **XNXX**
  - **Xstream**
@@ -793,6 +862,7 @@
  - **Ynet**
  - **YouJizz**
  - **youku**: 优酷
+ - **youku:show**
  - **YouPorn**
  - **YourUpload**
  - **youtube**: YouTube.com
@@ -806,6 +876,7 @@
  - **youtube:search**: YouTube.com searches
  - **youtube:search:date**: YouTube.com searches, newest videos first
  - **youtube:search_url**: YouTube.com search URLs
+ - **youtube:shared**
  - **youtube:show**: YouTube.com (multi-season) shows
  - **youtube:subscriptions**: YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)
  - **youtube:user**: YouTube.com user videos (URL or "ytuser" keyword)

setup.py

@@ -21,25 +21,37 @@ try:
     import py2exe
 except ImportError:
     if len(sys.argv) >= 2 and sys.argv[1] == 'py2exe':
-        print("Cannot import py2exe", file=sys.stderr)
+        print('Cannot import py2exe', file=sys.stderr)
         exit(1)
 
 py2exe_options = {
-    "bundle_files": 1,
-    "compressed": 1,
-    "optimize": 2,
-    "dist_dir": '.',
-    "dll_excludes": ['w9xpopen.exe', 'crypt32.dll'],
+    'bundle_files': 1,
+    'compressed': 1,
+    'optimize': 2,
+    'dist_dir': '.',
+    'dll_excludes': ['w9xpopen.exe', 'crypt32.dll'],
 }
 
+# Get the version from youtube_dl/version.py without importing the package
+exec(compile(open('youtube_dl/version.py').read(),
+             'youtube_dl/version.py', 'exec'))
+
+DESCRIPTION = 'YouTube video downloader'
+LONG_DESCRIPTION = 'Command-line program to download videos from YouTube.com and other video sites'
+
 py2exe_console = [{
-    "script": "./youtube_dl/__main__.py",
-    "dest_base": "youtube-dl",
+    'script': './youtube_dl/__main__.py',
+    'dest_base': 'youtube-dl',
+    'version': __version__,
+    'description': DESCRIPTION,
+    'comments': LONG_DESCRIPTION,
+    'product_name': 'youtube-dl',
+    'product_version': __version__,
 }]
 
 py2exe_params = {
     'console': py2exe_console,
-    'options': {"py2exe": py2exe_options},
+    'options': {'py2exe': py2exe_options},
     'zipfile': None
 }
 
@@ -72,7 +84,7 @@ else:
         params['scripts'] = ['bin/youtube-dl']
 
 class build_lazy_extractors(Command):
-    description = "Build the extractor lazy loading module"
+    description = 'Build the extractor lazy loading module'
     user_options = []
 
     def initialize_options(self):
@@ -87,16 +99,11 @@ class build_lazy_extractors(Command):
             dry_run=self.dry_run,
         )
 
-# Get the version from youtube_dl/version.py without importing the package
-exec(compile(open('youtube_dl/version.py').read(),
-             'youtube_dl/version.py', 'exec'))
-
 setup(
     name='youtube_dl',
     version=__version__,
-    description='YouTube video downloader',
-    long_description='Small command-line program to download videos from'
-    ' YouTube.com and other video sites.',
+    description=DESCRIPTION,
+    long_description=LONG_DESCRIPTION,
     url='https://github.com/rg3/youtube-dl',
     author='Ricardo Garcia',
     author_email='ytdl@yt-dl.org',
@@ -112,16 +119,17 @@ setup(
     # test_requires = ['nosetest'],
 
     classifiers=[
-        "Topic :: Multimedia :: Video",
-        "Development Status :: 5 - Production/Stable",
-        "Environment :: Console",
-        "License :: Public Domain",
-        "Programming Language :: Python :: 2.6",
-        "Programming Language :: Python :: 2.7",
-        "Programming Language :: Python :: 3",
-        "Programming Language :: Python :: 3.2",
-        "Programming Language :: Python :: 3.3",
-        "Programming Language :: Python :: 3.4",
+        'Topic :: Multimedia :: Video',
+        'Development Status :: 5 - Production/Stable',
+        'Environment :: Console',
+        'License :: Public Domain',
+        'Programming Language :: Python :: 2.6',
+        'Programming Language :: Python :: 2.7',
+        'Programming Language :: Python :: 3',
+        'Programming Language :: Python :: 3.2',
+        'Programming Language :: Python :: 3.3',
+        'Programming Language :: Python :: 3.4',
+        'Programming Language :: Python :: 3.5',
     ],
 
     cmdclass={'build_lazy_extractors': build_lazy_extractors},

test/helper.py

@@ -24,8 +24,13 @@ from youtube_dl.utils import (
 def get_params(override=None):
     PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)),
                                    "parameters.json")
+    LOCAL_PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)),
+                                         "local_parameters.json")
     with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
         parameters = json.load(pf)
+    if os.path.exists(LOCAL_PARAMETERS_FILE):
+        with io.open(LOCAL_PARAMETERS_FILE, encoding='utf-8') as pf:
+            parameters.update(json.load(pf))
     if override:
         parameters.update(override)
     return parameters

test/test_InfoExtractor.py

@@ -11,7 +11,7 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 from test.helper import FakeYDL
 from youtube_dl.extractor.common import InfoExtractor
 from youtube_dl.extractor import YoutubeIE, get_info_extractor
-from youtube_dl.utils import encode_data_uri, strip_jsonp, ExtractorError
+from youtube_dl.utils import encode_data_uri, strip_jsonp, ExtractorError, RegexNotFoundError
 
 
 class TestIE(InfoExtractor):
@@ -48,6 +48,9 @@ class TestInfoExtractor(unittest.TestCase):
         self.assertEqual(ie._og_search_property('foobar', html), 'Foo')
         self.assertEqual(ie._og_search_property('test1', html), 'foo > < bar')
         self.assertEqual(ie._og_search_property('test2', html), 'foo >//< bar')
+        self.assertEqual(ie._og_search_property(('test0', 'test1'), html), 'foo > < bar')
+        self.assertRaises(RegexNotFoundError, ie._og_search_property, 'test0', html, None, fatal=True)
+        self.assertRaises(RegexNotFoundError, ie._og_search_property, ('test0', 'test00'), html, None, fatal=True)
 
     def test_html_search_meta(self):
         ie = self.ie
@@ -66,6 +69,11 @@ class TestInfoExtractor(unittest.TestCase):
         self.assertEqual(ie._html_search_meta('d', html), '4')
         self.assertEqual(ie._html_search_meta('e', html), '5')
         self.assertEqual(ie._html_search_meta('f', html), '6')
+        self.assertEqual(ie._html_search_meta(('a', 'b', 'c'), html), '1')
+        self.assertEqual(ie._html_search_meta(('c', 'b', 'a'), html), '3')
+        self.assertEqual(ie._html_search_meta(('z', 'x', 'c'), html), '3')
+        self.assertRaises(RegexNotFoundError, ie._html_search_meta, 'z', html, None, fatal=True)
+        self.assertRaises(RegexNotFoundError, ie._html_search_meta, ('z', 'x'), html, None, fatal=True)
 
     def test_download_json(self):
         uri = encode_data_uri(b'{"foo": "blah"}', 'application/json')

test/test_YoutubeDL.py

@@ -335,6 +335,40 @@ class TestFormatSelection(unittest.TestCase):
             downloaded = ydl.downloaded_info_dicts[0]
             self.assertEqual(downloaded['format_id'], f1['format_id'])
 
+    def test_audio_only_extractor_format_selection(self):
+        # For extractors with incomplete formats (all formats are audio-only or
+        # video-only) best and worst should fallback to corresponding best/worst
+        # video-only or audio-only formats (as per
+        # https://github.com/rg3/youtube-dl/pull/5556)
+        formats = [
+            {'format_id': 'low', 'ext': 'mp3', 'preference': 1, 'vcodec': 'none', 'url': TEST_URL},
+            {'format_id': 'high', 'ext': 'mp3', 'preference': 2, 'vcodec': 'none', 'url': TEST_URL},
+        ]
+        info_dict = _make_result(formats)
+
+        ydl = YDL({'format': 'best'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'high')
+
+        ydl = YDL({'format': 'worst'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'low')
+
+    def test_format_not_available(self):
+        formats = [
+            {'format_id': 'regular', 'ext': 'mp4', 'height': 360, 'url': TEST_URL},
+            {'format_id': 'video', 'ext': 'mp4', 'height': 720, 'acodec': 'none', 'url': TEST_URL},
+        ]
+        info_dict = _make_result(formats)
+
+        # This must fail since complete video-audio format does not match filter
+        # and extractor does not provide incomplete only formats (i.e. only
+        # video-only or audio-only).
+        ydl = YDL({'format': 'best[height>360]'})
+        self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
+
     def test_invalid_format_specs(self):
         def assert_syntax_error(format_spec):
             ydl = YDL({'format': format_spec})

test/test_all_urls.py

@@ -6,6 +6,7 @@ from __future__ import unicode_literals
 import os
 import sys
 import unittest
+import collections
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 
@@ -100,8 +101,6 @@ class TestAllURLsMatching(unittest.TestCase):
         self.assertMatch(':ytsubs', ['youtube:subscriptions'])
         self.assertMatch(':ytsubscriptions', ['youtube:subscriptions'])
         self.assertMatch(':ythistory', ['youtube:history'])
-        self.assertMatch(':thedailyshow', ['ComedyCentralShows'])
-        self.assertMatch(':tds', ['ComedyCentralShows'])
 
     def test_vimeo_matching(self):
         self.assertMatch('https://vimeo.com/channels/tributes', ['vimeo:channel'])
@@ -130,6 +129,15 @@ class TestAllURLsMatching(unittest.TestCase):
             'https://screen.yahoo.com/smartwatches-latest-wearable-gadgets-163745379-cbs.html',
             ['Yahoo'])
 
+    def test_no_duplicated_ie_names(self):
+        name_accu = collections.defaultdict(list)
+        for ie in self.ies:
+            name_accu[ie.IE_NAME.lower()].append(type(ie).__name__)
+        for (ie_name, ie_list) in name_accu.items():
+            self.assertEqual(
+                len(ie_list), 1,
+                'Multiple extractors with the same IE_NAME "%s" (%s)' % (ie_name, ', '.join(ie_list)))
+
 
 if __name__ == '__main__':
     unittest.main()

test/test_compat.py

@@ -10,13 +10,14 @@ import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 
-from youtube_dl.utils import get_filesystem_encoding
 from youtube_dl.compat import (
     compat_getenv,
+    compat_setenv,
     compat_etree_fromstring,
     compat_expanduser,
     compat_shlex_split,
     compat_str,
+    compat_struct_unpack,
     compat_urllib_parse_unquote,
     compat_urllib_parse_unquote_plus,
     compat_urllib_parse_urlencode,
@@ -26,19 +27,22 @@ from youtube_dl.compat import (
 class TestCompat(unittest.TestCase):
     def test_compat_getenv(self):
         test_str = 'тест'
-        os.environ['YOUTUBE-DL-TEST'] = (
-            test_str if sys.version_info >= (3, 0)
-            else test_str.encode(get_filesystem_encoding()))
+        compat_setenv('YOUTUBE-DL-TEST', test_str)
         self.assertEqual(compat_getenv('YOUTUBE-DL-TEST'), test_str)
 
+    def test_compat_setenv(self):
+        test_var = 'YOUTUBE-DL-TEST'
+        test_str = 'тест'
+        compat_setenv(test_var, test_str)
+        compat_getenv(test_var)
+        self.assertEqual(compat_getenv(test_var), test_str)
+
     def test_compat_expanduser(self):
         old_home = os.environ.get('HOME')
         test_str = 'C:\Documents and Settings\тест\Application Data'
-        os.environ['HOME'] = (
-            test_str if sys.version_info >= (3, 0)
-            else test_str.encode(get_filesystem_encoding()))
+        compat_setenv('HOME', test_str)
         self.assertEqual(compat_expanduser('~'), test_str)
-        os.environ['HOME'] = old_home
+        compat_setenv('HOME', old_home or '')
 
     def test_all_present(self):
         import youtube_dl.compat
@@ -83,6 +87,8 @@ class TestCompat(unittest.TestCase):
 
     def test_compat_shlex_split(self):
         self.assertEqual(compat_shlex_split('-option "one two"'), ['-option', 'one two'])
+        self.assertEqual(compat_shlex_split('-option "one\ntwo" \n -flag'), ['-option', 'one\ntwo', '-flag'])
+        self.assertEqual(compat_shlex_split('-val 中文'), ['-val', '中文'])
 
     def test_compat_etree_fromstring(self):
         xml = '''
@@ -99,5 +105,15 @@ class TestCompat(unittest.TestCase):
         self.assertTrue(isinstance(doc.find('chinese').text, compat_str))
         self.assertTrue(isinstance(doc.find('foo/bar').text, compat_str))
 
+    def test_compat_etree_fromstring_doctype(self):
+        xml = '''<?xml version="1.0"?>
+<!DOCTYPE smil PUBLIC "-//W3C//DTD SMIL 2.0//EN" "http://www.w3.org/2001/SMIL20/SMIL20.dtd">
+<smil xmlns="http://www.w3.org/2001/SMIL20/Language"></smil>'''
+        compat_etree_fromstring(xml)
+
+    def test_struct_unpack(self):
+        self.assertEqual(compat_struct_unpack('!B', b'\x00'), (0,))
+
+
 if __name__ == '__main__':
     unittest.main()

test/test_http.py

@@ -16,6 +16,15 @@ import threading
 TEST_DIR = os.path.dirname(os.path.abspath(__file__))
 
 
+def http_server_port(httpd):
+    if os.name == 'java' and isinstance(httpd.socket, ssl.SSLSocket):
+        # In Jython SSLSocket is not a subclass of socket.socket
+        sock = httpd.socket.sock
+    else:
+        sock = httpd.socket
+    return sock.getsockname()[1]
+
+
 class HTTPTestRequestHandler(compat_http_server.BaseHTTPRequestHandler):
     def log_message(self, format, *args):
         pass
@@ -31,6 +40,22 @@ class HTTPTestRequestHandler(compat_http_server.BaseHTTPRequestHandler):
             self.send_header('Content-Type', 'video/mp4')
             self.end_headers()
             self.wfile.write(b'\x00\x00\x00\x00\x20\x66\x74[video]')
+        elif self.path == '/302':
+            if sys.version_info[0] == 3:
+                # XXX: Python 3 http server does not allow non-ASCII header values
+                self.send_response(404)
+                self.end_headers()
+                return
+
+            new_url = 'http://localhost:%d/中文.html' % http_server_port(self.server)
+            self.send_response(302)
+            self.send_header(b'Location', new_url.encode('utf-8'))
+            self.end_headers()
+        elif self.path == '/%E4%B8%AD%E6%96%87.html':
+            self.send_response(200)
+            self.send_header('Content-Type', 'text/html; charset=utf-8')
+            self.end_headers()
+            self.wfile.write(b'<html><video src="/vid.mp4" /></html>')
         else:
             assert False
 
@@ -47,18 +72,32 @@ class FakeLogger(object):
 
 
 class TestHTTP(unittest.TestCase):
+    def setUp(self):
+        self.httpd = compat_http_server.HTTPServer(
+            ('localhost', 0), HTTPTestRequestHandler)
+        self.port = http_server_port(self.httpd)
+        self.server_thread = threading.Thread(target=self.httpd.serve_forever)
+        self.server_thread.daemon = True
+        self.server_thread.start()
+
+    def test_unicode_path_redirection(self):
+        # XXX: Python 3 http server does not allow non-ASCII header values
+        if sys.version_info[0] == 3:
+            return
+
+        ydl = YoutubeDL({'logger': FakeLogger()})
+        r = ydl.extract_info('http://localhost:%d/302' % self.port)
+        self.assertEqual(r['url'], 'http://localhost:%d/vid.mp4' % self.port)
+
+
+class TestHTTPS(unittest.TestCase):
     def setUp(self):
         certfn = os.path.join(TEST_DIR, 'testcert.pem')
         self.httpd = compat_http_server.HTTPServer(
             ('localhost', 0), HTTPTestRequestHandler)
         self.httpd.socket = ssl.wrap_socket(
             self.httpd.socket, certfile=certfn, server_side=True)
-        if os.name == 'java':
-            # In Jython SSLSocket is not a subclass of socket.socket
-            sock = self.httpd.socket.sock
-        else:
-            sock = self.httpd.socket
-        self.port = sock.getsockname()[1]
+        self.port = http_server_port(self.httpd)
         self.server_thread = threading.Thread(target=self.httpd.serve_forever)
         self.server_thread.daemon = True
         self.server_thread.start()
@@ -94,32 +133,32 @@ class TestProxy(unittest.TestCase):
     def setUp(self):
         self.proxy = compat_http_server.HTTPServer(
             ('localhost', 0), _build_proxy_handler('normal'))
-        self.port = self.proxy.socket.getsockname()[1]
+        self.port = http_server_port(self.proxy)
         self.proxy_thread = threading.Thread(target=self.proxy.serve_forever)
         self.proxy_thread.daemon = True
         self.proxy_thread.start()
 
-        self.cn_proxy = compat_http_server.HTTPServer(
-            ('localhost', 0), _build_proxy_handler('cn'))
-        self.cn_port = self.cn_proxy.socket.getsockname()[1]
-        self.cn_proxy_thread = threading.Thread(target=self.cn_proxy.serve_forever)
-        self.cn_proxy_thread.daemon = True
-        self.cn_proxy_thread.start()
+        self.geo_proxy = compat_http_server.HTTPServer(
+            ('localhost', 0), _build_proxy_handler('geo'))
+        self.geo_port = http_server_port(self.geo_proxy)
+        self.geo_proxy_thread = threading.Thread(target=self.geo_proxy.serve_forever)
+        self.geo_proxy_thread.daemon = True
+        self.geo_proxy_thread.start()
 
     def test_proxy(self):
-        cn_proxy = 'localhost:{0}'.format(self.cn_port)
+        geo_proxy = 'localhost:{0}'.format(self.geo_port)
         ydl = YoutubeDL({
             'proxy': 'localhost:{0}'.format(self.port),
-            'cn_verification_proxy': cn_proxy,
+            'geo_verification_proxy': geo_proxy,
         })
         url = 'http://foo.com/bar'
         response = ydl.urlopen(url).read().decode('utf-8')
         self.assertEqual(response, 'normal: {0}'.format(url))
 
         req = compat_urllib_request.Request(url)
-        req.add_header('Ytdl-request-proxy', cn_proxy)
+        req.add_header('Ytdl-request-proxy', geo_proxy)
         response = ydl.urlopen(req).read().decode('utf-8')
-        self.assertEqual(response, 'cn: {0}'.format(url))
+        self.assertEqual(response, 'geo: {0}'.format(url))
 
     def test_proxy_with_idn(self):
         ydl = YoutubeDL({

test/test_socks.py

@@ -0,0 +1,118 @@
+#!/usr/bin/env python
+# coding: utf-8
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+import random
+import subprocess
+
+from test.helper import (
+    FakeYDL,
+    get_params,
+)
+from youtube_dl.compat import (
+    compat_str,
+    compat_urllib_request,
+)
+
+
+class TestMultipleSocks(unittest.TestCase):
+    @staticmethod
+    def _check_params(attrs):
+        params = get_params()
+        for attr in attrs:
+            if attr not in params:
+                print('Missing %s. Skipping.' % attr)
+                return
+        return params
+
+    def test_proxy_http(self):
+        params = self._check_params(['primary_proxy', 'primary_server_ip'])
+        if params is None:
+            return
+        ydl = FakeYDL({
+            'proxy': params['primary_proxy']
+        })
+        self.assertEqual(
+            ydl.urlopen('http://yt-dl.org/ip').read().decode('utf-8'),
+            params['primary_server_ip'])
+
+    def test_proxy_https(self):
+        params = self._check_params(['primary_proxy', 'primary_server_ip'])
+        if params is None:
+            return
+        ydl = FakeYDL({
+            'proxy': params['primary_proxy']
+        })
+        self.assertEqual(
+            ydl.urlopen('https://yt-dl.org/ip').read().decode('utf-8'),
+            params['primary_server_ip'])
+
+    def test_secondary_proxy_http(self):
+        params = self._check_params(['secondary_proxy', 'secondary_server_ip'])
+        if params is None:
+            return
+        ydl = FakeYDL()
+        req = compat_urllib_request.Request('http://yt-dl.org/ip')
+        req.add_header('Ytdl-request-proxy', params['secondary_proxy'])
+        self.assertEqual(
+            ydl.urlopen(req).read().decode('utf-8'),
+            params['secondary_server_ip'])
+
+    def test_secondary_proxy_https(self):
+        params = self._check_params(['secondary_proxy', 'secondary_server_ip'])
+        if params is None:
+            return
+        ydl = FakeYDL()
+        req = compat_urllib_request.Request('https://yt-dl.org/ip')
+        req.add_header('Ytdl-request-proxy', params['secondary_proxy'])
+        self.assertEqual(
+            ydl.urlopen(req).read().decode('utf-8'),
+            params['secondary_server_ip'])
+
+
+class TestSocks(unittest.TestCase):
+    _SKIP_SOCKS_TEST = True
+
+    def setUp(self):
+        if self._SKIP_SOCKS_TEST:
+            return
+
+        self.port = random.randint(20000, 30000)
+        self.server_process = subprocess.Popen([
+            'srelay', '-f', '-i', '127.0.0.1:%d' % self.port],
+            stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+
+    def tearDown(self):
+        if self._SKIP_SOCKS_TEST:
+            return
+
+        self.server_process.terminate()
+        self.server_process.communicate()
+
+    def _get_ip(self, protocol):
+        if self._SKIP_SOCKS_TEST:
+            return '127.0.0.1'
+
+        ydl = FakeYDL({
+            'proxy': '%s://127.0.0.1:%d' % (protocol, self.port),
+        })
+        return ydl.urlopen('http://yt-dl.org/ip').read().decode('utf-8')
+
+    def test_socks4(self):
+        self.assertTrue(isinstance(self._get_ip('socks4'), compat_str))
+
+    def test_socks4a(self):
+        self.assertTrue(isinstance(self._get_ip('socks4a'), compat_str))
+
+    def test_socks5(self):
+        self.assertTrue(isinstance(self._get_ip('socks5'), compat_str))
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_utils.py

@@ -33,6 +33,7 @@ from youtube_dl.utils import (
     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,
+    get_element_by_class,
     InAdvancePagedList,
     intlist_to_bytes,
     is_html,
@@ -41,6 +42,7 @@ from youtube_dl.utils import (
     ohdave_rsa_encrypt,
     OnDemandPagedList,
     orderedSet,
+    parse_age_limit,
     parse_duration,
     parse_filesize,
     parse_count,
@@ -50,20 +52,23 @@ from youtube_dl.utils import (
     sanitize_path,
     prepend_extension,
     replace_extension,
+    remove_start,
+    remove_end,
     remove_quotes,
     shell_quote,
     smuggle_url,
     str_to_int,
     strip_jsonp,
-    struct_unpack,
     timeconvert,
     unescapeHTML,
     unified_strdate,
+    unified_timestamp,
     unsmuggle_url,
     uppercase_escape,
     lowercase_escape,
     url_basename,
     urlencode_postdata,
+    urshift,
     update_url_query,
     version_tuple,
     xpath_with_ns,
@@ -77,6 +82,7 @@ from youtube_dl.utils import (
     cli_option,
     cli_valueless_option,
     cli_bool_option,
+    parse_codecs,
 )
 from youtube_dl.compat import (
     compat_chr,
@@ -139,8 +145,8 @@ class TestUtil(unittest.TestCase):
         self.assertEqual('yes_no', sanitize_filename('yes? no', restricted=True))
         self.assertEqual('this_-_that', sanitize_filename('this: that', restricted=True))
 
-        tests = 'a\xe4b\u4e2d\u56fd\u7684c'
-        self.assertEqual(sanitize_filename(tests, restricted=True), 'a_b_c')
+        tests = 'aäb\u4e2d\u56fd\u7684c'
+        self.assertEqual(sanitize_filename(tests, restricted=True), 'aab_c')
         self.assertTrue(sanitize_filename('\xf6', restricted=True) != '')  # No empty filename
 
         forbidden = '"\0\\/&!: \'\t\n()[]{}$;`^,#'
@@ -155,6 +161,10 @@ class TestUtil(unittest.TestCase):
         self.assertTrue(sanitize_filename('-', restricted=True) != '')
         self.assertTrue(sanitize_filename(':', restricted=True) != '')
 
+        self.assertEqual(sanitize_filename(
+            'ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ', restricted=True),
+            'AAAAAAAECEEEEIIIIDNOOOOOOOOEUUUUUYPssaaaaaaaeceeeeiiiionooooooooeuuuuuypy')
+
     def test_sanitize_ids(self):
         self.assertEqual(sanitize_filename('_n_cd26wFpw', is_id=True), '_n_cd26wFpw')
         self.assertEqual(sanitize_filename('_BD_eEpuzXw', is_id=True), '_BD_eEpuzXw')
@@ -212,6 +222,16 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(replace_extension('.abc', 'temp'), '.abc.temp')
         self.assertEqual(replace_extension('.abc.ext', 'temp'), '.abc.temp')
 
+    def test_remove_start(self):
+        self.assertEqual(remove_start(None, 'A - '), None)
+        self.assertEqual(remove_start('A - B', 'A - '), 'B')
+        self.assertEqual(remove_start('B - A', 'A - '), 'B - A')
+
+    def test_remove_end(self):
+        self.assertEqual(remove_end(None, ' - B'), None)
+        self.assertEqual(remove_end('A - B', ' - B'), 'A')
+        self.assertEqual(remove_end('B - A', ' - B'), 'B - A')
+
     def test_remove_quotes(self):
         self.assertEqual(remove_quotes(None), None)
         self.assertEqual(remove_quotes('"'), '"')
@@ -234,6 +254,8 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(unescapeHTML('/'), '/')
         self.assertEqual(unescapeHTML('é'), 'é')
         self.assertEqual(unescapeHTML('�'), '�')
+        # HTML5 entities
+        self.assertEqual(unescapeHTML('.''), '.\'')
 
     def test_date_from_str(self):
         self.assertEqual(date_from_str('yesterday'), date_from_str('now-1day'))
@@ -266,8 +288,29 @@ class TestUtil(unittest.TestCase):
             '20150202')
         self.assertEqual(unified_strdate('Feb 14th 2016 5:45PM'), '20160214')
         self.assertEqual(unified_strdate('25-09-2014'), '20140925')
+        self.assertEqual(unified_strdate('27.02.2016 17:30'), '20160227')
         self.assertEqual(unified_strdate('UNKNOWN DATE FORMAT'), None)
 
+    def test_unified_timestamps(self):
+        self.assertEqual(unified_timestamp('December 21, 2010'), 1292889600)
+        self.assertEqual(unified_timestamp('8/7/2009'), 1247011200)
+        self.assertEqual(unified_timestamp('Dec 14, 2012'), 1355443200)
+        self.assertEqual(unified_timestamp('2012/10/11 01:56:38 +0000'), 1349920598)
+        self.assertEqual(unified_timestamp('1968 12 10'), -33436800)
+        self.assertEqual(unified_timestamp('1968-12-10'), -33436800)
+        self.assertEqual(unified_timestamp('28/01/2014 21:00:00 +0100'), 1390939200)
+        self.assertEqual(
+            unified_timestamp('11/26/2014 11:30:00 AM PST', day_first=False),
+            1417001400)
+        self.assertEqual(
+            unified_timestamp('2/2/2015 6:47:40 PM', day_first=False),
+            1422902860)
+        self.assertEqual(unified_timestamp('Feb 14th 2016 5:45PM'), 1455471900)
+        self.assertEqual(unified_timestamp('25-09-2014'), 1411603200)
+        self.assertEqual(unified_timestamp('27.02.2016 17:30'), 1456594200)
+        self.assertEqual(unified_timestamp('UNKNOWN DATE FORMAT'), None)
+        self.assertEqual(unified_timestamp('May 16, 2016 11:15 PM'), 1463440500)
+
     def test_determine_ext(self):
         self.assertEqual(determine_ext('http://example.com/foo/bar.mp4/?download'), 'mp4')
         self.assertEqual(determine_ext('http://example.com/foo/bar/?download', None), None)
@@ -366,6 +409,12 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(res_url, url)
         self.assertEqual(res_data, None)
 
+        smug_url = smuggle_url(url, {'a': 'b'})
+        smug_smug_url = smuggle_url(smug_url, {'c': 'd'})
+        res_url, res_data = unsmuggle_url(smug_smug_url)
+        self.assertEqual(res_url, url)
+        self.assertEqual(res_data, {'a': 'b', 'c': 'd'})
+
     def test_shell_quote(self):
         args = ['ffmpeg', '-i', encodeFilename('ñ€ß\'.mp4')]
         self.assertEqual(shell_quote(args), """ffmpeg -i 'ñ€ß'"'"'.mp4'""")
@@ -384,6 +433,20 @@ class TestUtil(unittest.TestCase):
             url_basename('http://media.w3.org/2010/05/sintel/trailer.mp4'),
             'trailer.mp4')
 
+    def test_parse_age_limit(self):
+        self.assertEqual(parse_age_limit(None), None)
+        self.assertEqual(parse_age_limit(False), None)
+        self.assertEqual(parse_age_limit('invalid'), None)
+        self.assertEqual(parse_age_limit(0), 0)
+        self.assertEqual(parse_age_limit(18), 18)
+        self.assertEqual(parse_age_limit(21), 21)
+        self.assertEqual(parse_age_limit(22), None)
+        self.assertEqual(parse_age_limit('18'), 18)
+        self.assertEqual(parse_age_limit('18+'), 18)
+        self.assertEqual(parse_age_limit('PG-13'), 13)
+        self.assertEqual(parse_age_limit('TV-14'), 14)
+        self.assertEqual(parse_age_limit('TV-MA'), 17)
+
     def test_parse_duration(self):
         self.assertEqual(parse_duration(None), None)
         self.assertEqual(parse_duration(False), None)
@@ -453,9 +516,6 @@ class TestUtil(unittest.TestCase):
         testPL(5, 2, (2, 99), [2, 3, 4])
         testPL(5, 2, (20, 99), [])
 
-    def test_struct_unpack(self):
-        self.assertEqual(struct_unpack('!B', b'\x00'), (0,))
-
     def test_read_batch_urls(self):
         f = io.StringIO('''\xef\xbb\xbf foo
             bar\r
@@ -565,6 +625,29 @@ class TestUtil(unittest.TestCase):
             limit_length('foo bar baz asd', 12).startswith('foo bar'))
         self.assertTrue('...' in limit_length('foo bar baz asd', 12))
 
+    def test_parse_codecs(self):
+        self.assertEqual(parse_codecs(''), {})
+        self.assertEqual(parse_codecs('avc1.77.30, mp4a.40.2'), {
+            'vcodec': 'avc1.77.30',
+            'acodec': 'mp4a.40.2',
+        })
+        self.assertEqual(parse_codecs('mp4a.40.2'), {
+            'vcodec': 'none',
+            'acodec': 'mp4a.40.2',
+        })
+        self.assertEqual(parse_codecs('mp4a.40.5,avc1.42001e'), {
+            'vcodec': 'avc1.42001e',
+            'acodec': 'mp4a.40.5',
+        })
+        self.assertEqual(parse_codecs('avc3.640028'), {
+            'vcodec': 'avc3.640028',
+            'acodec': 'none',
+        })
+        self.assertEqual(parse_codecs(', h264,,newcodec,aac'), {
+            'vcodec': 'h264',
+            'acodec': 'aac',
+        })
+
     def test_escape_rfc3986(self):
         reserved = "!*'();:@&=+$,/?#[]"
         unreserved = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_.~'
@@ -617,6 +700,18 @@ class TestUtil(unittest.TestCase):
         json_code = js_to_json(inp)
         self.assertEqual(json.loads(json_code), json.loads(inp))
 
+        inp = '''{
+            0:{src:'skipped', type: 'application/dash+xml'},
+            1:{src:'skipped', type: 'application/vnd.apple.mpegURL'},
+        }'''
+        self.assertEqual(js_to_json(inp), '''{
+            "0":{"src":"skipped", "type": "application/dash+xml"},
+            "1":{"src":"skipped", "type": "application/vnd.apple.mpegURL"}
+        }''')
+
+        inp = '''{"foo":101}'''
+        self.assertEqual(js_to_json(inp), '''{"foo":101}''')
+
     def test_js_to_json_edgecases(self):
         on = js_to_json("{abc_def:'1\\'\\\\2\\\\\\'3\"4'}")
         self.assertEqual(json.loads(on), {"abc_def": "1'\\2\\'3\"4"})
@@ -640,6 +735,27 @@ class TestUtil(unittest.TestCase):
         on = js_to_json('{"abc": "def",}')
         self.assertEqual(json.loads(on), {'abc': 'def'})
 
+        on = js_to_json('{ 0: /* " \n */ ",]" , }')
+        self.assertEqual(json.loads(on), {'0': ',]'})
+
+        on = js_to_json(r'["<p>x<\/p>"]')
+        self.assertEqual(json.loads(on), ['<p>x</p>'])
+
+        on = js_to_json(r'["\xaa"]')
+        self.assertEqual(json.loads(on), ['\u00aa'])
+
+        on = js_to_json("['a\\\nb']")
+        self.assertEqual(json.loads(on), ['ab'])
+
+        on = js_to_json('{0xff:0xff}')
+        self.assertEqual(json.loads(on), {'255': 255})
+
+        on = js_to_json('{077:077}')
+        self.assertEqual(json.loads(on), {'63': 63})
+
+        on = js_to_json('{42:42}')
+        self.assertEqual(json.loads(on), {'42': 42})
+
     def test_extract_attributes(self):
         self.assertEqual(extract_attributes('<e x="y">'), {'x': 'y'})
         self.assertEqual(extract_attributes("<e x='y'>"), {'x': 'y'})
@@ -912,5 +1028,17 @@ The first line
         self.assertRaises(ValueError, encode_base_n, 0, 70)
         self.assertRaises(ValueError, encode_base_n, 0, 60, custom_table)
 
+    def test_urshift(self):
+        self.assertEqual(urshift(3, 1), 1)
+        self.assertEqual(urshift(-3, 1), 2147483646)
+
+    def test_get_element_by_class(self):
+        html = '''
+            <span class="foo bar">nice</span>
+        '''
+
+        self.assertEqual(get_element_by_class('foo', html), 'nice')
+        self.assertEqual(get_element_by_class('no-such-class', html), None)
+
 if __name__ == '__main__':
     unittest.main()

test/test_verbose_output.py

@@ -0,0 +1,70 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+from __future__ import unicode_literals
+
+import unittest
+
+import sys
+import os
+import subprocess
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+rootDir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+
+
+class TestVerboseOutput(unittest.TestCase):
+    def test_private_info_arg(self):
+        outp = subprocess.Popen(
+            [
+                sys.executable, 'youtube_dl/__main__.py', '-v',
+                '--username', 'johnsmith@gmail.com',
+                '--password', 'secret',
+            ], cwd=rootDir, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+        sout, serr = outp.communicate()
+        self.assertTrue('--username' in serr)
+        self.assertTrue('johnsmith' not in serr)
+        self.assertTrue('--password' in serr)
+        self.assertTrue('secret' not in serr)
+
+    def test_private_info_shortarg(self):
+        outp = subprocess.Popen(
+            [
+                sys.executable, 'youtube_dl/__main__.py', '-v',
+                '-u', 'johnsmith@gmail.com',
+                '-p', 'secret',
+            ], cwd=rootDir, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+        sout, serr = outp.communicate()
+        self.assertTrue('-u' in serr)
+        self.assertTrue('johnsmith' not in serr)
+        self.assertTrue('-p' in serr)
+        self.assertTrue('secret' not in serr)
+
+    def test_private_info_eq(self):
+        outp = subprocess.Popen(
+            [
+                sys.executable, 'youtube_dl/__main__.py', '-v',
+                '--username=johnsmith@gmail.com',
+                '--password=secret',
+            ], cwd=rootDir, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+        sout, serr = outp.communicate()
+        self.assertTrue('--username' in serr)
+        self.assertTrue('johnsmith' not in serr)
+        self.assertTrue('--password' in serr)
+        self.assertTrue('secret' not in serr)
+
+    def test_private_info_shortarg_eq(self):
+        outp = subprocess.Popen(
+            [
+                sys.executable, 'youtube_dl/__main__.py', '-v',
+                '-u=johnsmith@gmail.com',
+                '-p=secret',
+            ], cwd=rootDir, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+        sout, serr = outp.communicate()
+        self.assertTrue('-u' in serr)
+        self.assertTrue('johnsmith' not in serr)
+        self.assertTrue('-p' in serr)
+        self.assertTrue('secret' not in serr)
+
+if __name__ == '__main__':
+    unittest.main()

tox.ini

@@ -9,5 +9,6 @@ passenv = HOME
 defaultargs = test --exclude test_download.py --exclude test_age_restriction.py
     --exclude test_subtitles.py --exclude test_write_annotations.py
     --exclude test_youtube_lists.py --exclude test_iqiyi_sdk_interpreter.py
+    --exclude test_socks.py
 commands = nosetests --verbose {posargs:{[testenv]defaultargs}}  # --with-coverage --cover-package=youtube_dl --cover-html
                                                # test.test_download:TestDownload.test_NowVideo

youtube_dl/YoutubeDL.py

@@ -5,6 +5,7 @@ from __future__ import absolute_import, unicode_literals
 
 import collections
 import contextlib
+import copy
 import datetime
 import errno
 import fileinput
@@ -64,6 +65,7 @@ from .utils import (
     PostProcessingError,
     preferredencoding,
     prepend_extension,
+    register_socks_protocols,
     render_table,
     replace_extension,
     SameFileError,
@@ -195,8 +197,8 @@ class YoutubeDL(object):
     prefer_insecure:   Use HTTP instead of HTTPS to retrieve information.
                        At the moment, this is only supported by YouTube.
     proxy:             URL of the proxy server to use
-    cn_verification_proxy:  URL of the proxy to use for IP address verification
-                       on Chinese sites. (Experimental)
+    geo_verification_proxy:  URL of the proxy to use for IP address verification
+                       on geo-restricted sites. (Experimental)
     socket_timeout:    Time to wait for unresponsive hosts, in seconds
     bidi_workaround:   Work around buggy terminals without bidirectional text
                        support, using fridibi
@@ -303,6 +305,11 @@ class YoutubeDL(object):
         self.params.update(params)
         self.cache = Cache(self)
 
+        if self.params.get('cn_verification_proxy') is not None:
+            self.report_warning('--cn-verification-proxy is deprecated. Use --geo-verification-proxy instead.')
+            if self.params.get('geo_verification_proxy') is None:
+                self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
+
         if params.get('bidi_workaround', False):
             try:
                 import pty
@@ -325,7 +332,7 @@ class YoutubeDL(object):
                         ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
                 self._output_channel = os.fdopen(master, 'rb')
             except OSError as ose:
-                if ose.errno == 2:
+                if ose.errno == errno.ENOENT:
                     self.report_warning('Could not find fribidi executable, ignoring --bidi-workaround . Make sure that  fribidi  is an executable file in one of the directories in your $PATH.')
                 else:
                     raise
@@ -361,6 +368,8 @@ class YoutubeDL(object):
         for ph in self.params.get('progress_hooks', []):
             self.add_progress_hook(ph)
 
+        register_socks_protocols()
+
     def warn_if_short_id(self, argv):
         # short YouTube ID starting with dash?
         idxs = [
@@ -580,7 +589,7 @@ class YoutubeDL(object):
                 is_id=(k == 'id'))
             template_dict = dict((k, sanitize(k, v))
                                  for k, v in template_dict.items()
-                                 if v is not None)
+                                 if v is not None and not isinstance(v, (list, tuple, dict)))
             template_dict = collections.defaultdict(lambda: 'NA', template_dict)
 
             outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
@@ -717,6 +726,7 @@ class YoutubeDL(object):
         result_type = ie_result.get('_type', 'video')
 
         if result_type in ('url', 'url_transparent'):
+            ie_result['url'] = sanitize_url(ie_result['url'])
             extract_flat = self.params.get('extract_flat', False)
             if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) or
                     extract_flat is True):
@@ -1042,9 +1052,9 @@ class YoutubeDL(object):
             if isinstance(selector, list):
                 fs = [_build_selector_function(s) for s in selector]
 
-                def selector_function(formats):
+                def selector_function(ctx):
                     for f in fs:
-                        for format in f(formats):
+                        for format in f(ctx):
                             yield format
                 return selector_function
             elif selector.type == GROUP:
@@ -1052,17 +1062,17 @@ class YoutubeDL(object):
             elif selector.type == PICKFIRST:
                 fs = [_build_selector_function(s) for s in selector.selector]
 
-                def selector_function(formats):
+                def selector_function(ctx):
                     for f in fs:
-                        picked_formats = list(f(formats))
+                        picked_formats = list(f(ctx))
                         if picked_formats:
                             return picked_formats
                     return []
             elif selector.type == SINGLE:
                 format_spec = selector.selector
 
-                def selector_function(formats):
-                    formats = list(formats)
+                def selector_function(ctx):
+                    formats = list(ctx['formats'])
                     if not formats:
                         return
                     if format_spec == 'all':
@@ -1075,9 +1085,10 @@ class YoutubeDL(object):
                             if f.get('vcodec') != 'none' and f.get('acodec') != 'none']
                         if audiovideo_formats:
                             yield audiovideo_formats[format_idx]
-                        # for audio only (soundcloud) or video only (imgur) urls, select the best/worst audio format
-                        elif (all(f.get('acodec') != 'none' for f in formats) or
-                              all(f.get('vcodec') != 'none' for f in formats)):
+                        # for extractors with incomplete formats (audio only (soundcloud)
+                        # or video only (imgur)) we will fallback to best/worst
+                        # {video,audio}-only format
+                        elif ctx['incomplete_formats']:
                             yield formats[format_idx]
                     elif format_spec == 'bestaudio':
                         audio_formats = [
@@ -1151,17 +1162,18 @@ class YoutubeDL(object):
                     }
                 video_selector, audio_selector = map(_build_selector_function, selector.selector)
 
-                def selector_function(formats):
-                    formats = list(formats)
-                    for pair in itertools.product(video_selector(formats), audio_selector(formats)):
+                def selector_function(ctx):
+                    for pair in itertools.product(
+                            video_selector(copy.deepcopy(ctx)), audio_selector(copy.deepcopy(ctx))):
                         yield _merge(pair)
 
             filters = [self._build_format_filter(f) for f in selector.filters]
 
-            def final_selector(formats):
+            def final_selector(ctx):
+                ctx_copy = copy.deepcopy(ctx)
                 for _filter in filters:
-                    formats = list(filter(_filter, formats))
-                return selector_function(formats)
+                    ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
+                return selector_function(ctx_copy)
             return final_selector
 
         stream = io.BytesIO(format_spec.encode('utf-8'))
@@ -1219,6 +1231,10 @@ class YoutubeDL(object):
         if 'title' not in info_dict:
             raise ExtractorError('Missing "title" field in extractor result')
 
+        if not isinstance(info_dict['id'], compat_str):
+            self.report_warning('"id" field is not a string - forcing string conversion')
+            info_dict['id'] = compat_str(info_dict['id'])
+
         if 'playlist' not in info_dict:
             # It isn't part of a playlist
             info_dict['playlist'] = None
@@ -1364,7 +1380,34 @@ class YoutubeDL(object):
             req_format_list.append('best')
             req_format = '/'.join(req_format_list)
         format_selector = self.build_format_selector(req_format)
-        formats_to_download = list(format_selector(formats))
+
+        # While in format selection we may need to have an access to the original
+        # format set in order to calculate some metrics or do some processing.
+        # For now we need to be able to guess whether original formats provided
+        # by extractor are incomplete or not (i.e. whether extractor provides only
+        # video-only or audio-only formats) for proper formats selection for
+        # extractors with such incomplete formats (see
+        # https://github.com/rg3/youtube-dl/pull/5556).
+        # Since formats may be filtered during format selection and may not match
+        # the original formats the results may be incorrect. Thus original formats
+        # or pre-calculated metrics should be passed to format selection routines
+        # as well.
+        # We will pass a context object containing all necessary additional data
+        # instead of just formats.
+        # This fixes incorrect format selection issue (see
+        # https://github.com/rg3/youtube-dl/issues/10083).
+        incomplete_formats = (
+            # All formats are video-only or
+            all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats) or
+            # all formats are audio-only
+            all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
+
+        ctx = {
+            'formats': formats,
+            'incomplete_formats': incomplete_formats,
+        }
+
+        formats_to_download = list(format_selector(ctx))
         if not formats_to_download:
             raise ExtractorError('requested format not available',
                                  expected=True)
@@ -1639,7 +1682,7 @@ class YoutubeDL(object):
                     # Just a single file
                     success = dl(filename, info_dict)
             except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-                self.report_error('unable to download video data: %s' % str(err))
+                self.report_error('unable to download video data: %s' % error_to_compat_str(err))
                 return
             except (OSError, IOError) as err:
                 raise UnavailableVideoError(err)
@@ -2018,6 +2061,7 @@ class YoutubeDL(object):
         if opts_cookiefile is None:
             self.cookiejar = compat_cookiejar.CookieJar()
         else:
+            opts_cookiefile = compat_expanduser(opts_cookiefile)
             self.cookiejar = compat_cookiejar.MozillaCookieJar(
                 opts_cookiefile)
             if os.access(opts_cookiefile, os.R_OK):

youtube_dl/__init__.py

@@ -18,7 +18,6 @@ from .options import (
 from .compat import (
     compat_expanduser,
     compat_getpass,
-    compat_print,
     compat_shlex_split,
     workaround_optparse_bug9161,
 )
@@ -67,16 +66,16 @@ def _real_main(argv=None):
     # Custom HTTP headers
     if opts.headers is not None:
         for h in opts.headers:
-            if h.find(':', 1) < 0:
+            if ':' not in h:
                 parser.error('wrong header formatting, it should be key:value, not "%s"' % h)
-            key, value = h.split(':', 2)
+            key, value = h.split(':', 1)
             if opts.verbose:
                 write_string('[debug] Adding header from command line option %s:%s\n' % (key, value))
             std_headers[key] = value
 
     # Dump user agent
     if opts.dump_user_agent:
-        compat_print(std_headers['User-Agent'])
+        write_string(std_headers['User-Agent'] + '\n', out=sys.stdout)
         sys.exit(0)
 
     # Batch file verification
@@ -86,7 +85,9 @@ def _real_main(argv=None):
             if opts.batchfile == '-':
                 batchfd = sys.stdin
             else:
-                batchfd = io.open(opts.batchfile, 'r', encoding='utf-8', errors='ignore')
+                batchfd = io.open(
+                    compat_expanduser(opts.batchfile),
+                    'r', encoding='utf-8', errors='ignore')
             batch_urls = read_batch_urls(batchfd)
             if opts.verbose:
                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
@@ -99,10 +100,10 @@ def _real_main(argv=None):
 
     if opts.list_extractors:
         for ie in list_extractors(opts.age_limit):
-            compat_print(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else ''))
+            write_string(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else '') + '\n', out=sys.stdout)
             matchedUrls = [url for url in all_urls if ie.suitable(url)]
             for mu in matchedUrls:
-                compat_print('  ' + mu)
+                write_string('  ' + mu + '\n', out=sys.stdout)
         sys.exit(0)
     if opts.list_extractor_descriptions:
         for ie in list_extractors(opts.age_limit):
@@ -115,7 +116,7 @@ def _real_main(argv=None):
                 _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
                 _COUNTS = ('', '5', '10', 'all')
                 desc += ' (Example: "%s%s:%s" )' % (ie.SEARCH_KEY, random.choice(_COUNTS), random.choice(_SEARCHES))
-            compat_print(desc)
+            write_string(desc + '\n', out=sys.stdout)
         sys.exit(0)
 
     # Conflicting, missing and erroneous options
@@ -381,6 +382,8 @@ def _real_main(argv=None):
         'external_downloader_args': external_downloader_args,
         'postprocessor_args': postprocessor_args,
         'cn_verification_proxy': opts.cn_verification_proxy,
+        'geo_verification_proxy': opts.geo_verification_proxy,
+
     }
 
     with YoutubeDL(ydl_opts) as ydl:
@@ -404,7 +407,7 @@ def _real_main(argv=None):
 
         try:
             if opts.load_info_filename is not None:
-                retcode = ydl.download_with_info_file(opts.load_info_filename)
+                retcode = ydl.download_with_info_file(compat_expanduser(opts.load_info_filename))
             else:
                 retcode = ydl.download(all_urls)
         except MaxDownloadsReached:

youtube_dl/downloader/external.py

@@ -6,6 +6,7 @@ import sys
 import re
 
 from .common import FileDownloader
+from ..compat import compat_setenv
 from ..postprocessor.ffmpeg import FFmpegPostProcessor, EXT_TO_OUT_FORMATS
 from ..utils import (
     cli_option,
@@ -84,7 +85,7 @@ class ExternalFD(FileDownloader):
             cmd, stderr=subprocess.PIPE)
         _, stderr = p.communicate()
         if p.returncode != 0:
-            self.to_stderr(stderr)
+            self.to_stderr(stderr.decode('utf-8', 'replace'))
         return p.returncode
 
 
@@ -198,6 +199,19 @@ class FFmpegFD(ExternalFD):
                 '-headers',
                 ''.join('%s: %s\r\n' % (key, val) for key, val in headers.items())]
 
+        env = None
+        proxy = self.params.get('proxy')
+        if proxy:
+            if not re.match(r'^[\da-zA-Z]+://', proxy):
+                proxy = 'http://%s' % proxy
+            # Since December 2015 ffmpeg supports -http_proxy option (see
+            # http://git.videolan.org/?p=ffmpeg.git;a=commit;h=b4eb1f29ebddd60c41a2eb39f5af701e38e0d3fd)
+            # We could switch to the following code if we are able to detect version properly
+            # args += ['-http_proxy', proxy]
+            env = os.environ.copy()
+            compat_setenv('HTTP_PROXY', proxy, env=env)
+            compat_setenv('http_proxy', proxy, env=env)
+
         protocol = info_dict.get('protocol')
 
         if protocol == 'rtmp':
@@ -224,7 +238,7 @@ class FFmpegFD(ExternalFD):
                 args += ['-rtmp_live', 'live']
 
         args += ['-i', url, '-c', 'copy']
-        if protocol == 'm3u8':
+        if protocol in ('m3u8', 'm3u8_native'):
             if self.params.get('hls_use_mpegts', False) or tmpfilename == '-':
                 args += ['-f', 'mpegts']
             else:
@@ -239,7 +253,7 @@ class FFmpegFD(ExternalFD):
 
         self._debug_cmd(args)
 
-        proc = subprocess.Popen(args, stdin=subprocess.PIPE)
+        proc = subprocess.Popen(args, stdin=subprocess.PIPE, env=env)
         try:
             retval = proc.wait()
         except KeyboardInterrupt:

youtube_dl/downloader/f4m.py

@@ -12,37 +12,49 @@ from ..compat import (
     compat_urlparse,
     compat_urllib_error,
     compat_urllib_parse_urlparse,
+    compat_struct_pack,
+    compat_struct_unpack,
 )
 from ..utils import (
     encodeFilename,
     fix_xml_ampersands,
     sanitize_open,
-    struct_pack,
-    struct_unpack,
     xpath_text,
 )
 
 
+class DataTruncatedError(Exception):
+    pass
+
+
 class FlvReader(io.BytesIO):
     """
     Reader for Flv files
     The file format is documented in https://www.adobe.com/devnet/f4v.html
     """
 
+    def read_bytes(self, n):
+        data = self.read(n)
+        if len(data) < n:
+            raise DataTruncatedError(
+                'FlvReader error: need %d bytes while only %d bytes got' % (
+                    n, len(data)))
+        return data
+
     # Utility functions for reading numbers and strings
     def read_unsigned_long_long(self):
-        return struct_unpack('!Q', self.read(8))[0]
+        return compat_struct_unpack('!Q', self.read_bytes(8))[0]
 
     def read_unsigned_int(self):
-        return struct_unpack('!I', self.read(4))[0]
+        return compat_struct_unpack('!I', self.read_bytes(4))[0]
 
     def read_unsigned_char(self):
-        return struct_unpack('!B', self.read(1))[0]
+        return compat_struct_unpack('!B', self.read_bytes(1))[0]
 
     def read_string(self):
         res = b''
         while True:
-            char = self.read(1)
+            char = self.read_bytes(1)
             if char == b'\x00':
                 break
             res += char
@@ -53,18 +65,18 @@ class FlvReader(io.BytesIO):
         Read a box and return the info as a tuple: (box_size, box_type, box_data)
         """
         real_size = size = self.read_unsigned_int()
-        box_type = self.read(4)
+        box_type = self.read_bytes(4)
         header_end = 8
         if size == 1:
             real_size = self.read_unsigned_long_long()
             header_end = 16
-        return real_size, box_type, self.read(real_size - header_end)
+        return real_size, box_type, self.read_bytes(real_size - header_end)
 
     def read_asrt(self):
         # version
         self.read_unsigned_char()
         # flags
-        self.read(3)
+        self.read_bytes(3)
         quality_entry_count = self.read_unsigned_char()
         # QualityEntryCount
         for i in range(quality_entry_count):
@@ -85,7 +97,7 @@ class FlvReader(io.BytesIO):
         # version
         self.read_unsigned_char()
         # flags
-        self.read(3)
+        self.read_bytes(3)
         # time scale
         self.read_unsigned_int()
 
@@ -119,7 +131,7 @@ class FlvReader(io.BytesIO):
         # version
         self.read_unsigned_char()
         # flags
-        self.read(3)
+        self.read_bytes(3)
 
         self.read_unsigned_int()  # BootstrapinfoVersion
         # Profile,Live,Update,Reserved
@@ -184,6 +196,11 @@ def build_fragments_list(boot_info):
     first_frag_number = fragment_run_entry_table[0]['first']
     fragments_counter = itertools.count(first_frag_number)
     for segment, fragments_count in segment_run_table['segment_run']:
+        # In some live HDS streams (for example Rai), `fragments_count` is
+        # abnormal and causing out-of-memory errors. It's OK to change the
+        # number of fragments for live streams as they are updated periodically
+        if fragments_count == 4294967295 and boot_info['live']:
+            fragments_count = 2
         for _ in range(fragments_count):
             res.append((segment, next(fragments_counter)))
 
@@ -194,11 +211,11 @@ def build_fragments_list(boot_info):
 
 
 def write_unsigned_int(stream, val):
-    stream.write(struct_pack('!I', val))
+    stream.write(compat_struct_pack('!I', val))
 
 
 def write_unsigned_int_24(stream, val):
-    stream.write(struct_pack('!I', val)[1:])
+    stream.write(compat_struct_pack('!I', val)[1:])
 
 
 def write_flv_header(stream):
@@ -307,7 +324,7 @@ class F4mFD(FragmentFD):
         doc = compat_etree_fromstring(manifest)
         formats = [(int(f.attrib.get('bitrate', -1)), f)
                    for f in self._get_unencrypted_media(doc)]
-        if requested_bitrate is None:
+        if requested_bitrate is None or len(formats) == 1:
             # get the best format
             formats = sorted(formats, key=lambda f: f[0])
             rate, media = formats[-1]
@@ -317,7 +334,11 @@ class F4mFD(FragmentFD):
 
         base_url = compat_urlparse.urljoin(man_url, media.attrib['url'])
         bootstrap_node = doc.find(_add_ns('bootstrapInfo'))
-        boot_info, bootstrap_url = self._parse_bootstrap_node(bootstrap_node, base_url)
+        # From Adobe F4M 3.0 spec:
+        # The <baseURL> element SHALL be the base URL for all relative
+        # (HTTP-based) URLs in the manifest. If <baseURL> is not present, said
+        # URLs should be relative to the location of the containing document.
+        boot_info, bootstrap_url = self._parse_bootstrap_node(bootstrap_node, man_url)
         live = boot_info['live']
         metadata_node = media.find(_add_ns('metadata'))
         if metadata_node is not None:
@@ -374,7 +395,17 @@ class F4mFD(FragmentFD):
                 down.close()
                 reader = FlvReader(down_data)
                 while True:
-                    _, box_type, box_data = reader.read_box_info()
+                    try:
+                        _, box_type, box_data = reader.read_box_info()
+                    except DataTruncatedError:
+                        if test:
+                            # In tests, segments may be truncated, and thus
+                            # FlvReader may not be able to parse the whole
+                            # chunk. If so, write the segment as is
+                            # See https://github.com/rg3/youtube-dl/issues/9214
+                            dest_stream.write(down_data)
+                            break
+                        raise
                     if box_type == b'mdat':
                         dest_stream.write(box_data)
                         break

youtube_dl/downloader/hls.py

@@ -2,13 +2,24 @@ from __future__ import unicode_literals
 
 import os.path
 import re
+import binascii
+try:
+    from Crypto.Cipher import AES
+    can_decrypt_frag = True
+except ImportError:
+    can_decrypt_frag = False
 
 from .fragment import FragmentFD
+from .external import FFmpegFD
 
-from ..compat import compat_urlparse
+from ..compat import (
+    compat_urlparse,
+    compat_struct_pack,
+)
 from ..utils import (
     encodeFilename,
     sanitize_open,
+    parse_m3u8_attributes,
 )
 
 
@@ -17,42 +28,101 @@ class HlsFD(FragmentFD):
 
     FD_NAME = 'hlsnative'
 
+    @staticmethod
+    def can_download(manifest):
+        UNSUPPORTED_FEATURES = (
+            r'#EXT-X-KEY:METHOD=(?!NONE|AES-128)',  # encrypted streams [1]
+            r'#EXT-X-BYTERANGE',  # playlists composed of byte ranges of media files [2]
+
+            # Live streams heuristic does not always work (e.g. geo restricted to Germany
+            # http://hls-geo.daserste.de/i/videoportal/Film/c_620000/622873/format,716451,716457,716450,716458,716459,.mp4.csmil/index_4_av.m3u8?null=0)
+            # r'#EXT-X-MEDIA-SEQUENCE:(?!0$)',  # live streams [3]
+
+            # This heuristic also is not correct since segments may not be appended as well.
+            # Twitch vods of finished streams have EXT-X-PLAYLIST-TYPE:EVENT despite
+            # no segments will definitely be appended to the end of the playlist.
+            # r'#EXT-X-PLAYLIST-TYPE:EVENT',  # media segments may be appended to the end of
+            #                                 # event media playlists [4]
+
+            # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.2.4
+            # 2. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.2.2
+            # 3. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.2
+            # 4. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.5
+        )
+        check_results = [not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES]
+        check_results.append(can_decrypt_frag or '#EXT-X-KEY:METHOD=AES-128' not in manifest)
+        return all(check_results)
+
     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
         self.to_screen('[%s] Downloading m3u8 manifest' % self.FD_NAME)
         manifest = self.ydl.urlopen(man_url).read()
 
         s = manifest.decode('utf-8', 'ignore')
-        fragment_urls = []
+
+        if not self.can_download(s):
+            self.report_warning(
+                'hlsnative has detected features it does not support, '
+                'extraction will be delegated to ffmpeg')
+            fd = FFmpegFD(self.ydl, self.params)
+            for ph in self._progress_hooks:
+                fd.add_progress_hook(ph)
+            return fd.real_download(filename, info_dict)
+
+        total_frags = 0
         for line in s.splitlines():
             line = line.strip()
             if line and not line.startswith('#'):
-                segment_url = (
-                    line
-                    if re.match(r'^https?://', line)
-                    else compat_urlparse.urljoin(man_url, line))
-                fragment_urls.append(segment_url)
-                # We only download the first fragment during the test
-                if self.params.get('test', False):
-                    break
+                total_frags += 1
 
         ctx = {
             'filename': filename,
-            'total_frags': len(fragment_urls),
+            'total_frags': total_frags,
         }
 
         self._prepare_and_start_frag_download(ctx)
 
+        i = 0
+        media_sequence = 0
+        decrypt_info = {'METHOD': 'NONE'}
         frags_filenames = []
-        for i, frag_url in enumerate(fragment_urls):
-            frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i)
-            success = ctx['dl'].download(frag_filename, {'url': frag_url})
-            if not success:
-                return False
-            down, frag_sanitized = sanitize_open(frag_filename, 'rb')
-            ctx['dest_stream'].write(down.read())
-            down.close()
-            frags_filenames.append(frag_sanitized)
+        for line in s.splitlines():
+            line = line.strip()
+            if line:
+                if not line.startswith('#'):
+                    frag_url = (
+                        line
+                        if re.match(r'^https?://', line)
+                        else compat_urlparse.urljoin(man_url, line))
+                    frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i)
+                    success = ctx['dl'].download(frag_filename, {'url': frag_url})
+                    if not success:
+                        return False
+                    down, frag_sanitized = sanitize_open(frag_filename, 'rb')
+                    frag_content = down.read()
+                    down.close()
+                    if decrypt_info['METHOD'] == 'AES-128':
+                        iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', media_sequence)
+                        frag_content = AES.new(
+                            decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
+                    ctx['dest_stream'].write(frag_content)
+                    frags_filenames.append(frag_sanitized)
+                    # We only download the first fragment during the test
+                    if self.params.get('test', False):
+                        break
+                    i += 1
+                    media_sequence += 1
+                elif line.startswith('#EXT-X-KEY'):
+                    decrypt_info = parse_m3u8_attributes(line[11:])
+                    if decrypt_info['METHOD'] == 'AES-128':
+                        if 'IV' in decrypt_info:
+                            decrypt_info['IV'] = binascii.unhexlify(decrypt_info['IV'][2:])
+                        if not re.match(r'^https?://', decrypt_info['URI']):
+                            decrypt_info['URI'] = compat_urlparse.urljoin(
+                                man_url, decrypt_info['URI'])
+                        decrypt_info['KEY'] = self.ydl.urlopen(decrypt_info['URI']).read()
+                elif line.startswith('#EXT-X-MEDIA-SEQUENCE'):
+                    media_sequence = int(line[22:])
 
         self._finish_frag_download(ctx)
 

youtube_dl/extractor/abcnews.py

@@ -0,0 +1,135 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import calendar
+import re
+import time
+
+from .amp import AMPIE
+from .common import InfoExtractor
+from ..compat import compat_urlparse
+
+
+class AbcNewsVideoIE(AMPIE):
+    IE_NAME = 'abcnews:video'
+    _VALID_URL = 'http://abcnews.go.com/[^/]+/video/(?P<display_id>[0-9a-z-]+)-(?P<id>\d+)'
+
+    _TESTS = [{
+        'url': 'http://abcnews.go.com/ThisWeek/video/week-exclusive-irans-foreign-minister-zarif-20411932',
+        'info_dict': {
+            'id': '20411932',
+            'ext': 'mp4',
+            'display_id': 'week-exclusive-irans-foreign-minister-zarif',
+            'title': '\'This Week\' Exclusive: Iran\'s Foreign Minister Zarif',
+            'description': 'George Stephanopoulos goes one-on-one with Iranian Foreign Minister Dr. Javad Zarif.',
+            'duration': 180,
+            'thumbnail': 're:^https?://.*\.jpg$',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://abcnews.go.com/2020/video/2020-husband-stands-teacher-jail-student-affairs-26119478',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('display_id')
+        video_id = mobj.group('id')
+        info_dict = self._extract_feed_info(
+            'http://abcnews.go.com/video/itemfeed?id=%s' % video_id)
+        info_dict.update({
+            'id': video_id,
+            'display_id': display_id,
+        })
+        return info_dict
+
+
+class AbcNewsIE(InfoExtractor):
+    IE_NAME = 'abcnews'
+    _VALID_URL = 'https?://abcnews\.go\.com/(?:[^/]+/)+(?P<display_id>[0-9a-z-]+)/story\?id=(?P<id>\d+)'
+
+    _TESTS = [{
+        'url': 'http://abcnews.go.com/Blotter/News/dramatic-video-rare-death-job-america/story?id=10498713#.UIhwosWHLjY',
+        'info_dict': {
+            'id': '10498713',
+            'ext': 'flv',
+            'display_id': 'dramatic-video-rare-death-job-america',
+            'title': 'Occupational Hazards',
+            'description': 'Nightline investigates the dangers that lurk at various jobs.',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'upload_date': '20100428',
+            'timestamp': 1272412800,
+        },
+        'add_ie': ['AbcNewsVideo'],
+    }, {
+        'url': 'http://abcnews.go.com/Entertainment/justin-timberlake-performs-stop-feeling-eurovision-2016/story?id=39125818',
+        'info_dict': {
+            'id': '39125818',
+            'ext': 'mp4',
+            'display_id': 'justin-timberlake-performs-stop-feeling-eurovision-2016',
+            'title': 'Justin Timberlake Drops Hints For Secret Single',
+            'description': 'Lara Spencer reports the buzziest stories of the day in "GMA" Pop News.',
+            'upload_date': '20160515',
+            'timestamp': 1463329500,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+            # The embedded YouTube video is blocked due to copyright issues
+            'playlist_items': '1',
+        },
+        'add_ie': ['AbcNewsVideo'],
+    }, {
+        'url': 'http://abcnews.go.com/Technology/exclusive-apple-ceo-tim-cook-iphone-cracking-software/story?id=37173343',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('display_id')
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        video_url = self._search_regex(
+            r'window\.abcnvideo\.url\s*=\s*"([^"]+)"', webpage, 'video URL')
+        full_video_url = compat_urlparse.urljoin(url, video_url)
+
+        youtube_url = self._html_search_regex(
+            r'<iframe[^>]+src="(https://www\.youtube\.com/embed/[^"]+)"',
+            webpage, 'YouTube URL', default=None)
+
+        timestamp = None
+        date_str = self._html_search_regex(
+            r'<span[^>]+class="timestamp">([^<]+)</span>',
+            webpage, 'timestamp', fatal=False)
+        if date_str:
+            tz_offset = 0
+            if date_str.endswith(' ET'):  # Eastern Time
+                tz_offset = -5
+                date_str = date_str[:-3]
+            date_formats = ['%b. %d, %Y', '%b %d, %Y, %I:%M %p']
+            for date_format in date_formats:
+                try:
+                    timestamp = calendar.timegm(time.strptime(date_str.strip(), date_format))
+                except ValueError:
+                    continue
+            if timestamp is not None:
+                timestamp -= tz_offset * 3600
+
+        entry = {
+            '_type': 'url_transparent',
+            'ie_key': AbcNewsVideoIE.ie_key(),
+            'url': full_video_url,
+            'id': video_id,
+            'display_id': display_id,
+            'timestamp': timestamp,
+        }
+
+        if youtube_url:
+            entries = [entry, self.url_result(youtube_url, 'Youtube')]
+            return self.playlist_result(entries)
+
+        return entry

youtube_dl/extractor/adobetv.py

@@ -156,7 +156,10 @@ class AdobeTVVideoIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        video_data = self._download_json(url + '?format=json', video_id)
+        webpage = self._download_webpage(url, video_id)
+
+        video_data = self._parse_json(self._search_regex(
+            r'var\s+bridge\s*=\s*([^;]+);', webpage, 'bridged data'), video_id)
 
         formats = [{
             'format_id': '%s-%s' % (determine_ext(source['src']), source.get('height')),

youtube_dl/extractor/adultswim.py

@@ -83,6 +83,20 @@ class AdultSwimIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         }
+    }, {
+        # heroMetadata.trailer
+        'url': 'http://www.adultswim.com/videos/decker/inside-decker-a-new-hero/',
+        'info_dict': {
+            'id': 'I0LQFQkaSUaFp8PnAWHhoQ',
+            'ext': 'mp4',
+            'title': 'Decker - Inside Decker: A New Hero',
+            'description': 'md5:c916df071d425d62d70c86d4399d3ee0',
+            'duration': 249.008,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        }
     }]
 
     @staticmethod
@@ -133,20 +147,26 @@ class AdultSwimIE(InfoExtractor):
             if video_info is None:
                 if bootstrapped_data.get('slugged_video', {}).get('slug') == episode_path:
                     video_info = bootstrapped_data['slugged_video']
-                else:
-                    raise ExtractorError('Unable to find video info')
+            if not video_info:
+                video_info = bootstrapped_data.get('heroMetadata', {}).get('trailer').get('video')
+            if not video_info:
+                raise ExtractorError('Unable to find video info')
 
             show = bootstrapped_data['show']
             show_title = show['title']
             stream = video_info.get('stream')
-            clips = [stream] if stream else video_info.get('clips')
-            if not clips:
+            if stream and stream.get('videoPlaybackID'):
+                segment_ids = [stream['videoPlaybackID']]
+            elif video_info.get('clips'):
+                segment_ids = [clip['videoPlaybackID'] for clip in video_info['clips']]
+            elif video_info.get('videoPlaybackID'):
+                segment_ids = [video_info['videoPlaybackID']]
+            else:
                 raise ExtractorError(
                     'This video is only available via cable service provider subscription that'
                     ' is not currently supported. You may want to use --cookies.'
                     if video_info.get('auth') is True else 'Unable to find stream or clips',
                     expected=True)
-            segment_ids = [clip['videoPlaybackID'] for clip in clips]
 
         episode_id = video_info['id']
         episode_title = video_info['title']

youtube_dl/extractor/aenetworks.py

@@ -2,23 +2,137 @@ from __future__ import unicode_literals
 
 import re
 
-from .common import InfoExtractor
+from .theplatform import ThePlatformIE
 from ..utils import (
     smuggle_url,
     update_url_query,
     unescapeHTML,
+    extract_attributes,
+    get_element_by_attribute,
+)
+from ..compat import (
+    compat_urlparse,
 )
 
 
-class AENetworksIE(InfoExtractor):
+class AENetworksBaseIE(ThePlatformIE):
+    _THEPLATFORM_KEY = 'crazyjava'
+    _THEPLATFORM_SECRET = 's3cr3t'
+
+
+class AENetworksIE(AENetworksBaseIE):
     IE_NAME = 'aenetworks'
     IE_DESC = 'A+E Networks: A&E, Lifetime, History.com, FYI Network'
-    _VALID_URL = r'https?://(?:www\.)?(?:(?:history|aetv|mylifetime)\.com|fyi\.tv)/(?P<type>[^/]+)/(?:[^/]+/)+(?P<id>[^/]+?)(?:$|[?#])'
+    _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:history|aetv|mylifetime)\.com|fyi\.tv)/(?:shows/(?P<show_path>[^/]+(?:/[^/]+){0,2})|movies/(?P<movie_display_id>[^/]+)/full-movie)'
+    _TESTS = [{
+        'url': 'http://www.history.com/shows/mountain-men/season-1/episode-1',
+        'md5': '8ff93eb073449f151d6b90c0ae1ef0c7',
+        'info_dict': {
+            'id': '22253814',
+            'ext': 'mp4',
+            'title': 'Winter Is Coming',
+            'description': 'md5:641f424b7a19d8e24f26dea22cf59d74',
+            'timestamp': 1338306241,
+            'upload_date': '20120529',
+            'uploader': 'AENE-NEW',
+        },
+        'add_ie': ['ThePlatform'],
+    }, {
+        'url': 'http://www.history.com/shows/ancient-aliens/season-1',
+        'info_dict': {
+            'id': '71889446852',
+        },
+        'playlist_mincount': 5,
+    }, {
+        'url': 'http://www.mylifetime.com/shows/atlanta-plastic',
+        'info_dict': {
+            'id': 'SERIES4317',
+            'title': 'Atlanta Plastic',
+        },
+        'playlist_mincount': 2,
+    }, {
+        'url': 'http://www.aetv.com/shows/duck-dynasty/season-9/episode-1',
+        'only_matching': True
+    }, {
+        'url': 'http://www.fyi.tv/shows/tiny-house-nation/season-1/episode-8',
+        'only_matching': True
+    }, {
+        'url': 'http://www.mylifetime.com/shows/project-runway-junior/season-1/episode-6',
+        'only_matching': True
+    }, {
+        'url': 'http://www.mylifetime.com/movies/center-stage-on-pointe/full-movie',
+        'only_matching': True
+    }]
+    _DOMAIN_TO_REQUESTOR_ID = {
+        'history.com': 'HISTORY',
+        'aetv.com': 'AETV',
+        'mylifetime.com': 'LIFETIME',
+        'fyi.tv': 'FYI',
+    }
 
+    def _real_extract(self, url):
+        domain, show_path, movie_display_id = re.match(self._VALID_URL, url).groups()
+        display_id = show_path or movie_display_id
+        webpage = self._download_webpage(url, display_id)
+        if show_path:
+            url_parts = show_path.split('/')
+            url_parts_len = len(url_parts)
+            if url_parts_len == 1:
+                entries = []
+                for season_url_path in re.findall(r'(?s)<li[^>]+data-href="(/shows/%s/season-\d+)"' % url_parts[0], webpage):
+                    entries.append(self.url_result(
+                        compat_urlparse.urljoin(url, season_url_path), 'AENetworks'))
+                return self.playlist_result(
+                    entries, self._html_search_meta('aetn:SeriesId', webpage),
+                    self._html_search_meta('aetn:SeriesTitle', webpage))
+            elif url_parts_len == 2:
+                entries = []
+                for episode_item in re.findall(r'(?s)<div[^>]+class="[^"]*episode-item[^"]*"[^>]*>', webpage):
+                    episode_attributes = extract_attributes(episode_item)
+                    episode_url = compat_urlparse.urljoin(
+                        url, episode_attributes['data-canonical'])
+                    entries.append(self.url_result(
+                        episode_url, 'AENetworks',
+                        episode_attributes['data-videoid']))
+                return self.playlist_result(
+                    entries, self._html_search_meta('aetn:SeasonId', webpage))
+
+        query = {
+            'mbr': 'true',
+            'assetTypes': 'medium_video_s3'
+        }
+        video_id = self._html_search_meta('aetn:VideoID', webpage)
+        media_url = self._search_regex(
+            r"media_url\s*=\s*'([^']+)'", webpage, 'video url')
+        theplatform_metadata = self._download_theplatform_metadata(self._search_regex(
+            r'https?://link.theplatform.com/s/([^?]+)', media_url, 'theplatform_path'), video_id)
+        info = self._parse_theplatform_metadata(theplatform_metadata)
+        if theplatform_metadata.get('AETN$isBehindWall'):
+            requestor_id = self._DOMAIN_TO_REQUESTOR_ID[domain]
+            resource = '<rss version="2.0" xmlns:media="http://search.yahoo.com/mrss/"><channel><title>%s</title><item><title>%s</title><guid>%s</guid><media:rating scheme="urn:v-chip">%s</media:rating></item></channel></rss>' % (requestor_id, theplatform_metadata['title'], theplatform_metadata['AETN$PPL_pplProgramId'], theplatform_metadata['ratings'][0]['rating'])
+            query['auth'] = self._extract_mvpd_auth(
+                url, video_id, requestor_id, resource)
+        info.update(self._search_json_ld(webpage, video_id, fatal=False))
+        media_url = update_url_query(media_url, query)
+        media_url = self._sign_url(media_url, self._THEPLATFORM_KEY, self._THEPLATFORM_SECRET)
+        formats, subtitles = self._extract_theplatform_smil(media_url, video_id)
+        self._sort_formats(formats)
+        info.update({
+            'id': video_id,
+            'formats': formats,
+            'subtitles': subtitles,
+        })
+        return info
+
+
+class HistoryTopicIE(AENetworksBaseIE):
+    IE_NAME = 'history:topic'
+    IE_DESC = 'History.com Topic'
+    _VALID_URL = r'https?://(?:www\.)?history\.com/topics/(?:[^/]+/)?(?P<topic_id>[^/]+)(?:/[^/]+(?:/(?P<video_display_id>[^/?#]+))?)?'
     _TESTS = [{
         'url': 'http://www.history.com/topics/valentines-day/history-of-valentines-day/videos/bet-you-didnt-know-valentines-day?m=528e394da93ae&s=undefined&f=1&free=false',
         'info_dict': {
-            'id': 'g12m5Gyt3fdR',
+            'id': '40700995724',
             'ext': 'mp4',
             'title': "Bet You Didn't Know: Valentine's Day",
             'description': 'md5:7b57ea4829b391995b405fa60bd7b5f7',
@@ -31,57 +145,61 @@ class AENetworksIE(InfoExtractor):
             'skip_download': True,
         },
         'add_ie': ['ThePlatform'],
-        'expected_warnings': ['JSON-LD'],
     }, {
-        'url': 'http://www.history.com/shows/mountain-men/season-1/episode-1',
-        'md5': '8ff93eb073449f151d6b90c0ae1ef0c7',
-        'info_dict': {
-            'id': 'eg47EERs_JsZ',
-            'ext': 'mp4',
-            'title': 'Winter Is Coming',
-            'description': 'md5:641f424b7a19d8e24f26dea22cf59d74',
-            'timestamp': 1338306241,
-            'upload_date': '20120529',
-            'uploader': 'AENE-NEW',
+        'url': 'http://www.history.com/topics/world-war-i/world-war-i-history/videos',
+        'info_dict':
+        {
+            'id': 'world-war-i-history',
+            'title': 'World War I History',
         },
-        'add_ie': ['ThePlatform'],
+        'playlist_mincount': 24,
     }, {
-        'url': 'http://www.aetv.com/shows/duck-dynasty/video/inlawful-entry',
-        'only_matching': True
+        'url': 'http://www.history.com/topics/world-war-i-history/videos',
+        'only_matching': True,
     }, {
-        'url': 'http://www.fyi.tv/shows/tiny-house-nation/videos/207-sq-ft-minnesota-prairie-cottage',
-        'only_matching': True
+        'url': 'http://www.history.com/topics/world-war-i/world-war-i-history',
+        'only_matching': True,
     }, {
-        'url': 'http://www.mylifetime.com/shows/project-runway-junior/video/season-1/episode-6/superstar-clients',
-        'only_matching': True
+        'url': 'http://www.history.com/topics/world-war-i/world-war-i-history/speeches',
+        'only_matching': True,
     }]
 
-    def _real_extract(self, url):
-        page_type, video_id = re.match(self._VALID_URL, url).groups()
-
-        webpage = self._download_webpage(url, video_id)
-
-        video_url_re = [
-            r'data-href="[^"]*/%s"[^>]+data-release-url="([^"]+)"' % video_id,
-            r"media_url\s*=\s*'([^']+)'"
-        ]
-        video_url = unescapeHTML(self._search_regex(video_url_re, webpage, 'video url'))
-        query = {'mbr': 'true'}
-        if page_type == 'shows':
-            query['assetTypes'] = 'medium_video_s3'
-        if 'switch=hds' in video_url:
-            query['switch'] = 'hls'
-
-        info = self._search_json_ld(webpage, video_id, fatal=False)
-        info.update({
+    def theplatform_url_result(self, theplatform_url, video_id, query):
+        return {
             '_type': 'url_transparent',
+            'id': video_id,
             'url': smuggle_url(
-                update_url_query(video_url, query),
+                update_url_query(theplatform_url, query),
                 {
                     'sig': {
-                        'key': 'crazyjava',
-                        'secret': 's3cr3t'},
+                        'key': self._THEPLATFORM_KEY,
+                        'secret': self._THEPLATFORM_SECRET,
+                    },
                     'force_smil_url': True
                 }),
-        })
-        return info
+            'ie_key': 'ThePlatform',
+        }
+
+    def _real_extract(self, url):
+        topic_id, video_display_id = re.match(self._VALID_URL, url).groups()
+        if video_display_id:
+            webpage = self._download_webpage(url, video_display_id)
+            release_url, video_id = re.search(r"_videoPlayer.play\('([^']+)'\s*,\s*'[^']+'\s*,\s*'(\d+)'\)", webpage).groups()
+            release_url = unescapeHTML(release_url)
+
+            return self.theplatform_url_result(
+                release_url, video_id, {
+                    'mbr': 'true',
+                    'switch': 'hls'
+                })
+        else:
+            webpage = self._download_webpage(url, topic_id)
+            entries = []
+            for episode_item in re.findall(r'<a.+?data-release-url="[^"]+"[^>]*>', webpage):
+                video_attributes = extract_attributes(episode_item)
+                entries.append(self.theplatform_url_result(
+                    video_attributes['data-release-url'], video_attributes['data-id'], {
+                        'mbr': 'true',
+                        'switch': 'hls'
+                    }))
+            return self.playlist_result(entries, topic_id, get_element_by_attribute('class', 'show-title', webpage))

youtube_dl/extractor/afreecatv.py

@@ -0,0 +1,133 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_parse_urlparse,
+    compat_urlparse,
+)
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    xpath_element,
+    xpath_text,
+)
+
+
+class AfreecaTVIE(InfoExtractor):
+    IE_DESC = 'afreecatv.com'
+    _VALID_URL = r'''(?x)^
+        https?://(?:(live|afbbs|www)\.)?afreeca(?:tv)?\.com(?::\d+)?
+        (?:
+            /app/(?:index|read_ucc_bbs)\.cgi|
+            /player/[Pp]layer\.(?:swf|html))
+        \?.*?\bnTitleNo=(?P<id>\d+)'''
+    _TESTS = [{
+        'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=',
+        'md5': 'f72c89fe7ecc14c1b5ce506c4996046e',
+        'info_dict': {
+            'id': '36164052',
+            'ext': 'mp4',
+            'title': '데일리 에이프릴 요정들의 시상식!',
+            'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
+            'uploader': 'dailyapril',
+            'uploader_id': 'dailyapril',
+            'upload_date': '20160503',
+        }
+    }, {
+        'url': 'http://afbbs.afreecatv.com:8080/app/read_ucc_bbs.cgi?nStationNo=16711924&nTitleNo=36153164&szBjId=dailyapril&nBbsNo=18605867',
+        'info_dict': {
+            'id': '36153164',
+            'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
+            'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
+            'uploader': 'dailyapril',
+            'uploader_id': 'dailyapril',
+        },
+        'playlist_count': 2,
+        'playlist': [{
+            'md5': 'd8b7c174568da61d774ef0203159bf97',
+            'info_dict': {
+                'id': '36153164_1',
+                'ext': 'mp4',
+                'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
+                'upload_date': '20160502',
+            },
+        }, {
+            'md5': '58f2ce7f6044e34439ab2d50612ab02b',
+            'info_dict': {
+                'id': '36153164_2',
+                'ext': 'mp4',
+                'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
+                'upload_date': '20160502',
+            },
+        }],
+    }, {
+        'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def parse_video_key(key):
+        video_key = {}
+        m = re.match(r'^(?P<upload_date>\d{8})_\w+_(?P<part>\d+)$', key)
+        if m:
+            video_key['upload_date'] = m.group('upload_date')
+            video_key['part'] = m.group('part')
+        return video_key
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        parsed_url = compat_urllib_parse_urlparse(url)
+        info_url = compat_urlparse.urlunparse(parsed_url._replace(
+            netloc='afbbs.afreecatv.com:8080',
+            path='/api/video/get_video_info.php'))
+        video_xml = self._download_xml(info_url, video_id)
+
+        if xpath_element(video_xml, './track/video/file') is None:
+            raise ExtractorError('Specified AfreecaTV video does not exist',
+                                 expected=True)
+
+        title = xpath_text(video_xml, './track/title', 'title')
+        uploader = xpath_text(video_xml, './track/nickname', 'uploader')
+        uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
+        duration = int_or_none(xpath_text(video_xml, './track/duration',
+                                          'duration'))
+        thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
+
+        entries = []
+        for i, video_file in enumerate(video_xml.findall('./track/video/file')):
+            video_key = self.parse_video_key(video_file.get('key', ''))
+            if not video_key:
+                continue
+            entries.append({
+                'id': '%s_%s' % (video_id, video_key.get('part', i + 1)),
+                'title': title,
+                'upload_date': video_key.get('upload_date'),
+                'duration': int_or_none(video_file.get('duration')),
+                'url': video_file.text,
+            })
+
+        info = {
+            'id': video_id,
+            'title': title,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'duration': duration,
+            'thumbnail': thumbnail,
+        }
+
+        if len(entries) > 1:
+            info['_type'] = 'multi_video'
+            info['entries'] = entries
+        elif len(entries) == 1:
+            info['url'] = entries[0]['url']
+            info['upload_date'] = entries[0].get('upload_date')
+        else:
+            raise ExtractorError(
+                'No files found for the specified AfreecaTV video, either'
+                ' the URL is incorrect or the video has been made private.',
+                expected=True)
+
+        return info

youtube_dl/extractor/aftonbladet.py

@@ -24,10 +24,10 @@ class AftonbladetIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         # find internal video meta data
-        meta_url = 'http://aftonbladet-play.drlib.aptoma.no/video/%s.json'
+        meta_url = 'http://aftonbladet-play-metadata.cdn.drvideo.aptoma.no/video/%s.json'
         player_config = self._parse_json(self._html_search_regex(
             r'data-player-config="([^"]+)"', webpage, 'player config'), video_id)
-        internal_meta_id = player_config['videoId']
+        internal_meta_id = player_config['aptomaVideoId']
         internal_meta_url = meta_url % internal_meta_id
         internal_meta_json = self._download_json(
             internal_meta_url, video_id, 'Downloading video meta data')

youtube_dl/extractor/amp.py

@@ -5,6 +5,8 @@ from .common import InfoExtractor
 from ..utils import (
     int_or_none,
     parse_iso8601,
+    mimetype2ext,
+    determine_ext,
 )
 
 
@@ -50,21 +52,25 @@ class AMPIE(InfoExtractor):
         if isinstance(media_content, dict):
             media_content = [media_content]
         for media_data in media_content:
-            media = media_data['@attributes']
-            media_type = media['type']
-            if media_type == 'video/f4m':
+            media = media_data.get('@attributes', {})
+            media_url = media.get('url')
+            if not media_url:
+                continue
+            ext = mimetype2ext(media.get('type')) or determine_ext(media_url)
+            if ext == 'f4m':
                 formats.extend(self._extract_f4m_formats(
-                    media['url'] + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124',
+                    media_url + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124',
                     video_id, f4m_id='hds', fatal=False))
-            elif media_type == 'application/x-mpegURL':
+            elif ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
-                    media['url'], video_id, 'mp4', m3u8_id='hls', fatal=False))
+                    media_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
             else:
                 formats.append({
-                    'format_id': media_data['media-category']['@attributes']['label'],
+                    'format_id': media_data.get('media-category', {}).get('@attributes', {}).get('label'),
                     'url': media['url'],
                     'tbr': int_or_none(media.get('bitrate')),
                     'filesize': int_or_none(media.get('fileSize')),
+                    'ext': ext,
                 })
 
         self._sort_formats(formats)

youtube_dl/extractor/animeondemand.py

@@ -22,6 +22,7 @@ class AnimeOnDemandIE(InfoExtractor):
     _APPLY_HTML5_URL = 'https://www.anime-on-demand.de/html5apply'
     _NETRC_MACHINE = 'animeondemand'
     _TESTS = [{
+        # jap, OmU
         'url': 'https://www.anime-on-demand.de/anime/161',
         'info_dict': {
             'id': '161',
@@ -30,17 +31,21 @@ class AnimeOnDemandIE(InfoExtractor):
         },
         'playlist_mincount': 4,
     }, {
-        # Film wording is used instead of Episode
+        # Film wording is used instead of Episode, ger/jap, Dub/OmU
         'url': 'https://www.anime-on-demand.de/anime/39',
         'only_matching': True,
     }, {
-        # Episodes without titles
+        # Episodes without titles, jap, OmU
         'url': 'https://www.anime-on-demand.de/anime/162',
         'only_matching': True,
     }, {
         # ger/jap, Dub/OmU, account required
         'url': 'https://www.anime-on-demand.de/anime/169',
         'only_matching': True,
+    }, {
+        # Full length film, non-series, ger/jap, Dub/OmU, account required
+        'url': 'https://www.anime-on-demand.de/anime/185',
+        'only_matching': True,
     }]
 
     def _login(self):
@@ -110,35 +115,12 @@ class AnimeOnDemandIE(InfoExtractor):
 
         entries = []
 
-        for num, episode_html in enumerate(re.findall(
-                r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', webpage), 1):
-            episodebox_title = self._search_regex(
-                (r'class="episodebox-title"[^>]+title=(["\'])(?P<title>.+?)\1',
-                 r'class="episodebox-title"[^>]+>(?P<title>.+?)<'),
-                episode_html, 'episodebox title', default=None, group='title')
-            if not episodebox_title:
-                continue
-
-            episode_number = int(self._search_regex(
-                r'(?:Episode|Film)\s*(\d+)',
-                episodebox_title, 'episode number', default=num))
-            episode_title = self._search_regex(
-                r'(?:Episode|Film)\s*\d+\s*-\s*(.+)',
-                episodebox_title, 'episode title', default=None)
-
-            video_id = 'episode-%d' % episode_number
-
-            common_info = {
-                'id': video_id,
-                'series': anime_title,
-                'episode': episode_title,
-                'episode_number': episode_number,
-            }
-
+        def extract_info(html, video_id, num=None):
+            title, description = [None] * 2
             formats = []
 
             for input_ in re.findall(
-                    r'<input[^>]+class=["\'].*?streamstarter_html5[^>]+>', episode_html):
+                    r'<input[^>]+class=["\'].*?streamstarter_html5[^>]+>', html):
                 attributes = extract_attributes(input_)
                 playlist_urls = []
                 for playlist_key in ('data-playlist', 'data-otherplaylist'):
@@ -161,7 +143,7 @@ class AnimeOnDemandIE(InfoExtractor):
                         format_id_list.append(lang)
                     if kind:
                         format_id_list.append(kind)
-                    if not format_id_list:
+                    if not format_id_list and num is not None:
                         format_id_list.append(compat_str(num))
                     format_id = '-'.join(format_id_list)
                     format_note = ', '.join(filter(None, (kind, lang_note)))
@@ -215,28 +197,74 @@ class AnimeOnDemandIE(InfoExtractor):
                             })
                         formats.extend(file_formats)
 
-            if formats:
-                self._sort_formats(formats)
+            return {
+                'title': title,
+                'description': description,
+                'formats': formats,
+            }
+
+        def extract_entries(html, video_id, common_info, num=None):
+            info = extract_info(html, video_id, num)
+
+            if info['formats']:
+                self._sort_formats(info['formats'])
                 f = common_info.copy()
-                f.update({
-                    'title': title,
-                    'description': description,
-                    'formats': formats,
-                })
+                f.update(info)
                 entries.append(f)
 
-            # Extract teaser only when full episode is not available
-            if not formats:
+            # Extract teaser/trailer only when full episode is not available
+            if not info['formats']:
                 m = re.search(
-                    r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>Teaser<',
-                    episode_html)
+                    r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>(?P<kind>Teaser|Trailer)<',
+                    html)
                 if m:
                     f = common_info.copy()
                     f.update({
-                        'id': '%s-teaser' % f['id'],
+                        'id': '%s-%s' % (f['id'], m.group('kind').lower()),
                         'title': m.group('title'),
                         'url': compat_urlparse.urljoin(url, m.group('href')),
                     })
                     entries.append(f)
 
+        def extract_episodes(html):
+            for num, episode_html in enumerate(re.findall(
+                    r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', html), 1):
+                episodebox_title = self._search_regex(
+                    (r'class="episodebox-title"[^>]+title=(["\'])(?P<title>.+?)\1',
+                     r'class="episodebox-title"[^>]+>(?P<title>.+?)<'),
+                    episode_html, 'episodebox title', default=None, group='title')
+                if not episodebox_title:
+                    continue
+
+                episode_number = int(self._search_regex(
+                    r'(?:Episode|Film)\s*(\d+)',
+                    episodebox_title, 'episode number', default=num))
+                episode_title = self._search_regex(
+                    r'(?:Episode|Film)\s*\d+\s*-\s*(.+)',
+                    episodebox_title, 'episode title', default=None)
+
+                video_id = 'episode-%d' % episode_number
+
+                common_info = {
+                    'id': video_id,
+                    'series': anime_title,
+                    'episode': episode_title,
+                    'episode_number': episode_number,
+                }
+
+                extract_entries(episode_html, video_id, common_info)
+
+        def extract_film(html, video_id):
+            common_info = {
+                'id': anime_id,
+                'title': anime_title,
+                'description': anime_description,
+            }
+            extract_entries(html, video_id, common_info)
+
+        extract_episodes(webpage)
+
+        if not entries:
+            extract_film(webpage, anime_id)
+
         return self.playlist_result(entries, anime_id, anime_title, anime_description)

youtube_dl/extractor/anvato.py

@@ -0,0 +1,224 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import base64
+import hashlib
+import json
+import random
+import time
+
+from .common import InfoExtractor
+from ..aes import aes_encrypt
+from ..compat import compat_str
+from ..utils import (
+    bytes_to_intlist,
+    determine_ext,
+    intlist_to_bytes,
+    int_or_none,
+    strip_jsonp,
+)
+
+
+def md5_text(s):
+    if not isinstance(s, compat_str):
+        s = compat_str(s)
+    return hashlib.md5(s.encode('utf-8')).hexdigest()
+
+
+class AnvatoIE(InfoExtractor):
+    # Copied from anvplayer.min.js
+    _ANVACK_TABLE = {
+        'nbcu_nbcd_desktop_web_prod_93d8ead38ce2024f8f544b78306fbd15895ae5e6': 'NNemUkySjxLyPTKvZRiGntBIjEyK8uqicjMakIaQ',
+        'nbcu_nbcd_desktop_web_qa_1a6f01bdd0dc45a439043b694c8a031d': 'eSxJUbA2UUKBTXryyQ2d6NuM8oEqaPySvaPzfKNA',
+        'nbcu_nbcd_desktop_web_acc_eb2ff240a5d4ae9a63d4c297c32716b6c523a129': '89JR3RtUGbvKuuJIiKOMK0SoarLb5MUx8v89RcbP',
+        'nbcu_nbcd_watchvod_web_prod_e61107507180976724ec8e8319fe24ba5b4b60e1': 'Uc7dFt7MJ9GsBWB5T7iPvLaMSOt8BBxv4hAXk5vv',
+        'nbcu_nbcd_watchvod_web_qa_42afedba88a36203db5a4c09a5ba29d045302232': 'T12oDYVFP2IaFvxkmYMy5dKxswpLHtGZa4ZAXEi7',
+        'nbcu_nbcd_watchvod_web_acc_9193214448e2e636b0ffb78abacfd9c4f937c6ca': 'MmobcxUxMedUpohNWwXaOnMjlbiyTOBLL6d46ZpR',
+        'nbcu_local_monitor_web_acc_f998ad54eaf26acd8ee033eb36f39a7b791c6335': 'QvfIoPYrwsjUCcASiw3AIkVtQob2LtJHfidp9iWg',
+        'nbcu_cable_monitor_web_acc_a413759603e8bedfcd3c61b14767796e17834077': 'uwVPJLShvJWSs6sWEIuVem7MTF8A4IknMMzIlFto',
+        'nbcu_nbcd_mcpstage_web_qa_4c43a8f6e95a88dbb40276c0630ba9f693a63a4e': 'PxVYZVwjhgd5TeoPRxL3whssb5OUPnM3zyAzq8GY',
+        'nbcu_comcast_comcast_web_prod_074080762ad4ce956b26b43fb22abf153443a8c4': 'afnaRZfDyg1Z3WZHdupKfy6xrbAG2MHqe3VfuSwh',
+        'nbcu_comcast_comcast_web_qa_706103bb93ead3ef70b1de12a0e95e3c4481ade0': 'DcjsVbX9b3uoPlhdriIiovgFQZVxpISZwz0cx1ZK',
+        'nbcu_comcast_comcastcable_web_prod_669f04817536743563d7331c9293e59fbdbe3d07': '0RwMN2cWy10qhAhOscq3eK7aEe0wqnKt3vJ0WS4D',
+        'nbcu_comcast_comcastcable_web_qa_3d9d2d66219094127f0f6b09cc3c7bb076e3e1ca': '2r8G9DEya7PCqBceKZgrn2XkXgASjwLMuaFE1Aad',
+        'hearst_hearst_demo_web_stage_960726dfef3337059a01a78816e43b29ec04dfc7': 'cuZBPXTR6kSdoTCVXwk5KGA8rk3NrgGn4H6e9Dsp',
+        'anvato_mcpqa_demo_web_stage_18b55e00db5a13faa8d03ae6e41f6f5bcb15b922': 'IOaaLQ8ymqVyem14QuAvE5SndQynTcH5CrLkU2Ih',
+        'anvato_nextmedia_demo_web_stage_9787d56a02ff6b9f43e9a2b0920d8ca88beb5818': 'Pqu9zVzI1ApiIzbVA3VkGBEQHvdKSUuKpD6s2uaR',
+        'anvato_scripps_app_web_prod_0837996dbe373629133857ae9eb72e740424d80a': 'du1ccmn7RxzgizwbWU7hyUaGodNlJn7HtXI0WgXW',
+        'anvato_scripps_app_web_stage_360797e00fe2826be142155c4618cc52fce6c26c': '2PMrQ0BRoqCWl7nzphj0GouIMEh2mZYivAT0S1Su',
+        'fs2go_fs2go_go_all_prod_21934911ccfafc03a075894ead2260d11e2ddd24': 'RcuHlKikW2IJw6HvVoEkqq2UsuEJlbEl11pWXs4Q',
+        'fs2go_fs2go_go_web_prod_ead4b0eec7460c1a07783808db21b49cf1f2f9a7': '4K0HTT2u1zkQA2MaGaZmkLa1BthGSBdr7jllrhk5',
+        'fs2go_fs2go_go_web_stage_407585454a4400355d4391691c67f361': 'ftnc37VKRJBmHfoGGi3kT05bHyeJzilEzhKJCyl3',
+        'fs2go_fs2go_go_android_stage_44b714db6f8477f29afcba15a41e1d30': 'CtxpPvVpo6AbZGomYUhkKs7juHZwNml9b9J0J2gI',
+        'anvato_cbslocal_app_web_prod_547f3e49241ef0e5d30c79b2efbca5d92c698f67': 'Pw0XX5KBDsyRnPS0R2JrSrXftsy8Jnz5pAjaYC8s',
+        'anvato_cbslocal_app_web_stage_547a5f096594cd3e00620c6f825cad1096d28c80': '37OBUhX2uwNyKhhrNzSSNHSRPZpApC3trdqDBpuz',
+        'fs2go_att_att_web_prod_1042dddd089a05438b6a08f972941176f699ffd8': 'JLcF20JwYvpv6uAGcLWIaV12jKwaL1R8us4b6Zkg',
+        'fs2go_att_att_web_stage_807c5001955fc114a3331fe027ddc76e': 'gbu1oO1y0JiOFh4SUipt86P288JHpyjSqolrrT1x',
+        'fs2go_fs2go_tudor_web_prod_a7dd8e5a7cdc830cae55eae6f3e9fee5ee49eb9b': 'ipcp87VCEZXPPe868j3orLqzc03oTy7DXsGkAXXH',
+        'anvato_mhz_app_web_prod_b808218b30de7fdf60340cbd9831512bc1bf6d37': 'Stlm5Gs6BEhJLRTZHcNquyzxGqr23EuFmE5DCgjX',
+        'fs2go_charter_charter_web_stage_c2c6e5a68375a1bf00fff213d3ff8f61a835a54c': 'Lz4hbJp1fwL6jlcz4M2PMzghM4jp4aAmybtT5dPc',
+        'fs2go_charter_charter_web_prod_ebfe3b10f1af215a7321cd3d629e0b81dfa6fa8c': 'vUJsK345A1bVmyYDRhZX0lqFIgVXuqhmuyp1EtPK',
+        'anvato_epfox_app_web_prod_b3373168e12f423f41504f207000188daf88251b': 'GDKq1ixvX3MoBNdU5IOYmYa2DTUXYOozPjrCJnW7',
+        'anvato_epfox_app_web_stage_a3c2ce60f8f83ef374a88b68ee73a950f8ab87ce': '2jz2NH4BsXMaDsoJ5qkHMbcczAfIReo2eFYuVC1C',
+        'fs2go_verizon_verizon_web_stage_08e6df0354a4803f1b1f2428b5a9a382e8dbcd62': 'rKTVapNaAcmnUbGL4ZcuOoY4SE7VmZSQsblPFr7e',
+        'fs2go_verizon_verizon_web_prod_f909564cb606eff1f731b5e22e0928676732c445': 'qLSUuHerM3u9eNPzaHyUK52obai5MvE4XDJfqYe1',
+        'fs2go_foxcom_synd_web_stage_f7b9091f00ea25a4fdaaae77fca5b54cdc7e7043': '96VKF2vLd24fFiDfwPFpzM5llFN4TiIGAlodE0Re',
+        'fs2go_foxcom_synd_web_prod_0f2cdd64d87e4ab6a1d54aada0ff7a7c8387a064': 'agiPjbXEyEZUkbuhcnmVPhe9NNVbDjCFq2xkcx51',
+        'anvato_own_app_web_stage_1214ade5d28422c4dae9d03c1243aba0563c4dba': 'mzhamNac3swG4WsJAiUTacnGIODi6SWeVWk5D7ho',
+        'anvato_own_app_web_prod_944e162ed927ec3e9ed13eb68ed2f1008ee7565e': '9TSxh6G2TXOLBoYm9ro3LdNjjvnXpKb8UR8KoIP9',
+        'anvato_scripps_app_ftv_prod_a10a10468edd5afb16fb48171c03b956176afad1': 'COJ2i2UIPK7xZqIWswxe7FaVBOVgRkP1F6O6qGoH',
+        'anvato_scripps_app_ftv_stage_77d3ad2bdb021ec37ca2e35eb09acd396a974c9a': 'Q7nnopNLe2PPfGLOTYBqxSaRpl209IhqaEuDZi1F',
+        'anvato_univision_app_web_stage_551236ef07a0e17718c3995c35586b5ed8cb5031': 'D92PoLS6UitwxDRA191HUGT9OYcOjV6mPMa5wNyo',
+        'anvato_univision_app_web_prod_039a5c0a6009e637ae8ac906718a79911e0e65e1': '5mVS5u4SQjtw6NGw2uhMbKEIONIiLqRKck5RwQLR',
+        'nbcu_cnbc_springfield_ios_prod_670207fae43d6e9a94c351688851a2ce': 'M7fqCCIP9lW53oJbHs19OlJlpDrVyc2OL8gNeuTa',
+        'nbcu_cnbc_springfieldvod_ios_prod_7a5f04b1ceceb0e9c9e2264a44aa236e08e034c2': 'Yia6QbJahW0S7K1I0drksimhZb4UFq92xLBmmMvk',
+        'anvato_cox_app_web_prod_ce45cda237969f93e7130f50ee8bb6280c1484ab': 'cc0miZexpFtdoqZGvdhfXsLy7FXjRAOgb9V0f5fZ',
+        'anvato_cox_app_web_stage_c23dbe016a8e9d8c7101d10172b92434f6088bf9': 'yivU3MYHd2eDZcOfmLbINVtqxyecKTOp8OjOuoGJ',
+        'anvato_chnzero_app_web_stage_b1164d1352b579e792e542fddf13ee34c0eeb46b': 'A76QkXMmVH8lTCfU15xva1mZnSVcqeY4Xb22Kp7m',
+        'anvato_chnzero_app_web_prod_253d358928dc08ec161eda2389d53707288a730c': 'OA5QI3ZWZZkdtUEDqh28AH8GedsF6FqzJI32596b',
+        'anvato_discovery_vodpoc_web_stage_9fa7077b5e8af1f8355f65d4fb8d2e0e9d54e2b7': 'q3oT191tTQ5g3JCP67PkjLASI9s16DuWZ6fYmry3',
+        'anvato_discovery_vodpoc_web_prod_688614983167a1af6cdf6d76343fda10a65223c1': 'qRvRQCTVHd0VVOHsMvvfidyWmlYVrTbjby7WqIuK',
+        'nbcu_cnbc_springfieldvod_ftv_stage_826040aad1925a46ac5dfb4b3c5143e648c6a30d': 'JQaSb5a8Tz0PT4ti329DNmzDO30TnngTHmvX8Vua',
+        'nbcu_cnbc_springfield_ftv_stage_826040aad1925a46ac5dfb4b3c5143e648c6a30d': 'JQaSb5a8Tz0PT4ti329DNmzDO30TnngTHmvX8Vua',
+        'nbcu_nbcd_capture_web_stage_4dd9d585bfb984ebf856dee35db027b2465cc4ae': '0j1Ov4Vopyi2HpBZJYdL2m8ERJVGYh3nNpzPiO8F',
+        'nbcu_nbcd_watch3_android_prod_7712ca5fcf1c22f19ec1870a9650f9c37db22dcf': '3LN2UB3rPUAMu7ZriWkHky9vpLMXYha8JbSnxBlx',
+        'nbcu_nbcd_watchvod3_android_prod_0910a3a4692d57c0b5ff4316075bc5d096be45b9': 'mJagcQ2II30vUOAauOXne7ERwbf5S9nlB3IP17lQ',
+        'anvato_scripps_app_atv_prod_790deda22e16e71e83df58f880cd389908a45d52': 'CB6trI1mpoDIM5o54DNTsji90NDBQPZ4z4RqBNSH',
+        'nbcu_nbcd_watchv4_android_prod_ff67cef9cb409158c6f8c3533edddadd0b750507': 'j8CHQCUWjlYERj4NFRmUYOND85QNbHViH09UwuKm',
+        'nbcu_nbcd_watchvodv4_android_prod_a814d781609989dea6a629d50ae4c7ad8cc8e907': 'rkVnUXxdA9rawVLUlDQtMue9Y4Q7lFEaIotcUhjt',
+        'rvVKpA50qlOPLFxMjrCGf5pdkdQDm7qn': '1J7ZkY5Qz5lMLi93QOH9IveE7EYB3rLl',
+        'nbcu_dtv_local_web_prod_b266cf49defe255fd4426a97e27c09e513e9f82f': 'HuLnJDqzLa4saCzYMJ79zDRSQpEduw1TzjMNQu2b',
+        'nbcu_att_local_web_prod_4cef038b2d969a6b7d700a56a599040b6a619f67': 'Q0Em5VDc2KpydUrVwzWRXAwoNBulWUxCq2faK0AV',
+        'nbcu_dish_local_web_prod_c56dcaf2da2e9157a4266c82a78195f1dd570f6b': 'bC1LWmRz9ayj2AlzizeJ1HuhTfIaJGsDBnZNgoRg',
+        'nbcu_verizon_local_web_prod_88bebd2ce006d4ed980de8133496f9a74cb9b3e1': 'wzhDKJZpgvUSS1EQvpCQP8Q59qVzcPixqDGJefSk',
+        'nbcu_charter_local_web_prod_9ad90f7fc4023643bb718f0fe0fd5beea2382a50': 'PyNbxNhEWLzy1ZvWEQelRuIQY88Eub7xbSVRMdfT',
+        'nbcu_suddenlink_local_web_prod_20fb711725cac224baa1c1cb0b1c324d25e97178': '0Rph41lPXZbb3fqeXtHjjbxfSrNbtZp1Ygq7Jypa',
+        'nbcu_wow_local_web_prod_652d9ce4f552d9c2e7b5b1ed37b8cb48155174ad': 'qayIBZ70w1dItm2zS42AptXnxW15mkjRrwnBjMPv',
+        'nbcu_centurylink_local_web_prod_2034402b029bf3e837ad46814d9e4b1d1345ccd5': 'StePcPMkjsX51PcizLdLRMzxMEl5k2FlsMLUNV4k',
+        'nbcu_atlanticbrd_local_web_prod_8d5f5ecbf7f7b2f5e6d908dd75d90ae3565f682e': 'NtYLb4TFUS0pRs3XTkyO5sbVGYjVf17bVbjaGscI',
+        'nbcu_nbcd_watchvod_web_dev_08bc05699be47c4f31d5080263a8cfadc16d0f7c': 'hwxi2dgDoSWgfmVVXOYZm14uuvku4QfopstXckhr',
+        'anvato_nextmedia_app_web_prod_a4fa8c7204aa65e71044b57aaf63711980cfe5a0': 'tQN1oGPYY1nM85rJYePWGcIb92TG0gSqoVpQTWOw',
+        'anvato_mcp_lin_web_prod_4c36fbfd4d8d8ecae6488656e21ac6d1ac972749': 'GUXNf5ZDX2jFUpu4WT2Go4DJ5nhUCzpnwDRRUx1K',
+        'anvato_mcp_univision_web_prod_37fe34850c99a3b5cdb71dab10a417dd5cdecafa': 'bLDYF8JqfG42b7bwKEgQiU9E2LTIAtnKzSgYpFUH',
+        'anvato_mcp_fs2go_web_prod_c7b90a93e171469cdca00a931211a2f556370d0a': 'icgGoYGipQMMSEvhplZX1pwbN69srwKYWksz3xWK',
+        'anvato_mcp_sps_web_prod_54bdc90dd6ba21710e9f7074338365bba28da336': 'fA2iQdI7RDpynqzQYIpXALVS83NTPr8LLFK4LFsu',
+        'anvato_mcp_anv_web_prod_791407490f4c1ef2a4bcb21103e0cb1bcb3352b3': 'rMOUZqe9lwcGq2mNgG3EDusm6lKgsUnczoOX3mbg',
+        'anvato_mcp_gray_web_prod_4c10f067c393ed8fc453d3930f8ab2b159973900': 'rMOUZqe9lwcGq2mNgG3EDusm6lKgsUnczoOX3mbg',
+        'anvato_mcp_hearst_web_prod_5356c3de0fc7c90a3727b4863ca7fec3a4524a99': 'P3uXJ0fXXditBPCGkfvlnVScpPEfKmc64Zv7ZgbK',
+        'anvato_mcp_cbs_web_prod_02f26581ff80e5bda7aad28226a8d369037f2cbe': 'mGPvo5ZA5SgjOFAPEPXv7AnOpFUICX8hvFQVz69n',
+        'anvato_mcp_telemundo_web_prod_c5278d51ad46fda4b6ca3d0ea44a7846a054f582': 'qyT6PXXLjVNCrHaRVj0ugAhalNRS7Ee9BP7LUokD',
+        'nbcu_nbcd_watchvodv4_web_stage_4108362fba2d4ede21f262fea3c4162cbafd66c7': 'DhaU5lj0W2gEdcSSsnxURq8t7KIWtJfD966crVDk',
+        'anvato_scripps_app_ios_prod_409c41960c60b308db43c3cc1da79cab9f1c3d93': 'WPxj5GraLTkYCyj3M7RozLqIycjrXOEcDGFMIJPn',
+        'EZqvRyKBJLrgpClDPDF8I7Xpdp40Vx73': '4OxGd2dEakylntVKjKF0UK9PDPYB6A9W',
+        'M2v78QkpleXm9hPp9jUXI63x5vA6BogR': 'ka6K32k7ZALmpINkjJUGUo0OE42Md1BQ',
+        'nbcu_nbcd_desktop_web_prod_93d8ead38ce2024f8f544b78306fbd15895ae5e6_secure': 'NNemUkySjxLyPTKvZRiGntBIjEyK8uqicjMakIaQ'
+    }
+
+    _AUTH_KEY = b'\x31\xc2\x42\x84\x9e\x73\xa0\xce'
+
+    def __init__(self, *args, **kwargs):
+        super(AnvatoIE, self).__init__(*args, **kwargs)
+        self.__server_time = None
+
+    def _server_time(self, access_key, video_id):
+        if self.__server_time is not None:
+            return self.__server_time
+
+        self.__server_time = int(self._download_json(
+            self._api_prefix(access_key) + 'server_time?anvack=' + access_key, video_id,
+            note='Fetching server time')['server_time'])
+
+        return self.__server_time
+
+    def _api_prefix(self, access_key):
+        return 'https://tkx2-%s.anvato.net/rest/v2/' % ('prod' if 'prod' in access_key else 'stage')
+
+    def _get_video_json(self, access_key, video_id):
+        # See et() in anvplayer.min.js, which is an alias of getVideoJSON()
+        video_data_url = self._api_prefix(access_key) + 'mcp/video/%s?anvack=%s' % (video_id, access_key)
+        server_time = self._server_time(access_key, video_id)
+        input_data = '%d~%s~%s' % (server_time, md5_text(video_data_url), md5_text(server_time))
+
+        auth_secret = intlist_to_bytes(aes_encrypt(
+            bytes_to_intlist(input_data[:64]), bytes_to_intlist(self._AUTH_KEY)))
+
+        video_data_url += '&X-Anvato-Adst-Auth=' + base64.b64encode(auth_secret).decode('ascii')
+        anvrid = md5_text(time.time() * 1000 * random.random())[:30]
+        payload = {
+            'api': {
+                'anvrid': anvrid,
+                'anvstk': md5_text('%s|%s|%d|%s' % (
+                    access_key, anvrid, server_time, self._ANVACK_TABLE[access_key])),
+                'anvts': server_time,
+            },
+        }
+
+        return self._download_json(
+            video_data_url, video_id, transform_source=strip_jsonp,
+            data=json.dumps(payload).encode('utf-8'))
+
+    def _extract_anvato_videos(self, webpage, video_id):
+        anvplayer_data = self._parse_json(self._html_search_regex(
+            r'<script[^>]+data-anvp=\'([^\']+)\'', webpage,
+            'Anvato player data'), video_id)
+
+        video_id = anvplayer_data['video']
+        access_key = anvplayer_data['accessKey']
+
+        video_data = self._get_video_json(access_key, video_id)
+
+        formats = []
+        for published_url in video_data['published_urls']:
+            video_url = published_url['embed_url']
+            ext = determine_ext(video_url)
+
+            if ext == 'smil':
+                formats.extend(self._extract_smil_formats(video_url, video_id))
+                continue
+
+            tbr = int_or_none(published_url.get('kbps'))
+            a_format = {
+                'url': video_url,
+                'format_id': ('-'.join(filter(None, ['http', published_url.get('cdn_name')]))).lower(),
+                'tbr': tbr if tbr != 0 else None,
+            }
+
+            if ext == 'm3u8':
+                # Not using _extract_m3u8_formats here as individual media
+                # playlists are also included in published_urls.
+                if tbr is None:
+                    formats.append(self._m3u8_meta_format(video_url, ext='mp4', m3u8_id='hls'))
+                    continue
+                else:
+                    a_format.update({
+                        'format_id': '-'.join(filter(None, ['hls', compat_str(tbr)])),
+                        'ext': 'mp4',
+                    })
+            elif ext == 'mp3':
+                a_format['vcodec'] = 'none'
+            else:
+                a_format.update({
+                    'width': int_or_none(published_url.get('width')),
+                    'height': int_or_none(published_url.get('height')),
+                })
+            formats.append(a_format)
+
+        self._sort_formats(formats)
+
+        subtitles = {}
+        for caption in video_data.get('captions', []):
+            a_caption = {
+                'url': caption['url'],
+                'ext': 'tt' if caption.get('format') == 'SMPTE-TT' else None
+            }
+            subtitles.setdefault(caption['language'], []).append(a_caption)
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': video_data.get('def_title'),
+            'description': video_data.get('def_description'),
+            'categories': video_data.get('categories'),
+            'thumbnail': video_data.get('thumbnail'),
+            'subtitles': subtitles,
+        }

youtube_dl/extractor/aol.py

@@ -12,7 +12,7 @@ from ..utils import (
 
 class AolIE(InfoExtractor):
     IE_NAME = 'on.aol.com'
-    _VALID_URL = r'(?:aol-video:|https?://on\.aol\.com/.*-)(?P<id>[^/?-]+)'
+    _VALID_URL = r'(?:aol-video:|https?://on\.aol\.com/(?:[^/]+/)*(?:[^/?#&]+-)?)(?P<id>[^/?#&]+)'
 
     _TESTS = [{
         # video with 5min ID
@@ -53,6 +53,12 @@ class AolIE(InfoExtractor):
     }, {
         'url': 'http://on.aol.com/shows/park-bench-shw518173474-559a1b9be4b0c3bfad3357a7?context=SH:SHW518173474:PL4327:1460619712763',
         'only_matching': True,
+    }, {
+        'url': 'http://on.aol.com/video/519442220',
+        'only_matching': True,
+    }, {
+        'url': 'aol-video:5707d6b8e4b090497b04f706',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/appletrailers.py

@@ -7,6 +7,8 @@ from .common import InfoExtractor
 from ..compat import compat_urlparse
 from ..utils import (
     int_or_none,
+    parse_duration,
+    unified_strdate,
 )
 
 
@@ -16,7 +18,8 @@ class AppleTrailersIE(InfoExtractor):
     _TESTS = [{
         'url': 'http://trailers.apple.com/trailers/wb/manofsteel/',
         'info_dict': {
-            'id': 'manofsteel',
+            'id': '5111',
+            'title': 'Man of Steel',
         },
         'playlist': [
             {
@@ -70,6 +73,15 @@ class AppleTrailersIE(InfoExtractor):
             'id': 'blackthorn',
         },
         'playlist_mincount': 2,
+        'expected_warnings': ['Unable to download JSON metadata'],
+    }, {
+        # json data only available from http://trailers.apple.com/trailers/feeds/data/15881.json
+        'url': 'http://trailers.apple.com/trailers/fox/kungfupanda3/',
+        'info_dict': {
+            'id': '15881',
+            'title': 'Kung Fu Panda 3',
+        },
+        'playlist_mincount': 4,
     }, {
         'url': 'http://trailers.apple.com/ca/metropole/autrui/',
         'only_matching': True,
@@ -85,6 +97,45 @@ class AppleTrailersIE(InfoExtractor):
         movie = mobj.group('movie')
         uploader_id = mobj.group('company')
 
+        webpage = self._download_webpage(url, movie)
+        film_id = self._search_regex(r"FilmId\s*=\s*'(\d+)'", webpage, 'film id')
+        film_data = self._download_json(
+            'http://trailers.apple.com/trailers/feeds/data/%s.json' % film_id,
+            film_id, fatal=False)
+
+        if film_data:
+            entries = []
+            for clip in film_data.get('clips', []):
+                clip_title = clip['title']
+
+                formats = []
+                for version, version_data in clip.get('versions', {}).items():
+                    for size, size_data in version_data.get('sizes', {}).items():
+                        src = size_data.get('src')
+                        if not src:
+                            continue
+                        formats.append({
+                            'format_id': '%s-%s' % (version, size),
+                            'url': re.sub(r'_(\d+p.mov)', r'_h\1', src),
+                            'width': int_or_none(size_data.get('width')),
+                            'height': int_or_none(size_data.get('height')),
+                            'language': version[:2],
+                        })
+                self._sort_formats(formats)
+
+                entries.append({
+                    'id': movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', clip_title).lower(),
+                    'formats': formats,
+                    'title': clip_title,
+                    'thumbnail': clip.get('screen') or clip.get('thumb'),
+                    'duration': parse_duration(clip.get('runtime') or clip.get('faded')),
+                    'upload_date': unified_strdate(clip.get('posted')),
+                    'uploader_id': uploader_id,
+                })
+
+            page_data = film_data.get('page', {})
+            return self.playlist_result(entries, film_id, page_data.get('movie_title'))
+
         playlist_url = compat_urlparse.urljoin(url, 'includes/playlists/itunes.inc')
 
         def fix_html(s):

youtube_dl/extractor/archiveorg.py

@@ -1,67 +1,65 @@
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
-from ..utils import unified_strdate
+from .jwplatform import JWPlatformBaseIE
+from ..utils import (
+    unified_strdate,
+    clean_html,
+)
 
 
-class ArchiveOrgIE(InfoExtractor):
+class ArchiveOrgIE(JWPlatformBaseIE):
     IE_NAME = 'archive.org'
     IE_DESC = 'archive.org videos'
-    _VALID_URL = r'https?://(?:www\.)?archive\.org/details/(?P<id>[^?/]+)(?:[?].*)?$'
+    _VALID_URL = r'https?://(?:www\.)?archive\.org/(?:details|embed)/(?P<id>[^/?#]+)(?:[?].*)?$'
     _TESTS = [{
         'url': 'http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect',
         'md5': '8af1d4cf447933ed3c7f4871162602db',
         'info_dict': {
             'id': 'XD300-23_68HighlightsAResearchCntAugHumanIntellect',
-            'ext': 'ogv',
+            'ext': 'ogg',
             'title': '1968 Demo - FJCC Conference Presentation Reel #1',
-            'description': 'md5:1780b464abaca9991d8968c877bb53ed',
+            'description': 'md5:da45c349df039f1cc8075268eb1b5c25',
             'upload_date': '19681210',
             'uploader': 'SRI International'
         }
     }, {
         'url': 'https://archive.org/details/Cops1922',
-        'md5': '18f2a19e6d89af8425671da1cf3d4e04',
+        'md5': 'bc73c8ab3838b5a8fc6c6651fa7b58ba',
         'info_dict': {
             'id': 'Cops1922',
-            'ext': 'ogv',
+            'ext': 'mp4',
             'title': 'Buster Keaton\'s "Cops" (1922)',
-            'description': 'md5:70f72ee70882f713d4578725461ffcc3',
+            'description': 'md5:b4544662605877edd99df22f9620d858',
         }
+    }, {
+        'url': 'http://archive.org/embed/XD300-23_68HighlightsAResearchCntAugHumanIntellect',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+        webpage = self._download_webpage(
+            'http://archive.org/embed/' + video_id, video_id)
+        jwplayer_playlist = self._parse_json(self._search_regex(
+            r"(?s)Play\('[^']+'\s*,\s*(\[.+\])\s*,\s*{.*?}\);",
+            webpage, 'jwplayer playlist'), video_id)
+        info = self._parse_jwplayer_data(
+            {'playlist': jwplayer_playlist}, video_id, base_url=url)
 
-        json_url = url + ('&' if '?' in url else '?') + 'output=json'
-        data = self._download_json(json_url, video_id)
+        def get_optional(metadata, field):
+            return metadata.get(field, [None])[0]
 
-        def get_optional(data_dict, field):
-            return data_dict['metadata'].get(field, [None])[0]
-
-        title = get_optional(data, 'title')
-        description = get_optional(data, 'description')
-        uploader = get_optional(data, 'creator')
-        upload_date = unified_strdate(get_optional(data, 'date'))
-
-        formats = [
-            {
-                'format': fdata['format'],
-                'url': 'http://' + data['server'] + data['dir'] + fn,
-                'file_size': int(fdata['size']),
-            }
-            for fn, fdata in data['files'].items()
-            if 'Video' in fdata['format']]
-
-        self._sort_formats(formats)
-
-        return {
-            '_type': 'video',
-            'id': video_id,
-            'title': title,
-            'formats': formats,
-            'description': description,
-            'uploader': uploader,
-            'upload_date': upload_date,
-            'thumbnail': data.get('misc', {}).get('image'),
-        }
+        metadata = self._download_json(
+            'http://archive.org/details/' + video_id, video_id, query={
+                'output': 'json',
+            })['metadata']
+        info.update({
+            'title': get_optional(metadata, 'title') or info.get('title'),
+            'description': clean_html(get_optional(metadata, 'description')),
+        })
+        if info.get('_type') != 'playlist':
+            info.update({
+                'uploader': get_optional(metadata, 'creator'),
+                'upload_date': unified_strdate(get_optional(metadata, 'date')),
+            })
+        return info

youtube_dl/extractor/ard.py

@@ -8,19 +8,19 @@ from .generic import GenericIE
 from ..utils import (
     determine_ext,
     ExtractorError,
-    get_element_by_attribute,
     qualities,
     int_or_none,
     parse_duration,
     unified_strdate,
     xpath_text,
+    update_url_query,
 )
 from ..compat import compat_etree_fromstring
 
 
 class ARDMediathekIE(InfoExtractor):
     IE_NAME = 'ARD:mediathek'
-    _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
+    _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.(?:daserste|rbb-online)\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
 
     _TESTS = [{
         'url': 'http://www.ardmediathek.de/tv/Dokumentation-und-Reportage/Ich-liebe-das-Leben-trotzdem/rbb-Fernsehen/Video?documentId=29582122&bcastId=3822114',
@@ -35,6 +35,7 @@ class ARDMediathekIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         },
+        'skip': 'HTTP Error 404: Not Found',
     }, {
         'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916',
         'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e',
@@ -45,6 +46,7 @@ class ARDMediathekIE(InfoExtractor):
             'description': 'md5:196392e79876d0ac94c94e8cdb2875f1',
             'duration': 5252,
         },
+        'skip': 'HTTP Error 404: Not Found',
     }, {
         # audio
         'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
@@ -56,9 +58,22 @@ class ARDMediathekIE(InfoExtractor):
             'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef',
             'duration': 3240,
         },
+        'skip': 'HTTP Error 404: Not Found',
     }, {
         'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
         'only_matching': True,
+    }, {
+        # audio
+        'url': 'http://mediathek.rbb-online.de/radio/Hörspiel/Vor-dem-Fest/kulturradio/Audio?documentId=30796318&topRessort=radio&bcastId=9839158',
+        'md5': '4e8f00631aac0395fee17368ac0e9867',
+        'info_dict': {
+            'id': '30796318',
+            'ext': 'mp3',
+            'title': 'Vor dem Fest',
+            'description': 'md5:c0c1c8048514deaed2a73b3a60eecacb',
+            'duration': 3287,
+        },
+        'skip': 'Video is no longer available',
     }]
 
     def _extract_media_info(self, media_info_url, webpage, video_id):
@@ -114,11 +129,14 @@ class ARDMediathekIE(InfoExtractor):
                         continue
                     if ext == 'f4m':
                         formats.extend(self._extract_f4m_formats(
-                            stream_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124',
-                            video_id, preference=-1, f4m_id='hds', fatal=False))
+                            update_url_query(stream_url, {
+                                'hdcore': '3.1.1',
+                                'plugin': 'aasp-3.1.1.69.124'
+                            }),
+                            video_id, f4m_id='hds', fatal=False))
                     elif ext == 'm3u8':
                         formats.extend(self._extract_m3u8_formats(
-                            stream_url, video_id, 'mp4', preference=1, m3u8_id='hls', fatal=False))
+                            stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
                     else:
                         if server and server.startswith('rtmp'):
                             f = {
@@ -232,7 +250,8 @@ class ARDIE(InfoExtractor):
             'title': 'Die Story im Ersten: Mission unter falscher Flagge',
             'upload_date': '20140804',
             'thumbnail': 're:^https?://.*\.jpg$',
-        }
+        },
+        'skip': 'HTTP Error 404: Not Found',
     }
 
     def _real_extract(self, url):
@@ -274,41 +293,3 @@ class ARDIE(InfoExtractor):
             'upload_date': upload_date,
             'thumbnail': thumbnail,
         }
-
-
-class SportschauIE(ARDMediathekIE):
-    IE_NAME = 'Sportschau'
-    _VALID_URL = r'(?P<baseurl>https?://(?:www\.)?sportschau\.de/(?:[^/]+/)+video(?P<id>[^/#?]+))\.html'
-    _TESTS = [{
-        'url': 'http://www.sportschau.de/tourdefrance/videoseppeltkokainhatnichtsmitklassischemdopingzutun100.html',
-        'info_dict': {
-            'id': 'seppeltkokainhatnichtsmitklassischemdopingzutun100',
-            'ext': 'mp4',
-            'title': 'Seppelt: "Kokain hat nichts mit klassischem Doping zu tun"',
-            'thumbnail': 're:^https?://.*\.jpg$',
-            'description': 'Der ARD-Doping Experte Hajo Seppelt gibt seine Einschätzung zum ersten Dopingfall der diesjährigen Tour de France um den Italiener Luca Paolini ab.',
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        },
-    }]
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        base_url = mobj.group('baseurl')
-
-        webpage = self._download_webpage(url, video_id)
-        title = get_element_by_attribute('class', 'headline', webpage)
-        description = self._html_search_meta('description', webpage, 'description')
-
-        info = self._extract_media_info(
-            base_url + '-mc_defaultQuality-h.json', webpage, video_id)
-
-        info.update({
-            'title': title,
-            'description': description,
-        })
-
-        return info

youtube_dl/extractor/arkena.py

@@ -0,0 +1,115 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    float_or_none,
+    int_or_none,
+    mimetype2ext,
+    parse_iso8601,
+    strip_jsonp,
+)
+
+
+class ArkenaIE(InfoExtractor):
+    _VALID_URL = r'https?://play\.arkena\.com/(?:config|embed)/avp/v\d/player/media/(?P<id>[^/]+)/[^/]+/(?P<account_id>\d+)'
+    _TESTS = [{
+        'url': 'https://play.arkena.com/embed/avp/v2/player/media/b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe/1/129411',
+        'md5': 'b96f2f71b359a8ecd05ce4e1daa72365',
+        'info_dict': {
+            'id': 'b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe',
+            'ext': 'mp4',
+            'title': 'Big Buck Bunny',
+            'description': 'Royalty free test video',
+            'timestamp': 1432816365,
+            'upload_date': '20150528',
+            'is_live': False,
+        },
+    }, {
+        'url': 'https://play.arkena.com/config/avp/v2/player/media/b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe/1/129411/?callbackMethod=jQuery1111023664739129262213_1469227693893',
+        'only_matching': True,
+    }, {
+        'url': 'http://play.arkena.com/config/avp/v1/player/media/327336/darkmatter/131064/?callbackMethod=jQuery1111002221189684892677_1469227595972',
+        'only_matching': True,
+    }, {
+        'url': 'http://play.arkena.com/embed/avp/v1/player/media/327336/darkmatter/131064/',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_url(webpage):
+        # See https://support.arkena.com/display/PLAY/Ways+to+embed+your+video
+        mobj = re.search(
+            r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//play\.arkena\.com/embed/avp/.+?)\1',
+            webpage)
+        if mobj:
+            return mobj.group('url')
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        account_id = mobj.group('account_id')
+
+        playlist = self._download_json(
+            'https://play.arkena.com/config/avp/v2/player/media/%s/0/%s/?callbackMethod=_'
+            % (video_id, account_id),
+            video_id, transform_source=strip_jsonp)['Playlist'][0]
+
+        media_info = playlist['MediaInfo']
+        title = media_info['Title']
+        media_files = playlist['MediaFiles']
+
+        is_live = False
+        formats = []
+        for kind_case, kind_formats in media_files.items():
+            kind = kind_case.lower()
+            for f in kind_formats:
+                f_url = f.get('Url')
+                if not f_url:
+                    continue
+                is_live = f.get('Live') == 'true'
+                exts = (mimetype2ext(f.get('Type')), determine_ext(f_url, None))
+                if kind == 'm3u8' or 'm3u8' in exts:
+                    formats.extend(self._extract_m3u8_formats(
+                        f_url, video_id, 'mp4',
+                        entry_protocol='m3u8' if is_live else 'm3u8_native',
+                        m3u8_id=kind, fatal=False, live=is_live))
+                elif kind == 'flash' or 'f4m' in exts:
+                    formats.extend(self._extract_f4m_formats(
+                        f_url, video_id, f4m_id=kind, fatal=False))
+                elif kind == 'dash' or 'mpd' in exts:
+                    formats.extend(self._extract_mpd_formats(
+                        f_url, video_id, mpd_id=kind, fatal=False))
+                elif kind == 'silverlight':
+                    # TODO: process when ism is supported (see
+                    # https://github.com/rg3/youtube-dl/issues/8118)
+                    continue
+                else:
+                    tbr = float_or_none(f.get('Bitrate'), 1000)
+                    formats.append({
+                        'url': f_url,
+                        'format_id': '%s-%d' % (kind, tbr) if tbr else kind,
+                        'tbr': tbr,
+                    })
+        self._sort_formats(formats)
+
+        description = media_info.get('Description')
+        video_id = media_info.get('VideoId') or video_id
+        timestamp = parse_iso8601(media_info.get('PublishDate'))
+        thumbnails = [{
+            'url': thumbnail['Url'],
+            'width': int_or_none(thumbnail.get('Size')),
+        } for thumbnail in (media_info.get('Poster') or []) if thumbnail.get('Url')]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'timestamp': timestamp,
+            'is_live': is_live,
+            'thumbnails': thumbnails,
+            'formats': formats,
+        }

youtube_dl/extractor/arte.py

@@ -61,10 +61,7 @@ class ArteTvIE(InfoExtractor):
         }
 
 
-class ArteTVPlus7IE(InfoExtractor):
-    IE_NAME = 'arte.tv:+7'
-    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&+])'
-
+class ArteTVBaseIE(InfoExtractor):
     @classmethod
     def _extract_url_info(cls, url):
         mobj = re.match(cls._VALID_URL, url)
@@ -78,6 +75,125 @@ class ArteTVPlus7IE(InfoExtractor):
             video_id = mobj.group('id')
         return video_id, lang
 
+    def _extract_from_json_url(self, json_url, video_id, lang, title=None):
+        info = self._download_json(json_url, video_id)
+        player_info = info['videoJsonPlayer']
+
+        upload_date_str = player_info.get('shootingDate')
+        if not upload_date_str:
+            upload_date_str = (player_info.get('VRA') or player_info.get('VDA') or '').split(' ')[0]
+
+        title = (player_info.get('VTI') or title or player_info['VID']).strip()
+        subtitle = player_info.get('VSU', '').strip()
+        if subtitle:
+            title += ' - %s' % subtitle
+
+        info_dict = {
+            'id': player_info['VID'],
+            'title': title,
+            'description': player_info.get('VDE'),
+            'upload_date': unified_strdate(upload_date_str),
+            'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
+        }
+        qfunc = qualities(['HQ', 'MQ', 'EQ', 'SQ'])
+
+        LANGS = {
+            'fr': 'F',
+            'de': 'A',
+            'en': 'E[ANG]',
+            'es': 'E[ESP]',
+        }
+
+        langcode = LANGS.get(lang, lang)
+
+        formats = []
+        for format_id, format_dict in player_info['VSR'].items():
+            f = dict(format_dict)
+            versionCode = f.get('versionCode')
+            l = re.escape(langcode)
+
+            # Language preference from most to least priority
+            # Reference: section 5.6.3 of
+            # http://www.arte.tv/sites/en/corporate/files/complete-technical-guidelines-arte-geie-v1-05.pdf
+            PREFERENCES = (
+                # original version in requested language, without subtitles
+                r'VO{0}$'.format(l),
+                # original version in requested language, with partial subtitles in requested language
+                r'VO{0}-ST{0}$'.format(l),
+                # original version in requested language, with subtitles for the deaf and hard-of-hearing in requested language
+                r'VO{0}-STM{0}$'.format(l),
+                # non-original (dubbed) version in requested language, without subtitles
+                r'V{0}$'.format(l),
+                # non-original (dubbed) version in requested language, with subtitles partial subtitles in requested language
+                r'V{0}-ST{0}$'.format(l),
+                # non-original (dubbed) version in requested language, with subtitles for the deaf and hard-of-hearing in requested language
+                r'V{0}-STM{0}$'.format(l),
+                # original version in requested language, with partial subtitles in different language
+                r'VO{0}-ST(?!{0}).+?$'.format(l),
+                # original version in requested language, with subtitles for the deaf and hard-of-hearing in different language
+                r'VO{0}-STM(?!{0}).+?$'.format(l),
+                # original version in different language, with partial subtitles in requested language
+                r'VO(?:(?!{0}).+?)?-ST{0}$'.format(l),
+                # original version in different language, with subtitles for the deaf and hard-of-hearing in requested language
+                r'VO(?:(?!{0}).+?)?-STM{0}$'.format(l),
+                # original version in different language, without subtitles
+                r'VO(?:(?!{0}))?$'.format(l),
+                # original version in different language, with partial subtitles in different language
+                r'VO(?:(?!{0}).+?)?-ST(?!{0}).+?$'.format(l),
+                # original version in different language, with subtitles for the deaf and hard-of-hearing in different language
+                r'VO(?:(?!{0}).+?)?-STM(?!{0}).+?$'.format(l),
+            )
+
+            for pref, p in enumerate(PREFERENCES):
+                if re.match(p, versionCode):
+                    lang_pref = len(PREFERENCES) - pref
+                    break
+            else:
+                lang_pref = -1
+
+            format = {
+                'format_id': format_id,
+                'preference': -10 if f.get('videoFormat') == 'M3U8' else None,
+                'language_preference': lang_pref,
+                'format_note': '%s, %s' % (f.get('versionCode'), f.get('versionLibelle')),
+                'width': int_or_none(f.get('width')),
+                'height': int_or_none(f.get('height')),
+                'tbr': int_or_none(f.get('bitrate')),
+                'quality': qfunc(f.get('quality')),
+            }
+
+            if f.get('mediaType') == 'rtmp':
+                format['url'] = f['streamer']
+                format['play_path'] = 'mp4:' + f['url']
+                format['ext'] = 'flv'
+            else:
+                format['url'] = f['url']
+
+            formats.append(format)
+
+        self._check_formats(formats, video_id)
+        self._sort_formats(formats)
+
+        info_dict['formats'] = formats
+        return info_dict
+
+
+class ArteTVPlus7IE(ArteTVBaseIE):
+    IE_NAME = 'arte.tv:+7'
+    _VALID_URL = r'https?://(?:(?:www|sites)\.)?arte\.tv/[^/]+/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+
+    _TESTS = [{
+        'url': 'http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D',
+        'only_matching': True,
+    }, {
+        'url': 'http://sites.arte.tv/karambolage/de/video/karambolage-22',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if ArteTVPlaylistIE.suitable(url) else super(ArteTVPlus7IE, cls).suitable(url)
+
     def _real_extract(self, url):
         video_id, lang = self._extract_url_info(url)
         webpage = self._download_webpage(url, video_id)
@@ -127,84 +243,10 @@ class ArteTVPlus7IE(InfoExtractor):
             return self._extract_from_json_url(json_url, video_id, lang, title=title)
         # Different kind of embed URL (e.g.
         # http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium)
-        embed_url = self._search_regex(
-            r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1',
-            webpage, 'embed url', group='url')
-        return self.url_result(embed_url)
-
-    def _extract_from_json_url(self, json_url, video_id, lang, title=None):
-        info = self._download_json(json_url, video_id)
-        player_info = info['videoJsonPlayer']
-
-        upload_date_str = player_info.get('shootingDate')
-        if not upload_date_str:
-            upload_date_str = (player_info.get('VRA') or player_info.get('VDA') or '').split(' ')[0]
-
-        title = (player_info.get('VTI') or title or player_info['VID']).strip()
-        subtitle = player_info.get('VSU', '').strip()
-        if subtitle:
-            title += ' - %s' % subtitle
-
-        info_dict = {
-            'id': player_info['VID'],
-            'title': title,
-            'description': player_info.get('VDE'),
-            'upload_date': unified_strdate(upload_date_str),
-            'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
-        }
-        qfunc = qualities(['HQ', 'MQ', 'EQ', 'SQ'])
-
-        LANGS = {
-            'fr': 'F',
-            'de': 'A',
-            'en': 'E[ANG]',
-            'es': 'E[ESP]',
-        }
-
-        formats = []
-        for format_id, format_dict in player_info['VSR'].items():
-            f = dict(format_dict)
-            versionCode = f.get('versionCode')
-            langcode = LANGS.get(lang, lang)
-            lang_rexs = [r'VO?%s-' % re.escape(langcode), r'VO?.-ST%s$' % re.escape(langcode)]
-            lang_pref = None
-            if versionCode:
-                matched_lang_rexs = [r for r in lang_rexs if re.match(r, versionCode)]
-                lang_pref = -10 if not matched_lang_rexs else 10 * len(matched_lang_rexs)
-            source_pref = 0
-            if versionCode is not None:
-                # The original version with subtitles has lower relevance
-                if re.match(r'VO-ST(F|A|E)', versionCode):
-                    source_pref -= 10
-                # The version with sourds/mal subtitles has also lower relevance
-                elif re.match(r'VO?(F|A|E)-STM\1', versionCode):
-                    source_pref -= 9
-            format = {
-                'format_id': format_id,
-                'preference': -10 if f.get('videoFormat') == 'M3U8' else None,
-                'language_preference': lang_pref,
-                'format_note': '%s, %s' % (f.get('versionCode'), f.get('versionLibelle')),
-                'width': int_or_none(f.get('width')),
-                'height': int_or_none(f.get('height')),
-                'tbr': int_or_none(f.get('bitrate')),
-                'quality': qfunc(f.get('quality')),
-                'source_preference': source_pref,
-            }
-
-            if f.get('mediaType') == 'rtmp':
-                format['url'] = f['streamer']
-                format['play_path'] = 'mp4:' + f['url']
-                format['ext'] = 'flv'
-            else:
-                format['url'] = f['url']
-
-            formats.append(format)
-
-        self._check_formats(formats, video_id)
-        self._sort_formats(formats)
-
-        info_dict['formats'] = formats
-        return info_dict
+        entries = [
+            self.url_result(url)
+            for _, url in re.findall(r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1', webpage)]
+        return self.playlist_result(entries)
 
 
 # It also uses the arte_vp_url url from the webpage to extract the information
@@ -213,22 +255,17 @@ class ArteTVCreativeIE(ArteTVPlus7IE):
     _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 
     _TESTS = [{
-        'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
+        'url': 'http://creative.arte.tv/fr/episode/osmosis-episode-1',
         'info_dict': {
-            'id': '72176',
+            'id': '057405-001-A',
             'ext': 'mp4',
-            'title': 'Folge 2 - Corporate Design',
-            'upload_date': '20131004',
+            'title': 'OSMOSIS - N\'AYEZ PLUS PEUR D\'AIMER (1)',
+            'upload_date': '20150716',
         },
     }, {
         'url': 'http://creative.arte.tv/fr/Monty-Python-Reunion',
-        'info_dict': {
-            'id': '160676',
-            'ext': 'mp4',
-            'title': 'Monty Python live (mostly)',
-            'description': 'Événement ! Quarante-cinq ans après leurs premiers succès, les légendaires Monty Python remontent sur scène.\n',
-            'upload_date': '20140805',
-        }
+        'playlist_count': 11,
+        'add_ie': ['Youtube'],
     }, {
         'url': 'http://creative.arte.tv/de/episode/agentur-amateur-4-der-erste-kunde',
         'only_matching': True,
@@ -239,7 +276,7 @@ class ArteTVInfoIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:info'
     _VALID_URL = r'https?://info\.arte\.tv/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://info.arte.tv/fr/service-civique-un-cache-misere',
         'info_dict': {
             'id': '067528-000-A',
@@ -247,7 +284,7 @@ class ArteTVInfoIE(ArteTVPlus7IE):
             'title': 'Service civique, un cache misère ?',
             'upload_date': '20160403',
         },
-    }
+    }]
 
 
 class ArteTVFutureIE(ArteTVPlus7IE):
@@ -272,6 +309,8 @@ class ArteTVDDCIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:ddc'
     _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>[^/?#&]+)'
 
+    _TESTS = []
+
     def _real_extract(self, url):
         video_id, lang = self._extract_url_info(url)
         if lang == 'folge':
@@ -290,7 +329,7 @@ class ArteTVConcertIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:concert'
     _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde',
         'md5': '9ea035b7bd69696b67aa2ccaaa218161',
         'info_dict': {
@@ -300,24 +339,23 @@ class ArteTVConcertIE(ArteTVPlus7IE):
             'upload_date': '20140128',
             'description': 'md5:486eb08f991552ade77439fe6d82c305',
         },
-    }
+    }]
 
 
 class ArteTVCinemaIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:cinema'
     _VALID_URL = r'https?://cinema\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>.+)'
 
-    _TEST = {
-        'url': 'http://cinema.arte.tv/de/node/38291',
-        'md5': '6b275511a5107c60bacbeeda368c3aa1',
+    _TESTS = [{
+        'url': 'http://cinema.arte.tv/fr/article/les-ailes-du-desir-de-julia-reck',
+        'md5': 'a5b9dd5575a11d93daf0e3f404f45438',
         'info_dict': {
-            'id': '055876-000_PWA12025-D',
+            'id': '062494-000-A',
             'ext': 'mp4',
-            'title': 'Tod auf dem Nil',
-            'upload_date': '20160122',
-            'description': 'md5:7f749bbb77d800ef2be11d54529b96bc',
+            'title': 'Film lauréat du concours web - "Les ailes du désir" de Julia Reck',
+            'upload_date': '20150807',
         },
-    }
+    }]
 
 
 class ArteTVMagazineIE(ArteTVPlus7IE):
@@ -362,9 +400,42 @@ class ArteTVEmbedIE(ArteTVPlus7IE):
         )
     '''
 
+    _TESTS = []
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         lang = mobj.group('lang')
         json_url = mobj.group('json_url')
         return self._extract_from_json_url(json_url, video_id, lang)
+
+
+class ArteTVPlaylistIE(ArteTVBaseIE):
+    IE_NAME = 'arte.tv:playlist'
+    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/[^#]*#collection/(?P<id>PL-\d+)'
+
+    _TESTS = [{
+        'url': 'http://www.arte.tv/guide/de/plus7/?country=DE#collection/PL-013263/ARTETV',
+        'info_dict': {
+            'id': 'PL-013263',
+            'title': 'Areva & Uramin',
+            'description': 'md5:a1dc0312ce357c262259139cfd48c9bf',
+        },
+        'playlist_mincount': 6,
+    }, {
+        'url': 'http://www.arte.tv/guide/de/playlists?country=DE#collection/PL-013190/ARTETV',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        playlist_id, lang = self._extract_url_info(url)
+        collection = self._download_json(
+            'https://api.arte.tv/api/player/v1/collectionData/%s/%s?source=videos'
+            % (lang, playlist_id), playlist_id)
+        title = collection.get('title')
+        description = collection.get('shortDescription') or collection.get('teaserText')
+        entries = [
+            self._extract_from_json_url(
+                video['jsonUrl'], video.get('programId') or playlist_id, lang)
+            for video in collection['videos'] if video.get('jsonUrl')]
+        return self.playlist_result(entries, playlist_id, title, description)

youtube_dl/extractor/audiomack.py

@@ -6,6 +6,7 @@ import time
 
 from .common import InfoExtractor
 from .soundcloud import SoundcloudIE
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
     url_basename,
@@ -136,7 +137,7 @@ class AudiomackAlbumIE(InfoExtractor):
                         result[resultkey] = api_response[apikey]
                 song_id = url_basename(api_response['url']).rpartition('.')[0]
                 result['entries'].append({
-                    'id': api_response.get('id', song_id),
+                    'id': compat_str(api_response.get('id', song_id)),
                     'uploader': api_response.get('artist'),
                     'title': api_response.get('title', song_id),
                     'url': api_response['url'],

youtube_dl/extractor/azubu.py

@@ -46,6 +46,7 @@ class AzubuIE(InfoExtractor):
                 'uploader_id': 272749,
                 'view_count': int,
             },
+            'skip': 'Channel offline',
         },
     ]
 
@@ -56,22 +57,26 @@ class AzubuIE(InfoExtractor):
             'http://www.azubu.tv/api/video/%s' % video_id, video_id)['data']
 
         title = data['title'].strip()
-        description = data['description']
-        thumbnail = data['thumbnail']
-        view_count = data['view_count']
-        uploader = data['user']['username']
-        uploader_id = data['user']['id']
+        description = data.get('description')
+        thumbnail = data.get('thumbnail')
+        view_count = data.get('view_count')
+        user = data.get('user', {})
+        uploader = user.get('username')
+        uploader_id = user.get('id')
 
         stream_params = json.loads(data['stream_params'])
 
-        timestamp = float_or_none(stream_params['creationDate'], 1000)
-        duration = float_or_none(stream_params['length'], 1000)
+        timestamp = float_or_none(stream_params.get('creationDate'), 1000)
+        duration = float_or_none(stream_params.get('length'), 1000)
 
         renditions = stream_params.get('renditions') or []
         video = stream_params.get('FLVFullLength') or stream_params.get('videoFullLength')
         if video:
             renditions.append(video)
 
+        if not renditions and not user.get('channel', {}).get('is_live', True):
+            raise ExtractorError('%s said: channel is offline.' % self.IE_NAME, expected=True)
+
         formats = [{
             'url': fmt['url'],
             'width': fmt['frameWidth'],

youtube_dl/extractor/bandcamp.py

@@ -29,7 +29,7 @@ class BandcampIE(InfoExtractor):
         '_skip': 'There is a limit of 200 free downloads / month for the test song'
     }, {
         'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
-        'md5': '2b68e5851514c20efdff2afc5603b8b4',
+        'md5': '73d0b3171568232574e45652f8720b5c',
         'info_dict': {
             'id': '2650410135',
             'ext': 'mp3',
@@ -48,6 +48,10 @@ class BandcampIE(InfoExtractor):
             if m_trackinfo:
                 json_code = m_trackinfo.group(1)
                 data = json.loads(json_code)[0]
+                track_id = compat_str(data['id'])
+
+                if not data.get('file'):
+                    raise ExtractorError('Not streamable', video_id=track_id, expected=True)
 
                 formats = []
                 for format_id, format_url in data['file'].items():
@@ -64,7 +68,7 @@ class BandcampIE(InfoExtractor):
                 self._sort_formats(formats)
 
                 return {
-                    'id': compat_str(data['id']),
+                    'id': track_id,
                     'title': data['title'],
                     'formats': formats,
                     'duration': float_or_none(data.get('duration')),

youtube_dl/extractor/bbc.py

@@ -5,11 +5,13 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
+    dict_get,
     ExtractorError,
     float_or_none,
     int_or_none,
     parse_duration,
     parse_iso8601,
+    try_get,
     unescapeHTML,
 )
 from ..compat import (
@@ -31,7 +33,7 @@ class BBCCoUkIE(InfoExtractor):
                             music/clips[/#]|
                             radio/player/
                         )
-                        (?P<id>%s)
+                        (?P<id>%s)(?!/(?:episodes|broadcasts|clips))
                     ''' % _ID_REGEX
 
     _MEDIASELECTOR_URLS = [
@@ -192,6 +194,7 @@ class BBCCoUkIE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
+            'skip': 'Now it\'s really geo-restricted',
         }, {
             # compact player (https://github.com/rg3/youtube-dl/issues/8147)
             'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player',
@@ -228,51 +231,6 @@ class BBCCoUkIE(InfoExtractor):
         asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
         return [ref.get('href') for ref in asx.findall('./Entry/ref')]
 
-    def _extract_connection(self, connection, programme_id):
-        formats = []
-        kind = connection.get('kind')
-        protocol = connection.get('protocol')
-        supplier = connection.get('supplier')
-        if protocol == 'http':
-            href = connection.get('href')
-            transfer_format = connection.get('transferFormat')
-            # ASX playlist
-            if supplier == 'asx':
-                for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
-                    formats.append({
-                        'url': ref,
-                        'format_id': 'ref%s_%s' % (i, supplier),
-                    })
-            # Skip DASH until supported
-            elif transfer_format == 'dash':
-                pass
-            elif transfer_format == 'hls':
-                formats.extend(self._extract_m3u8_formats(
-                    href, programme_id, ext='mp4', entry_protocol='m3u8_native',
-                    m3u8_id=supplier, fatal=False))
-            # Direct link
-            else:
-                formats.append({
-                    'url': href,
-                    'format_id': supplier or kind or protocol,
-                })
-        elif protocol == 'rtmp':
-            application = connection.get('application', 'ondemand')
-            auth_string = connection.get('authString')
-            identifier = connection.get('identifier')
-            server = connection.get('server')
-            formats.append({
-                'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
-                'play_path': identifier,
-                'app': '%s?%s' % (application, auth_string),
-                'page_url': 'http://www.bbc.co.uk',
-                'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
-                'rtmp_live': False,
-                'ext': 'flv',
-                'format_id': supplier,
-            })
-        return formats
-
     def _extract_items(self, playlist):
         return playlist.findall('./{%s}item' % self._EMP_PLAYLIST_NS)
 
@@ -293,46 +251,6 @@ class BBCCoUkIE(InfoExtractor):
     def _extract_connections(self, media):
         return self._findall_ns(media, './{%s}connection')
 
-    def _extract_video(self, media, programme_id):
-        formats = []
-        vbr = int_or_none(media.get('bitrate'))
-        vcodec = media.get('encoding')
-        service = media.get('service')
-        width = int_or_none(media.get('width'))
-        height = int_or_none(media.get('height'))
-        file_size = int_or_none(media.get('media_file_size'))
-        for connection in self._extract_connections(media):
-            conn_formats = self._extract_connection(connection, programme_id)
-            for format in conn_formats:
-                format.update({
-                    'width': width,
-                    'height': height,
-                    'vbr': vbr,
-                    'vcodec': vcodec,
-                    'filesize': file_size,
-                })
-                if service:
-                    format['format_id'] = '%s_%s' % (service, format['format_id'])
-            formats.extend(conn_formats)
-        return formats
-
-    def _extract_audio(self, media, programme_id):
-        formats = []
-        abr = int_or_none(media.get('bitrate'))
-        acodec = media.get('encoding')
-        service = media.get('service')
-        for connection in self._extract_connections(media):
-            conn_formats = self._extract_connection(connection, programme_id)
-            for format in conn_formats:
-                format.update({
-                    'format_id': '%s_%s' % (service, format['format_id']),
-                    'abr': abr,
-                    'acodec': acodec,
-                    'vcodec': 'none',
-                })
-            formats.extend(conn_formats)
-        return formats
-
     def _get_subtitles(self, media, programme_id):
         subtitles = {}
         for connection in self._extract_connections(media):
@@ -378,13 +296,87 @@ class BBCCoUkIE(InfoExtractor):
     def _process_media_selector(self, media_selection, programme_id):
         formats = []
         subtitles = None
+        urls = []
 
         for media in self._extract_medias(media_selection):
             kind = media.get('kind')
-            if kind == 'audio':
-                formats.extend(self._extract_audio(media, programme_id))
-            elif kind == 'video':
-                formats.extend(self._extract_video(media, programme_id))
+            if kind in ('video', 'audio'):
+                bitrate = int_or_none(media.get('bitrate'))
+                encoding = media.get('encoding')
+                service = media.get('service')
+                width = int_or_none(media.get('width'))
+                height = int_or_none(media.get('height'))
+                file_size = int_or_none(media.get('media_file_size'))
+                for connection in self._extract_connections(media):
+                    href = connection.get('href')
+                    if href in urls:
+                        continue
+                    if href:
+                        urls.append(href)
+                    conn_kind = connection.get('kind')
+                    protocol = connection.get('protocol')
+                    supplier = connection.get('supplier')
+                    transfer_format = connection.get('transferFormat')
+                    format_id = supplier or conn_kind or protocol
+                    if service:
+                        format_id = '%s_%s' % (service, format_id)
+                    # ASX playlist
+                    if supplier == 'asx':
+                        for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
+                            formats.append({
+                                'url': ref,
+                                'format_id': 'ref%s_%s' % (i, format_id),
+                            })
+                    elif transfer_format == 'dash':
+                        formats.extend(self._extract_mpd_formats(
+                            href, programme_id, mpd_id=format_id, fatal=False))
+                    elif transfer_format == 'hls':
+                        formats.extend(self._extract_m3u8_formats(
+                            href, programme_id, ext='mp4', entry_protocol='m3u8_native',
+                            m3u8_id=format_id, fatal=False))
+                    elif transfer_format == 'hds':
+                        formats.extend(self._extract_f4m_formats(
+                            href, programme_id, f4m_id=format_id, fatal=False))
+                    else:
+                        if not service and not supplier and bitrate:
+                            format_id += '-%d' % bitrate
+                        fmt = {
+                            'format_id': format_id,
+                            'filesize': file_size,
+                        }
+                        if kind == 'video':
+                            fmt.update({
+                                'width': width,
+                                'height': height,
+                                'vbr': bitrate,
+                                'vcodec': encoding,
+                            })
+                        else:
+                            fmt.update({
+                                'abr': bitrate,
+                                'acodec': encoding,
+                                'vcodec': 'none',
+                            })
+                        if protocol == 'http':
+                            # Direct link
+                            fmt.update({
+                                'url': href,
+                            })
+                        elif protocol == 'rtmp':
+                            application = connection.get('application', 'ondemand')
+                            auth_string = connection.get('authString')
+                            identifier = connection.get('identifier')
+                            server = connection.get('server')
+                            fmt.update({
+                                'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
+                                'play_path': identifier,
+                                'app': '%s?%s' % (application, auth_string),
+                                'page_url': 'http://www.bbc.co.uk',
+                                'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
+                                'rtmp_live': False,
+                                'ext': 'flv',
+                            })
+                        formats.append(fmt)
             elif kind == 'captions':
                 subtitles = self.extract_subtitles(media, programme_id)
         return formats, subtitles
@@ -589,6 +581,7 @@ class BBCIE(BBCCoUkIE):
             'id': '150615_telabyad_kentin_cogu',
             'ext': 'mp4',
             'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
+            'description': 'md5:33a4805a855c9baf7115fcbde57e7025',
             'timestamp': 1434397334,
             'upload_date': '20150615',
         },
@@ -602,6 +595,7 @@ class BBCIE(BBCCoUkIE):
             'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
             'ext': 'mp4',
             'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
+            'description': 'md5:1525f17448c4ee262b64b8f0c9ce66c8',
             'timestamp': 1434713142,
             'upload_date': '20150619',
         },
@@ -651,6 +645,23 @@ class BBCIE(BBCCoUkIE):
             # rtmp download
             'skip_download': True,
         }
+    }, {
+        # single video embedded with Morph
+        'url': 'http://www.bbc.co.uk/sport/live/olympics/36895975',
+        'info_dict': {
+            'id': 'p041vhd0',
+            'ext': 'mp4',
+            'title': "Nigeria v Japan - Men's First Round",
+            'description': 'Live coverage of the first round from Group B at the Amazonia Arena.',
+            'duration': 7980,
+            'uploader': 'BBC Sport',
+            'uploader_id': 'bbc_sport',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        'skip': 'Georestricted to UK',
     }, {
         # single video with playlist.sxml URL in playlist param
         'url': 'http://www.bbc.com/sport/0/football/33653409',
@@ -698,7 +709,9 @@ class BBCIE(BBCCoUkIE):
 
     @classmethod
     def suitable(cls, url):
-        return False if BBCCoUkIE.suitable(url) or BBCCoUkArticleIE.suitable(url) else super(BBCIE, cls).suitable(url)
+        EXCLUDE_IE = (BBCCoUkIE, BBCCoUkArticleIE, BBCCoUkIPlayerPlaylistIE, BBCCoUkPlaylistIE)
+        return (False if any(ie.suitable(url) for ie in EXCLUDE_IE)
+                else super(BBCIE, cls).suitable(url))
 
     def _extract_from_media_meta(self, media_meta, video_id):
         # Direct links to media in media metadata (e.g.
@@ -815,8 +828,29 @@ class BBCIE(BBCCoUkIE):
                         # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
                         playlist = data_playable.get('otherSettings', {}).get('playlist', {})
                         if playlist:
-                            entries.append(self._extract_from_playlist_sxml(
-                                playlist.get('progressiveDownloadUrl'), playlist_id, timestamp))
+                            entry = None
+                            for key in ('streaming', 'progressiveDownload'):
+                                playlist_url = playlist.get('%sUrl' % key)
+                                if not playlist_url:
+                                    continue
+                                try:
+                                    info = self._extract_from_playlist_sxml(
+                                        playlist_url, playlist_id, timestamp)
+                                    if not entry:
+                                        entry = info
+                                    else:
+                                        entry['title'] = info['title']
+                                        entry['formats'].extend(info['formats'])
+                                except Exception as e:
+                                    # Some playlist URL may fail with 500, at the same time
+                                    # the other one may work fine (e.g.
+                                    # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
+                                    if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
+                                        continue
+                                    raise
+                            if entry:
+                                self._sort_formats(entry['formats'])
+                                entries.append(entry)
 
         if entries:
             return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
@@ -849,6 +883,50 @@ class BBCIE(BBCCoUkIE):
                 'subtitles': subtitles,
             }
 
+        # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
+        # There are several setPayload calls may be present but the video
+        # seems to be always related to the first one
+        morph_payload = self._parse_json(
+            self._search_regex(
+                r'Morph\.setPayload\([^,]+,\s*({.+?})\);',
+                webpage, 'morph payload', default='{}'),
+            playlist_id, fatal=False)
+        if morph_payload:
+            components = try_get(morph_payload, lambda x: x['body']['components'], list) or []
+            for component in components:
+                if not isinstance(component, dict):
+                    continue
+                lead_media = try_get(component, lambda x: x['props']['leadMedia'], dict)
+                if not lead_media:
+                    continue
+                identifiers = lead_media.get('identifiers')
+                if not identifiers or not isinstance(identifiers, dict):
+                    continue
+                programme_id = identifiers.get('vpid') or identifiers.get('playablePid')
+                if not programme_id:
+                    continue
+                title = lead_media.get('title') or self._og_search_title(webpage)
+                formats, subtitles = self._download_media_selector(programme_id)
+                self._sort_formats(formats)
+                description = lead_media.get('summary')
+                uploader = lead_media.get('masterBrand')
+                uploader_id = lead_media.get('mid')
+                duration = None
+                duration_d = lead_media.get('duration')
+                if isinstance(duration_d, dict):
+                    duration = parse_duration(dict_get(
+                        duration_d, ('rawDuration', 'formattedDuration', 'spokenDuration')))
+                return {
+                    'id': programme_id,
+                    'title': title,
+                    'description': description,
+                    'duration': duration,
+                    'uploader': uploader,
+                    'uploader_id': uploader_id,
+                    'formats': formats,
+                    'subtitles': subtitles,
+                }
+
         def extract_all(pattern):
             return list(filter(None, map(
                 lambda s: self._parse_json(s, playlist_id, fatal=False),
@@ -866,7 +944,7 @@ class BBCIE(BBCCoUkIE):
             r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
         if entries:
             return self.playlist_result(
-                [self.url_result(entry, 'BBCCoUk') for entry in entries],
+                [self.url_result(entry_, 'BBCCoUk') for entry_ in entries],
                 playlist_id, playlist_title, playlist_description)
 
         # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511)
@@ -975,3 +1053,82 @@ class BBCCoUkArticleIE(InfoExtractor):
             r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)]
 
         return self.playlist_result(entries, playlist_id, title, description)
+
+
+class BBCCoUkPlaylistBaseIE(InfoExtractor):
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        entries = [
+            self.url_result(self._URL_TEMPLATE % video_id, BBCCoUkIE.ie_key())
+            for video_id in re.findall(
+                self._VIDEO_ID_TEMPLATE % BBCCoUkIE._ID_REGEX, webpage)]
+
+        title, description = self._extract_title_and_description(webpage)
+
+        return self.playlist_result(entries, playlist_id, title, description)
+
+
+class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE):
+    IE_NAME = 'bbc.co.uk:iplayer:playlist'
+    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/(?:episodes|group)/(?P<id>%s)' % BBCCoUkIE._ID_REGEX
+    _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s'
+    _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)'
+    _TESTS = [{
+        'url': 'http://www.bbc.co.uk/iplayer/episodes/b05rcz9v',
+        'info_dict': {
+            'id': 'b05rcz9v',
+            'title': 'The Disappearance',
+            'description': 'French thriller serial about a missing teenager.',
+        },
+        'playlist_mincount': 6,
+        'skip': 'This programme is not currently available on BBC iPlayer',
+    }, {
+        # Available for over a year unlike 30 days for most other programmes
+        'url': 'http://www.bbc.co.uk/iplayer/group/p02tcc32',
+        'info_dict': {
+            'id': 'p02tcc32',
+            'title': 'Bohemian Icons',
+            'description': 'md5:683e901041b2fe9ba596f2ab04c4dbe7',
+        },
+        'playlist_mincount': 10,
+    }]
+
+    def _extract_title_and_description(self, webpage):
+        title = self._search_regex(r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
+        description = self._search_regex(
+            r'<p[^>]+class=(["\'])subtitle\1[^>]*>(?P<value>[^<]+)</p>',
+            webpage, 'description', fatal=False, group='value')
+        return title, description
+
+
+class BBCCoUkPlaylistIE(BBCCoUkPlaylistBaseIE):
+    IE_NAME = 'bbc.co.uk:playlist'
+    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/(?P<id>%s)/(?:episodes|broadcasts|clips)' % BBCCoUkIE._ID_REGEX
+    _URL_TEMPLATE = 'http://www.bbc.co.uk/programmes/%s'
+    _VIDEO_ID_TEMPLATE = r'data-pid=["\'](%s)'
+    _TESTS = [{
+        'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
+        'info_dict': {
+            'id': 'b05rcz9v',
+            'title': 'The Disappearance - Clips - BBC Four',
+            'description': 'French thriller serial about a missing teenager.',
+        },
+        'playlist_mincount': 7,
+    }, {
+        'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/broadcasts/2016/06',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.bbc.co.uk/programmes/b055jkys/episodes/player',
+        'only_matching': True,
+    }]
+
+    def _extract_title_and_description(self, webpage):
+        title = self._og_search_title(webpage, fatal=False)
+        description = self._og_search_description(webpage)
+        return title, description

youtube_dl/extractor/bet.py

@@ -1,31 +1,27 @@
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
-from ..compat import compat_urllib_parse_unquote
-from ..utils import (
-    xpath_text,
-    xpath_with_ns,
-    int_or_none,
-    parse_iso8601,
-)
+from .mtv import MTVServicesInfoExtractor
+from ..utils import unified_strdate
+from ..compat import compat_urllib_parse_urlencode
 
 
-class BetIE(InfoExtractor):
+class BetIE(MTVServicesInfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?bet\.com/(?:[^/]+/)+(?P<id>.+?)\.html'
     _TESTS = [
         {
             'url': 'http://www.bet.com/news/politics/2014/12/08/in-bet-exclusive-obama-talks-race-and-racism.html',
             'info_dict': {
-                'id': 'news/national/2014/a-conversation-with-president-obama',
+                'id': '07e96bd3-8850-3051-b856-271b457f0ab8',
                 'display_id': 'in-bet-exclusive-obama-talks-race-and-racism',
                 'ext': 'flv',
                 'title': 'A Conversation With President Obama',
-                'description': 'md5:699d0652a350cf3e491cd15cc745b5da',
+                'description': 'President Obama urges persistence in confronting racism and bias.',
                 'duration': 1534,
-                'timestamp': 1418075340,
                 'upload_date': '20141208',
-                'uploader': 'admin',
                 'thumbnail': 're:(?i)^https?://.*\.jpg$',
+                'subtitles': {
+                    'en': 'mincount:2',
+                }
             },
             'params': {
                 # rtmp download
@@ -35,16 +31,17 @@ class BetIE(InfoExtractor):
         {
             'url': 'http://www.bet.com/video/news/national/2014/justice-for-ferguson-a-community-reacts.html',
             'info_dict': {
-                'id': 'news/national/2014/justice-for-ferguson-a-community-reacts',
+                'id': '9f516bf1-7543-39c4-8076-dd441b459ba9',
                 'display_id': 'justice-for-ferguson-a-community-reacts',
                 'ext': 'flv',
                 'title': 'Justice for Ferguson: A Community Reacts',
                 'description': 'A BET News special.',
                 'duration': 1696,
-                'timestamp': 1416942360,
                 'upload_date': '20141125',
-                'uploader': 'admin',
                 'thumbnail': 're:(?i)^https?://.*\.jpg$',
+                'subtitles': {
+                    'en': 'mincount:2',
+                }
             },
             'params': {
                 # rtmp download
@@ -53,57 +50,32 @@ class BetIE(InfoExtractor):
         }
     ]
 
+    _FEED_URL = "http://feeds.mtvnservices.com/od/feed/bet-mrss-player"
+
+    def _get_feed_query(self, uri):
+        return compat_urllib_parse_urlencode({
+            'uuid': uri,
+        })
+
+    def _extract_mgid(self, webpage):
+        return self._search_regex(r'data-uri="([^"]+)', webpage, 'mgid')
+
     def _real_extract(self, url):
         display_id = self._match_id(url)
+
         webpage = self._download_webpage(url, display_id)
+        mgid = self._extract_mgid(webpage)
+        videos_info = self._get_videos_info(mgid)
 
-        media_url = compat_urllib_parse_unquote(self._search_regex(
-            [r'mediaURL\s*:\s*"([^"]+)"', r"var\s+mrssMediaUrl\s*=\s*'([^']+)'"],
-            webpage, 'media URL'))
+        info_dict = videos_info['entries'][0]
 
-        video_id = self._search_regex(
-            r'/video/(.*)/_jcr_content/', media_url, 'video id')
+        upload_date = unified_strdate(self._html_search_meta('date', webpage))
+        description = self._html_search_meta('description', webpage)
 
-        mrss = self._download_xml(media_url, display_id)
-
-        item = mrss.find('./channel/item')
-
-        NS_MAP = {
-            'dc': 'http://purl.org/dc/elements/1.1/',
-            'media': 'http://search.yahoo.com/mrss/',
-            'ka': 'http://kickapps.com/karss',
-        }
-
-        title = xpath_text(item, './title', 'title')
-        description = xpath_text(
-            item, './description', 'description', fatal=False)
-
-        timestamp = parse_iso8601(xpath_text(
-            item, xpath_with_ns('./dc:date', NS_MAP),
-            'upload date', fatal=False))
-        uploader = xpath_text(
-            item, xpath_with_ns('./dc:creator', NS_MAP),
-            'uploader', fatal=False)
-
-        media_content = item.find(
-            xpath_with_ns('./media:content', NS_MAP))
-        duration = int_or_none(media_content.get('duration'))
-        smil_url = media_content.get('url')
-
-        thumbnail = media_content.find(
-            xpath_with_ns('./media:thumbnail', NS_MAP)).get('url')
-
-        formats = self._extract_smil_formats(smil_url, display_id)
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
+        info_dict.update({
             'display_id': display_id,
-            'title': title,
             'description': description,
-            'thumbnail': thumbnail,
-            'timestamp': timestamp,
-            'uploader': uploader,
-            'duration': duration,
-            'formats': formats,
-        }
+            'upload_date': upload_date,
+        })
+
+        return info_dict

youtube_dl/extractor/bigflix.py

@@ -12,7 +12,7 @@ class BigflixIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?bigflix\.com/.+/(?P<id>[0-9]+)'
     _TESTS = [{
         'url': 'http://www.bigflix.com/Hindi-movies/Action-movies/Singham-Returns/16537',
-        'md5': 'ec76aa9b1129e2e5b301a474e54fab74',
+        'md5': 'dc1b4aebb46e3a7077ecc0d9f43f61e3',
         'info_dict': {
             'id': '16537',
             'ext': 'mp4',
@@ -26,7 +26,7 @@ class BigflixIE(InfoExtractor):
             'id': '16070',
             'ext': 'mp4',
             'title': 'Madarasapatinam',
-            'description': 'md5:63b9b8ed79189c6f0418c26d9a3452ca',
+            'description': 'md5:9f0470b26a4ba8e824c823b5d95c2f6b',
             'formats': 'mincount:2',
         },
         'params': {

youtube_dl/extractor/bilibili.py

@@ -1,34 +1,42 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import calendar
+import datetime
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import (
+    compat_etree_fromstring,
+    compat_str,
+    compat_parse_qs,
+    compat_xml_parse_error,
+)
 from ..utils import (
-    int_or_none,
-    unescapeHTML,
     ExtractorError,
+    int_or_none,
+    float_or_none,
     xpath_text,
 )
 
 
 class BiliBiliIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.bilibili\.(?:tv|com)/video/av(?P<id>\d+)(?:/index_(?P<page_num>\d+).html)?'
+    _VALID_URL = r'https?://www\.bilibili\.(?:tv|com)/video/av(?P<id>\d+)'
 
     _TESTS = [{
         'url': 'http://www.bilibili.tv/video/av1074402/',
-        'md5': '2c301e4dab317596e837c3e7633e7d86',
+        'md5': '5f7d29e1a2872f3df0cf76b1f87d3788',
         'info_dict': {
             'id': '1554319',
             'ext': 'flv',
             'title': '【金坷垃】金泡沫',
-            'duration': 308313,
+            'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923',
+            'duration': 308.067,
+            'timestamp': 1398012660,
             'upload_date': '20140420',
             'thumbnail': 're:^https?://.+\.jpg',
-            'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923',
-            'timestamp': 1397983878,
             'uploader': '菊子桑',
+            'uploader_id': '156160',
         },
     }, {
         'url': 'http://www.bilibili.com/video/av1041170/',
@@ -36,75 +44,186 @@ class BiliBiliIE(InfoExtractor):
             'id': '1041170',
             'title': '【BD1080P】刀语【诸神&异域】',
             'description': '这是个神奇的故事~每个人不留弹幕不给走哦~切利哦！~',
-            'uploader': '枫叶逝去',
-            'timestamp': 1396501299,
         },
         'playlist_count': 9,
+    }, {
+        'url': 'http://www.bilibili.com/video/av4808130/',
+        'info_dict': {
+            'id': '4808130',
+            'title': '【长篇】哆啦A梦443【钉铛】',
+            'description': '(2016.05.27)来组合客人的脸吧&amp;amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;amp;illust_id=56912929',
+        },
+        'playlist': [{
+            'md5': '55cdadedf3254caaa0d5d27cf20a8f9c',
+            'info_dict': {
+                'id': '4808130_part1',
+                'ext': 'flv',
+                'title': '【长篇】哆啦A梦443【钉铛】',
+                'description': '(2016.05.27)来组合客人的脸吧&amp;amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;amp;illust_id=56912929',
+                'timestamp': 1464564180,
+                'upload_date': '20160529',
+                'uploader': '喜欢拉面',
+                'uploader_id': '151066',
+            },
+        }, {
+            'md5': '926f9f67d0c482091872fbd8eca7ea3d',
+            'info_dict': {
+                'id': '4808130_part2',
+                'ext': 'flv',
+                'title': '【长篇】哆啦A梦443【钉铛】',
+                'description': '(2016.05.27)来组合客人的脸吧&amp;amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;amp;illust_id=56912929',
+                'timestamp': 1464564180,
+                'upload_date': '20160529',
+                'uploader': '喜欢拉面',
+                'uploader_id': '151066',
+            },
+        }, {
+            'md5': '4b7b225b968402d7c32348c646f1fd83',
+            'info_dict': {
+                'id': '4808130_part3',
+                'ext': 'flv',
+                'title': '【长篇】哆啦A梦443【钉铛】',
+                'description': '(2016.05.27)来组合客人的脸吧&amp;amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;amp;illust_id=56912929',
+                'timestamp': 1464564180,
+                'upload_date': '20160529',
+                'uploader': '喜欢拉面',
+                'uploader_id': '151066',
+            },
+        }, {
+            'md5': '7b795e214166501e9141139eea236e91',
+            'info_dict': {
+                'id': '4808130_part4',
+                'ext': 'flv',
+                'title': '【长篇】哆啦A梦443【钉铛】',
+                'description': '(2016.05.27)来组合客人的脸吧&amp;amp;寻母六千里锭 抱歉，又轮到周日上班现在才到家 封面www.pixiv.net/member_illust.php?mode=medium&amp;amp;illust_id=56912929',
+                'timestamp': 1464564180,
+                'upload_date': '20160529',
+                'uploader': '喜欢拉面',
+                'uploader_id': '151066',
+            },
+        }],
+    }, {
+        # Missing upload time
+        'url': 'http://www.bilibili.com/video/av1867637/',
+        'info_dict': {
+            'id': '2880301',
+            'ext': 'flv',
+            'title': '【HDTV】【喜剧】岳父岳母真难当 （2014）【法国票房冠军】',
+            'description': '一个信奉天主教的法国旧式传统资产阶级家庭中有四个女儿。三个女儿却分别找了阿拉伯、犹太、中国丈夫，老夫老妻唯独期盼剩下未嫁的小女儿能找一个信奉天主教的法国白人，结果没想到小女儿找了一位非裔黑人……【这次应该不会跳帧了】',
+            'uploader': '黑夜为猫',
+            'uploader_id': '610729',
+        },
+        'params': {
+            # Just to test metadata extraction
+            'skip_download': True,
+        },
+        'expected_warnings': ['upload time'],
     }]
 
+    # BiliBili blocks keys from time to time. The current key is extracted from
+    # the Android client
+    # TODO: find the sign algorithm used in the flash player
+    _APP_KEY = '86385cdc024c0f6c'
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
-        page_num = mobj.group('page_num') or '1'
 
-        view_data = self._download_json(
-            'http://api.bilibili.com/view?type=json&appkey=8e9fc618fbd41e28&id=%s&page=%s' % (video_id, page_num),
-            video_id)
-        if 'error' in view_data:
-            raise ExtractorError('%s said: %s' % (self.IE_NAME, view_data['error']), expected=True)
+        webpage = self._download_webpage(url, video_id)
 
-        cid = view_data['cid']
-        title = unescapeHTML(view_data['title'])
+        params = compat_parse_qs(self._search_regex(
+            [r'EmbedPlayer\([^)]+,\s*"([^"]+)"\)',
+             r'<iframe[^>]+src="https://secure\.bilibili\.com/secure,([^"]+)"'],
+            webpage, 'player parameters'))
+        cid = params['cid'][0]
 
-        doc = self._download_xml(
-            'http://interface.bilibili.com/v_cdn_play?appkey=8e9fc618fbd41e28&cid=%s' % cid,
-            cid,
-            'Downloading page %s/%s' % (page_num, view_data['pages'])
-        )
+        info_xml_str = self._download_webpage(
+            'http://interface.bilibili.com/v_cdn_play',
+            cid, query={'appkey': self._APP_KEY, 'cid': cid},
+            note='Downloading video info page')
 
-        if xpath_text(doc, './result') == 'error':
-            raise ExtractorError('%s said: %s' % (self.IE_NAME, xpath_text(doc, './message')), expected=True)
+        err_msg = None
+        durls = None
+        info_xml = None
+        try:
+            info_xml = compat_etree_fromstring(info_xml_str.encode('utf-8'))
+        except compat_xml_parse_error:
+            info_json = self._parse_json(info_xml_str, video_id, fatal=False)
+            err_msg = (info_json or {}).get('error_text')
+        else:
+            err_msg = xpath_text(info_xml, './message')
+
+        if info_xml is not None:
+            durls = info_xml.findall('./durl')
+        if not durls:
+            if err_msg:
+                raise ExtractorError('%s said: %s' % (self.IE_NAME, err_msg), expected=True)
+            else:
+                raise ExtractorError('No videos found!')
 
         entries = []
 
-        for durl in doc.findall('./durl'):
+        for durl in durls:
             size = xpath_text(durl, ['./filesize', './size'])
             formats = [{
                 'url': durl.find('./url').text,
                 'filesize': int_or_none(size),
-                'ext': 'flv',
             }]
-            backup_urls = durl.find('./backup_url')
-            if backup_urls is not None:
-                for backup_url in backup_urls.findall('./url'):
-                    formats.append({'url': backup_url.text})
-            formats.reverse()
+            for backup_url in durl.findall('./backup_url/url'):
+                formats.append({
+                    'url': backup_url.text,
+                    # backup URLs have lower priorities
+                    'preference': -2 if 'hd.mp4' in backup_url.text else -3,
+                })
+
+            self._sort_formats(formats)
 
             entries.append({
                 'id': '%s_part%s' % (cid, xpath_text(durl, './order')),
-                'title': title,
                 'duration': int_or_none(xpath_text(durl, './length'), 1000),
                 'formats': formats,
             })
 
+        title = self._html_search_regex('<h1[^>]+title="([^"]+)">', webpage, 'title')
+        description = self._html_search_meta('description', webpage)
+        datetime_str = self._html_search_regex(
+            r'<time[^>]+datetime="([^"]+)"', webpage, 'upload time', fatal=False)
+        timestamp = None
+        if datetime_str:
+            timestamp = calendar.timegm(datetime.datetime.strptime(datetime_str, '%Y-%m-%dT%H:%M').timetuple())
+
+        # TODO 'view_count' requires deobfuscating Javascript
         info = {
             'id': compat_str(cid),
             'title': title,
-            'description': view_data.get('description'),
-            'thumbnail': view_data.get('pic'),
-            'uploader': view_data.get('author'),
-            'timestamp': int_or_none(view_data.get('created')),
-            'view_count': int_or_none(view_data.get('play')),
-            'duration': int_or_none(xpath_text(doc, './timelength')),
+            'description': description,
+            'timestamp': timestamp,
+            'thumbnail': self._html_search_meta('thumbnailUrl', webpage),
+            'duration': float_or_none(xpath_text(info_xml, './timelength'), scale=1000),
         }
 
+        uploader_mobj = re.search(
+            r'<a[^>]+href="https?://space\.bilibili\.com/(?P<id>\d+)"[^>]+title="(?P<name>[^"]+)"',
+            webpage)
+        if uploader_mobj:
+            info.update({
+                'uploader': uploader_mobj.group('name'),
+                'uploader_id': uploader_mobj.group('id'),
+            })
+
+        for entry in entries:
+            entry.update(info)
+
         if len(entries) == 1:
-            entries[0].update(info)
             return entries[0]
         else:
-            info.update({
+            for idx, entry in enumerate(entries):
+                entry['id'] = '%s_part%d' % (video_id, (idx + 1))
+
+            return {
                 '_type': 'multi_video',
                 'id': video_id,
+                'title': title,
+                'description': description,
                 'entries': entries,
-            })
-            return info
+            }

youtube_dl/extractor/biobiochiletv.py

@@ -2,11 +2,15 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import remove_end
+from ..utils import (
+    ExtractorError,
+    remove_end,
+)
+from .rudo import RudoIE
 
 
 class BioBioChileTVIE(InfoExtractor):
-    _VALID_URL = r'https?://tv\.biobiochile\.cl/notas/(?:[^/]+/)+(?P<id>[^/]+)\.shtml'
+    _VALID_URL = r'https?://(?:tv|www)\.biobiochile\.cl/(?:notas|noticias)/(?:[^/]+/)+(?P<id>[^/]+)\.shtml'
 
     _TESTS = [{
         'url': 'http://tv.biobiochile.cl/notas/2015/10/21/sobre-camaras-y-camarillas-parlamentarias.shtml',
@@ -18,6 +22,7 @@ class BioBioChileTVIE(InfoExtractor):
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader': 'Fernando Atria',
         },
+        'skip': 'URL expired and redirected to http://www.biobiochile.cl/portada/bbtv/index.html',
     }, {
         # different uploader layout
         'url': 'http://tv.biobiochile.cl/notas/2016/03/18/natalia-valdebenito-repasa-a-diputado-hasbun-paso-a-la-categoria-de-hablar-brutalidades.shtml',
@@ -32,6 +37,16 @@ class BioBioChileTVIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+        'skip': 'URL expired and redirected to http://www.biobiochile.cl/portada/bbtv/index.html',
+    }, {
+        'url': 'http://www.biobiochile.cl/noticias/bbtv/comentarios-bio-bio/2016/07/08/edecanes-del-congreso-figuras-decorativas-que-le-cuestan-muy-caro-a-los-chilenos.shtml',
+        'info_dict': {
+            'id': 'edecanes-del-congreso-figuras-decorativas-que-le-cuestan-muy-caro-a-los-chilenos',
+            'ext': 'mp4',
+            'uploader': '(none)',
+            'upload_date': '20160708',
+            'title': 'Edecanes del Congreso: Figuras decorativas que le cuestan muy caro a los chilenos',
+        },
     }, {
         'url': 'http://tv.biobiochile.cl/notas/2015/10/22/ninos-transexuales-de-quien-es-la-decision.shtml',
         'only_matching': True,
@@ -45,42 +60,22 @@ class BioBioChileTVIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
+        rudo_url = RudoIE._extract_url(webpage)
+        if not rudo_url:
+            raise ExtractorError('No videos found')
+
         title = remove_end(self._og_search_title(webpage), ' - BioBioChile TV')
 
-        file_url = self._search_regex(
-            r'loadFWPlayerVideo\([^,]+,\s*(["\'])(?P<url>.+?)\1',
-            webpage, 'file url', group='url')
-
-        base_url = self._search_regex(
-            r'file\s*:\s*(["\'])(?P<url>.+?)\1\s*\+\s*fileURL', webpage,
-            'base url', default='http://unlimited2-cl.digitalproserver.com/bbtv/',
-            group='url')
-
-        formats = self._extract_m3u8_formats(
-            '%s%s/playlist.m3u8' % (base_url, file_url), video_id, 'mp4',
-            entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)
-        f = {
-            'url': '%s%s' % (base_url, file_url),
-            'format_id': 'http',
-            'protocol': 'http',
-            'preference': 1,
-        }
-        if formats:
-            f_copy = formats[-1].copy()
-            f_copy.update(f)
-            f = f_copy
-        formats.append(f)
-        self._sort_formats(formats)
-
         thumbnail = self._og_search_thumbnail(webpage)
         uploader = self._html_search_regex(
-            r'<a[^>]+href=["\']https?://busca\.biobiochile\.cl/author[^>]+>(.+?)</a>',
+            r'<a[^>]+href=["\']https?://(?:busca|www)\.biobiochile\.cl/(?:lista/)?(?:author|autor)[^>]+>(.+?)</a>',
             webpage, 'uploader', fatal=False)
 
         return {
+            '_type': 'url_transparent',
+            'url': rudo_url,
             'id': video_id,
             'title': title,
             'thumbnail': thumbnail,
             'uploader': uploader,
-            'formats': formats,
         }

youtube_dl/extractor/biqle.py

@@ -0,0 +1,39 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+
+class BIQLEIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?biqle\.(?:com|org|ru)/watch/(?P<id>-?\d+_\d+)'
+    _TESTS = [{
+        'url': 'http://www.biqle.ru/watch/847655_160197695',
+        'md5': 'ad5f746a874ccded7b8f211aeea96637',
+        'info_dict': {
+            'id': '160197695',
+            'ext': 'mp4',
+            'title': 'Foo Fighters - The Pretender (Live at Wembley Stadium)',
+            'uploader': 'Andrey Rogozin',
+            'upload_date': '20110605',
+        }
+    }, {
+        'url': 'https://biqle.org/watch/-44781847_168547604',
+        'md5': '7f24e72af1db0edf7c1aaba513174f97',
+        'info_dict': {
+            'id': '168547604',
+            'ext': 'mp4',
+            'title': 'Ребенок в шоке от автоматической мойки',
+            'uploader': 'Dmitry Kotov',
+        }
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        embed_url = self._proto_relative_url(self._search_regex(
+            r'<iframe.+?src="((?:http:)?//daxab\.com/[^"]+)".*?></iframe>', webpage, 'embed url'))
+
+        return {
+            '_type': 'url_transparent',
+            'url': embed_url,
+        }

youtube_dl/extractor/bloomberg.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -17,6 +18,21 @@ class BloombergIE(InfoExtractor):
             'title': 'Shah\'s Presentation on Foreign-Exchange Strategies',
             'description': 'md5:a8ba0302912d03d246979735c17d2761',
         },
+        'params': {
+            'format': 'best[format_id^=hds]',
+        },
+    }, {
+        # video ID in BPlayer(...)
+        'url': 'http://www.bloomberg.com/features/2016-hello-world-new-zealand/',
+        'info_dict': {
+            'id': '938c7e72-3f25-4ddb-8b85-a9be731baa74',
+            'ext': 'flv',
+            'title': 'Meet the Real-Life Tech Wizards of Middle Earth',
+            'description': 'Hello World, Episode 1: New Zealand’s freaky AI babies, robot exoskeletons, and a virtual you.',
+        },
+        'params': {
+            'format': 'best[format_id^=hds]',
+        },
     }, {
         'url': 'http://www.bloomberg.com/news/articles/2015-11-12/five-strange-things-that-have-been-happening-in-financial-markets',
         'only_matching': True,
@@ -30,7 +46,11 @@ class BloombergIE(InfoExtractor):
         webpage = self._download_webpage(url, name)
         video_id = self._search_regex(
             r'["\']bmmrId["\']\s*:\s*(["\'])(?P<url>.+?)\1',
-            webpage, 'id', group='url')
+            webpage, 'id', group='url', default=None)
+        if not video_id:
+            bplayer_data = self._parse_json(self._search_regex(
+                r'BPlayer\(null,\s*({[^;]+})\);', webpage, 'id'), name)
+            video_id = bplayer_data['id']
         title = re.sub(': Video$', '', self._og_search_title(webpage))
 
         embed_info = self._download_json(

youtube_dl/extractor/br.py

@@ -29,7 +29,8 @@ class BRIE(InfoExtractor):
                 'duration': 180,
                 'uploader': 'Reinhard Weber',
                 'upload_date': '20150422',
-            }
+            },
+            'skip': '404 not found',
         },
         {
             'url': 'http://www.br.de/nachrichten/oberbayern/inhalt/muenchner-polizeipraesident-schreiber-gestorben-100.html',
@@ -40,7 +41,8 @@ class BRIE(InfoExtractor):
                 'title': 'Manfred Schreiber ist tot',
                 'description': 'md5:b454d867f2a9fc524ebe88c3f5092d97',
                 'duration': 26,
-            }
+            },
+            'skip': '404 not found',
         },
         {
             'url': 'https://www.br-klassik.de/audio/peeping-tom-premierenkritik-dance-festival-muenchen-100.html',
@@ -51,7 +53,8 @@ class BRIE(InfoExtractor):
                 'title': 'Kurzweilig und sehr bewegend',
                 'description': 'md5:0351996e3283d64adeb38ede91fac54e',
                 'duration': 296,
-            }
+            },
+            'skip': '404 not found',
         },
         {
             'url': 'http://www.br.de/radio/bayern1/service/team/videos/team-video-erdelt100.html',

youtube_dl/extractor/brightcove.py

@@ -26,6 +26,8 @@ from ..utils import (
     unescapeHTML,
     unsmuggle_url,
     update_url_query,
+    clean_html,
+    mimetype2ext,
 )
 
 
@@ -90,6 +92,7 @@ class BrightcoveLegacyIE(InfoExtractor):
                 'description': 'md5:363109c02998fee92ec02211bd8000df',
                 'uploader': 'National Ballet of Canada',
             },
+            'skip': 'Video gone',
         },
         {
             # test flv videos served by akamaihd.net
@@ -108,7 +111,7 @@ class BrightcoveLegacyIE(InfoExtractor):
             },
         },
         {
-            # playlist test
+            # playlist with 'videoList'
             # from http://support.brightcove.com/en/video-cloud/docs/playlist-support-single-video-players
             'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=3550052898001&playerKey=AQ%7E%7E%2CAAABmA9XpXk%7E%2C-Kp7jNgisre1fG5OdqpAFUTcs0lP_ZoL',
             'info_dict': {
@@ -117,6 +120,15 @@ class BrightcoveLegacyIE(InfoExtractor):
             },
             'playlist_mincount': 7,
         },
+        {
+            # playlist with 'playlistTab' (https://github.com/rg3/youtube-dl/issues/9965)
+            'url': 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=AQ%7E%7E,AAABXlLMdok%7E,NJ4EoMlZ4rZdx9eU1rkMVd8EaYPBBUlg',
+            'info_dict': {
+                'id': '1522758701001',
+                'title': 'Lesson 08',
+            },
+            'playlist_mincount': 10,
+        },
     ]
     FLV_VCODECS = {
         1: 'SORENSON',
@@ -298,18 +310,25 @@ class BrightcoveLegacyIE(InfoExtractor):
             info_url, player_key, 'Downloading playlist information')
 
         json_data = json.loads(playlist_info)
-        if 'videoList' not in json_data:
+        if 'videoList' in json_data:
+            playlist_info = json_data['videoList']
+            playlist_dto = playlist_info['mediaCollectionDTO']
+        elif 'playlistTabs' in json_data:
+            playlist_info = json_data['playlistTabs']
+            playlist_dto = playlist_info['lineupListDTO']['playlistDTOs'][0]
+        else:
             raise ExtractorError('Empty playlist')
-        playlist_info = json_data['videoList']
-        videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
+
+        videos = [self._extract_video_info(video_info) for video_info in playlist_dto['videoDTOs']]
 
         return self.playlist_result(videos, playlist_id='%s' % playlist_info['id'],
-                                    playlist_title=playlist_info['mediaCollectionDTO']['displayName'])
+                                    playlist_title=playlist_dto['displayName'])
 
     def _extract_video_info(self, video_info):
+        video_id = compat_str(video_info['id'])
         publisher_id = video_info.get('publisherId')
         info = {
-            'id': compat_str(video_info['id']),
+            'id': video_id,
             'title': video_info['displayName'].strip(),
             'description': video_info.get('shortDescription'),
             'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
@@ -331,7 +350,8 @@ class BrightcoveLegacyIE(InfoExtractor):
                     url_comp = compat_urllib_parse_urlparse(url)
                     if url_comp.path.endswith('.m3u8'):
                         formats.extend(
-                            self._extract_m3u8_formats(url, info['id'], 'mp4'))
+                            self._extract_m3u8_formats(
+                                url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
                         continue
                     elif 'akamaihd.net' in url_comp.netloc:
                         # This type of renditions are served through
@@ -365,7 +385,7 @@ class BrightcoveLegacyIE(InfoExtractor):
                     a_format.update({
                         'format_id': 'hls%s' % ('-%s' % tbr if tbr else ''),
                         'ext': 'mp4',
-                        'protocol': 'm3u8',
+                        'protocol': 'm3u8_native',
                     })
 
                 formats.append(a_format)
@@ -395,7 +415,7 @@ class BrightcoveLegacyIE(InfoExtractor):
                     return ad_info
 
         if 'url' not in info and not info.get('formats'):
-            raise ExtractorError('Unable to extract video url for %s' % info['id'])
+            raise ExtractorError('Unable to extract video url for %s' % video_id)
         return info
 
 
@@ -442,6 +462,10 @@ class BrightcoveNewIE(InfoExtractor):
         # non numeric ref: prefixed video id
         'url': 'http://players.brightcove.net/710858724001/default_default/index.html?videoId=ref:event-stream-356',
         'only_matching': True,
+    }, {
+        # unavailable video without message but with error_code
+        'url': 'http://players.brightcove.net/1305187701/c832abfb-641b-44eb-9da0-2fe76786505f_default/index.html?videoId=4377407326001',
+        'only_matching': True,
     }]
 
     @staticmethod
@@ -512,8 +536,9 @@ class BrightcoveNewIE(InfoExtractor):
             })
         except ExtractorError as e:
             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
-                json_data = self._parse_json(e.cause.read().decode(), video_id)
-                raise ExtractorError(json_data[0]['message'], expected=True)
+                json_data = self._parse_json(e.cause.read().decode(), video_id)[0]
+                raise ExtractorError(
+                    json_data.get('message') or json_data['error_code'], expected=True)
             raise
 
         title = json_data['name'].strip()
@@ -521,14 +546,16 @@ class BrightcoveNewIE(InfoExtractor):
         formats = []
         for source in json_data.get('sources', []):
             container = source.get('container')
-            source_type = source.get('type')
+            ext = mimetype2ext(source.get('type'))
             src = source.get('src')
-            if source_type == 'application/x-mpegURL' or container == 'M2TS':
+            if ext == 'ism':
+                continue
+            elif ext == 'm3u8' or container == 'M2TS':
                 if not src:
                     continue
                 formats.extend(self._extract_m3u8_formats(
-                    src, video_id, 'mp4', m3u8_id='hls', fatal=False))
-            elif source_type == 'application/dash+xml':
+                    src, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+            elif ext == 'mpd':
                 if not src:
                     continue
                 formats.extend(self._extract_mpd_formats(src, video_id, 'dash', fatal=False))
@@ -544,7 +571,7 @@ class BrightcoveNewIE(InfoExtractor):
                     'tbr': tbr,
                     'filesize': int_or_none(source.get('size')),
                     'container': container,
-                    'ext': container.lower(),
+                    'ext': ext or container.lower(),
                 }
                 if width == 0 and height == 0:
                     f.update({
@@ -578,6 +605,13 @@ class BrightcoveNewIE(InfoExtractor):
                         'format_id': build_format_id('rtmp'),
                     })
                 formats.append(f)
+
+        errors = json_data.get('errors')
+        if not formats and errors:
+            error = errors[0]
+            raise ExtractorError(
+                error.get('message') or error.get('error_subcode') or error['error_code'], expected=True)
+
         self._sort_formats(formats)
 
         subtitles = {}
@@ -590,7 +624,7 @@ class BrightcoveNewIE(InfoExtractor):
         return {
             'id': video_id,
             'title': title,
-            'description': json_data.get('description'),
+            'description': clean_html(json_data.get('description')),
             'thumbnail': json_data.get('thumbnail') or json_data.get('poster'),
             'duration': float_or_none(json_data.get('duration'), 1000),
             'timestamp': parse_iso8601(json_data.get('published_at')),

youtube_dl/extractor/buzzfeed.py

@@ -5,6 +5,7 @@ import json
 import re
 
 from .common import InfoExtractor
+from .facebook import FacebookIE
 
 
 class BuzzFeedIE(InfoExtractor):
@@ -20,11 +21,11 @@ class BuzzFeedIE(InfoExtractor):
             'info_dict': {
                 'id': 'aVCR29aE_OQ',
                 'ext': 'mp4',
+                'title': 'Angry Ram destroys a punching bag..',
+                'description': 'md5:c59533190ef23fd4458a5e8c8c872345',
                 'upload_date': '20141024',
                 'uploader_id': 'Buddhanz1',
-                'description': 'He likes to stay in shape with his heavy bag, he wont stop until its on the ground\n\nFollow Angry Ram on Facebook for regular updates -\nhttps://www.facebook.com/pages/Angry-Ram/1436897249899558?ref=hl',
-                'uploader': 'Buddhanz',
-                'title': 'Angry Ram destroys a punching bag',
+                'uploader': 'Angry Ram',
             }
         }]
     }, {
@@ -41,13 +42,30 @@ class BuzzFeedIE(InfoExtractor):
             'info_dict': {
                 'id': 'mVmBL8B-In0',
                 'ext': 'mp4',
+                'title': 're:Munchkin the Teddy Bear gets her exercise',
+                'description': 'md5:28faab95cda6e361bcff06ec12fc21d8',
                 'upload_date': '20141124',
                 'uploader_id': 'CindysMunchkin',
-                'description': 're:© 2014 Munchkin the',
                 'uploader': 're:^Munchkin the',
-                'title': 're:Munchkin the Teddy Bear gets her exercise',
             },
         }]
+    }, {
+        'url': 'http://www.buzzfeed.com/craigsilverman/the-most-adorable-crash-landing-ever#.eq7pX0BAmK',
+        'info_dict': {
+            'id': 'the-most-adorable-crash-landing-ever',
+            'title': 'Watch This Baby Goose Make The Most Adorable Crash Landing',
+            'description': 'This gosling knows how to stick a landing.',
+        },
+        'playlist': [{
+            'md5': '763ca415512f91ca62e4621086900a23',
+            'info_dict': {
+                'id': '971793786185728',
+                'ext': 'mp4',
+                'title': 'We set up crash pads so that the goslings on our roof would have a safe landi...',
+                'uploader': 'Calgary Outdoor Centre-University of Calgary',
+            },
+        }],
+        'add_ie': ['Facebook'],
     }]
 
     def _real_extract(self, url):
@@ -66,6 +84,10 @@ class BuzzFeedIE(InfoExtractor):
                 continue
             entries.append(self.url_result(video['url']))
 
+        facebook_url = FacebookIE._extract_url(webpage)
+        if facebook_url:
+            entries.append(self.url_result(facebook_url))
+
         return {
             '_type': 'playlist',
             'id': playlist_id,

youtube_dl/extractor/byutv.py

@@ -11,6 +11,7 @@ class BYUtvIE(InfoExtractor):
     _VALID_URL = r'^https?://(?:www\.)?byutv.org/watch/[0-9a-f-]+/(?P<video_id>[^/?#]+)'
     _TEST = {
         'url': 'http://www.byutv.org/watch/6587b9a3-89d2-42a6-a7f7-fd2f81840a7d/studio-c-season-5-episode-5',
+        'md5': '05850eb8c749e2ee05ad5a1c34668493',
         'info_dict': {
             'id': 'studio-c-season-5-episode-5',
             'ext': 'mp4',
@@ -21,7 +22,8 @@ class BYUtvIE(InfoExtractor):
         },
         'params': {
             'skip_download': True,
-        }
+        },
+        'add_ie': ['Ooyala'],
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/camdemy.py

@@ -1,7 +1,6 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import datetime
 import re
 
 from .common import InfoExtractor
@@ -10,8 +9,10 @@ from ..compat import (
     compat_urlparse,
 )
 from ..utils import (
-    parse_iso8601,
+    clean_html,
+    parse_duration,
     str_to_int,
+    unified_strdate,
 )
 
 
@@ -26,14 +27,14 @@ class CamdemyIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Ch1-1 Introduction, Signals (02-23-2012)',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'description': '',
             'creator': 'ss11spring',
+            'duration': 1591,
             'upload_date': '20130114',
-            'timestamp': 1358154556,
             'view_count': int,
         }
     }, {
         # With non-empty description
+        # webpage returns "No permission or not login"
         'url': 'http://www.camdemy.com/media/13885',
         'md5': '4576a3bb2581f86c61044822adbd1249',
         'info_dict': {
@@ -41,64 +42,71 @@ class CamdemyIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'EverCam + Camdemy QuickStart',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'description': 'md5:050b62f71ed62928f8a35f1a41e186c9',
+            'description': 'md5:2a9f989c2b153a2342acee579c6e7db6',
             'creator': 'evercam',
-            'upload_date': '20140620',
-            'timestamp': 1403271569,
+            'duration': 318,
         }
     }, {
-        # External source
+        # External source (YouTube)
         'url': 'http://www.camdemy.com/media/14842',
-        'md5': '50e1c3c3aa233d3d7b7daa2fa10b1cf7',
         'info_dict': {
             'id': '2vsYQzNIsJo',
             'ext': 'mp4',
+            'title': 'Excel 2013 Tutorial - How to add Password Protection',
+            'description': 'Excel 2013 Tutorial for Beginners - How to add Password Protection',
             'upload_date': '20130211',
             'uploader': 'Hun Kim',
-            'description': 'Excel 2013 Tutorial for Beginners - How to add Password Protection',
             'uploader_id': 'hunkimtutorials',
-            'title': 'Excel 2013 Tutorial - How to add Password Protection',
-        }
+        },
+        'params': {
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        page = self._download_webpage(url, video_id)
+
+        webpage = self._download_webpage(url, video_id)
 
         src_from = self._html_search_regex(
-            r"<div class='srcFrom'>Source: <a title='([^']+)'", page,
-            'external source', default=None)
+            r"class=['\"]srcFrom['\"][^>]*>Sources?(?:\s+from)?\s*:\s*<a[^>]+(?:href|title)=(['\"])(?P<url>(?:(?!\1).)+)\1",
+            webpage, 'external source', default=None, group='url')
         if src_from:
             return self.url_result(src_from)
 
         oembed_obj = self._download_json(
             'http://www.camdemy.com/oembed/?format=json&url=' + url, video_id)
 
+        title = oembed_obj['title']
         thumb_url = oembed_obj['thumbnail_url']
         video_folder = compat_urlparse.urljoin(thumb_url, 'video/')
         file_list_doc = self._download_xml(
             compat_urlparse.urljoin(video_folder, 'fileList.xml'),
-            video_id, 'Filelist XML')
+            video_id, 'Downloading filelist XML')
         file_name = file_list_doc.find('./video/item/fileName').text
         video_url = compat_urlparse.urljoin(video_folder, file_name)
 
-        timestamp = parse_iso8601(self._html_search_regex(
-            r"<div class='title'>Posted\s*:</div>\s*<div class='value'>([^<>]+)<",
-            page, 'creation time', fatal=False),
-            delimiter=' ', timezone=datetime.timedelta(hours=8))
-        view_count = str_to_int(self._html_search_regex(
-            r"<div class='title'>Views\s*:</div>\s*<div class='value'>([^<>]+)<",
-            page, 'view count', fatal=False))
+        # Some URLs return "No permission or not login" in a webpage despite being
+        # freely available via oembed JSON URL (e.g. http://www.camdemy.com/media/13885)
+        upload_date = unified_strdate(self._search_regex(
+            r'>published on ([^<]+)<', webpage,
+            'upload date', default=None))
+        view_count = str_to_int(self._search_regex(
+            r'role=["\']viewCnt["\'][^>]*>([\d,.]+) views',
+            webpage, 'view count', default=None))
+        description = self._html_search_meta(
+            'description', webpage, default=None) or clean_html(
+            oembed_obj.get('description'))
 
         return {
             'id': video_id,
             'url': video_url,
-            'title': oembed_obj['title'],
+            'title': title,
             'thumbnail': thumb_url,
-            'description': self._html_search_meta('description', page),
-            'creator': oembed_obj['author_name'],
-            'duration': oembed_obj['duration'],
-            'timestamp': timestamp,
+            'description': description,
+            'creator': oembed_obj.get('author_name'),
+            'duration': parse_duration(oembed_obj.get('duration')),
+            'upload_date': upload_date,
             'view_count': view_count,
         }
 

youtube_dl/extractor/canalplus.py

@@ -4,11 +4,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urllib_parse_urlparse
 from ..utils import (
     ExtractorError,
     HEADRequest,
     unified_strdate,
-    url_basename,
     qualities,
     int_or_none,
 )
@@ -16,24 +16,38 @@ from ..utils import (
 
 class CanalplusIE(InfoExtractor):
     IE_DESC = 'canalplus.fr, piwiplus.fr and d8.tv'
-    _VALID_URL = r'https?://(?:www\.(?P<site>canalplus\.fr|piwiplus\.fr|d8\.tv|itele\.fr)/.*?/(?P<path>.*)|player\.canalplus\.fr/#/(?P<id>[0-9]+))'
+    _VALID_URL = r'''(?x)
+                        https?://
+                            (?:
+                                (?:
+                                    (?:(?:www|m)\.)?canalplus\.fr|
+                                    (?:www\.)?piwiplus\.fr|
+                                    (?:www\.)?d8\.tv|
+                                    (?:www\.)?d17\.tv|
+                                    (?:www\.)?itele\.fr
+                                )/(?:(?:[^/]+/)*(?P<display_id>[^/?#&]+))?(?:\?.*\bvid=(?P<vid>\d+))?|
+                                player\.canalplus\.fr/#/(?P<id>\d+)
+                            )
+
+                    '''
     _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/%s/%s?format=json'
     _SITE_ID_MAP = {
-        'canalplus.fr': 'cplus',
-        'piwiplus.fr': 'teletoon',
-        'd8.tv': 'd8',
-        'itele.fr': 'itele',
+        'canalplus': 'cplus',
+        'piwiplus': 'teletoon',
+        'd8': 'd8',
+        'd17': 'd17',
+        'itele': 'itele',
     }
 
     _TESTS = [{
-        'url': 'http://www.canalplus.fr/c-emissions/pid1830-c-zapping.html?vid=1263092',
-        'md5': '12164a6f14ff6df8bd628e8ba9b10b78',
+        'url': 'http://www.canalplus.fr/c-emissions/pid1830-c-zapping.html?vid=1192814',
+        'md5': '41f438a4904f7664b91b4ed0dec969dc',
         'info_dict': {
-            'id': '1263092',
+            'id': '1192814',
             'ext': 'mp4',
-            'title': 'Le Zapping - 13/05/15',
-            'description': 'md5:09738c0d06be4b5d06a0940edb0da73f',
-            'upload_date': '20150513',
+            'title': "L'Année du Zapping 2014 - L'Année du Zapping 2014",
+            'description': "Toute l'année 2014 dans un Zapping exceptionnel !",
+            'upload_date': '20150105',
         },
     }, {
         'url': 'http://www.piwiplus.fr/videos-piwi/pid1405-le-labyrinthe-boing-super-ranger.html?vid=1108190',
@@ -46,35 +60,45 @@ class CanalplusIE(InfoExtractor):
         },
         'skip': 'Only works from France',
     }, {
-        'url': 'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html',
+        'url': 'http://www.d8.tv/d8-docs-mags/pid5198-d8-en-quete-d-actualite.html?vid=1390231',
         'info_dict': {
-            'id': '966289',
-            'ext': 'flv',
-            'title': 'Campagne intime - Documentaire exceptionnel',
-            'description': 'md5:d2643b799fb190846ae09c61e59a859f',
-            'upload_date': '20131108',
-        },
-        'skip': 'videos get deleted after a while',
-    }, {
-        'url': 'http://www.itele.fr/france/video/aubervilliers-un-lycee-en-colere-111559',
-        'md5': '38b8f7934def74f0d6f3ba6c036a5f82',
-        'info_dict': {
-            'id': '1213714',
+            'id': '1390231',
             'ext': 'mp4',
-            'title': 'Aubervilliers : un lycée en colère - Le 11/02/2015 à 06h45',
-            'description': 'md5:8216206ec53426ea6321321f3b3c16db',
-            'upload_date': '20150211',
+            'title': "Vacances pas chères : prix discount ou grosses dépenses ? - En quête d'actualité",
+            'description': 'md5:edb6cf1cb4a1e807b5dd089e1ac8bfc6',
+            'upload_date': '20160512',
         },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.itele.fr/chroniques/invite-bruce-toussaint/thierry-solere-nicolas-sarkozy-officialisera-sa-candidature-a-la-primaire-quand-il-le-voudra-167224',
+        'info_dict': {
+            'id': '1398334',
+            'ext': 'mp4',
+            'title': "L'invité de Bruce Toussaint du 07/06/2016 - ",
+            'description': 'md5:40ac7c9ad0feaeb6f605bad986f61324',
+            'upload_date': '20160607',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://m.canalplus.fr/?vid=1398231',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.d17.tv/emissions/pid8303-lolywood.html?vid=1397061',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.groupdict().get('id')
+        video_id = mobj.groupdict().get('id') or mobj.groupdict().get('vid')
 
-        site_id = self._SITE_ID_MAP[mobj.group('site') or 'canal']
+        site_id = self._SITE_ID_MAP[compat_urllib_parse_urlparse(url).netloc.rsplit('.', 2)[-2]]
 
         # Beware, some subclasses do not define an id group
-        display_id = url_basename(mobj.group('path'))
+        display_id = mobj.group('display_id') or video_id
 
         if video_id is None:
             webpage = self._download_webpage(url, display_id)

youtube_dl/extractor/carambatv.py

@@ -0,0 +1,88 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    float_or_none,
+    int_or_none,
+    try_get,
+)
+
+
+class CarambaTVIE(InfoExtractor):
+    _VALID_URL = r'(?:carambatv:|https?://video1\.carambatv\.ru/v/)(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://video1.carambatv.ru/v/191910501',
+        'md5': '2f4a81b7cfd5ab866ee2d7270cb34a2a',
+        'info_dict': {
+            'id': '191910501',
+            'ext': 'mp4',
+            'title': '[BadComedian] - Разборка в Маниле (Абсолютный обзор)',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'duration': 2678.31,
+        },
+    }, {
+        'url': 'carambatv:191910501',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        video = self._download_json(
+            'http://video1.carambatv.ru/v/%s/videoinfo.js' % video_id,
+            video_id)
+
+        title = video['title']
+
+        base_url = video.get('video') or 'http://video1.carambatv.ru/v/%s/' % video_id
+
+        formats = [{
+            'url': base_url + f['fn'],
+            'height': int_or_none(f.get('height')),
+            'format_id': '%sp' % f['height'] if f.get('height') else None,
+        } for f in video['qualities'] if f.get('fn')]
+        self._sort_formats(formats)
+
+        thumbnail = video.get('splash')
+        duration = float_or_none(try_get(
+            video, lambda x: x['annotations'][0]['end_time'], compat_str))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,
+        }
+
+
+class CarambaTVPageIE(InfoExtractor):
+    _VALID_URL = r'https?://carambatv\.ru/(?:[^/]+/)+(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'http://carambatv.ru/movie/bad-comedian/razborka-v-manile/',
+        'md5': '',
+        'info_dict': {
+            'id': '191910501',
+            'ext': 'mp4',
+            'title': '[BadComedian] - Разборка в Маниле (Абсолютный обзор)',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 2678.31,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        video_url = self._og_search_property('video:iframe', webpage, default=None)
+
+        if not video_url:
+            video_id = self._search_regex(
+                r'(?:video_id|crmb_vuid)\s*[:=]\s*["\']?(\d+)',
+                webpage, 'video id')
+            video_url = 'carambatv:%s' % video_id
+
+        return self.url_result(video_url, CarambaTVIE.ie_key())

youtube_dl/extractor/cbc.py

@@ -4,65 +4,83 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import js_to_json
+from ..compat import compat_str
+from ..utils import (
+    js_to_json,
+    smuggle_url,
+    try_get,
+)
 
 
 class CBCIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?:[^/]+/)+(?P<id>[^/?#]+)'
+    _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?!player/)(?:[^/]+/)+(?P<id>[^/?#]+)'
     _TESTS = [{
         # with mediaId
         'url': 'http://www.cbc.ca/22minutes/videos/clips-season-23/don-cherry-play-offs',
+        'md5': '97e24d09672fc4cf56256d6faa6c25bc',
         'info_dict': {
             'id': '2682904050',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Don Cherry – All-Stars',
             'description': 'Don Cherry has a bee in his bonnet about AHL player John Scott because that guy’s got heart.',
-            'timestamp': 1454475540,
+            'timestamp': 1454463000,
             'upload_date': '20160203',
-        },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        },
-    }, {
-        # with clipId
-        'url': 'http://www.cbc.ca/archives/entry/1978-robin-williams-freestyles-on-90-minutes-live',
-        'info_dict': {
-            'id': '2487345465',
-            'ext': 'flv',
-            'title': 'Robin Williams freestyles on 90 Minutes Live',
-            'description': 'Wacky American comedian Robin Williams shows off his infamous "freestyle" comedic talents while being interviewed on CBC\'s 90 Minutes Live.',
-            'upload_date': '19700101',
             'uploader': 'CBCC-NEW',
         },
-        'params': {
-            # rtmp download
-            'skip_download': True,
+        'skip': 'Geo-restricted to Canada',
+    }, {
+        # with clipId, feed available via tpfeed.cbc.ca and feed.theplatform.com
+        'url': 'http://www.cbc.ca/22minutes/videos/22-minutes-update/22-minutes-update-episode-4',
+        'md5': '162adfa070274b144f4fdc3c3b8207db',
+        'info_dict': {
+            'id': '2414435309',
+            'ext': 'mp4',
+            'title': '22 Minutes Update: What Not To Wear Quebec',
+            'description': "This week's latest Canadian top political story is What Not To Wear Quebec.",
+            'upload_date': '20131025',
+            'uploader': 'CBCC-NEW',
+            'timestamp': 1382717907,
+        },
+    }, {
+        # with clipId, feed only available via tpfeed.cbc.ca
+        'url': 'http://www.cbc.ca/archives/entry/1978-robin-williams-freestyles-on-90-minutes-live',
+        'md5': '0274a90b51a9b4971fe005c63f592f12',
+        'info_dict': {
+            'id': '2487345465',
+            'ext': 'mp4',
+            'title': 'Robin Williams freestyles on 90 Minutes Live',
+            'description': 'Wacky American comedian Robin Williams shows off his infamous "freestyle" comedic talents while being interviewed on CBC\'s 90 Minutes Live.',
+            'upload_date': '19780210',
+            'uploader': 'CBCC-NEW',
+            'timestamp': 255977160,
         },
     }, {
         # multiple iframes
         'url': 'http://www.cbc.ca/natureofthings/blog/birds-eye-view-from-vancouvers-burrard-street-bridge-how-we-got-the-shot',
         'playlist': [{
+            'md5': '377572d0b49c4ce0c9ad77470e0b96b4',
             'info_dict': {
                 'id': '2680832926',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'An Eagle\'s-Eye View Off Burrard Bridge',
                 'description': 'Hercules the eagle flies from Vancouver\'s Burrard Bridge down to a nearby park with a mini-camera strapped to his back.',
-                'upload_date': '19700101',
+                'upload_date': '20160201',
+                'timestamp': 1454342820,
+                'uploader': 'CBCC-NEW',
             },
         }, {
+            'md5': '415a0e3f586113894174dfb31aa5bb1a',
             'info_dict': {
                 'id': '2658915080',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Fly like an eagle!',
                 'description': 'Eagle equipped with a mini camera flies from the world\'s tallest tower',
-                'upload_date': '19700101',
+                'upload_date': '20150315',
+                'timestamp': 1426443984,
+                'uploader': 'CBCC-NEW',
             },
         }],
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        },
+        'skip': 'Geo-restricted to Canada',
     }]
 
     @classmethod
@@ -80,9 +98,15 @@ class CBCIE(InfoExtractor):
             media_id = player_info.get('mediaId')
             if not media_id:
                 clip_id = player_info['clipId']
-                media_id = self._download_json(
-                    'http://feed.theplatform.com/f/h9dtGB/punlNGjMlc1F?fields=id&byContent=byReleases%3DbyId%253D' + clip_id,
-                    clip_id)['entries'][0]['id'].split('/')[-1]
+                feed = self._download_json(
+                    'http://tpfeed.cbc.ca/f/ExhSPC/vms_5akSXx4Ng_Zn?byCustomValue={:mpsReleases}{%s}' % clip_id,
+                    clip_id, fatal=False)
+                if feed:
+                    media_id = try_get(feed, lambda x: x['entries'][0]['guid'], compat_str)
+                if not media_id:
+                    media_id = self._download_json(
+                        'http://feed.theplatform.com/f/h9dtGB/punlNGjMlc1F?fields=id&byContent=byReleases%3DbyId%253D' + clip_id,
+                        clip_id)['entries'][0]['id'].split('/')[-1]
             return self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id)
         else:
             entries = [self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id) for media_id in re.findall(r'<iframe[^>]+src="[^"]+?mediaId=(\d+)"', webpage)]
@@ -91,24 +115,55 @@ class CBCIE(InfoExtractor):
 
 class CBCPlayerIE(InfoExtractor):
     _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/|i/caffeine/syndicate/\?mediaId=))(?P<id>\d+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.cbc.ca/player/play/2683190193',
+        'md5': '64d25f841ddf4ddb28a235338af32e2c',
         'info_dict': {
             'id': '2683190193',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Gerry Runs a Sweat Shop',
             'description': 'md5:b457e1c01e8ff408d9d801c1c2cd29b0',
-            'timestamp': 1455067800,
+            'timestamp': 1455071400,
             'upload_date': '20160210',
+            'uploader': 'CBCC-NEW',
         },
-        'params': {
-            # rtmp download
-            'skip_download': True,
+        'skip': 'Geo-restricted to Canada',
+    }, {
+        # Redirected from http://www.cbc.ca/player/AudioMobile/All%20in%20a%20Weekend%20Montreal/ID/2657632011/
+        'url': 'http://www.cbc.ca/player/play/2657631896',
+        'md5': 'e5e708c34ae6fca156aafe17c43e8b75',
+        'info_dict': {
+            'id': '2657631896',
+            'ext': 'mp3',
+            'title': 'CBC Montreal is organizing its first ever community hackathon!',
+            'description': 'The modern technology we tend to depend on so heavily, is never without it\'s share of hiccups and headaches. Next weekend - CBC Montreal will be getting members of the public for its first Hackathon.',
+            'timestamp': 1425704400,
+            'upload_date': '20150307',
+            'uploader': 'CBCC-NEW',
         },
-    }
+    }, {
+        # available only when we add `formats=MPEG4,FLV,MP3` to theplatform url
+        'url': 'http://www.cbc.ca/player/play/2164402062',
+        'md5': '17a61eb813539abea40618d6323a7f82',
+        'info_dict': {
+            'id': '2164402062',
+            'ext': 'flv',
+            'title': 'Cancer survivor four times over',
+            'description': 'Tim Mayer has beaten three different forms of cancer four times in five years.',
+            'timestamp': 1320410746,
+            'upload_date': '20111104',
+            'uploader': 'CBCC-NEW',
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        return self.url_result(
-            'http://feed.theplatform.com/f/ExhSPC/vms_5akSXx4Ng_Zn?byGuid=%s' % video_id,
-            'ThePlatformFeed', video_id)
+        return {
+            '_type': 'url_transparent',
+            'ie_key': 'ThePlatform',
+            'url': smuggle_url(
+                'http://link.theplatform.com/s/ExhSPC/media/guid/2655402169/%s?mbr=true&formats=MPEG4,FLV,MP3' % video_id, {
+                    'force_smil_url': True
+                }),
+            'id': video_id,
+        }

youtube_dl/extractor/cbs.py

@@ -1,15 +1,13 @@
 from __future__ import unicode_literals
 
-from .theplatform import ThePlatformIE
+from .theplatform import ThePlatformFeedIE
 from ..utils import (
-    xpath_text,
-    xpath_element,
     int_or_none,
     find_xpath_attr,
 )
 
 
-class CBSBaseIE(ThePlatformIE):
+class CBSBaseIE(ThePlatformFeedIE):
     def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
         closed_caption_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', 'ClosedCaptionURL')
         return {
@@ -19,9 +17,22 @@ class CBSBaseIE(ThePlatformIE):
             }]
         } if closed_caption_e is not None and closed_caption_e.attrib.get('value') else []
 
+    def _extract_video_info(self, filter_query, video_id):
+        return self._extract_feed_info(
+            'dJ5BDC', 'VxxJg8Ymh8sE', filter_query, video_id, lambda entry: {
+                'series': entry.get('cbs$SeriesTitle'),
+                'season_number': int_or_none(entry.get('cbs$SeasonNumber')),
+                'episode': entry.get('cbs$EpisodeTitle'),
+                'episode_number': int_or_none(entry.get('cbs$EpisodeNumber')),
+            }, {
+                'StreamPack': {
+                    'manifest': 'm3u',
+                }
+            })
+
 
 class CBSIE(CBSBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/(?:video|artist)|colbertlateshow\.com/(?:video|podcasts))/[^/]+/(?P<id>[^/]+)'
+    _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)'
 
     _TESTS = [{
         'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
@@ -36,25 +47,7 @@ class CBSIE(CBSBaseIE):
             'upload_date': '20131127',
             'uploader': 'CBSI-NEW',
         },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        },
-        '_skip': 'Blocked outside the US',
-    }, {
-        'url': 'http://www.cbs.com/shows/liveonletterman/artist/221752/st-vincent/',
-        'info_dict': {
-            'id': 'WWF_5KqY3PK1',
-            'display_id': 'st-vincent',
-            'ext': 'flv',
-            'title': 'Live on Letterman - St. Vincent',
-            'description': 'Live On Letterman: St. Vincent in concert from New York\'s Ed Sullivan Theater on Tuesday, July 16, 2014.',
-            'duration': 3221,
-        },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        },
+        'expected_warnings': ['Failed to download m3u8 information'],
         '_skip': 'Blocked outside the US',
     }, {
         'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
@@ -66,43 +59,5 @@ class CBSIE(CBSBaseIE):
     TP_RELEASE_URL_TEMPLATE = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true'
 
     def _real_extract(self, url):
-        display_id = self._match_id(url)
-        webpage = self._download_webpage(url, display_id)
-        content_id = self._search_regex(
-            [r"video\.settings\.content_id\s*=\s*'([^']+)';", r"cbsplayer\.contentId\s*=\s*'([^']+)';"],
-            webpage, 'content id')
-        items_data = self._download_xml(
-            'http://can.cbs.com/thunder/player/videoPlayerService.php',
-            content_id, query={'partner': 'cbs', 'contentId': content_id})
-        video_data = xpath_element(items_data, './/item')
-        title = xpath_text(video_data, 'videoTitle', 'title', True)
-
-        subtitles = {}
-        formats = []
-        for item in items_data.findall('.//item'):
-            pid = xpath_text(item, 'pid')
-            if not pid:
-                continue
-            tp_release_url = self.TP_RELEASE_URL_TEMPLATE % pid
-            if '.m3u8' in xpath_text(item, 'contentUrl', default=''):
-                tp_release_url += '&manifest=m3u'
-            tp_formats, tp_subtitles = self._extract_theplatform_smil(
-                tp_release_url, content_id, 'Downloading %s SMIL data' % pid)
-            formats.extend(tp_formats)
-            subtitles = self._merge_subtitles(subtitles, tp_subtitles)
-        self._sort_formats(formats)
-
-        info = self.get_metadata('dJ5BDC/media/guid/2198311517/%s' % content_id, content_id)
-        info.update({
-            'id': content_id,
-            'display_id': display_id,
-            'title': title,
-            'series': xpath_text(video_data, 'seriesTitle'),
-            'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
-            'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),
-            'duration': int_or_none(xpath_text(video_data, 'videoLength'), 1000),
-            'thumbnail': xpath_text(video_data, 'previewImageURL'),
-            'formats': formats,
-            'subtitles': subtitles,
-        })
-        return info
+        content_id = self._match_id(url)
+        return self._extract_video_info('byGuid=%s' % content_id, content_id)

youtube_dl/extractor/cbsinteractive.py

@@ -80,9 +80,6 @@ class CBSInteractiveIE(ThePlatformIE):
 
         media_guid_path = 'media/guid/%d/%s' % (self.MPX_ACCOUNTS[site], vdata['mpxRefId'])
         formats, subtitles = [], {}
-        if site == 'cnet':
-            formats, subtitles = self._extract_theplatform_smil(
-                self.TP_RELEASE_URL_TEMPLATE % media_guid_path, video_id)
         for (fkey, vid) in vdata['files'].items():
             if fkey == 'hls_phone' and 'hls_tablet' in vdata['files']:
                 continue
@@ -94,7 +91,7 @@ class CBSInteractiveIE(ThePlatformIE):
             subtitles = self._merge_subtitles(subtitles, tp_subtitles)
         self._sort_formats(formats)
 
-        info = self.get_metadata('kYEXFC/%s' % media_guid_path, video_id)
+        info = self._extract_theplatform_metadata('kYEXFC/%s' % media_guid_path, video_id)
         info.update({
             'id': video_id,
             'display_id': display_id,

youtube_dl/extractor/cbslocal.py

@@ -0,0 +1,79 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .anvato import AnvatoIE
+from .sendtonews import SendtoNewsIE
+from ..compat import compat_urlparse
+from ..utils import unified_timestamp
+
+
+class CBSLocalIE(AnvatoIE):
+    _VALID_URL = r'https?://[a-z]+\.cbslocal\.com/\d+/\d+/\d+/(?P<id>[0-9a-z-]+)'
+
+    _TESTS = [{
+        # Anvato backend
+        'url': 'http://losangeles.cbslocal.com/2016/05/16/safety-advocates-say-fatal-car-seat-failures-are-public-health-crisis',
+        'md5': 'f0ee3081e3843f575fccef901199b212',
+        'info_dict': {
+            'id': '3401037',
+            'ext': 'mp4',
+            'title': 'Safety Advocates Say Fatal Car Seat Failures Are \'Public Health Crisis\'',
+            'description': 'Collapsing seats have been the focus of scrutiny for decades, though experts say remarkably little has been done to address the issue. Randy Paige reports.',
+            'thumbnail': 're:^https?://.*',
+            'timestamp': 1463440500,
+            'upload_date': '20160516',
+            'subtitles': {
+                'en': 'mincount:5',
+            },
+            'categories': [
+                'Stations\\Spoken Word\\KCBSTV',
+                'Syndication\\MSN',
+                'Syndication\\NDN',
+                'Syndication\\AOL',
+                'Syndication\\Yahoo',
+                'Syndication\\Tribune',
+                'Syndication\\Curb.tv',
+                'Content\\News'
+            ],
+        },
+    }, {
+        # SendtoNews embed
+        'url': 'http://cleveland.cbslocal.com/2016/05/16/indians-score-season-high-15-runs-in-blowout-win-over-reds-rapid-reaction/',
+        'info_dict': {
+            'id': 'GxfCe0Zo7D-175909-5588',
+            'ext': 'mp4',
+            'title': 'Recap: CLE 15, CIN 6',
+            'description': '5/16/16: Indians\' bats explode for 15 runs in a win',
+            'upload_date': '20160516',
+            'timestamp': 1463433840,
+            'duration': 49,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        sendtonews_url = SendtoNewsIE._extract_url(webpage)
+        if sendtonews_url:
+            info_dict = {
+                '_type': 'url_transparent',
+                'url': compat_urlparse.urljoin(url, sendtonews_url),
+            }
+        else:
+            info_dict = self._extract_anvato_videos(webpage, display_id)
+
+        time_str = self._html_search_regex(
+            r'class="entry-date">([^<]+)<', webpage, 'released date', fatal=False)
+        timestamp = unified_timestamp(time_str)
+
+        info_dict.update({
+            'display_id': display_id,
+            'timestamp': timestamp,
+        })
+
+        return info_dict

youtube_dl/extractor/cbsnews.py

@@ -26,13 +26,17 @@ class CBSNewsIE(CBSBaseIE):
                 # rtmp download
                 'skip_download': True,
             },
+            'skip': 'Subscribers only',
         },
         {
             'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
             'info_dict': {
-                'id': 'fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack',
+                'id': 'SNJBOYzXiWBOvaLsdzwH8fmtP1SCd91Y',
                 'ext': 'mp4',
                 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
+                'description': 'md5:4a6983e480542d8b333a947bfc64ddc7',
+                'upload_date': '19700101',
+                'uploader': 'CBSI-NEW',
                 'thumbnail': 're:^https?://.*\.jpg$',
                 'duration': 205,
                 'subtitles': {
@@ -58,37 +62,15 @@ class CBSNewsIE(CBSBaseIE):
             webpage, 'video JSON info'), video_id)
 
         item = video_info['item'] if 'item' in video_info else video_info
-        title = item.get('articleTitle') or item.get('hed')
-        duration = item.get('duration')
-        thumbnail = item.get('mediaImage') or item.get('thumbnail')
-
-        subtitles = {}
-        formats = []
-        for format_id in ['RtmpMobileLow', 'RtmpMobileHigh', 'Hls', 'RtmpDesktop']:
-            pid = item.get('media' + format_id)
-            if not pid:
-                continue
-            release_url = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true' % pid
-            tp_formats, tp_subtitles = self._extract_theplatform_smil(release_url, video_id, 'Downloading %s SMIL data' % pid)
-            formats.extend(tp_formats)
-            subtitles = self._merge_subtitles(subtitles, tp_subtitles)
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'formats': formats,
-            'subtitles': subtitles,
-        }
+        guid = item['mpxRefId']
+        return self._extract_video_info('byGuid=%s' % guid, guid)
 
 
 class CBSNewsLiveVideoIE(InfoExtractor):
     IE_DESC = 'CBS News Live Videos'
     _VALID_URL = r'https?://(?:www\.)?cbsnews\.com/live/video/(?P<id>[\da-z_-]+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.cbsnews.com/live/video/clinton-sanders-prepare-to-face-off-in-nh/',
         'info_dict': {
             'id': 'clinton-sanders-prepare-to-face-off-in-nh',
@@ -96,7 +78,15 @@ class CBSNewsLiveVideoIE(InfoExtractor):
             'title': 'Clinton, Sanders Prepare To Face Off In NH',
             'duration': 334,
         },
-    }
+        'skip': 'Video gone, redirected to http://www.cbsnews.com/live/',
+    }, {
+        'url': 'http://www.cbsnews.com/live/video/video-shows-intense-paragliding-accident/',
+        'info_dict': {
+            'id': 'video-shows-intense-paragliding-accident',
+            'ext': 'flv',
+            'title': 'Video Shows Intense Paragliding Accident',
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

youtube_dl/extractor/cbssports.py

@@ -1,30 +1,28 @@
 from __future__ import unicode_literals
 
-import re
-
-from .common import InfoExtractor
+from .cbs import CBSBaseIE
 
 
-class CBSSportsIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.cbssports\.com/video/player/(?P<section>[^/]+)/(?P<id>[^/]+)'
+class CBSSportsIE(CBSBaseIE):
+    _VALID_URL = r'https?://www\.cbssports\.com/video/player/[^/]+/(?P<id>\d+)'
 
-    _TEST = {
-        'url': 'http://www.cbssports.com/video/player/tennis/318462531970/0/us-open-flashbacks-1990s',
+    _TESTS = [{
+        'url': 'http://www.cbssports.com/video/player/videos/708337219968/0/ben-simmons-the-next-lebron?-not-so-fast',
         'info_dict': {
-            'id': '_d5_GbO8p1sT',
-            'ext': 'flv',
-            'title': 'US Open flashbacks: 1990s',
-            'description': 'Bill Macatee relives the best moments in US Open history from the 1990s.',
+            'id': '708337219968',
+            'ext': 'mp4',
+            'title': 'Ben Simmons the next LeBron? Not so fast',
+            'description': 'md5:854294f627921baba1f4b9a990d87197',
+            'timestamp': 1466293740,
+            'upload_date': '20160618',
+            'uploader': 'CBSI-NEW',
         },
-    }
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        }
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        section = mobj.group('section')
-        video_id = mobj.group('id')
-        all_videos = self._download_json(
-            'http://www.cbssports.com/data/video/player/getVideos/%s?as=json' % section,
-            video_id)
-        # The json file contains the info of all the videos in the section
-        video_info = next(v for v in all_videos if v['pcid'] == video_id)
-        return self.url_result('theplatform:%s' % video_info['pid'], 'ThePlatform')
+        video_id = self._match_id(url)
+        return self._extract_video_info('byId=%s' % video_id, video_id)

youtube_dl/extractor/ccc.py

@@ -1,13 +1,9 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
-    parse_duration,
-    qualities,
-    unified_strdate,
+    parse_iso8601,
 )
 
 
@@ -19,14 +15,14 @@ class CCCIE(InfoExtractor):
         'url': 'https://media.ccc.de/v/30C3_-_5443_-_en_-_saal_g_-_201312281830_-_introduction_to_processor_design_-_byterazor#video',
         'md5': '3a1eda8f3a29515d27f5adb967d7e740',
         'info_dict': {
-            'id': '30C3_-_5443_-_en_-_saal_g_-_201312281830_-_introduction_to_processor_design_-_byterazor',
+            'id': '1839',
             'ext': 'mp4',
             'title': 'Introduction to Processor Design',
-            'description': 'md5:80be298773966f66d56cb11260b879af',
+            'description': 'md5:df55f6d073d4ceae55aae6f2fd98a0ac',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'view_count': int,
             'upload_date': '20131228',
-            'duration': 3660,
+            'timestamp': 1388188800,
+            'duration': 3710,
         }
     }, {
         'url': 'https://media.ccc.de/v/32c3-7368-shopshifting#download',
@@ -34,79 +30,48 @@ class CCCIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        event_id = self._search_regex("data-id='(\d+)'", webpage, 'event id')
+        event_data = self._download_json('https://media.ccc.de/public/events/%s' % event_id, event_id)
 
-        if self._downloader.params.get('prefer_free_formats'):
-            preference = qualities(['mp3', 'opus', 'mp4-lq', 'webm-lq', 'h264-sd', 'mp4-sd', 'webm-sd', 'mp4', 'webm', 'mp4-hd', 'h264-hd', 'webm-hd'])
-        else:
-            preference = qualities(['opus', 'mp3', 'webm-lq', 'mp4-lq', 'webm-sd', 'h264-sd', 'mp4-sd', 'webm', 'mp4', 'webm-hd', 'mp4-hd', 'h264-hd'])
-
-        title = self._html_search_regex(
-            r'(?s)<h1>(.*?)</h1>', webpage, 'title')
-        description = self._html_search_regex(
-            r'(?s)<h3>About</h3>(.+?)<h3>',
-            webpage, 'description', fatal=False)
-        upload_date = unified_strdate(self._html_search_regex(
-            r"(?s)<span[^>]+class='[^']*fa-calendar-o'[^>]*>(.+?)</span>",
-            webpage, 'upload date', fatal=False))
-        view_count = int_or_none(self._html_search_regex(
-            r"(?s)<span class='[^']*fa-eye'></span>(.*?)</li>",
-            webpage, 'view count', fatal=False))
-        duration = parse_duration(self._html_search_regex(
-            r'(?s)<span[^>]+class=(["\']).*?fa-clock-o.*?\1[^>]*></span>(?P<duration>.+?)</li',
-            webpage, 'duration', fatal=False, group='duration'))
-
-        matches = re.finditer(r'''(?xs)
-            <(?:span|div)\s+class='label\s+filetype'>(?P<format>[^<]*)</(?:span|div)>\s*
-            <(?:span|div)\s+class='label\s+filetype'>(?P<lang>[^<]*)</(?:span|div)>\s*
-            <a\s+download\s+href='(?P<http_url>[^']+)'>\s*
-            (?:
-                .*?
-                <a\s+(?:download\s+)?href='(?P<torrent_url>[^']+\.torrent)'
-            )?''', webpage)
         formats = []
-        for m in matches:
-            format = m.group('format')
-            format_id = self._search_regex(
-                r'.*/([a-z0-9_-]+)/[^/]*$',
-                m.group('http_url'), 'format id', default=None)
-            if format_id:
-                format_id = m.group('lang') + '-' + format_id
-            vcodec = 'h264' if 'h264' in format_id else (
-                'none' if format_id in ('mp3', 'opus') else None
+        for recording in event_data.get('recordings', []):
+            recording_url = recording.get('recording_url')
+            if not recording_url:
+                continue
+            language = recording.get('language')
+            folder = recording.get('folder')
+            format_id = None
+            if language:
+                format_id = language
+            if folder:
+                if language:
+                    format_id += '-' + folder
+                else:
+                    format_id = folder
+            vcodec = 'h264' if 'h264' in folder else (
+                'none' if folder in ('mp3', 'opus') else None
             )
             formats.append({
                 'format_id': format_id,
-                'format': format,
-                'language': m.group('lang'),
-                'url': m.group('http_url'),
+                'url': recording_url,
+                'width': int_or_none(recording.get('width')),
+                'height': int_or_none(recording.get('height')),
+                'filesize': int_or_none(recording.get('size'), invscale=1024 * 1024),
+                'language': language,
                 'vcodec': vcodec,
-                'preference': preference(format_id),
             })
-
-            if m.group('torrent_url'):
-                formats.append({
-                    'format_id': 'torrent-%s' % (format if format_id is None else format_id),
-                    'format': '%s (torrent)' % format,
-                    'proto': 'torrent',
-                    'format_note': '(unsupported; will just download the .torrent file)',
-                    'vcodec': vcodec,
-                    'preference': -100 + preference(format_id),
-                    'url': m.group('torrent_url'),
-                })
         self._sort_formats(formats)
 
-        thumbnail = self._html_search_regex(
-            r"<video.*?poster='([^']+)'", webpage, 'thumbnail', fatal=False)
-
         return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'view_count': view_count,
-            'upload_date': upload_date,
-            'duration': duration,
+            'id': event_id,
+            'display_id': display_id,
+            'title': event_data['title'],
+            'description': event_data.get('description'),
+            'thumbnail': event_data.get('thumb_url'),
+            'timestamp': parse_iso8601(event_data.get('date')),
+            'duration': int_or_none(event_data.get('length')),
+            'tags': event_data.get('tags'),
             'formats': formats,
         }

youtube_dl/extractor/cda.py

@@ -58,7 +58,8 @@ class CDAIE(InfoExtractor):
         def extract_format(page, version):
             unpacked = decode_packed_codes(page)
             format_url = self._search_regex(
-                r"url:\\'(.+?)\\'", unpacked, '%s url' % version, fatal=False)
+                r"(?:file|url)\s*:\s*(\\?[\"'])(?P<url>http.+?)\1", unpacked,
+                '%s url' % version, fatal=False, group='url')
             if not format_url:
                 return
             f = {
@@ -75,7 +76,8 @@ class CDAIE(InfoExtractor):
             info_dict['formats'].append(f)
             if not info_dict['duration']:
                 info_dict['duration'] = parse_duration(self._search_regex(
-                    r"duration:\\'(.+?)\\'", unpacked, 'duration', fatal=False))
+                    r"duration\s*:\s*(\\?[\"'])(?P<duration>.+?)\1",
+                    unpacked, 'duration', fatal=False, group='duration'))
 
         extract_format(webpage, 'default')
 

youtube_dl/extractor/ceskatelevize.py

@@ -33,19 +33,33 @@ class CeskaTelevizeIE(InfoExtractor):
             'skip_download': True,
         },
     }, {
-        'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina',
+        'url': 'http://www.ceskatelevize.cz/ivysilani/10441294653-hyde-park-civilizace/215411058090502/bonus/20641-bonus-01-en',
         'info_dict': {
-            'id': '61924494876844374',
+            'id': '61924494877028507',
             'ext': 'mp4',
-            'title': 'První republika: Zpěvačka z Dupárny Bobina',
-            'description': 'Sága mapující atmosféru první republiky od r. 1918 do r. 1945.',
+            'title': 'Hyde Park Civilizace: Bonus 01 - En',
+            'description': 'English Subtittles',
             'thumbnail': 're:^https?://.*\.jpg',
-            'duration': 88.4,
+            'duration': 81.3,
         },
         'params': {
             # m3u8 download
             'skip_download': True,
         },
+    }, {
+        # live stream
+        'url': 'http://www.ceskatelevize.cz/ivysilani/zive/ct4/',
+        'info_dict': {
+            'id': 402,
+            'ext': 'mp4',
+            'title': 're:^ČT Sport \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
+            'is_live': True,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        'skip': 'Georestricted to Czech Republic',
     }, {
         # video with 18+ caution trailer
         'url': 'http://www.ceskatelevize.cz/porady/10520528904-queer/215562210900007-bogotart/',
@@ -118,19 +132,21 @@ class CeskaTelevizeIE(InfoExtractor):
         req = sanitized_Request(compat_urllib_parse_unquote(playlist_url))
         req.add_header('Referer', url)
 
-        playlist_title = self._og_search_title(webpage)
-        playlist_description = self._og_search_description(webpage)
+        playlist_title = self._og_search_title(webpage, default=None)
+        playlist_description = self._og_search_description(webpage, default=None)
 
         playlist = self._download_json(req, playlist_id)['playlist']
         playlist_len = len(playlist)
 
         entries = []
         for item in playlist:
+            is_live = item.get('type') == 'LIVE'
             formats = []
             for format_id, stream_url in item['streamUrls'].items():
                 formats.extend(self._extract_m3u8_formats(
                     stream_url, playlist_id, 'mp4',
-                    entry_protocol='m3u8_native', fatal=False))
+                    entry_protocol='m3u8' if is_live else 'm3u8_native',
+                    fatal=False))
             self._sort_formats(formats)
 
             item_id = item.get('id') or item['assetId']
@@ -145,14 +161,22 @@ class CeskaTelevizeIE(InfoExtractor):
                 if subs:
                     subtitles = self.extract_subtitles(episode_id, subs)
 
+            if playlist_len == 1:
+                final_title = playlist_title or title
+                if is_live:
+                    final_title = self._live_title(final_title)
+            else:
+                final_title = '%s (%s)' % (playlist_title, title)
+
             entries.append({
                 'id': item_id,
-                'title': playlist_title if playlist_len == 1 else '%s (%s)' % (playlist_title, title),
+                'title': final_title,
                 'description': playlist_description if playlist_len == 1 else None,
                 'thumbnail': thumbnail,
                 'duration': duration,
                 'formats': formats,
                 'subtitles': subtitles,
+                'is_live': is_live,
             })
 
         return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)

youtube_dl/extractor/channel9.py

@@ -20,54 +20,64 @@ class Channel9IE(InfoExtractor):
     '''
     IE_DESC = 'Channel 9'
     IE_NAME = 'channel9'
-    _VALID_URL = r'https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+)/?'
+    _VALID_URL = r'https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+?)(?P<rss>/RSS)?/?(?:[?#&]|$)'
 
-    _TESTS = [
-        {
-            'url': 'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002',
-            'md5': 'bbd75296ba47916b754e73c3a4bbdf10',
-            'info_dict': {
-                'id': 'Events/TechEd/Australia/2013/KOS002',
-                'ext': 'mp4',
-                'title': 'Developer Kick-Off Session: Stuff We Love',
-                'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
-                'duration': 4576,
-                'thumbnail': 're:http://.*\.jpg',
-                'session_code': 'KOS002',
-                'session_day': 'Day 1',
-                'session_room': 'Arena 1A',
-                'session_speakers': ['Ed Blankenship', 'Andrew Coates', 'Brady Gaster', 'Patrick Klug', 'Mads Kristensen'],
-            },
+    _TESTS = [{
+        'url': 'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002',
+        'md5': 'bbd75296ba47916b754e73c3a4bbdf10',
+        'info_dict': {
+            'id': 'Events/TechEd/Australia/2013/KOS002',
+            'ext': 'mp4',
+            'title': 'Developer Kick-Off Session: Stuff We Love',
+            'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
+            'duration': 4576,
+            'thumbnail': 're:http://.*\.jpg',
+            'session_code': 'KOS002',
+            'session_day': 'Day 1',
+            'session_room': 'Arena 1A',
+            'session_speakers': ['Ed Blankenship', 'Andrew Coates', 'Brady Gaster', 'Patrick Klug',
+                                 'Mads Kristensen'],
         },
-        {
-            'url': 'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing',
-            'md5': 'b43ee4529d111bc37ba7ee4f34813e68',
-            'info_dict': {
-                'id': 'posts/Self-service-BI-with-Power-BI-nuclear-testing',
-                'ext': 'mp4',
-                'title': 'Self-service BI with Power BI - nuclear testing',
-                'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
-                'duration': 1540,
-                'thumbnail': 're:http://.*\.jpg',
-                'authors': ['Mike Wilmot'],
-            },
+    }, {
+        'url': 'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing',
+        'md5': 'b43ee4529d111bc37ba7ee4f34813e68',
+        'info_dict': {
+            'id': 'posts/Self-service-BI-with-Power-BI-nuclear-testing',
+            'ext': 'mp4',
+            'title': 'Self-service BI with Power BI - nuclear testing',
+            'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
+            'duration': 1540,
+            'thumbnail': 're:http://.*\.jpg',
+            'authors': ['Mike Wilmot'],
         },
-        {
-            # low quality mp4 is best
-            'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
-            'info_dict': {
-                'id': 'Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
-                'ext': 'mp4',
-                'title': 'Ranges for the Standard Library',
-                'description': 'md5:2e6b4917677af3728c5f6d63784c4c5d',
-                'duration': 5646,
-                'thumbnail': 're:http://.*\.jpg',
-            },
-            'params': {
-                'skip_download': True,
-            },
-        }
-    ]
+    }, {
+        # low quality mp4 is best
+        'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
+        'info_dict': {
+            'id': 'Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
+            'ext': 'mp4',
+            'title': 'Ranges for the Standard Library',
+            'description': 'md5:2e6b4917677af3728c5f6d63784c4c5d',
+            'duration': 5646,
+            'thumbnail': 're:http://.*\.jpg',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://channel9.msdn.com/Niners/Splendid22/Queue/76acff796e8f411184b008028e0d492b/RSS',
+        'info_dict': {
+            'id': 'Niners/Splendid22/Queue/76acff796e8f411184b008028e0d492b',
+            'title': 'Channel 9',
+        },
+        'playlist_count': 2,
+    }, {
+        'url': 'https://channel9.msdn.com/Events/DEVintersection/DEVintersection-2016/RSS',
+        'only_matching': True,
+    }, {
+        'url': 'https://channel9.msdn.com/Events/Speakers/scott-hanselman/RSS?UrlSafeName=scott-hanselman',
+        'only_matching': True,
+    }]
 
     _RSS_URL = 'http://channel9.msdn.com/%s/RSS'
 
@@ -254,22 +264,30 @@ class Channel9IE(InfoExtractor):
 
         return self.playlist_result(contents)
 
-    def _extract_list(self, content_path):
-        rss = self._download_xml(self._RSS_URL % content_path, content_path, 'Downloading RSS')
+    def _extract_list(self, video_id, rss_url=None):
+        if not rss_url:
+            rss_url = self._RSS_URL % video_id
+        rss = self._download_xml(rss_url, video_id, '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)
+        return self.playlist_result(entries, video_id, title_text)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         content_path = mobj.group('contentpath')
+        rss = mobj.group('rss')
 
-        webpage = self._download_webpage(url, content_path, 'Downloading web page')
+        if rss:
+            return self._extract_list(content_path, url)
 
-        page_type_m = re.search(r'<meta name="WT.entryid" content="(?P<pagetype>[^:]+)[^"]+"/>', webpage)
-        if page_type_m is not None:
-            page_type = page_type_m.group('pagetype')
+        webpage = self._download_webpage(
+            url, content_path, 'Downloading web page')
+
+        page_type = self._search_regex(
+            r'<meta[^>]+name=(["\'])WT\.entryid\1[^>]+content=(["\'])(?P<pagetype>[^:]+).+?\2',
+            webpage, 'page type', default=None, group='pagetype')
+        if page_type:
             if page_type == 'Entry':      # 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
@@ -278,6 +296,5 @@ class Channel9IE(InfoExtractor):
                 return self._extract_list(content_path)
             else:
                 raise ExtractorError('Unexpected WT.entryid %s' % page_type, expected=True)
-
         else:  # Assuming list
             return self._extract_list(content_path)

youtube_dl/extractor/cinemassacre.py

@@ -1,119 +0,0 @@
-# encoding: utf-8
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..utils import ExtractorError
-from .screenwavemedia import ScreenwaveMediaIE
-
-
-class CinemassacreIE(InfoExtractor):
-    _VALID_URL = 'https?://(?:www\.)?cinemassacre\.com/(?P<date_y>[0-9]{4})/(?P<date_m>[0-9]{2})/(?P<date_d>[0-9]{2})/(?P<display_id>[^?#/]+)'
-    _TESTS = [
-        {
-            'url': 'http://cinemassacre.com/2012/11/10/avgn-the-movie-trailer/',
-            'md5': 'fde81fbafaee331785f58cd6c0d46190',
-            'info_dict': {
-                'id': 'Cinemassacre-19911',
-                'ext': 'mp4',
-                'upload_date': '20121110',
-                'title': '“Angry Video Game Nerd: The Movie” – Trailer',
-                'description': 'md5:fb87405fcb42a331742a0dce2708560b',
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
-        },
-        {
-            'url': 'http://cinemassacre.com/2013/10/02/the-mummys-hand-1940',
-            'md5': 'd72f10cd39eac4215048f62ab477a511',
-            'info_dict': {
-                'id': 'Cinemassacre-521be8ef82b16',
-                'ext': 'mp4',
-                'upload_date': '20131002',
-                'title': 'The Mummy’s Hand (1940)',
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
-        },
-        {
-            # Youtube embedded video
-            'url': 'http://cinemassacre.com/2006/12/07/chronologically-confused-about-bad-movie-and-video-game-sequel-titles/',
-            'md5': 'ec9838a5520ef5409b3e4e42fcb0a3b9',
-            'info_dict': {
-                'id': 'OEVzPCY2T-g',
-                'ext': 'webm',
-                'title': 'AVGN: Chronologically Confused about Bad Movie and Video Game Sequel Titles',
-                'upload_date': '20061207',
-                'uploader': 'Cinemassacre',
-                'uploader_id': 'JamesNintendoNerd',
-                'description': 'md5:784734696c2b8b7f4b8625cc799e07f6',
-            }
-        },
-        {
-            # Youtube embedded video
-            'url': 'http://cinemassacre.com/2006/09/01/mckids/',
-            'md5': '7393c4e0f54602ad110c793eb7a6513a',
-            'info_dict': {
-                'id': 'FnxsNhuikpo',
-                'ext': 'webm',
-                'upload_date': '20060901',
-                'uploader': 'Cinemassacre Extra',
-                'description': 'md5:de9b751efa9e45fbaafd9c8a1123ed53',
-                'uploader_id': 'Cinemassacre',
-                'title': 'AVGN: McKids',
-            }
-        },
-        {
-            'url': 'http://cinemassacre.com/2015/05/25/mario-kart-64-nintendo-64-james-mike-mondays/',
-            'md5': '1376908e49572389e7b06251a53cdd08',
-            'info_dict': {
-                'id': 'Cinemassacre-555779690c440',
-                'ext': 'mp4',
-                'description': 'Let’s Play Mario Kart 64 !! Mario Kart 64 is a classic go-kart racing game released for the Nintendo 64 (N64). Today James & Mike do 4 player Battle Mode with Kyle and Bootsy!',
-                'title': 'Mario Kart 64 (Nintendo 64) James & Mike Mondays',
-                'upload_date': '20150525',
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
-        }
-    ]
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('display_id')
-        video_date = mobj.group('date_y') + mobj.group('date_m') + mobj.group('date_d')
-
-        webpage = self._download_webpage(url, display_id)
-
-        playerdata_url = self._search_regex(
-            [
-                ScreenwaveMediaIE.EMBED_PATTERN,
-                r'<iframe[^>]+src="(?P<url>(?:https?:)?//(?:[^.]+\.)?youtube\.com/.+?)"',
-            ],
-            webpage, 'player data URL', default=None, group='url')
-        if not playerdata_url:
-            raise ExtractorError('Unable to find player data')
-
-        video_title = self._html_search_regex(
-            r'<title>(?P<title>.+?)\|', webpage, 'title')
-        video_description = self._html_search_regex(
-            r'<div class="entry-content">(?P<description>.+?)</div>',
-            webpage, 'description', flags=re.DOTALL, fatal=False)
-        video_thumbnail = self._og_search_thumbnail(webpage)
-
-        return {
-            '_type': 'url_transparent',
-            'display_id': display_id,
-            'title': video_title,
-            'description': video_description,
-            'upload_date': video_date,
-            'thumbnail': video_thumbnail,
-            'url': playerdata_url,
-        }

youtube_dl/extractor/cliphunter.py

@@ -23,7 +23,7 @@ class CliphunterIE(InfoExtractor):
         (?P<id>[0-9]+)/
         (?P<seo>.+?)(?:$|[#\?])
     '''
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.cliphunter.com/w/1012420/Fun_Jynx_Maze_solo',
         'md5': 'b7c9bbd4eb3a226ab91093714dcaa480',
         'info_dict': {
@@ -32,8 +32,19 @@ class CliphunterIE(InfoExtractor):
             'title': 'Fun Jynx Maze solo',
             'thumbnail': 're:^https?://.*\.jpg$',
             'age_limit': 18,
-        }
-    }
+        },
+        'skip': 'Video gone',
+    }, {
+        'url': 'http://www.cliphunter.com/w/2019449/ShesNew__My_booty_girlfriend_Victoria_Paradices_pussy_filled_with_jizz',
+        'md5': '55a723c67bfc6da6b0cfa00d55da8a27',
+        'info_dict': {
+            'id': '2019449',
+            'ext': 'mp4',
+            'title': 'ShesNew - My booty girlfriend, Victoria Paradice\'s pussy filled with jizz',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'age_limit': 18,
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

youtube_dl/extractor/cliprs.py

@@ -1,16 +1,10 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    float_or_none,
-    int_or_none,
-    parse_iso8601,
-)
+from .onet import OnetBaseIE
 
 
-class ClipRsIE(InfoExtractor):
+class ClipRsIE(OnetBaseIE):
     _VALID_URL = r'https?://(?:www\.)?clip\.rs/(?P<id>[^/]+)/\d+'
     _TEST = {
         'url': 'http://www.clip.rs/premijera-frajle-predstavljaju-novi-spot-za-pesmu-moli-me-moli/3732',
@@ -27,64 +21,13 @@ class ClipRsIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        display_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(url, display_id)
 
-        video_id = self._search_regex(
-            r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id')
+        mvp_id = self._search_mvp_id(webpage)
 
-        response = self._download_json(
-            'http://qi.ckm.onetapi.pl/', video_id,
-            query={
-                'body[id]': video_id,
-                'body[jsonrpc]': '2.0',
-                'body[method]': 'get_asset_detail',
-                'body[params][ID_Publikacji]': video_id,
-                'body[params][Service]': 'www.onet.pl',
-                'content-type': 'application/jsonp',
-                'x-onet-app': 'player.front.onetapi.pl',
-            })
+        info_dict = self._extract_from_id(mvp_id, webpage)
+        info_dict['display_id'] = display_id
 
-        error = response.get('error')
-        if error:
-            raise ExtractorError(
-                '%s said: %s' % (self.IE_NAME, error['message']), expected=True)
-
-        video = response['result'].get('0')
-
-        formats = []
-        for _, formats_dict in video['formats'].items():
-            if not isinstance(formats_dict, dict):
-                continue
-            for format_id, format_list in formats_dict.items():
-                if not isinstance(format_list, list):
-                    continue
-                for f in format_list:
-                    if not f.get('url'):
-                        continue
-                    formats.append({
-                        'url': f['url'],
-                        'format_id': format_id,
-                        'height': int_or_none(f.get('vertical_resolution')),
-                        'width': int_or_none(f.get('horizontal_resolution')),
-                        'abr': float_or_none(f.get('audio_bitrate')),
-                        'vbr': float_or_none(f.get('video_bitrate')),
-                    })
-        self._sort_formats(formats)
-
-        meta = video.get('meta', {})
-
-        title = self._og_search_title(webpage, default=None) or meta['title']
-        description = self._og_search_description(webpage, default=None) or meta.get('description')
-        duration = meta.get('length') or meta.get('lenght')
-        timestamp = parse_iso8601(meta.get('addDate'), ' ')
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'duration': duration,
-            'timestamp': timestamp,
-            'formats': formats,
-        }
+        return info_dict

youtube_dl/extractor/closertotruth.py

@@ -0,0 +1,92 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class CloserToTruthIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?closertotruth\.com/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'http://closertotruth.com/series/solutions-the-mind-body-problem#video-3688',
+        'info_dict': {
+            'id': '0_zof1ktre',
+            'display_id': 'solutions-the-mind-body-problem',
+            'ext': 'mov',
+            'title': 'Solutions to the Mind-Body Problem?',
+            'upload_date': '20140221',
+            'timestamp': 1392956007,
+            'uploader_id': 'CTTXML'
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://closertotruth.com/episodes/how-do-brains-work',
+        'info_dict': {
+            'id': '0_iuxai6g6',
+            'display_id': 'how-do-brains-work',
+            'ext': 'mov',
+            'title': 'How do Brains Work?',
+            'upload_date': '20140221',
+            'timestamp': 1392956024,
+            'uploader_id': 'CTTXML'
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://closertotruth.com/interviews/1725',
+        'info_dict': {
+            'id': '1725',
+            'title': 'AyaFr-002',
+        },
+        'playlist_mincount': 2,
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        partner_id = self._search_regex(
+            r'<script[^>]+src=["\'].*?\b(?:partner_id|p)/(\d+)',
+            webpage, 'kaltura partner_id')
+
+        title = self._search_regex(
+            r'<title>(.+?)\s*\|\s*.+?</title>', webpage, 'video title')
+
+        select = self._search_regex(
+            r'(?s)<select[^>]+id="select-version"[^>]*>(.+?)</select>',
+            webpage, 'select version', default=None)
+        if select:
+            entry_ids = set()
+            entries = []
+            for mobj in re.finditer(
+                    r'<option[^>]+value=(["\'])(?P<id>[0-9a-z_]+)(?:#.+?)?\1[^>]*>(?P<title>[^<]+)',
+                    webpage):
+                entry_id = mobj.group('id')
+                if entry_id in entry_ids:
+                    continue
+                entry_ids.add(entry_id)
+                entries.append({
+                    '_type': 'url_transparent',
+                    'url': 'kaltura:%s:%s' % (partner_id, entry_id),
+                    'ie_key': 'Kaltura',
+                    'title': mobj.group('title'),
+                })
+            if entries:
+                return self.playlist_result(entries, display_id, title)
+
+        entry_id = self._search_regex(
+            r'<a[^>]+id=(["\'])embed-kaltura\1[^>]+data-kaltura=(["\'])(?P<id>[0-9a-z_]+)\2',
+            webpage, 'kaltura entry_id', group='id')
+
+        return {
+            '_type': 'url_transparent',
+            'display_id': display_id,
+            'url': 'kaltura:%s:%s' % (partner_id, entry_id),
+            'ie_key': 'Kaltura',
+            'title': title
+        }

youtube_dl/extractor/cloudy.py

@@ -6,7 +6,6 @@ import re
 from .common import InfoExtractor
 from ..compat import (
     compat_parse_qs,
-    compat_urllib_parse_urlencode,
     compat_HTTPError,
 )
 from ..utils import (
@@ -17,37 +16,26 @@ from ..utils import (
 
 
 class CloudyIE(InfoExtractor):
-    _IE_DESC = 'cloudy.ec and videoraj.ch'
+    _IE_DESC = 'cloudy.ec'
     _VALID_URL = r'''(?x)
-        https?://(?:www\.)?(?P<host>cloudy\.ec|videoraj\.ch)/
+        https?://(?:www\.)?cloudy\.ec/
         (?:v/|embed\.php\?id=)
         (?P<id>[A-Za-z0-9]+)
         '''
-    _EMBED_URL = 'http://www.%s/embed.php?id=%s'
-    _API_URL = 'http://www.%s/api/player.api.php?%s'
+    _EMBED_URL = 'http://www.cloudy.ec/embed.php?id=%s'
+    _API_URL = 'http://www.cloudy.ec/api/player.api.php'
     _MAX_TRIES = 2
-    _TESTS = [
-        {
-            'url': 'https://www.cloudy.ec/v/af511e2527aac',
-            'md5': '5cb253ace826a42f35b4740539bedf07',
-            'info_dict': {
-                'id': 'af511e2527aac',
-                'ext': 'flv',
-                'title': 'Funny Cats and Animals Compilation june 2013',
-            }
-        },
-        {
-            'url': 'http://www.videoraj.ch/v/47f399fd8bb60',
-            'md5': '7d0f8799d91efd4eda26587421c3c3b0',
-            'info_dict': {
-                'id': '47f399fd8bb60',
-                'ext': 'flv',
-                'title': 'Burning a New iPhone 5 with Gasoline - Will it Survive?',
-            }
+    _TEST = {
+        'url': 'https://www.cloudy.ec/v/af511e2527aac',
+        'md5': '5cb253ace826a42f35b4740539bedf07',
+        'info_dict': {
+            'id': 'af511e2527aac',
+            'ext': 'flv',
+            'title': 'Funny Cats and Animals Compilation june 2013',
         }
-    ]
+    }
 
-    def _extract_video(self, video_host, video_id, file_key, error_url=None, try_num=0):
+    def _extract_video(self, video_id, file_key, error_url=None, try_num=0):
 
         if try_num > self._MAX_TRIES - 1:
             raise ExtractorError('Unable to extract video URL', expected=True)
@@ -64,9 +52,8 @@ class CloudyIE(InfoExtractor):
                 'errorUrl': error_url,
             })
 
-        data_url = self._API_URL % (video_host, compat_urllib_parse_urlencode(form))
         player_data = self._download_webpage(
-            data_url, video_id, 'Downloading player data')
+            self._API_URL, video_id, 'Downloading player data', query=form)
         data = compat_parse_qs(player_data)
 
         try_num += 1
@@ -88,7 +75,7 @@ class CloudyIE(InfoExtractor):
             except ExtractorError as e:
                 if isinstance(e.cause, compat_HTTPError) and e.cause.code in [404, 410]:
                     self.report_warning('Invalid video URL, requesting another', video_id)
-                    return self._extract_video(video_host, video_id, file_key, video_url, try_num)
+                    return self._extract_video(video_id, file_key, video_url, try_num)
 
         return {
             'id': video_id,
@@ -98,14 +85,13 @@ class CloudyIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_host = mobj.group('host')
         video_id = mobj.group('id')
 
-        url = self._EMBED_URL % (video_host, video_id)
+        url = self._EMBED_URL % video_id
         webpage = self._download_webpage(url, video_id)
 
         file_key = self._search_regex(
             [r'key\s*:\s*"([^"]+)"', r'filekey\s*=\s*"([^"]+)"'],
             webpage, 'file_key')
 
-        return self._extract_video(video_host, video_id, file_key)
+        return self._extract_video(video_id, file_key)

youtube_dl/extractor/cmt.py

@@ -1,5 +1,7 @@
 from __future__ import unicode_literals
+
 from .mtv import MTVIE
+from ..utils import ExtractorError
 
 
 class CMTIE(MTVIE):
@@ -16,7 +18,27 @@ class CMTIE(MTVIE):
             'title': 'Garth Brooks - "The Call (featuring Trisha Yearwood)"',
             'description': 'Blame It All On My Roots',
         },
+        'skip': 'Video not available',
+    }, {
+        'url': 'http://www.cmt.com/videos/misc/1504699/still-the-king-ep-109-in-3-minutes.jhtml#id=1739908',
+        'md5': 'e61a801ca4a183a466c08bd98dccbb1c',
+        'info_dict': {
+            'id': '1504699',
+            'ext': 'mp4',
+            'title': 'Still The King Ep. 109 in 3 Minutes',
+            'description': 'Relive or catch up with Still The King by watching this recap of season 1, episode 9. New episodes Sundays 9/8c.',
+            'timestamp': 1469421000.0,
+            'upload_date': '20160725',
+        },
     }, {
         'url': 'http://www.cmt.com/shows/party-down-south/party-down-south-ep-407-gone-girl/1738172/playlist/#id=1738172',
         'only_matching': True,
     }]
+
+    @classmethod
+    def _transform_rtmp_url(cls, rtmp_video_url):
+        if 'error_not_available.swf' in rtmp_video_url:
+            raise ExtractorError(
+                '%s said: video is not available' % cls.IE_NAME, expected=True)
+
+        return super(CMTIE, cls)._transform_rtmp_url(rtmp_video_url)

youtube_dl/extractor/collegehumor.py

@@ -1,101 +0,0 @@
-from __future__ import unicode_literals
-
-import json
-import re
-
-from .common import InfoExtractor
-from ..utils import int_or_none
-
-
-class CollegeHumorIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/(video|embed|e)/(?P<videoid>[0-9]+)/?(?P<shorttitle>.*)$'
-
-    _TESTS = [
-        {
-            'url': 'http://www.collegehumor.com/video/6902724/comic-con-cosplay-catastrophe',
-            'md5': 'dcc0f5c1c8be98dc33889a191f4c26bd',
-            'info_dict': {
-                'id': '6902724',
-                'ext': 'mp4',
-                'title': 'Comic-Con Cosplay Catastrophe',
-                'description': "Fans get creative this year at San Diego.  Too creative.  And yes, that's really Joss Whedon.",
-                'age_limit': 13,
-                'duration': 187,
-            },
-        }, {
-            'url': 'http://www.collegehumor.com/video/3505939/font-conference',
-            'md5': '72fa701d8ef38664a4dbb9e2ab721816',
-            'info_dict': {
-                'id': '3505939',
-                'ext': 'mp4',
-                'title': 'Font Conference',
-                'description': "This video wasn't long enough, so we made it double-spaced.",
-                'age_limit': 10,
-                'duration': 179,
-            },
-        }, {
-            # embedded youtube video
-            'url': 'http://www.collegehumor.com/embed/6950306',
-            'info_dict': {
-                'id': 'Z-bao9fg6Yc',
-                'ext': 'mp4',
-                'title': 'Young Americans Think President John F. Kennedy Died THIS MORNING IN A CAR ACCIDENT!!!',
-                'uploader': 'Mark Dice',
-                'uploader_id': 'MarkDice',
-                'description': 'md5:62c3dab9351fac7bb44b53b69511d87f',
-                'upload_date': '20140127',
-            },
-            'params': {
-                'skip_download': True,
-            },
-            'add_ie': ['Youtube'],
-        },
-    ]
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('videoid')
-
-        jsonUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id + '.json'
-        data = json.loads(self._download_webpage(
-            jsonUrl, video_id, 'Downloading info JSON'))
-        vdata = data['video']
-        if vdata.get('youtubeId') is not None:
-            return {
-                '_type': 'url',
-                'url': vdata['youtubeId'],
-                'ie_key': 'Youtube',
-            }
-
-        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:
-            age_limit = None  # None = No idea
-
-        PREFS = {'high_quality': 2, 'low_quality': 0}
-        formats = []
-        for format_key in ('mp4', 'webm'):
-            for qname, qurl in vdata.get(format_key, {}).items():
-                formats.append({
-                    'format_id': format_key + '_' + qname,
-                    'url': qurl,
-                    'format': format_key,
-                    'preference': PREFS.get(qname),
-                })
-        self._sort_formats(formats)
-
-        duration = int_or_none(vdata.get('duration'), 1000)
-        like_count = int_or_none(vdata.get('likes'))
-
-        return {
-            'id': video_id,
-            'title': vdata['title'],
-            'description': vdata.get('description'),
-            'thumbnail': vdata.get('thumbnail'),
-            'formats': formats,
-            'age_limit': age_limit,
-            'duration': duration,
-            'like_count': like_count,
-        }

youtube_dl/extractor/comedycentral.py

@@ -1,17 +1,7 @@
 from __future__ import unicode_literals
 
-import re
-
 from .mtv import MTVServicesInfoExtractor
-from ..compat import (
-    compat_str,
-    compat_urllib_parse_urlencode,
-)
-from ..utils import (
-    ExtractorError,
-    float_or_none,
-    unified_strdate,
-)
+from .common import InfoExtractor
 
 
 class ComedyCentralIE(MTVServicesInfoExtractor):
@@ -26,8 +16,10 @@ class ComedyCentralIE(MTVServicesInfoExtractor):
         'info_dict': {
             'id': 'cef0cbb3-e776-4bc9-b62e-8016deccb354',
             'ext': 'mp4',
-            'title': 'CC:Stand-Up|Greg Fitzsimmons: Life on Stage|Uncensored - Too Good of a Mother',
+            'title': 'CC:Stand-Up|August 18, 2013|1|0101|Uncensored - Too Good of a Mother',
             'description': 'After a certain point, breastfeeding becomes c**kblocking.',
+            'timestamp': 1376798400,
+            'upload_date': '20130818',
         },
     }, {
         'url': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/interviews/6yx39d/exclusive-rand-paul-extended-interview',
@@ -35,238 +27,92 @@ class ComedyCentralIE(MTVServicesInfoExtractor):
     }]
 
 
-class ComedyCentralShowsIE(MTVServicesInfoExtractor):
-    IE_DESC = 'The Daily Show / The Colbert Report'
-    # urls can be abbreviations like :thedailyshow
-    # urls for episodes like:
-    # or urls for clips like: http://www.thedailyshow.com/watch/mon-december-10-2012/any-given-gun-day
-    #                     or: http://www.colbertnation.com/the-colbert-report-videos/421667/november-29-2012/moon-shattering-news
-    #                     or: http://www.colbertnation.com/the-colbert-report-collections/422008/festival-of-lights/79524
-    _VALID_URL = r'''(?x)^(:(?P<shortname>tds|thedailyshow)
-                      |https?://(:www\.)?
-                          (?P<showname>thedailyshow|thecolbertreport)\.(?:cc\.)?com/
-                         ((?:full-)?episodes/(?:[0-9a-z]{6}/)?(?P<episode>.*)|
-                          (?P<clip>
-                              (?:(?:guests/[^/]+|videos|video-playlists|special-editions|news-team/[^/]+)/[^/]+/(?P<videotitle>[^/?#]+))
-                              |(the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
-                              |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*))
-                          )|
-                          (?P<interview>
-                              extended-interviews/(?P<interID>[0-9a-z]+)/
-                              (?:playlist_tds_extended_)?(?P<interview_title>[^/?#]*?)
-                              (?:/[^/?#]?|[?#]|$))))
-                     '''
+class ToshIE(MTVServicesInfoExtractor):
+    IE_DESC = 'Tosh.0'
+    _VALID_URL = r'^https?://tosh\.cc\.com/video-(?:clips|collections)/[^/]+/(?P<videotitle>[^/?#]+)'
+    _FEED_URL = 'http://tosh.cc.com/feeds/mrss'
+
     _TESTS = [{
-        'url': 'http://thedailyshow.cc.com/watch/thu-december-13-2012/kristen-stewart',
-        'md5': '4e2f5cb088a83cd8cdb7756132f9739d',
+        'url': 'http://tosh.cc.com/video-clips/68g93d/twitter-users-share-summer-plans',
         'info_dict': {
-            'id': 'ab9ab3e7-5a98-4dbe-8b21-551dc0523d55',
-            'ext': 'mp4',
-            'upload_date': '20121213',
-            'description': 'Kristen Stewart learns to let loose in "On the Road."',
-            'uploader': 'thedailyshow',
-            'title': 'thedailyshow kristen-stewart part 1',
-        }
-    }, {
-        'url': 'http://thedailyshow.cc.com/extended-interviews/b6364d/sarah-chayes-extended-interview',
-        'info_dict': {
-            'id': 'sarah-chayes-extended-interview',
-            'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
-            'title': 'thedailyshow Sarah Chayes Extended Interview',
+            'description': 'Tosh asked fans to share their summer plans.',
+            'title': 'Twitter Users Share Summer Plans',
         },
-        'playlist': [
-            {
-                'info_dict': {
-                    'id': '0baad492-cbec-4ec1-9e50-ad91c291127f',
-                    'ext': 'mp4',
-                    'upload_date': '20150129',
-                    'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
-                    'uploader': 'thedailyshow',
-                    'title': 'thedailyshow sarah-chayes-extended-interview part 1',
+        'playlist': [{
+            'md5': 'f269e88114c1805bb6d7653fecea9e06',
+            'info_dict': {
+                'id': '90498ec2-ed00-11e0-aca6-0026b9414f30',
+                'ext': 'mp4',
+                'title': 'Tosh.0|June 9, 2077|2|211|Twitter Users Share Summer Plans',
+                'description': 'Tosh asked fans to share their summer plans.',
+                'thumbnail': 're:^https?://.*\.jpg',
+                # It's really reported to be published on year 2077
+                'upload_date': '20770610',
+                'timestamp': 3390510600,
+                'subtitles': {
+                    'en': 'mincount:3',
                 },
             },
-            {
-                'info_dict': {
-                    'id': '1e4fb91b-8ce7-4277-bd7c-98c9f1bbd283',
-                    'ext': 'mp4',
-                    'upload_date': '20150129',
-                    'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
-                    'uploader': 'thedailyshow',
-                    'title': 'thedailyshow sarah-chayes-extended-interview part 2',
-                },
-            },
-        ],
-        'params': {
-            'skip_download': True,
-        },
+        }]
     }, {
-        'url': 'http://thedailyshow.cc.com/extended-interviews/xm3fnq/andrew-napolitano-extended-interview',
-        'only_matching': True,
-    }, {
-        'url': 'http://thecolbertreport.cc.com/videos/29w6fx/-realhumanpraise-for-fox-news',
-        'only_matching': True,
-    }, {
-        'url': 'http://thecolbertreport.cc.com/videos/gh6urb/neil-degrasse-tyson-pt--1?xrs=eml_col_031114',
-        'only_matching': True,
-    }, {
-        'url': 'http://thedailyshow.cc.com/guests/michael-lewis/3efna8/exclusive---michael-lewis-extended-interview-pt--3',
-        'only_matching': True,
-    }, {
-        'url': 'http://thedailyshow.cc.com/episodes/sy7yv0/april-8--2014---denis-leary',
-        'only_matching': True,
-    }, {
-        'url': 'http://thecolbertreport.cc.com/episodes/8ase07/april-8--2014---jane-goodall',
-        'only_matching': True,
-    }, {
-        'url': 'http://thedailyshow.cc.com/video-playlists/npde3s/the-daily-show-19088-highlights',
-        'only_matching': True,
-    }, {
-        'url': 'http://thedailyshow.cc.com/video-playlists/t6d9sg/the-daily-show-20038-highlights/be3cwo',
-        'only_matching': True,
-    }, {
-        'url': 'http://thedailyshow.cc.com/special-editions/2l8fdb/special-edition---a-look-back-at-food',
-        'only_matching': True,
-    }, {
-        'url': 'http://thedailyshow.cc.com/news-team/michael-che/7wnfel/we-need-to-talk-about-israel',
+        'url': 'http://tosh.cc.com/video-collections/x2iz7k/just-plain-foul/m5q4fp',
         'only_matching': True,
     }]
 
-    _available_formats = ['3500', '2200', '1700', '1200', '750', '400']
+    @classmethod
+    def _transform_rtmp_url(cls, rtmp_video_url):
+        new_urls = super(ToshIE, cls)._transform_rtmp_url(rtmp_video_url)
+        new_urls['rtmp'] = rtmp_video_url.replace('viacomccstrm', 'viacommtvstrm')
+        return new_urls
 
-    _video_extensions = {
-        '3500': 'mp4',
-        '2200': 'mp4',
-        '1700': 'mp4',
-        '1200': 'mp4',
-        '750': 'mp4',
-        '400': 'mp4',
-    }
-    _video_dimensions = {
-        '3500': (1280, 720),
-        '2200': (960, 540),
-        '1700': (768, 432),
-        '1200': (640, 360),
-        '750': (512, 288),
-        '400': (384, 216),
-    }
+
+class ComedyCentralTVIE(MTVServicesInfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?comedycentral\.tv/(?:staffeln|shows)/(?P<id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'http://www.comedycentral.tv/staffeln/7436-the-mindy-project-staffel-4',
+        'info_dict': {
+            'id': 'local_playlist-f99b626bdfe13568579a',
+            'ext': 'flv',
+            'title': 'Episode_the-mindy-project_shows_season-4_episode-3_full-episode_part1',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.comedycentral.tv/shows/1074-workaholics',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.comedycentral.tv/shows/1727-the-mindy-project/bonus',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
 
-        if mobj.group('shortname'):
-            return self.url_result('http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes')
+        webpage = self._download_webpage(url, video_id)
 
-        if mobj.group('clip'):
-            if mobj.group('videotitle'):
-                epTitle = mobj.group('videotitle')
-            elif mobj.group('showname') == 'thedailyshow':
-                epTitle = mobj.group('tdstitle')
-            else:
-                epTitle = mobj.group('cntitle')
-            dlNewest = False
-        elif mobj.group('interview'):
-            epTitle = mobj.group('interview_title')
-            dlNewest = False
-        else:
-            dlNewest = not mobj.group('episode')
-            if dlNewest:
-                epTitle = mobj.group('showname')
-            else:
-                epTitle = mobj.group('episode')
-        show_name = mobj.group('showname')
+        mrss_url = self._search_regex(
+            r'data-mrss=(["\'])(?P<url>(?:(?!\1).)+)\1',
+            webpage, 'mrss url', group='url')
 
-        webpage, htmlHandle = self._download_webpage_handle(url, epTitle)
-        if dlNewest:
-            url = htmlHandle.geturl()
-            mobj = re.match(self._VALID_URL, url, re.VERBOSE)
-            if mobj is None:
-                raise ExtractorError('Invalid redirected URL: ' + url)
-            if mobj.group('episode') == '':
-                raise ExtractorError('Redirected URL is still not specific: ' + url)
-            epTitle = (mobj.group('episode') or mobj.group('videotitle')).rpartition('/')[-1]
+        return self._get_videos_info_from_url(mrss_url, video_id)
 
-        mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
-        if len(mMovieParams) == 0:
-            # The Colbert Report embeds the information in a without
-            # a URL prefix; so extract the alternate reference
-            # and then add the URL prefix manually.
 
-            altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video|playlist).*?:.*?)"', webpage)
-            if len(altMovieParams) == 0:
-                raise ExtractorError('unable to find Flash URL in webpage ' + url)
-            else:
-                mMovieParams = [('http://media.mtvnservices.com/' + altMovieParams[0], altMovieParams[0])]
+class ComedyCentralShortnameIE(InfoExtractor):
+    _VALID_URL = r'^:(?P<id>tds|thedailyshow)$'
+    _TESTS = [{
+        'url': ':tds',
+        'only_matching': True,
+    }, {
+        'url': ':thedailyshow',
+        'only_matching': True,
+    }]
 
-        uri = mMovieParams[0][1]
-        # Correct cc.com in uri
-        uri = re.sub(r'(episode:[^.]+)(\.cc)?\.com', r'\1.com', uri)
-
-        index_url = 'http://%s.cc.com/feeds/mrss?%s' % (show_name, compat_urllib_parse_urlencode({'uri': uri}))
-        idoc = self._download_xml(
-            index_url, epTitle,
-            'Downloading show index', 'Unable to download episode index')
-
-        title = idoc.find('./channel/title').text
-        description = idoc.find('./channel/description').text
-
-        entries = []
-        item_els = idoc.findall('.//item')
-        for part_num, itemEl in enumerate(item_els):
-            upload_date = unified_strdate(itemEl.findall('./pubDate')[0].text)
-            thumbnail = itemEl.find('.//{http://search.yahoo.com/mrss/}thumbnail').attrib.get('url')
-
-            content = itemEl.find('.//{http://search.yahoo.com/mrss/}content')
-            duration = float_or_none(content.attrib.get('duration'))
-            mediagen_url = content.attrib['url']
-            guid = itemEl.find('./guid').text.rpartition(':')[-1]
-
-            cdoc = self._download_xml(
-                mediagen_url, epTitle,
-                'Downloading configuration for segment %d / %d' % (part_num + 1, len(item_els)))
-
-            turls = []
-            for rendition in cdoc.findall('.//rendition'):
-                finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
-                turls.append(finfo)
-
-            formats = []
-            for format, rtmp_video_url in turls:
-                w, h = self._video_dimensions.get(format, (None, None))
-                formats.append({
-                    'format_id': 'vhttp-%s' % format,
-                    'url': self._transform_rtmp_url(rtmp_video_url),
-                    'ext': self._video_extensions.get(format, 'mp4'),
-                    'height': h,
-                    'width': w,
-                })
-                formats.append({
-                    'format_id': 'rtmp-%s' % format,
-                    'url': rtmp_video_url.replace('viacomccstrm', 'viacommtvstrm'),
-                    'ext': self._video_extensions.get(format, 'mp4'),
-                    'height': h,
-                    'width': w,
-                })
-                self._sort_formats(formats)
-
-            subtitles = self._extract_subtitles(cdoc, guid)
-
-            virtual_id = show_name + ' ' + epTitle + ' part ' + compat_str(part_num + 1)
-            entries.append({
-                'id': guid,
-                'title': virtual_id,
-                'formats': formats,
-                'uploader': show_name,
-                'upload_date': upload_date,
-                'duration': duration,
-                'thumbnail': thumbnail,
-                'description': description,
-                'subtitles': subtitles,
-            })
-
-        return {
-            '_type': 'playlist',
-            'id': epTitle,
-            'entries': entries,
-            'title': show_name + ' ' + title,
-            'description': description,
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        shortcut_map = {
+            'tds': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes',
+            'thedailyshow': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes',
         }
+        return self.url_result(shortcut_map[video_id])

youtube_dl/extractor/common.py

@@ -44,7 +44,9 @@ from ..utils import (
     sanitized_Request,
     unescapeHTML,
     unified_strdate,
+    unified_timestamp,
     url_basename,
+    xpath_element,
     xpath_text,
     xpath_with_ns,
     determine_protocol,
@@ -52,6 +54,9 @@ from ..utils import (
     mimetype2ext,
     update_Request,
     update_url_query,
+    parse_m3u8_attributes,
+    extract_attributes,
+    parse_codecs,
 )
 
 
@@ -159,11 +164,12 @@ class InfoExtractor(object):
                         * "height" (optional, int)
                         * "resolution" (optional, string "{width}x{height"},
                                         deprecated)
+                        * "filesize" (optional, int)
     thumbnail:      Full URL to a video thumbnail image.
     description:    Full video description.
     uploader:       Full name of the video uploader.
     license:        License name the video is licensed under.
-    creator:        The main artist who created the video.
+    creator:        The creator of the video.
     release_date:   The date (YYYYMMDD) when the video was released.
     timestamp:      UNIX timestamp of the moment the video became available.
     upload_date:    Video upload date (YYYYMMDD).
@@ -721,9 +727,14 @@ class InfoExtractor(object):
                     [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(prop)
 
     def _og_search_property(self, prop, html, name=None, **kargs):
+        if not isinstance(prop, (list, tuple)):
+            prop = [prop]
         if name is None:
-            name = 'OpenGraph %s' % prop
-        escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs)
+            name = 'OpenGraph %s' % prop[0]
+        og_regexes = []
+        for p in prop:
+            og_regexes.extend(self._og_regexes(p))
+        escaped = self._search_regex(og_regexes, html, name, flags=re.DOTALL, **kargs)
         if escaped is None:
             return None
         return unescapeHTML(escaped)
@@ -747,10 +758,12 @@ class InfoExtractor(object):
         return self._og_search_property('url', html, **kargs)
 
     def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs):
+        if not isinstance(name, (list, tuple)):
+            name = [name]
         if display_name is None:
-            display_name = name
+            display_name = name[0]
         return self._html_search_regex(
-            self._meta_regex(name),
+            [self._meta_regex(n) for n in name],
             html, display_name, fatal=fatal, group='content', **kwargs)
 
     def _dc_search_uploader(self, html):
@@ -799,40 +812,63 @@ class InfoExtractor(object):
         return self._html_search_meta('twitter:player', html,
                                       'twitter card player')
 
-    def _search_json_ld(self, html, video_id, **kwargs):
+    def _search_json_ld(self, html, video_id, expected_type=None, **kwargs):
         json_ld = self._search_regex(
             r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>',
             html, 'JSON-LD', group='json_ld', **kwargs)
         if not json_ld:
             return {}
-        return self._json_ld(json_ld, video_id, fatal=kwargs.get('fatal', True))
+        return self._json_ld(
+            json_ld, video_id, fatal=kwargs.get('fatal', True),
+            expected_type=expected_type)
 
-    def _json_ld(self, json_ld, video_id, fatal=True):
+    def _json_ld(self, json_ld, video_id, fatal=True, expected_type=None):
         if isinstance(json_ld, compat_str):
             json_ld = self._parse_json(json_ld, video_id, fatal=fatal)
         if not json_ld:
             return {}
         info = {}
-        if json_ld.get('@context') == 'http://schema.org':
-            item_type = json_ld.get('@type')
-            if item_type == 'TVEpisode':
-                info.update({
-                    'episode': unescapeHTML(json_ld.get('name')),
-                    'episode_number': int_or_none(json_ld.get('episodeNumber')),
-                    'description': unescapeHTML(json_ld.get('description')),
-                })
-                part_of_season = json_ld.get('partOfSeason')
-                if isinstance(part_of_season, dict) and part_of_season.get('@type') == 'TVSeason':
-                    info['season_number'] = int_or_none(part_of_season.get('seasonNumber'))
-                part_of_series = json_ld.get('partOfSeries')
-                if isinstance(part_of_series, dict) and part_of_series.get('@type') == 'TVSeries':
-                    info['series'] = unescapeHTML(part_of_series.get('name'))
-            elif item_type == 'Article':
-                info.update({
-                    'timestamp': parse_iso8601(json_ld.get('datePublished')),
-                    'title': unescapeHTML(json_ld.get('headline')),
-                    'description': unescapeHTML(json_ld.get('articleBody')),
-                })
+        if not isinstance(json_ld, (list, tuple, dict)):
+            return info
+        if isinstance(json_ld, dict):
+            json_ld = [json_ld]
+        for e in json_ld:
+            if e.get('@context') == 'http://schema.org':
+                item_type = e.get('@type')
+                if expected_type is not None and expected_type != item_type:
+                    return info
+                if item_type == 'TVEpisode':
+                    info.update({
+                        'episode': unescapeHTML(e.get('name')),
+                        'episode_number': int_or_none(e.get('episodeNumber')),
+                        'description': unescapeHTML(e.get('description')),
+                    })
+                    part_of_season = e.get('partOfSeason')
+                    if isinstance(part_of_season, dict) and part_of_season.get('@type') == 'TVSeason':
+                        info['season_number'] = int_or_none(part_of_season.get('seasonNumber'))
+                    part_of_series = e.get('partOfSeries') or e.get('partOfTVSeries')
+                    if isinstance(part_of_series, dict) and part_of_series.get('@type') == 'TVSeries':
+                        info['series'] = unescapeHTML(part_of_series.get('name'))
+                elif item_type == 'Article':
+                    info.update({
+                        'timestamp': parse_iso8601(e.get('datePublished')),
+                        'title': unescapeHTML(e.get('headline')),
+                        'description': unescapeHTML(e.get('articleBody')),
+                    })
+                elif item_type == 'VideoObject':
+                    info.update({
+                        'url': e.get('contentUrl'),
+                        'title': unescapeHTML(e.get('name')),
+                        'description': unescapeHTML(e.get('description')),
+                        'thumbnail': e.get('thumbnailUrl'),
+                        'duration': parse_duration(e.get('duration')),
+                        'timestamp': unified_timestamp(e.get('uploadDate')),
+                        'filesize': float_or_none(e.get('contentSize')),
+                        'tbr': int_or_none(e.get('bitrate')),
+                        'width': int_or_none(e.get('width')),
+                        'height': int_or_none(e.get('height')),
+                    })
+                break
         return dict((k, v) for k, v in info.items() if v is not None)
 
     @staticmethod
@@ -874,7 +910,11 @@ class InfoExtractor(object):
                 f['ext'] = determine_ext(f['url'])
 
             if isinstance(field_preference, (list, tuple)):
-                return tuple(f.get(field) if f.get(field) is not None else -1 for field in field_preference)
+                return tuple(
+                    f.get(field)
+                    if f.get(field) is not None
+                    else ('' if field == 'format_id' else -1)
+                    for field in field_preference)
 
             preference = f.get('preference')
             if preference is None:
@@ -882,7 +922,8 @@ class InfoExtractor(object):
                 if f.get('ext') in ['f4f', 'f4m']:  # Not yet supported
                     preference -= 0.5
 
-            proto_preference = 0 if determine_protocol(f) in ['http', 'https'] else -0.1
+            protocol = f.get('protocol') or determine_protocol(f)
+            proto_preference = 0 if protocol in ['http', 'https'] else (-0.5 if protocol == 'rtsp' else -0.1)
 
             if f.get('vcodec') == 'none':  # audio only
                 preference -= 50
@@ -987,7 +1028,7 @@ class InfoExtractor(object):
 
     def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None,
                              transform_source=lambda s: fix_xml_ampersands(s).strip(),
-                             fatal=True):
+                             fatal=True, m3u8_id=None):
         manifest = self._download_xml(
             manifest_url, video_id, 'Downloading f4m manifest',
             'Unable to download f4m manifest',
@@ -1001,11 +1042,18 @@ class InfoExtractor(object):
 
         return self._parse_f4m_formats(
             manifest, manifest_url, video_id, preference=preference, f4m_id=f4m_id,
-            transform_source=transform_source, fatal=fatal)
+            transform_source=transform_source, fatal=fatal, m3u8_id=m3u8_id)
 
     def _parse_f4m_formats(self, manifest, manifest_url, video_id, preference=None, f4m_id=None,
                            transform_source=lambda s: fix_xml_ampersands(s).strip(),
-                           fatal=True):
+                           fatal=True, m3u8_id=None):
+        # currently youtube-dl cannot decode the playerVerificationChallenge as Akamai uses Adobe Alchemy
+        akamai_pv = manifest.find('{http://ns.adobe.com/f4m/1.0}pv-2.0')
+        if akamai_pv is not None and ';' in akamai_pv.text:
+            playerVerificationChallenge = akamai_pv.text.split(';')[0]
+            if playerVerificationChallenge.strip() != '':
+                return []
+
         formats = []
         manifest_version = '1.0'
         media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media')
@@ -1022,9 +1070,26 @@ class InfoExtractor(object):
             'base URL', default=None)
         if base_url:
             base_url = base_url.strip()
+
+        bootstrap_info = xpath_element(
+            manifest, ['{http://ns.adobe.com/f4m/1.0}bootstrapInfo', '{http://ns.adobe.com/f4m/2.0}bootstrapInfo'],
+            'bootstrap info', default=None)
+
         for i, media_el in enumerate(media_nodes):
-            if manifest_version == '2.0':
-                media_url = media_el.attrib.get('href') or media_el.attrib.get('url')
+            tbr = int_or_none(media_el.attrib.get('bitrate'))
+            width = int_or_none(media_el.attrib.get('width'))
+            height = int_or_none(media_el.attrib.get('height'))
+            format_id = '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)]))
+            # If <bootstrapInfo> is present, the specified f4m is a
+            # stream-level manifest, and only set-level manifests may refer to
+            # external resources.  See section 11.4 and section 4 of F4M spec
+            if bootstrap_info is None:
+                media_url = None
+                # @href is introduced in 2.0, see section 11.6 of F4M spec
+                if manifest_version == '2.0':
+                    media_url = media_el.attrib.get('href')
+                if media_url is None:
+                    media_url = media_el.attrib.get('url')
                 if not media_url:
                     continue
                 manifest_url = (
@@ -1034,37 +1099,58 @@ class InfoExtractor(object):
                 # since bitrates in parent manifest (this one) and media_url manifest
                 # may differ leading to inability to resolve the format by requested
                 # bitrate in f4m downloader
-                if determine_ext(manifest_url) == 'f4m':
-                    formats.extend(self._extract_f4m_formats(
+                ext = determine_ext(manifest_url)
+                if ext == 'f4m':
+                    f4m_formats = self._extract_f4m_formats(
                         manifest_url, video_id, preference=preference, f4m_id=f4m_id,
-                        transform_source=transform_source, fatal=fatal))
+                        transform_source=transform_source, fatal=fatal)
+                    # Sometimes stream-level manifest contains single media entry that
+                    # does not contain any quality metadata (e.g. http://matchtv.ru/#live-player).
+                    # At the same time parent's media entry in set-level manifest may
+                    # contain it. We will copy it from parent in such cases.
+                    if len(f4m_formats) == 1:
+                        f = f4m_formats[0]
+                        f.update({
+                            'tbr': f.get('tbr') or tbr,
+                            'width': f.get('width') or width,
+                            'height': f.get('height') or height,
+                            'format_id': f.get('format_id') if not tbr else format_id,
+                        })
+                    formats.extend(f4m_formats)
+                    continue
+                elif ext == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        manifest_url, video_id, 'mp4', preference=preference,
+                        m3u8_id=m3u8_id, fatal=fatal))
                     continue
-            tbr = int_or_none(media_el.attrib.get('bitrate'))
             formats.append({
-                'format_id': '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)])),
+                'format_id': format_id,
                 'url': manifest_url,
-                'ext': 'flv',
+                'ext': 'flv' if bootstrap_info is not None else None,
                 'tbr': tbr,
-                'width': int_or_none(media_el.attrib.get('width')),
-                'height': int_or_none(media_el.attrib.get('height')),
+                'width': width,
+                'height': height,
                 'preference': preference,
             })
         return formats
 
-    def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
-                              entry_protocol='m3u8', preference=None,
-                              m3u8_id=None, note=None, errnote=None,
-                              fatal=True):
-
-        formats = [{
+    def _m3u8_meta_format(self, m3u8_url, ext=None, preference=None, m3u8_id=None):
+        return {
             'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])),
             'url': m3u8_url,
             'ext': ext,
             'protocol': 'm3u8',
-            'preference': preference - 1 if preference else -1,
+            'preference': preference - 100 if preference else -100,
             'resolution': 'multiple',
             'format_note': 'Quality selection URL',
-        }]
+        }
+
+    def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
+                              entry_protocol='m3u8', preference=None,
+                              m3u8_id=None, note=None, errnote=None,
+                              fatal=True, live=False):
+
+        formats = [self._m3u8_meta_format(m3u8_url, ext, preference, m3u8_id)]
 
         format_url = lambda u: (
             u
@@ -1104,23 +1190,11 @@ class InfoExtractor(object):
             }]
         last_info = None
         last_media = None
-        kv_rex = re.compile(
-            r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)')
         for line in m3u8_doc.splitlines():
             if line.startswith('#EXT-X-STREAM-INF:'):
-                last_info = {}
-                for m in kv_rex.finditer(line):
-                    v = m.group('val')
-                    if v.startswith('"'):
-                        v = v[1:-1]
-                    last_info[m.group('key')] = v
+                last_info = parse_m3u8_attributes(line)
             elif line.startswith('#EXT-X-MEDIA:'):
-                last_media = {}
-                for m in kv_rex.finditer(line):
-                    v = m.group('val')
-                    if v.startswith('"'):
-                        v = v[1:-1]
-                    last_media[m.group('key')] = v
+                last_media = parse_m3u8_attributes(line)
             elif line.startswith('#') or not line.strip():
                 continue
             else:
@@ -1131,13 +1205,21 @@ class InfoExtractor(object):
                 format_id = []
                 if m3u8_id:
                     format_id.append(m3u8_id)
-                last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') != 'SUBTITLES' else None
-                format_id.append(last_media_name if last_media_name else '%d' % (tbr if tbr else len(formats)))
+                last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') not in ('SUBTITLES', 'CLOSED-CAPTIONS') else None
+                # Despite specification does not mention NAME attribute for
+                # EXT-X-STREAM-INF it still sometimes may be present
+                stream_name = last_info.get('NAME') or last_media_name
+                # Bandwidth of live streams may differ over time thus making
+                # format_id unpredictable. So it's better to keep provided
+                # format_id intact.
+                if not live:
+                    format_id.append(stream_name if stream_name else '%d' % (tbr if tbr else len(formats)))
                 f = {
                     'format_id': '-'.join(format_id),
                     'url': format_url(line.strip()),
                     'tbr': tbr,
                     'ext': ext,
+                    'fps': float_or_none(last_info.get('FRAME-RATE')),
                     'protocol': entry_protocol,
                     'preference': preference,
                 }
@@ -1146,24 +1228,17 @@ class InfoExtractor(object):
                     width_str, height_str = resolution.split('x')
                     f['width'] = int(width_str)
                     f['height'] = int(height_str)
-                codecs = last_info.get('CODECS')
-                if codecs:
-                    vcodec, acodec = [None] * 2
-                    va_codecs = codecs.split(',')
-                    if len(va_codecs) == 1:
-                        # Audio only entries usually come with single codec and
-                        # no resolution. For more robustness we also check it to
-                        # be mp4 audio.
-                        if not resolution and va_codecs[0].startswith('mp4a'):
-                            vcodec, acodec = 'none', va_codecs[0]
-                        else:
-                            vcodec = va_codecs[0]
-                    else:
-                        vcodec, acodec = va_codecs[:2]
+                # Unified Streaming Platform
+                mobj = re.search(
+                    r'audio.*?(?:%3D|=)(\d+)(?:-video.*?(?:%3D|=)(\d+))?', f['url'])
+                if mobj:
+                    abr, vbr = mobj.groups()
+                    abr, vbr = float_or_none(abr, 1000), float_or_none(vbr, 1000)
                     f.update({
-                        'acodec': acodec,
-                        'vcodec': vcodec,
+                        'vbr': vbr,
+                        'abr': abr,
                     })
+                f.update(parse_codecs(last_info.get('CODECS')))
                 if last_media is not None:
                     f['m3u8_media'] = last_media
                     last_media = None
@@ -1264,21 +1339,21 @@ class InfoExtractor(object):
         m3u8_count = 0
 
         srcs = []
-        videos = smil.findall(self._xpath_ns('.//video', namespace))
-        for video in videos:
-            src = video.get('src')
+        media = smil.findall(self._xpath_ns('.//video', namespace)) + smil.findall(self._xpath_ns('.//audio', namespace))
+        for medium in media:
+            src = medium.get('src')
             if not src or src in srcs:
                 continue
             srcs.append(src)
 
-            bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
-            filesize = int_or_none(video.get('size') or video.get('fileSize'))
-            width = int_or_none(video.get('width'))
-            height = int_or_none(video.get('height'))
-            proto = video.get('proto')
-            ext = video.get('ext')
+            bitrate = float_or_none(medium.get('system-bitrate') or medium.get('systemBitrate'), 1000)
+            filesize = int_or_none(medium.get('size') or medium.get('fileSize'))
+            width = int_or_none(medium.get('width'))
+            height = int_or_none(medium.get('height'))
+            proto = medium.get('proto')
+            ext = medium.get('ext')
             src_ext = determine_ext(src)
-            streamer = video.get('streamer') or base
+            streamer = medium.get('streamer') or base
 
             if proto == 'rtmp' or streamer.startswith('rtmp'):
                 rtmp_count += 1
@@ -1418,6 +1493,13 @@ class InfoExtractor(object):
             compat_etree_fromstring(mpd.encode('utf-8')), mpd_id, mpd_base_url, formats_dict=formats_dict)
 
     def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', formats_dict={}):
+        """
+        Parse formats from MPD manifest.
+        References:
+         1. MPEG-DASH Standard, ISO/IEC 23009-1:2014(E),
+            http://standards.iso.org/ittf/PubliclyAvailableStandards/c065274_ISO_IEC_23009-1_2014.zip
+         2. https://en.wikipedia.org/wiki/Dynamic_Adaptive_Streaming_over_HTTP
+        """
         if mpd_doc.get('type') == 'dynamic':
             return []
 
@@ -1450,8 +1532,16 @@ class InfoExtractor(object):
                         s_e = segment_timeline.findall(_add_ns('S'))
                         if s_e:
                             ms_info['total_number'] = 0
+                            ms_info['s'] = []
                             for s in s_e:
-                                ms_info['total_number'] += 1 + int(s.get('r', '0'))
+                                r = int(s.get('r', 0))
+                                ms_info['total_number'] += 1 + r
+                                ms_info['s'].append({
+                                    't': int(s.get('t', 0)),
+                                    # @d is mandatory (see [1, 5.3.9.6.2, Table 17, page 60])
+                                    'd': int(s.attrib['d']),
+                                    'r': r,
+                                })
                     else:
                         timescale = segment_template.get('timescale')
                         if timescale:
@@ -1488,7 +1578,7 @@ class InfoExtractor(object):
                         continue
                     representation_attrib = adaptation_set.attrib.copy()
                     representation_attrib.update(representation.attrib)
-                    # According to page 41 of ISO/IEC 29001-1:2014, @mimeType is mandatory
+                    # According to [1, 5.3.7.2, Table 9, page 41], @mimeType is mandatory
                     mime_type = representation_attrib['mimeType']
                     content_type = mime_type.split('/')[0]
                     if content_type == 'text':
@@ -1532,16 +1622,40 @@ class InfoExtractor(object):
                                 representation_ms_info['total_number'] = int(math.ceil(float(period_duration) / segment_duration))
                             media_template = representation_ms_info['media_template']
                             media_template = media_template.replace('$RepresentationID$', representation_id)
-                            media_template = re.sub(r'\$(Number|Bandwidth)\$', r'%(\1)d', media_template)
-                            media_template = re.sub(r'\$(Number|Bandwidth)%([^$]+)\$', r'%(\1)\2', media_template)
+                            media_template = re.sub(r'\$(Number|Bandwidth|Time)\$', r'%(\1)d', media_template)
+                            media_template = re.sub(r'\$(Number|Bandwidth|Time)%([^$]+)\$', r'%(\1)\2', media_template)
                             media_template.replace('$$', '$')
-                            representation_ms_info['segment_urls'] = [
-                                media_template % {
-                                    'Number': segment_number,
-                                    'Bandwidth': representation_attrib.get('bandwidth')}
-                                for segment_number in range(
-                                    representation_ms_info['start_number'],
-                                    representation_ms_info['total_number'] + representation_ms_info['start_number'])]
+
+                            # As per [1, 5.3.9.4.4, Table 16, page 55] $Number$ and $Time$
+                            # can't be used at the same time
+                            if '%(Number' in media_template:
+                                representation_ms_info['segment_urls'] = [
+                                    media_template % {
+                                        'Number': segment_number,
+                                        'Bandwidth': representation_attrib.get('bandwidth'),
+                                    }
+                                    for segment_number in range(
+                                        representation_ms_info['start_number'],
+                                        representation_ms_info['total_number'] + representation_ms_info['start_number'])]
+                            else:
+                                representation_ms_info['segment_urls'] = []
+                                segment_time = 0
+
+                                def add_segment_url():
+                                    representation_ms_info['segment_urls'].append(
+                                        media_template % {
+                                            'Time': segment_time,
+                                            'Bandwidth': representation_attrib.get('bandwidth'),
+                                        }
+                                    )
+
+                                for num, s in enumerate(representation_ms_info['s']):
+                                    segment_time = s.get('t') or segment_time
+                                    add_segment_url()
+                                    for r in range(s.get('r', 0)):
+                                        segment_time += s['d']
+                                        add_segment_url()
+                                    segment_time += s['d']
                         if 'segment_urls' in representation_ms_info:
                             f.update({
                                 'segment_urls': representation_ms_info['segment_urls'],
@@ -1568,6 +1682,62 @@ class InfoExtractor(object):
                         self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
         return formats
 
+    def _parse_html5_media_entries(self, base_url, webpage):
+        def absolute_url(video_url):
+            return compat_urlparse.urljoin(base_url, video_url)
+
+        def parse_content_type(content_type):
+            if not content_type:
+                return {}
+            ctr = re.search(r'(?P<mimetype>[^/]+/[^;]+)(?:;\s*codecs="?(?P<codecs>[^"]+))?', content_type)
+            if ctr:
+                mimetype, codecs = ctr.groups()
+                f = parse_codecs(codecs)
+                f['ext'] = mimetype2ext(mimetype)
+                return f
+            return {}
+
+        entries = []
+        for media_tag, media_type, media_content in re.findall(r'(?s)(<(?P<tag>video|audio)[^>]*>)(.*?)</(?P=tag)>', webpage):
+            media_info = {
+                'formats': [],
+                'subtitles': {},
+            }
+            media_attributes = extract_attributes(media_tag)
+            src = media_attributes.get('src')
+            if src:
+                media_info['formats'].append({
+                    'url': absolute_url(src),
+                    'vcodec': 'none' if media_type == 'audio' else None,
+                })
+            media_info['thumbnail'] = media_attributes.get('poster')
+            if media_content:
+                for source_tag in re.findall(r'<source[^>]+>', media_content):
+                    source_attributes = extract_attributes(source_tag)
+                    src = source_attributes.get('src')
+                    if not src:
+                        continue
+                    f = parse_content_type(source_attributes.get('type'))
+                    f.update({
+                        'url': absolute_url(src),
+                        'vcodec': 'none' if media_type == 'audio' else None,
+                    })
+                    media_info['formats'].append(f)
+                for track_tag in re.findall(r'<track[^>]+>', media_content):
+                    track_attributes = extract_attributes(track_tag)
+                    kind = track_attributes.get('kind')
+                    if not kind or kind == 'subtitles':
+                        src = track_attributes.get('src')
+                        if not src:
+                            continue
+                        lang = track_attributes.get('srclang') or track_attributes.get('lang') or track_attributes.get('label')
+                        media_info['subtitles'].setdefault(lang, []).append({
+                            'url': absolute_url(src),
+                        })
+            if media_info['formats']:
+                entries.append(media_info)
+        return entries
+
     def _live_title(self, name):
         """ Generate the title for a live video """
         now = datetime.datetime.now()
@@ -1628,7 +1798,7 @@ class InfoExtractor(object):
 
         any_restricted = False
         for tc in self.get_testcases(include_onlymatching=False):
-            if 'playlist' in tc:
+            if tc.get('playlist', []):
                 tc = tc['playlist'][0]
             is_restricted = age_restricted(
                 tc.get('info_dict', {}).get('age_limit'), age_limit)
@@ -1681,6 +1851,13 @@ class InfoExtractor(object):
     def _mark_watched(self, *args, **kwargs):
         raise NotImplementedError('This method must be implemented by subclasses')
 
+    def geo_verification_headers(self):
+        headers = {}
+        geo_verification_proxy = self._downloader.params.get('geo_verification_proxy')
+        if geo_verification_proxy:
+            headers['Ytdl-request-proxy'] = geo_verification_proxy
+        return headers
+
 
 class SearchInfoExtractor(InfoExtractor):
     """

youtube_dl/extractor/condenast.py

@@ -5,13 +5,17 @@ import re
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_urllib_parse_urlencode,
     compat_urllib_parse_urlparse,
     compat_urlparse,
 )
 from ..utils import (
     orderedSet,
     remove_end,
+    extract_attributes,
+    mimetype2ext,
+    determine_ext,
+    int_or_none,
+    parse_iso8601,
 )
 
 
@@ -58,6 +62,9 @@ class CondeNastIE(InfoExtractor):
             'ext': 'mp4',
             'title': '3D Printed Speakers Lit With LED',
             'description': 'Check out these beautiful 3D printed LED speakers.  You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
+            'uploader': 'wired',
+            'upload_date': '20130314',
+            'timestamp': 1363219200,
         }
     }, {
         # JS embed
@@ -67,70 +74,92 @@ class CondeNastIE(InfoExtractor):
             'id': '55f9cf8b61646d1acf00000c',
             'ext': 'mp4',
             'title': '3D printed TSA Travel Sentry keys really do open TSA locks',
+            'uploader': 'arstechnica',
+            'upload_date': '20150916',
+            'timestamp': 1442434955,
         }
     }]
 
     def _extract_series(self, url, webpage):
-        title = self._html_search_regex(r'<div class="cne-series-info">.*?<h1>(.+?)</h1>',
-                                        webpage, 'series title', flags=re.DOTALL)
+        title = self._html_search_regex(
+            r'(?s)<div class="cne-series-info">.*?<h1>(.+?)</h1>',
+            webpage, 'series title')
         url_object = compat_urllib_parse_urlparse(url)
         base_url = '%s://%s' % (url_object.scheme, url_object.netloc)
-        m_paths = re.finditer(r'<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]',
-                              webpage, flags=re.DOTALL)
+        m_paths = re.finditer(
+            r'(?s)<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]', webpage)
         paths = orderedSet(m.group(1) for m in m_paths)
         build_url = lambda path: compat_urlparse.urljoin(base_url, path)
         entries = [self.url_result(build_url(path), 'CondeNast') for path in paths]
         return self.playlist_result(entries, playlist_title=title)
 
     def _extract_video(self, webpage, url_type):
-        if url_type != 'embed':
-            description = self._html_search_regex(
-                [
-                    r'<div class="cne-video-description">(.+?)</div>',
-                    r'<div class="video-post-content">(.+?)</div>',
-                ],
-                webpage, 'description', fatal=False, flags=re.DOTALL)
+        query = {}
+        params = self._search_regex(
+            r'(?s)var params = {(.+?)}[;,]', webpage, 'player params', default=None)
+        if params:
+            query.update({
+                'videoId': self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id'),
+                'playerId': self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id'),
+                'target': self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target'),
+            })
         else:
-            description = None
-        params = self._search_regex(r'var params = {(.+?)}[;,]', webpage,
-                                    'player params', flags=re.DOTALL)
-        video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id')
-        player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id')
-        target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target')
-        data = compat_urllib_parse_urlencode({'videoId': video_id,
-                                              'playerId': player_id,
-                                              'target': target,
-                                              })
-        base_info_url = self._search_regex(r'url = [\'"](.+?)[\'"][,;]',
-                                           webpage, 'base info url',
-                                           default='http://player.cnevids.com/player/loader.js?')
-        info_url = base_info_url + data
-        info_page = self._download_webpage(info_url, video_id,
-                                           'Downloading video info')
-        video_info = self._search_regex(r'var\s+video\s*=\s*({.+?});', info_page, 'video info')
-        video_info = self._parse_json(video_info, video_id)
+            params = extract_attributes(self._search_regex(
+                r'(<[^>]+data-js="video-player"[^>]+>)',
+                webpage, 'player params element'))
+            query.update({
+                'videoId': params['data-video'],
+                'playerId': params['data-player'],
+                'target': params['id'],
+            })
+        video_id = query['videoId']
+        video_info = None
+        info_page = self._download_webpage(
+            'http://player.cnevids.com/player/video.js',
+            video_id, 'Downloading video info', query=query, fatal=False)
+        if info_page:
+            video_info = self._parse_json(self._search_regex(
+                r'loadCallback\(({.+})\)', info_page, 'video info'), video_id)['video']
+        else:
+            info_page = self._download_webpage(
+                'http://player.cnevids.com/player/loader.js',
+                video_id, 'Downloading loader info', query=query)
+            video_info = self._parse_json(self._search_regex(
+                r'var\s+video\s*=\s*({.+?});', info_page, 'video info'), video_id)
+        title = video_info['title']
 
-        formats = [{
-            'format_id': '%s-%s' % (fdata['type'].split('/')[-1], fdata['quality']),
-            'url': fdata['src'],
-            'ext': fdata['type'].split('/')[-1],
-            'quality': 1 if fdata['quality'] == 'high' else 0,
-        } for fdata in video_info['sources'][0]]
+        formats = []
+        for fdata in video_info.get('sources', [{}])[0]:
+            src = fdata.get('src')
+            if not src:
+                continue
+            ext = mimetype2ext(fdata.get('type')) or determine_ext(src)
+            quality = fdata.get('quality')
+            formats.append({
+                'format_id': ext + ('-%s' % quality if quality else ''),
+                'url': src,
+                'ext': ext,
+                'quality': 1 if quality == 'high' else 0,
+            })
         self._sort_formats(formats)
 
-        return {
+        info = self._search_json_ld(webpage, video_id) if url_type != 'embed' else {}
+        info.update({
             'id': video_id,
             'formats': formats,
-            'title': video_info['title'],
-            'thumbnail': video_info['poster_frame'],
-            'description': description,
-        }
+            'title': title,
+            'thumbnail': video_info.get('poster_frame'),
+            'uploader': video_info.get('brand'),
+            'duration': int_or_none(video_info.get('duration')),
+            'tags': video_info.get('tags'),
+            'series': video_info.get('series_title'),
+            'season': video_info.get('season_title'),
+            'timestamp': parse_iso8601(video_info.get('premiere_date')),
+        })
+        return info
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        site = mobj.group('site')
-        url_type = mobj.group('type')
-        item_id = mobj.group('id')
+        site, url_type, item_id = re.match(self._VALID_URL, url).groups()
 
         # Convert JS embed to regular embed
         if url_type == 'embedjs':

youtube_dl/extractor/coub.py

@@ -0,0 +1,143 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    float_or_none,
+    int_or_none,
+    parse_iso8601,
+    qualities,
+)
+
+
+class CoubIE(InfoExtractor):
+    _VALID_URL = r'(?:coub:|https?://(?:coub\.com/(?:view|embed|coubs)/|c-cdn\.coub\.com/fb-player\.swf\?.*\bcoub(?:ID|id)=))(?P<id>[\da-z]+)'
+
+    _TESTS = [{
+        'url': 'http://coub.com/view/5u5n1',
+        'info_dict': {
+            'id': '5u5n1',
+            'ext': 'mp4',
+            'title': 'The Matrix Moonwalk',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 4.6,
+            'timestamp': 1428527772,
+            'upload_date': '20150408',
+            'uploader': 'Артём Лоскутников',
+            'uploader_id': 'artyom.loskutnikov',
+            'view_count': int,
+            'like_count': int,
+            'repost_count': int,
+            'comment_count': int,
+            'age_limit': 0,
+        },
+    }, {
+        'url': 'http://c-cdn.coub.com/fb-player.swf?bot_type=vk&coubID=7w5a4',
+        'only_matching': True,
+    }, {
+        'url': 'coub:5u5n1',
+        'only_matching': True,
+    }, {
+        # longer video id
+        'url': 'http://coub.com/view/237d5l5h',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        coub = self._download_json(
+            'http://coub.com/api/v2/coubs/%s.json' % video_id, video_id)
+
+        if coub.get('error'):
+            raise ExtractorError(
+                '%s said: %s' % (self.IE_NAME, coub['error']), expected=True)
+
+        title = coub['title']
+
+        file_versions = coub['file_versions']
+
+        QUALITIES = ('low', 'med', 'high')
+
+        MOBILE = 'mobile'
+        IPHONE = 'iphone'
+        HTML5 = 'html5'
+
+        SOURCE_PREFERENCE = (MOBILE, IPHONE, HTML5)
+
+        quality_key = qualities(QUALITIES)
+        preference_key = qualities(SOURCE_PREFERENCE)
+
+        formats = []
+
+        for kind, items in file_versions.get(HTML5, {}).items():
+            if kind not in ('video', 'audio'):
+                continue
+            if not isinstance(items, dict):
+                continue
+            for quality, item in items.items():
+                if not isinstance(item, dict):
+                    continue
+                item_url = item.get('url')
+                if not item_url:
+                    continue
+                formats.append({
+                    'url': item_url,
+                    'format_id': '%s-%s-%s' % (HTML5, kind, quality),
+                    'filesize': int_or_none(item.get('size')),
+                    'vcodec': 'none' if kind == 'audio' else None,
+                    'quality': quality_key(quality),
+                    'preference': preference_key(HTML5),
+                })
+
+        iphone_url = file_versions.get(IPHONE, {}).get('url')
+        if iphone_url:
+            formats.append({
+                'url': iphone_url,
+                'format_id': IPHONE,
+                'preference': preference_key(IPHONE),
+            })
+
+        mobile_url = file_versions.get(MOBILE, {}).get('audio_url')
+        if mobile_url:
+            formats.append({
+                'url': mobile_url,
+                'format_id': '%s-audio' % MOBILE,
+                'preference': preference_key(MOBILE),
+            })
+
+        self._sort_formats(formats)
+
+        thumbnail = coub.get('picture')
+        duration = float_or_none(coub.get('duration'))
+        timestamp = parse_iso8601(coub.get('published_at') or coub.get('created_at'))
+        uploader = coub.get('channel', {}).get('title')
+        uploader_id = coub.get('channel', {}).get('permalink')
+
+        view_count = int_or_none(coub.get('views_count') or coub.get('views_increase_count'))
+        like_count = int_or_none(coub.get('likes_count'))
+        repost_count = int_or_none(coub.get('recoubs_count'))
+        comment_count = int_or_none(coub.get('comments_count'))
+
+        age_restricted = coub.get('age_restricted', coub.get('age_restricted_by_admin'))
+        if age_restricted is not None:
+            age_limit = 18 if age_restricted is True else 0
+        else:
+            age_limit = None
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'timestamp': timestamp,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'view_count': view_count,
+            'like_count': like_count,
+            'repost_count': repost_count,
+            'comment_count': comment_count,
+            'age_limit': age_limit,
+            'formats': formats,
+        }

youtube_dl/extractor/crunchyroll.py

@@ -11,7 +11,6 @@ from math import pow, sqrt, floor
 from .common import InfoExtractor
 from ..compat import (
     compat_etree_fromstring,
-    compat_urllib_parse_unquote,
     compat_urllib_parse_urlencode,
     compat_urllib_request,
     compat_urlparse,
@@ -27,6 +26,7 @@ from ..utils import (
     unified_strdate,
     urlencode_postdata,
     xpath_text,
+    extract_attributes,
 )
 from ..aes import (
     aes_cbc_decrypt,
@@ -306,28 +306,36 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', webpage,
             'video_uploader', fatal=False)
 
-        playerdata_url = compat_urllib_parse_unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
-        playerdata_req = sanitized_Request(playerdata_url)
-        playerdata_req.data = urlencode_postdata({'current_page': webpage_url})
-        playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
-        playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info')
-
-        stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, 'stream_id')
-        video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
-
+        available_fmts = []
+        for a, fmt in re.findall(r'(<a[^>]+token=["\']showmedia\.([0-9]{3,4})p["\'][^>]+>)', webpage):
+            attrs = extract_attributes(a)
+            href = attrs.get('href')
+            if href and '/freetrial' in href:
+                continue
+            available_fmts.append(fmt)
+        if not available_fmts:
+            for p in (r'token=["\']showmedia\.([0-9]{3,4})p"', r'showmedia\.([0-9]{3,4})p'):
+                available_fmts = re.findall(p, webpage)
+                if available_fmts:
+                    break
+        video_encode_ids = []
         formats = []
-        for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage):
+        for fmt in available_fmts:
             stream_quality, stream_format = self._FORMAT_IDS[fmt]
             video_format = fmt + 'p'
             streamdata_req = sanitized_Request(
                 'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s'
-                % (stream_id, stream_format, stream_quality),
+                % (video_id, stream_format, stream_quality),
                 compat_urllib_parse_urlencode({'current_page': url}).encode('utf-8'))
             streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
             streamdata = self._download_xml(
                 streamdata_req, video_id,
                 note='Downloading media info for %s' % video_format)
             stream_info = streamdata.find('./{default}preload/stream_info')
+            video_encode_id = xpath_text(stream_info, './video_encode_id')
+            if video_encode_id in video_encode_ids:
+                continue
+            video_encode_ids.append(video_encode_id)
             video_url = xpath_text(stream_info, './host')
             video_play_path = xpath_text(stream_info, './file')
             if not video_url or not video_play_path:
@@ -359,6 +367,14 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
                 'ext': 'flv',
             })
             formats.append(format_info)
+        self._sort_formats(formats)
+
+        metadata = self._download_xml(
+            'http://www.crunchyroll.com/xml', video_id,
+            note='Downloading media info', query={
+                'req': 'RpcApiVideoPlayer_GetMediaMetadata',
+                'media_id': video_id,
+            })
 
         subtitles = self.extract_subtitles(video_id, webpage)
 
@@ -366,9 +382,12 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             'id': video_id,
             'title': video_title,
             'description': video_description,
-            'thumbnail': video_thumbnail,
+            'thumbnail': xpath_text(metadata, 'episode_image_url'),
             'uploader': video_uploader,
             'upload_date': video_upload_date,
+            'series': xpath_text(metadata, 'series_title'),
+            'episode': xpath_text(metadata, 'episode_title'),
+            'episode_number': int_or_none(xpath_text(metadata, 'episode_number')),
             'subtitles': subtitles,
             'formats': formats,
         }

youtube_dl/extractor/cspan.py

@@ -51,8 +51,11 @@ class CSpanIE(InfoExtractor):
         'url': 'http://www.c-span.org/video/?104517-1/immigration-reforms-needed-protect-skilled-american-workers',
         'info_dict': {
             'id': 'judiciary031715',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Immigration Reforms Needed to Protect Skilled American Workers',
+        },
+        'params': {
+            'skip_download': True,  # m3u8 downloads
         }
     }]
 

youtube_dl/extractor/ctv.py

@@ -0,0 +1,30 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+
+class CTVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?ctv\.ca/video/player\?vid=(?P<id>[0-9.]+)'
+    _TESTS = [{
+        'url': 'http://www.ctv.ca/video/player?vid=706966',
+        'md5': 'ff2ebbeae0aa2dcc32a830c3fd69b7b0',
+        'info_dict': {
+            'id': '706966',
+            'ext': 'mp4',
+            'title': 'Larry Day and Richard Jutras on the TIFF red carpet of \'Stonewall\'',
+            'description': 'etalk catches up with Larry Day and Richard Jutras on the TIFF red carpet of "Stonewall”.',
+            'upload_date': '20150919',
+            'timestamp': 1442624700,
+        },
+        'expected_warnings': ['HTTP Error 404'],
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        return {
+            '_type': 'url_transparent',
+            'id': video_id,
+            'url': '9c9media:ctv_web:%s' % video_id,
+            'ie_key': 'NineCNineMedia',
+        }

youtube_dl/extractor/ctvnews.py

@@ -0,0 +1,65 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import orderedSet
+
+
+class CTVNewsIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?ctvnews\.ca/(?:video\?(?:clip|playlist|bin)Id=|.*?)(?P<id>[0-9.]+)'
+    _TESTS = [{
+        'url': 'http://www.ctvnews.ca/video?clipId=901995',
+        'md5': '10deb320dc0ccb8d01d34d12fc2ea672',
+        'info_dict': {
+            'id': '901995',
+            'ext': 'mp4',
+            'title': 'Extended: \'That person cannot be me\' Johnson says',
+            'description': 'md5:958dd3b4f5bbbf0ed4d045c790d89285',
+            'timestamp': 1467286284,
+            'upload_date': '20160630',
+        }
+    }, {
+        'url': 'http://www.ctvnews.ca/video?playlistId=1.2966224',
+        'info_dict':
+        {
+            'id': '1.2966224',
+        },
+        'playlist_mincount': 19,
+    }, {
+        'url': 'http://www.ctvnews.ca/video?binId=1.2876780',
+        'info_dict':
+        {
+            'id': '1.2876780',
+        },
+        'playlist_mincount': 100,
+    }, {
+        'url': 'http://www.ctvnews.ca/1.810401',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.ctvnews.ca/canadiens-send-p-k-subban-to-nashville-in-blockbuster-trade-1.2967231',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        page_id = self._match_id(url)
+
+        def ninecninemedia_url_result(clip_id):
+            return {
+                '_type': 'url_transparent',
+                'id': clip_id,
+                'url': '9c9media:ctvnews_web:%s' % clip_id,
+                'ie_key': 'NineCNineMedia',
+            }
+
+        if page_id.isdigit():
+            return ninecninemedia_url_result(page_id)
+        else:
+            webpage = self._download_webpage('http://www.ctvnews.ca/%s' % page_id, page_id, query={
+                'ot': 'example.AjaxPageLayout.ot',
+                'maxItemsPerPage': 1000000,
+            })
+            entries = [ninecninemedia_url_result(clip_id) for clip_id in orderedSet(
+                re.findall(r'clip\.id\s*=\s*(\d+);', webpage))]
+            return self.playlist_result(entries, page_id)

youtube_dl/extractor/cwtv.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class CWTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?cw(?:tv|seed)\.com/shows/(?:[^/]+/){2}\?play=(?P<id>[a-z0-9]{8}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{12})'
+    _VALID_URL = r'https?://(?:www\.)?cw(?:tv(?:pr)?|seed)\.com/(?:shows/)?(?:[^/]+/)+[^?]*\?.*\b(?:play|watch)=(?P<id>[a-z0-9]{8}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{12})'
     _TESTS = [{
         'url': 'http://cwtv.com/shows/arrow/legends-of-yesterday/?play=6b15e985-9345-4f60-baf8-56e96be57c63',
         'info_dict': {
@@ -48,6 +48,15 @@ class CWTVIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         }
+    }, {
+        'url': 'http://cwtv.com/thecw/chroniclesofcisco/?play=8adebe35-f447-465f-ab52-e863506ff6d6',
+        'only_matching': True,
+    }, {
+        'url': 'http://cwtvpr.com/the-cw/video?watch=9eee3f60-ef4e-440b-b3b2-49428ac9c54e',
+        'only_matching': True,
+    }, {
+        'url': 'http://cwtv.com/shows/arrow/legends-of-yesterday/?watch=6b15e985-9345-4f60-baf8-56e96be57c63',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/dailymail.py

@@ -0,0 +1,62 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    determine_protocol,
+    unescapeHTML,
+)
+
+
+class DailyMailIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/video/[^/]+/video-(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html',
+        'md5': 'f6129624562251f628296c3a9ffde124',
+        'info_dict': {
+            'id': '1295863',
+            'ext': 'mp4',
+            'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'',
+            'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        video_data = self._parse_json(self._search_regex(
+            r"data-opts='({.+?})'", webpage, 'video data'), video_id)
+        title = unescapeHTML(video_data['title'])
+        video_sources = self._download_json(video_data.get(
+            'sources', {}).get('url') or 'http://www.dailymail.co.uk/api/player/%s/video-sources.json' % video_id, video_id)
+
+        formats = []
+        for rendition in video_sources['renditions']:
+            rendition_url = rendition.get('url')
+            if not rendition_url:
+                continue
+            tbr = int_or_none(rendition.get('encodingRate'), 1000)
+            container = rendition.get('videoContainer')
+            is_hls = container == 'M2TS'
+            protocol = 'm3u8_native' if is_hls else determine_protocol({'url': rendition_url})
+            formats.append({
+                'format_id': ('hls' if is_hls else protocol) + ('-%d' % tbr if tbr else ''),
+                'url': rendition_url,
+                'width': int_or_none(rendition.get('frameWidth')),
+                'height': int_or_none(rendition.get('frameHeight')),
+                'tbr': tbr,
+                'vcodec': rendition.get('videoCodec'),
+                'container': container,
+                'protocol': protocol,
+                'ext': 'mp4' if is_hls else None,
+            })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': unescapeHTML(video_data.get('descr')),
+            'thumbnail': video_data.get('poster') or video_data.get('thumbnail'),
+            'formats': formats,
+        }

youtube_dl/extractor/dailymotion.py

@@ -16,6 +16,7 @@ from ..utils import (
     sanitized_Request,
     str_to_int,
     unescapeHTML,
+    mimetype2ext,
 )
 
 
@@ -111,6 +112,13 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
         }
     ]
 
+    @staticmethod
+    def _extract_urls(webpage):
+        # Look for embedded Dailymotion player
+        matches = re.findall(
+            r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
+        return list(map(lambda m: unescapeHTML(m[1]), matches))
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
@@ -153,18 +161,19 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                     type_ = media.get('type')
                     if type_ == 'application/vnd.lumberjack.manifest':
                         continue
-                    ext = determine_ext(media_url)
-                    if type_ == 'application/x-mpegURL' or ext == 'm3u8':
+                    ext = mimetype2ext(type_) or determine_ext(media_url)
+                    if ext == 'm3u8':
                         formats.extend(self._extract_m3u8_formats(
                             media_url, video_id, 'mp4', preference=-1,
                             m3u8_id='hls', fatal=False))
-                    elif type_ == 'application/f4m' or ext == 'f4m':
+                    elif ext == 'f4m':
                         formats.extend(self._extract_f4m_formats(
                             media_url, video_id, preference=-1, f4m_id='hds', fatal=False))
                     else:
                         f = {
                             'url': media_url,
                             'format_id': 'http-%s' % quality,
+                            'ext': ext,
                         }
                         m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
                         if m:
@@ -322,7 +331,9 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 
             for video_id in re.findall(r'data-xid="(.+?)"', webpage):
                 if video_id not in video_ids:
-                    yield self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
+                    yield self.url_result(
+                        'http://www.dailymotion.com/video/%s' % video_id,
+                        DailymotionIE.ie_key(), video_id)
                     video_ids.add(video_id)
 
             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:

youtube_dl/extractor/daum.py

@@ -66,22 +66,32 @@ class DaumIE(InfoExtractor):
             'view_count': int,
             'comment_count': int,
         },
+    }, {
+        # Requires dte_type=WEB (#9972)
+        'url': 'http://tvpot.daum.net/v/s3794Uf1NZeZ1qMpGpeqeRU',
+        'md5': 'a8917742069a4dd442516b86e7d66529',
+        'info_dict': {
+            'id': 's3794Uf1NZeZ1qMpGpeqeRU',
+            'ext': 'mp4',
+            'title': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny) [쇼! 음악중심] 508회 20160611',
+            'description': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)\n\n[쇼! 음악중심] 20160611, 507회',
+            'upload_date': '20160611',
+        },
     }]
 
     def _real_extract(self, url):
         video_id = compat_urllib_parse_unquote(self._match_id(url))
-        query = compat_urllib_parse_urlencode({'vid': video_id})
         movie_data = self._download_json(
-            'http://videofarm.daum.net/controller/api/closed/v1_2/IntegratedMovieData.json?' + query,
-            video_id, 'Downloading video formats info')
+            'http://videofarm.daum.net/controller/api/closed/v1_2/IntegratedMovieData.json',
+            video_id, 'Downloading video formats info', query={'vid': video_id, 'dte_type': 'WEB'})
 
         # For urls like http://m.tvpot.daum.net/v/65139429, where the video_id is really a clipid
         if not movie_data.get('output_list', {}).get('output_list') and re.match(r'^\d+$', video_id):
             return self.url_result('http://tvpot.daum.net/clip/ClipView.do?clipid=%s' % video_id)
 
         info = self._download_xml(
-            'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id,
-            'Downloading video info')
+            'http://tvpot.daum.net/clip/ClipInfoXml.do', video_id,
+            'Downloading video info', query={'vid': video_id})
 
         formats = []
         for format_el in movie_data['output_list']['output_list']:

youtube_dl/extractor/dbtv.py

@@ -4,78 +4,47 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_str
-from ..utils import (
-    float_or_none,
-    int_or_none,
-    clean_html,
-)
 
 
 class DBTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dbtv\.no/(?:(?:lazyplayer|player)/)?(?P<id>[0-9]+)(?:#(?P<display_id>.+))?'
+    _VALID_URL = r'https?://(?:www\.)?dbtv\.no/(?:[^/]+/)?(?P<id>[0-9]+)(?:#(?P<display_id>.+))?'
     _TESTS = [{
         'url': 'http://dbtv.no/3649835190001#Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen',
-        'md5': 'b89953ed25dacb6edb3ef6c6f430f8bc',
+        'md5': '2e24f67936517b143a234b4cadf792ec',
         'info_dict': {
-            'id': '33100',
+            'id': '3649835190001',
             'display_id': 'Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen',
             'ext': 'mp4',
             'title': 'Skulle teste ut fornøyelsespark, men kollegaen var bare opptatt av bikinikroppen',
             'description': 'md5:1504a54606c4dde3e4e61fc97aa857e0',
-            'thumbnail': 're:https?://.*\.jpg$',
-            'timestamp': 1404039863.438,
+            'thumbnail': 're:https?://.*\.jpg',
+            'timestamp': 1404039863,
             'upload_date': '20140629',
             'duration': 69.544,
-            'view_count': int,
-            'categories': list,
-        }
+            'uploader_id': '1027729757001',
+        },
+        'add_ie': ['BrightcoveNew']
     }, {
         'url': 'http://dbtv.no/3649835190001',
         'only_matching': True,
     }, {
         'url': 'http://www.dbtv.no/lazyplayer/4631135248001',
         'only_matching': True,
+    }, {
+        'url': 'http://dbtv.no/vice/5000634109001',
+        'only_matching': True,
+    }, {
+        'url': 'http://dbtv.no/filmtrailer/3359293614001',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        display_id = mobj.group('display_id') or video_id
-
-        data = self._download_json(
-            'http://api.dbtv.no/discovery/%s' % video_id, display_id)
-
-        video = data['playlist'][0]
-
-        formats = [{
-            'url': f['URL'],
-            'vcodec': f.get('container'),
-            'width': int_or_none(f.get('width')),
-            'height': int_or_none(f.get('height')),
-            'vbr': float_or_none(f.get('rate'), 1000),
-            'filesize': int_or_none(f.get('size')),
-        } for f in video['renditions'] if 'URL' in f]
-
-        if not formats:
-            for url_key, format_id in [('URL', 'mp4'), ('HLSURL', 'hls')]:
-                if url_key in video:
-                    formats.append({
-                        'url': video[url_key],
-                        'format_id': format_id,
-                    })
-
-        self._sort_formats(formats)
+        video_id, display_id = re.match(self._VALID_URL, url).groups()
 
         return {
-            'id': compat_str(video['id']),
+            '_type': 'url_transparent',
+            'url': 'http://players.brightcove.net/1027729757001/default_default/index.html?videoId=%s' % video_id,
+            'id': video_id,
             'display_id': display_id,
-            'title': video['title'],
-            'description': clean_html(video['desc']),
-            'thumbnail': video.get('splash') or video.get('thumb'),
-            'timestamp': float_or_none(video.get('publishedAt'), 1000),
-            'duration': float_or_none(video.get('length'), 1000),
-            'view_count': int_or_none(video.get('views')),
-            'categories': video.get('tags'),
-            'formats': formats,
+            'ie_key': 'BrightcoveNew',
         }

youtube_dl/extractor/dcn.py

@@ -20,7 +20,7 @@ from ..utils import (
 
 
 class DCNIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?show/(?P<show_id>\d+)/[^/]+(?:/(?P<video_id>\d+)/(?P<season_id>\d+))?'
+    _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?show/(?P<show_id>\d+)/[^/]+(?:/(?P<video_id>\d+)/(?P<season_id>\d+))?'
 
     def _real_extract(self, url):
         show_id, video_id, season_id = re.match(self._VALID_URL, url).groups()
@@ -55,30 +55,30 @@ class DCNBaseIE(InfoExtractor):
             'is_live': is_live,
         }
 
-    def _extract_video_formats(self, webpage, video_id, entry_protocol):
+    def _extract_video_formats(self, webpage, video_id, m3u8_entry_protocol):
         formats = []
-        m3u8_url = self._html_search_regex(
-            r'file\s*:\s*"([^"]+)', webpage, 'm3u8 url', fatal=False)
-        if m3u8_url:
-            formats.extend(self._extract_m3u8_formats(
-                m3u8_url, video_id, 'mp4', entry_protocol, m3u8_id='hls', fatal=None))
-
-        rtsp_url = self._search_regex(
-            r'<a[^>]+href="(rtsp://[^"]+)"', webpage, 'rtsp url', fatal=False)
-        if rtsp_url:
-            formats.append({
-                'url': rtsp_url,
-                'format_id': 'rtsp',
-            })
-
+        format_url_base = 'http' + self._html_search_regex(
+            [
+                r'file\s*:\s*"https?(://[^"]+)/playlist.m3u8',
+                r'<a[^>]+href="rtsp(://[^"]+)"'
+            ], webpage, 'format url')
+        formats.extend(self._extract_mpd_formats(
+            format_url_base + '/manifest.mpd',
+            video_id, mpd_id='dash', fatal=False))
+        formats.extend(self._extract_m3u8_formats(
+            format_url_base + '/playlist.m3u8', video_id, 'mp4',
+            m3u8_entry_protocol, m3u8_id='hls', fatal=False))
+        formats.extend(self._extract_f4m_formats(
+            format_url_base + '/manifest.f4m',
+            video_id, f4m_id='hds', fatal=False))
         self._sort_formats(formats)
         return formats
 
 
 class DCNVideoIE(DCNBaseIE):
     IE_NAME = 'dcn:video'
-    _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?(?:video/[^/]+|media|catchup/[^/]+/[^/]+)/(?P<id>\d+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?(?:video(?:/[^/]+)?|media|catchup/[^/]+/[^/]+)/(?P<id>\d+)'
+    _TESTS = [{
         'url': 'http://www.dcndigital.ae/#/video/%D8%B1%D8%AD%D9%84%D8%A9-%D8%A7%D9%84%D8%B9%D9%85%D8%B1-%D8%A7%D9%84%D8%AD%D9%84%D9%82%D8%A9-1/17375',
         'info_dict':
         {
@@ -94,7 +94,10 @@ class DCNVideoIE(DCNBaseIE):
             # m3u8 download
             'skip_download': True,
         },
-    }
+    }, {
+        'url': 'http://awaan.ae/video/26723981/%D8%AF%D8%A7%D8%B1-%D8%A7%D9%84%D8%B3%D9%84%D8%A7%D9%85:-%D8%AE%D9%8A%D8%B1-%D8%AF%D9%88%D8%B1-%D8%A7%D9%84%D8%A3%D9%86%D8%B5%D8%A7%D8%B1',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -120,7 +123,7 @@ class DCNVideoIE(DCNBaseIE):
 
 class DCNLiveIE(DCNBaseIE):
     IE_NAME = 'dcn:live'
-    _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?live/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?live/(?P<id>\d+)'
 
     def _real_extract(self, url):
         channel_id = self._match_id(url)
@@ -147,7 +150,7 @@ class DCNLiveIE(DCNBaseIE):
 
 class DCNSeasonIE(InfoExtractor):
     IE_NAME = 'dcn:season'
-    _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?program/(?:(?P<show_id>\d+)|season/(?P<season_id>\d+))'
+    _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?program/(?:(?P<show_id>\d+)|season/(?P<season_id>\d+))'
     _TEST = {
         'url': 'http://dcndigital.ae/#/program/205024/%D9%85%D8%AD%D8%A7%D8%B6%D8%B1%D8%A7%D8%AA-%D8%A7%D9%84%D8%B4%D9%8A%D8%AE-%D8%A7%D9%84%D8%B4%D8%B9%D8%B1%D8%A7%D9%88%D9%8A',
         'info_dict':

youtube_dl/extractor/dfb.py

@@ -12,39 +12,46 @@ class DFBIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://tv.dfb.de/video/u-19-em-stimmen-zum-spiel-gegen-russland/11633/',
-        # The md5 is different each time
+        'md5': 'ac0f98a52a330f700b4b3034ad240649',
         'info_dict': {
             'id': '11633',
             'display_id': 'u-19-em-stimmen-zum-spiel-gegen-russland',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'U 19-EM: Stimmen zum Spiel gegen Russland',
             'upload_date': '20150714',
         },
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        display_id = mobj.group('display_id')
+        display_id, video_id = re.match(self._VALID_URL, url).groups()
 
-        webpage = self._download_webpage(url, display_id)
         player_info = self._download_xml(
             'http://tv.dfb.de/server/hd_video.php?play=%s' % video_id,
             display_id)
         video_info = player_info.find('video')
+        stream_access_url = self._proto_relative_url(video_info.find('url').text.strip())
 
-        f4m_info = self._download_xml(
-            self._proto_relative_url(video_info.find('url').text.strip()), display_id)
-        token_el = f4m_info.find('token')
-        manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth'] + '&hdcore=3.2.0'
-        formats = self._extract_f4m_formats(manifest_url, display_id)
+        formats = []
+        # see http://tv.dfb.de/player/js/ajax.js for the method to extract m3u8 formats
+        for sa_url in (stream_access_url, stream_access_url + '&area=&format=iphone'):
+            stream_access_info = self._download_xml(sa_url, display_id)
+            token_el = stream_access_info.find('token')
+            manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth']
+            if '.f4m' in manifest_url:
+                formats.extend(self._extract_f4m_formats(
+                    manifest_url + '&hdcore=3.2.0',
+                    display_id, f4m_id='hds', fatal=False))
+            else:
+                formats.extend(self._extract_m3u8_formats(
+                    manifest_url, display_id, 'mp4',
+                    'm3u8_native', m3u8_id='hls', fatal=False))
         self._sort_formats(formats)
 
         return {
             'id': video_id,
             'display_id': display_id,
             'title': video_info.find('title').text,
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'thumbnail': 'http://tv.dfb.de/images/%s_640x360.jpg' % video_id,
             'upload_date': unified_strdate(video_info.find('time_date').text),
             'formats': formats,
         }

youtube_dl/extractor/discovery.py

@@ -33,6 +33,7 @@ class DiscoveryIE(InfoExtractor):
             'duration': 156,
             'timestamp': 1302032462,
             'upload_date': '20110405',
+            'uploader_id': '103207',
         },
         'params': {
             'skip_download': True,  # requires ffmpeg
@@ -54,7 +55,11 @@ class DiscoveryIE(InfoExtractor):
             'upload_date': '20140725',
             'timestamp': 1406246400,
             'duration': 116,
+            'uploader_id': '103207',
         },
+        'params': {
+            'skip_download': True,  # requires ffmpeg
+        }
     }]
 
     def _real_extract(self, url):
@@ -66,13 +71,19 @@ class DiscoveryIE(InfoExtractor):
         entries = []
 
         for idx, video_info in enumerate(info['playlist']):
-            formats = self._extract_m3u8_formats(
-                video_info['src'], display_id, 'mp4', 'm3u8_native', m3u8_id='hls',
-                note='Download m3u8 information for video %d' % (idx + 1))
-            self._sort_formats(formats)
+            subtitles = {}
+            caption_url = video_info.get('captionsUrl')
+            if caption_url:
+                subtitles = {
+                    'en': [{
+                        'url': caption_url,
+                    }]
+                }
+
             entries.append({
+                '_type': 'url_transparent',
+                'url': 'http://players.brightcove.net/103207/default_default/index.html?videoId=ref:%s' % video_info['referenceId'],
                 'id': compat_str(video_info['id']),
-                'formats': formats,
                 'title': video_info['title'],
                 'description': video_info.get('description'),
                 'duration': parse_duration(video_info.get('video_length')),
@@ -80,6 +91,7 @@ class DiscoveryIE(InfoExtractor):
                 'thumbnail': video_info.get('thumbnailURL'),
                 'alt_title': video_info.get('secondary_title'),
                 'timestamp': parse_iso8601(video_info.get('publishedDate')),
+                'subtitles': subtitles,
             })
 
         return self.playlist_result(entries, display_id, video_title)

youtube_dl/extractor/discoverygo.py

@@ -0,0 +1,98 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    extract_attributes,
+    int_or_none,
+    parse_age_limit,
+    unescapeHTML,
+)
+
+
+class DiscoveryGoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?discoverygo\.com/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'https://www.discoverygo.com/love-at-first-kiss/kiss-first-ask-questions-later/',
+        'info_dict': {
+            'id': '57a33c536b66d1cd0345eeb1',
+            'ext': 'mp4',
+            'title': 'Kiss First, Ask Questions Later!',
+            'description': 'md5:fe923ba34050eae468bffae10831cb22',
+            'duration': 2579,
+            'series': 'Love at First Kiss',
+            'season_number': 1,
+            'episode_number': 1,
+            'age_limit': 14,
+        },
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        container = extract_attributes(
+            self._search_regex(
+                r'(<div[^>]+class=["\']video-player-container[^>]+>)',
+                webpage, 'video container'))
+
+        video = self._parse_json(
+            unescapeHTML(container.get('data-video') or container.get('data-json')),
+            display_id)
+
+        title = video['name']
+
+        stream = video['stream']
+        STREAM_URL_SUFFIX = 'streamUrl'
+        formats = []
+        for stream_kind in ('', 'hds'):
+            suffix = STREAM_URL_SUFFIX.capitalize() if stream_kind else STREAM_URL_SUFFIX
+            stream_url = stream.get('%s%s' % (stream_kind, suffix))
+            if not stream_url:
+                continue
+            if stream_kind == '':
+                formats.extend(self._extract_m3u8_formats(
+                    stream_url, display_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+            elif stream_kind == 'hds':
+                formats.extend(self._extract_f4m_formats(
+                    stream_url, display_id, f4m_id=stream_kind, fatal=False))
+        self._sort_formats(formats)
+
+        video_id = video.get('id') or display_id
+        description = video.get('description', {}).get('detailed')
+        duration = int_or_none(video.get('duration'))
+
+        series = video.get('show', {}).get('name')
+        season_number = int_or_none(video.get('season', {}).get('number'))
+        episode_number = int_or_none(video.get('episodeNumber'))
+
+        tags = video.get('tags')
+        age_limit = parse_age_limit(video.get('parental', {}).get('rating'))
+
+        subtitles = {}
+        captions = stream.get('captions')
+        if isinstance(captions, list):
+            for caption in captions:
+                subtitle_url = caption.get('fileUrl')
+                if (not subtitle_url or not isinstance(subtitle_url, compat_str) or
+                        not subtitle_url.startswith('http')):
+                    continue
+                lang = caption.get('fileLang', 'en')
+                subtitles.setdefault(lang, []).append({'url': subtitle_url})
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'duration': duration,
+            'series': series,
+            'season_number': season_number,
+            'episode_number': episode_number,
+            'tags': tags,
+            'age_limit': age_limit,
+            'formats': formats,
+            'subtitles': subtitles,
+        }