release 2016.08.13

There's no video anymore

.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 *2016.06.11.1*. 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 Issues with outdated version will be rejected.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 **2016.06.11.1**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.08.13*. 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 Issues with outdated version will be rejected.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 **2016.08.13**
 
 ### 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 2016.06.11.1
+[debug] youtube-dl version 2016.08.13
 [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.

.travis.yml

@@ -7,9 +7,6 @@ python:
   - "3.4"
   - "3.5"
 sudo: false
-install:
-  - bash ./devscripts/install_srelay.sh
-  - export PATH=$PATH:$(pwd)/tmp/srelay-0.4.8b6
 script: nosetests test --verbose
 notifications:
   email:

AUTHORS

@@ -173,3 +173,11 @@ Kevin Deldycke
 inondle
 Tomáš Čech
 Déstin Reed
+Roman Tsiupa
+Artur Krysiak
+Jakub Adam Wieczorek
+Aleksandar Topuzović
+Nehal Patel
+Rob van Bekkum
+Petr Zvoníček
+Pratyush Singh

CONTRIBUTING.md

@@ -46,7 +46,7 @@ Make sure that someone has not already opened the issue you're trying to open. S
 
 ###  Why are existing options not enough?
 
-Before requesting a new feature, please have a quick peek at [the list of supported options](https://github.com/rg3/youtube-dl/blob/master/README.md#synopsis). Many feature requests are for features that actually exist already! Please, absolutely do show off your work in the issue report and detail how the existing similar options do *not* solve your problem.
+Before requesting a new feature, please have a quick peek at [the list of supported options](https://github.com/rg3/youtube-dl/blob/master/README.md#options). Many feature requests are for features that actually exist already! Please, absolutely do show off your work in the issue report and detail how the existing similar options do *not* solve your problem.
 
 ###  Is there enough context in your bug report?
 
@@ -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 +1,394 @@
+version 2016.08.13
+
+Core
+* Show progress for curl external downloader
+* Forward more options to curl external downloader
+
+Extractors
+* [pbs] Fix description extraction
+* [franceculture] Fix extraction (#10324)
+* [pornotube] Fix extraction (#10322)
+* [4tube] Fix metadata extraction (#10321)
+* [imgur] Fix width and height extraction (#10325)
+* [expotv] Improve extraction
++ [vbox7] Fix extraction (#10309)
+- [tapely] Remove extractor (#10323)
+* [muenchentv] Fix extraction (#10313)
++ [24video] Add support for .me and .xxx TLDs
+* [24video] Fix comment count extraction
+* [sunporno] Add support for embed URLs
+* [sunporno] Fix metadata extraction (#10316)
++ [hgtv] Add extractor for hgtv.ca (#3999)
+- [pbs] Remove request to unavailable API
++ [pbs] Add support for high quality HTTP formats
++ [crunchyroll] Add support for HLS formats (#10301)
+
+version 2016.08.12
+
+Core
+* Subtitles are now written as is. Newline conversions are disabled. (#10268)
++ Recognize more formats in unified_timestamp
+
+Extractors
+- [goldenmoustache] Remove extractor (#10298)
+* [drtuber] Improve title extraction
+* [drtuber] Make dislike count optional (#10297)
+* [chirbit] Fix extraction (#10296)
+* [francetvinfo] Relax URL regular expression
+* [rtlnl] Relax URL regular expression (#10282)
+* [formula1] Relax URL regular expression (#10283)
+* [wat] Improve extraction (#10281)
+* [ctsnews] Fix extraction
+
+
+version 2016.08.10
+
+Core
+* Make --metadata-from-title non fatal when title does not match the pattern
+* Introduce options for randomized sleep before each download
+  --min-sleep-interval and --max-sleep-interval (#9930)
+* Respect default in _search_json_ld
+
+Extractors
++ [uol] Add extractor for uol.com.br (#4263)
+* [rbmaradio] Fix extraction and extract all formats (#10242)
++ [sonyliv] Add extractor for sonyliv.com (#10258)
+* [aparat] Fix extraction
+* [cwtv] Extract HTTP formats
++ [rozhlas] Add extractor for prehravac.rozhlas.cz (#10253)
+* [kuwo:singer] Fix extraction
+
+
+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

@@ -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 -L 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:
@@ -44,7 +44,7 @@ Or with [MacPorts](https://www.macports.org/):
 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,
@@ -103,9 +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)
 
@@ -330,7 +330,15 @@ which means you can modify it, redistribute it or use it however you like.
                                      bidirectional text support. Requires bidiv
                                      or fribidi executable in PATH
     --sleep-interval SECONDS         Number of seconds to sleep before each
-                                     download.
+                                     download when used alone or a lower bound
+                                     of a range for randomized sleep before each
+                                     download (minimum possible number of
+                                     seconds to sleep) when used along with
+                                     --max-sleep-interval.
+    --max-sleep-interval SECONDS     Upper bound of a range for randomized sleep
+                                     before each download (maximum possible
+                                     number of seconds to sleep). Must only be
+                                     used along with --min-sleep-interval.
 
 ## Video Format Options:
     -f, --format FORMAT              Video format code, see the "FORMAT
@@ -424,7 +432,7 @@ 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 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`.
+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:
 ```
@@ -432,6 +440,7 @@ 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`.
@@ -511,6 +520,9 @@ 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
@@ -550,6 +562,10 @@ 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.
@@ -883,9 +899,17 @@ 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
@@ -928,20 +952,152 @@ 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).
@@ -957,7 +1113,7 @@ 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:
 
@@ -1048,7 +1204,7 @@ Make sure that someone has not already opened the issue you're trying to open. S
 
 ###  Why are existing options not enough?
 
-Before requesting a new feature, please have a quick peek at [the list of supported options](https://github.com/rg3/youtube-dl/blob/master/README.md#synopsis). Many feature requests are for features that actually exist already! Please, absolutely do show off your work in the issue report and detail how the existing similar options do *not* solve your problem.
+Before requesting a new feature, please have a quick peek at [the list of supported options](https://github.com/rg3/youtube-dl/blob/master/README.md#options). Many feature requests are for features that actually exist already! Please, absolutely do show off your work in the issue report and detail how the existing similar options do *not* solve your problem.
 
 ###  Is there enough context in your bug report?
 

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/install_srelay.sh

@@ -1,8 +0,0 @@
-#!/bin/bash
-
-mkdir -p tmp && cd tmp
-wget -N http://downloads.sourceforge.net/project/socks-relay/socks-relay/srelay-0.4.8/srelay-0.4.8b6.tar.gz
-tar zxvf srelay-0.4.8b6.tar.gz
-cd srelay-0.4.8b6
-./configure
-make

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__:
@@ -47,11 +59,34 @@ 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)
+    if ie in _ALL_CLASSES:
         names.append(name)
 
 module_contents.append(

devscripts/prepare_manpage.py

@@ -54,7 +54,11 @@ def filter_options(readme):
 
         if in_options:
             if line.lstrip().startswith('-'):
-                option, description = re.split(r'\s{2,}', line.lstrip())
+                split = re.split(r'\s{2,}', line.lstrip())
+                # Description string may start with `-` as well. If there is
+                # only one piece then it's a description bit not an option.
+                if len(split) > 1:
+                    option, description = split
                     split_option = option.split(' ')
 
                     if not split_option[-1].startswith('-'):  # metavar
@@ -63,7 +67,7 @@ def filter_options(readme):
                     # Pandoc's definition_lists. See http://pandoc.org/README.html
                     # for more information.
                     ret += '\n%s\n:   %s\n' % (option, description)
-            else:
+                    continue
             ret += line.lstrip() + '\n'
         else:
             ret += line + '\n'

devscripts/release.sh

@@ -15,6 +15,7 @@
 set -e
 
 skip_tests=true
+gpg_sign_commits=""
 buildserver='localhost:8142'
 
 while true
@@ -24,6 +25,10 @@ case "$1" in
         skip_tests=false
         shift
     ;;
+    --gpg-sign-commits|-S)
+        gpg_sign_commits="-S"
+        shift
+    ;;
     --buildserver)
         buildserver="$2"
         shift 2
@@ -66,10 +71,13 @@ fi
 /bin/echo -e "\n### Changing version in version.py..."
 sed -i "s/__version__ = '.*'/__version__ = '$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"
@@ -116,7 +124,7 @@ git clone --branch gh-pages --single-branch . build/gh-pages
     "$ROOT/devscripts/gh-pages/update-copyright.py"
     "$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

@@ -14,6 +14,7 @@
  - **8tracks**
  - **91porn**
  - **9gag**
+ - **9now.com.au**
  - **abc.net.au**
  - **Abc7News**
  - **abcnews**
@@ -45,7 +46,7 @@
  - **archive.org**: archive.org videos
  - **ARD**
  - **ARD:mediathek**
- - **ARD:mediathek**: Saarländischer Rundfunk
+ - **Arkena**
  - **arte.tv**
  - **arte.tv:+7**
  - **arte.tv:cinema**
@@ -74,6 +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**
@@ -104,6 +107,8 @@
  - **canalc2.tv**
  - **Canalplus**: canalplus.fr, piwiplus.fr and d8.tv
  - **Canvas**
+ - **CarambaTV**
+ - **CarambaTVPage**
  - **CBC**
  - **CBCPlayer**
  - **CBS**
@@ -124,6 +129,7 @@
  - **cliphunter**
  - **ClipRs**
  - **Clipsyndicate**
+ - **CloserToTruth**
  - **cloudtime**: CloudTime
  - **Cloudy**
  - **Clubic**
@@ -136,7 +142,8 @@
  - **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**
@@ -148,6 +155,8 @@
  - **CSNNE**
  - **CSpan**: C-SPAN
  - **CtsNews**: 華視新聞
+ - **CTV**
+ - **CTVNews**
  - **culturebox.francetvinfo.fr**
  - **CultureUnplugged**
  - **CWTV**
@@ -173,6 +182,7 @@
  - **DigitallySpeaking**
  - **Digiteka**
  - **Discovery**
+ - **DiscoveryGo**
  - **Dotsub**
  - **DouyuTV**: 斗鱼
  - **DPlay**
@@ -218,6 +228,7 @@
  - **Firstpost**
  - **FiveTV**
  - **Flickr**
+ - **Flipagram**
  - **Folketinget**: Folketinget (ft.dk; Danish parliament)
  - **FootyRoom**
  - **Formula1**
@@ -227,7 +238,6 @@
  - **FoxSports**
  - **france2.fr:generation-quoi**
  - **FranceCulture**
- - **FranceCultureEmission**
  - **FranceInter**
  - **francetv**: France 2, 3, 4, 5 and Ô
  - **francetvinfo.fr**
@@ -236,14 +246,13 @@
  - **FreeVideo**
  - **Funimation**
  - **FunnyOrDie**
+ - **Fusion**
  - **GameInformer**
- - **Gamekings**
  - **GameOne**
  - **gameone:playlist**
  - **Gamersyde**
  - **GameSpot**
  - **GameStar**
- - **Gametrailers**
  - **Gazeta**
  - **GDCVault**
  - **generic**: Generic downloader that works on some sites
@@ -255,7 +264,6 @@
  - **GloboArticle**
  - **GodTube**
  - **GodTV**
- - **GoldenMoustache**
  - **Golem**
  - **GoogleDrive**
  - **Goshgay**
@@ -268,7 +276,9 @@
  - **HellPorno**
  - **Helsinki**: helsinki.fi
  - **HentaiStigma**
+ - **HGTV**
  - **HistoricFilms**
+ - **history:topic**: History.com Topic
  - **hitbox**
  - **hitbox:live**
  - **HornBunny**
@@ -276,6 +286,8 @@
  - **HotStar**
  - **Howcast**
  - **HowStuffWorks**
+ - **HRTi**
+ - **HRTiPlaylist**
  - **HuffPost**: Huffington Post
  - **Hypem**
  - **Iconosquare**
@@ -303,6 +315,7 @@
  - **jpopsuki.tv**
  - **JWPlatform**
  - **Kaltura**
+ - **Kamcord**
  - **KanalPlay**: Kanal 5/9/11 Play
  - **Kankan**
  - **Karaoketv**
@@ -322,8 +335,10 @@
  - **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**
@@ -355,6 +370,7 @@
  - **MatchTV**
  - **MDR**: MDR.DE and KiKA
  - **media.ccc.de**
+ - **META**
  - **metacafe**
  - **Metacritic**
  - **Mgoon**
@@ -381,10 +397,9 @@
  - **MovieFap**
  - **Moviezine**
  - **MPORA**
- - **MSNBC**
+ - **MSN**
  - **MTV**
  - **mtv.de**
- - **mtviggy.com**
  - **mtvservices:embedded**
  - **MuenchenTV**: münchen.tv
  - **MusicPlayOn**
@@ -400,7 +415,8 @@
  - **MyVidster**
  - **n-tv.de**
  - **natgeo**
- - **natgeo:channel**
+ - **natgeo:episodeguide**
+ - **natgeo:video**
  - **Naver**
  - **NBA**
  - **NBC**
@@ -424,7 +440,6 @@
  - **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**
@@ -432,8 +447,11 @@
  - **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**
@@ -461,9 +479,12 @@
  - **NYTimes**
  - **NYTimesArticle**
  - **ocw.mit.edu**
+ - **OdaTV**
  - **Odnoklassniki**
  - **OktoberfestTV**
  - **on.aol.com**
+ - **onet.tv**
+ - **onet.tv:channel**
  - **OnionStudios**
  - **Ooyala**
  - **OoyalaExternal**
@@ -497,8 +518,10 @@
  - **plus.google**: Google Plus
  - **pluzz.francetv.fr**
  - **podomatic**
+ - **Pokemon**
+ - **PolskieRadio**
  - **PornHd**
- - **PornHub**
+ - **PornHub**: PornHub and Thumbzilla
  - **PornHubPlaylist**
  - **PornHubUserVideos**
  - **Pornotube**
@@ -516,6 +539,7 @@
  - **qqmusic:singer**: QQ音乐 - 歌手
  - **qqmusic:toplist**: QQ音乐 - 排行榜
  - **R7**
+ - **R7Article**
  - **radio.de**
  - **radiobremen**
  - **radiocanada**
@@ -535,8 +559,11 @@
  - **revision3:embed**
  - **RICE**
  - **RingTV**
+ - **RockstarGames**
+ - **RoosterTeeth**
  - **RottenTomatoes**
  - **Roxwel**
+ - **Rozhlas**
  - **RTBF**
  - **rte**: Raidió Teilifís Éireann TV
  - **rte:radio**: Raidió Teilifís Éireann radio
@@ -547,7 +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
@@ -580,8 +609,10 @@
  - **Shared**: shared.sx and vivo.sx
  - **ShareSix**
  - **Sina**
+ - **SixPlay**
+ - **skynewsarabia:article**
  - **skynewsarabia:video**
- - **skynewsarabia:video**
+ - **SkySports**
  - **Slideshare**
  - **Slutload**
  - **smotri**: Smotri.com
@@ -590,6 +621,7 @@
  - **smotri:user**: Smotri.com user videos
  - **Snotr**
  - **Sohu**
+ - **SonyLIV**
  - **soundcloud**
  - **soundcloud:playlist**
  - **soundcloud:search**: Soundcloud search
@@ -613,12 +645,14 @@
  - **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**
@@ -630,7 +664,6 @@
  - **SztvHu**
  - **Tagesschau**
  - **tagesschau:player**
- - **Tapely**
  - **Tass**
  - **TDSLifeway**
  - **teachertube**: teachertube.com videos
@@ -647,6 +680,7 @@
  - **Telegraaf**
  - **TeleMB**
  - **TeleTask**
+ - **Telewebion**
  - **TF1**
  - **TheIntercept**
  - **ThePlatform**
@@ -665,6 +699,7 @@
  - **TNAFlix**
  - **TNAFlixNetworkEmbed**
  - **toggle**
+ - **Tosh**: Tosh.0
  - **tou.tv**
  - **Toypics**: Toypics user profile
  - **ToypicsUser**: Toypics user profile
@@ -694,10 +729,12 @@
  - **tvigle**: Интернет-телевидение Tvigle.ru
  - **tvland.com**
  - **tvp**: Telewizja Polska
+ - **tvp:embed**: Telewizja Polska
  - **tvp:series**
  - **TVPlay**: TV3Play and related services
  - **Tweakers**
  - **twitch:chapter**
+ - **twitch:clips**
  - **twitch:past_broadcasts**
  - **twitch:profile**
  - **twitch:stream**
@@ -710,7 +747,9 @@
  - **udemy:course**
  - **UDNEmbed**: 聯合影音
  - **Unistra**
+ - **uol.com.br**
  - **Urort**: NRK P3 Urørt
+ - **URPlay**
  - **USAToday**
  - **ustream**
  - **ustream:channel**
@@ -728,6 +767,7 @@
  - **vh1.com**
  - **Vice**
  - **ViceShow**
+ - **Vidbit**
  - **Viddler**
  - **video.google:search**: Google Video search
  - **video.mit.edu**
@@ -767,8 +807,10 @@
  - **vine:user**
  - **vk**: VK
  - **vk:uservideos**: VK - User's Videos
+ - **vk:wallpost**
  - **vlive**
  - **Vodlocker**
+ - **VODPlatform**
  - **VoiceRepublic**
  - **VoxMedia**
  - **Vporn**
@@ -792,10 +834,11 @@
  - **WNL**
  - **WorldStarHipHop**
  - **wrzuta.pl**
+ - **wrzuta.pl:playlist**
  - **WSJ**: Wall Street Journal
  - **XBef**
  - **XboxClips**
- - **XFileShare**: XFileShare based sites: DaClips, FileHoot, GorillaVid, MovPod, PowerWatch, Rapidvideo.ws, TheVideoBee, Vidto, Streamin.To
+ - **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
@@ -834,6 +877,7 @@
  - **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/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

@@ -87,6 +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 = '''

test/test_http.py

@@ -138,27 +138,27 @@ class TestProxy(unittest.TestCase):
         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 = http_server_port(self.cn_proxy)
-        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_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,
@@ -60,11 +62,13 @@ from youtube_dl.utils import (
     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,
@@ -78,6 +82,7 @@ from youtube_dl.utils import (
     cli_option,
     cli_valueless_option,
     cli_bool_option,
+    parse_codecs,
 )
 from youtube_dl.compat import (
     compat_chr,
@@ -283,8 +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)
@@ -383,6 +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'""")
@@ -401,6 +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)
@@ -579,6 +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-_.~'
@@ -640,6 +709,9 @@ class TestUtil(unittest.TestCase):
             "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"})
@@ -896,6 +968,7 @@ The first line
         self.assertEqual(cli_option({'proxy': '127.0.0.1:3128'}, '--proxy', 'proxy'), ['--proxy', '127.0.0.1:3128'])
         self.assertEqual(cli_option({'proxy': None}, '--proxy', 'proxy'), [])
         self.assertEqual(cli_option({}, '--proxy', 'proxy'), [])
+        self.assertEqual(cli_option({'retries': 10}, '--retries', 'retries'), ['--retries', '10'])
 
     def test_cli_valueless_option(self):
         self.assertEqual(cli_valueless_option(
@@ -956,5 +1029,17 @@ 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(b'--username' in serr)
+        self.assertTrue(b'johnsmith' not in serr)
+        self.assertTrue(b'--password' in serr)
+        self.assertTrue(b'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(b'-u' in serr)
+        self.assertTrue(b'johnsmith' not in serr)
+        self.assertTrue(b'-p' in serr)
+        self.assertTrue(b'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(b'--username' in serr)
+        self.assertTrue(b'johnsmith' not in serr)
+        self.assertTrue(b'--password' in serr)
+        self.assertTrue(b'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(b'-u' in serr)
+        self.assertTrue(b'johnsmith' not in serr)
+        self.assertTrue(b'-p' in serr)
+        self.assertTrue(b'secret' not in serr)
+
+if __name__ == '__main__':
+    unittest.main()

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
@@ -196,8 +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
@@ -248,7 +249,16 @@ class YoutubeDL(object):
     source_address:    (Experimental) Client-side IP address to bind to.
     call_home:         Boolean, true iff we are allowed to contact the
                        youtube-dl servers for debugging.
-    sleep_interval:    Number of seconds to sleep before each download.
+    sleep_interval:    Number of seconds to sleep before each download when
+                       used alone or a lower bound of a range for randomized
+                       sleep before each download (minimum possible number
+                       of seconds to sleep) when used along with
+                       max_sleep_interval.
+    max_sleep_interval:Upper bound of a range for randomized sleep before each
+                       download (maximum possible number of seconds to sleep).
+                       Must only be used along with sleep_interval.
+                       Actual sleep time will be a random float from range
+                       [sleep_interval; max_sleep_interval].
     listformats:       Print an overview of available video formats and exit.
     list_thumbnails:   Print a table of all thumbnails and exit.
     match_filter:      A function that gets called with the info_dict of
@@ -304,6 +314,11 @@ 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
@@ -1046,9 +1061,9 @@ 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:
@@ -1056,17 +1071,17 @@ 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':
@@ -1079,9 +1094,10 @@ 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 = [
@@ -1155,17 +1171,18 @@ 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'))
@@ -1372,7 +1389,34 @@ 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)
@@ -1559,7 +1603,9 @@ class YoutubeDL(object):
                         self.to_screen('[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
                     else:
                         self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
-                        with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
+                        # Use newline='' to prevent conversion of newline characters
+                        # See https://github.com/rg3/youtube-dl/issues/10268
+                        with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
                             subfile.write(sub_data)
                 except (OSError, IOError):
                     self.report_error('Cannot write subtitles file ' + sub_filename)

youtube_dl/__init__.py

@@ -145,6 +145,16 @@ def _real_main(argv=None):
         if numeric_limit is None:
             parser.error('invalid max_filesize specified')
         opts.max_filesize = numeric_limit
+    if opts.sleep_interval is not None:
+        if opts.sleep_interval < 0:
+            parser.error('sleep interval must be positive or 0')
+    if opts.max_sleep_interval is not None:
+        if opts.max_sleep_interval < 0:
+            parser.error('max sleep interval must be positive or 0')
+        if opts.max_sleep_interval < opts.sleep_interval:
+            parser.error('max sleep interval must be greater than or equal to min sleep interval')
+    else:
+        opts.max_sleep_interval = opts.sleep_interval
 
     def parse_retries(retries):
         if retries in ('inf', 'infinite'):
@@ -370,6 +380,7 @@ def _real_main(argv=None):
         'source_address': opts.source_address,
         'call_home': opts.call_home,
         'sleep_interval': opts.sleep_interval,
+        'max_sleep_interval': opts.max_sleep_interval,
         'external_downloader': opts.external_downloader,
         'list_thumbnails': opts.list_thumbnails,
         'playlist_items': opts.playlist_items,
@@ -382,6 +393,8 @@ 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:

youtube_dl/compat.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import binascii
@@ -2594,15 +2595,19 @@ except ImportError:  # Python < 3.3
             return "'" + s.replace("'", "'\"'\"'") + "'"
 
 
-if sys.version_info >= (2, 7, 3):
+try:
+    args = shlex.split('中文')
+    assert (isinstance(args, list) and
+            isinstance(args[0], compat_str) and
+            args[0] == '中文')
     compat_shlex_split = shlex.split
-else:
+except (AssertionError, UnicodeEncodeError):
     # Working around shlex issue with unicode strings on some python 2
     # versions (see http://bugs.python.org/issue1548891)
     def compat_shlex_split(s, comments=False, posix=True):
         if isinstance(s, compat_str):
             s = s.encode('utf-8')
-        return shlex.split(s, comments, posix)
+        return list(map(lambda s: s.decode('utf-8'), shlex.split(s, comments, posix)))
 
 
 def compat_ord(c):

youtube_dl/downloader/common.py

@@ -4,6 +4,7 @@ import os
 import re
 import sys
 import time
+import random
 
 from ..compat import compat_os_name
 from ..utils import (
@@ -342,8 +343,11 @@ class FileDownloader(object):
             })
             return True
 
-        sleep_interval = self.params.get('sleep_interval')
-        if sleep_interval:
+        min_sleep_interval = self.params.get('sleep_interval')
+        if min_sleep_interval:
+            max_sleep_interval = self.params.get('max_sleep_interval', min_sleep_interval)
+            print(min_sleep_interval, max_sleep_interval)
+            sleep_interval = random.uniform(min_sleep_interval, max_sleep_interval)
             self.to_screen('[download] Sleeping %s seconds...' % sleep_interval)
             time.sleep(sleep_interval)
 

youtube_dl/downloader/external.py

@@ -85,7 +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
 
 
@@ -96,6 +96,12 @@ class CurlFD(ExternalFD):
         cmd = [self.exe, '--location', '-o', tmpfilename]
         for key, val in info_dict['http_headers'].items():
             cmd += ['--header', '%s: %s' % (key, val)]
+        cmd += self._bool_option('--continue-at', 'continuedl', '-', '0')
+        cmd += self._valueless_option('--silent', 'noprogress')
+        cmd += self._valueless_option('--verbose', 'verbose')
+        cmd += self._option('--limit-rate', 'ratelimit')
+        cmd += self._option('--retry', 'retries')
+        cmd += self._option('--max-filesize', 'max_filesize')
         cmd += self._option('--interface', 'source_address')
         cmd += self._option('--proxy', 'proxy')
         cmd += self._valueless_option('--insecure', 'nocheckcertificate')
@@ -103,6 +109,16 @@ class CurlFD(ExternalFD):
         cmd += ['--', info_dict['url']]
         return cmd
 
+    def _call_downloader(self, tmpfilename, info_dict):
+        cmd = [encodeArgument(a) for a in self._make_cmd(tmpfilename, info_dict)]
+
+        self._debug_cmd(cmd)
+
+        # curl writes the progress to stderr so don't capture it.
+        p = subprocess.Popen(cmd)
+        p.communicate()
+        return p.returncode
+
 
 class AxelFD(ExternalFD):
     AVAILABLE_OPT = '-V'

youtube_dl/downloader/f4m.py

@@ -196,6 +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)))
 
@@ -329,7 +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:

youtube_dl/downloader/hls.py

@@ -2,14 +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,
 )
 
 
@@ -21,7 +31,7 @@ class HlsFD(FragmentFD):
     @staticmethod
     def can_download(manifest):
         UNSUPPORTED_FEATURES = (
-            r'#EXT-X-KEY:METHOD=(?!NONE)',  # encrypted streams [1]
+            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
@@ -39,7 +49,9 @@ class HlsFD(FragmentFD):
             # 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
         )
-        return all(not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES)
+        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']
@@ -57,36 +69,60 @@ class HlsFD(FragmentFD):
                 fd.add_progress_hook(ph)
             return fd.real_download(filename, info_dict)
 
-        fragment_urls = []
+        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):
+        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')
-            ctx['dest_stream'].write(down.read())
+                    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/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:
+            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
                 }),
+            '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'
                 })
-        return info
+        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/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 in ('video/f4m', 'application/f4m+xml'):
+            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.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)
-                f = common_info.copy()
-                f.update({
+            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(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/aol.py

@@ -123,6 +123,10 @@ class AolFeaturesIE(InfoExtractor):
             'title': 'What To Watch - February 17, 2016',
         },
         'add_ie': ['FiveMin'],
+        'params': {
+            # encrypted m3u8 download
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/aparat.py

@@ -1,8 +1,6 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
@@ -15,7 +13,7 @@ class AparatIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.aparat.com/v/wP8On',
-        'md5': '6714e0af7e0d875c5a39c4dc4ab46ad1',
+        'md5': '131aca2e14fe7c4dcb3c4877ba300c89',
         'info_dict': {
             'id': 'wP8On',
             'ext': 'mp4',
@@ -31,13 +29,13 @@ class AparatIE(InfoExtractor):
         # Note: There is an easier-to-parse configuration at
         # http://www.aparat.com/video/video/config/videohash/%video_id
         # but the URL in there does not work
-        embed_url = ('http://www.aparat.com/video/video/embed/videohash/' +
-                     video_id + '/vt/frame')
+        embed_url = 'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id
         webpage = self._download_webpage(embed_url, video_id)
 
-        video_urls = [video_url.replace('\\/', '/') for video_url in re.findall(
-            r'(?:fileList\[[0-9]+\]\s*=|"file"\s*:)\s*"([^"]+)"', webpage)]
-        for i, video_url in enumerate(video_urls):
+        file_list = self._parse_json(self._search_regex(
+            r'fileList\s*=\s*JSON\.parse\(\'([^\']+)\'\)', webpage, 'file list'), video_id)
+        for i, item in enumerate(file_list[0]):
+            video_url = item['file']
             req = HEADRequest(video_url)
             res = self._request_webpage(
                 req, video_id, note='Testing video URL %d' % i, errnote=False)

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

@@ -180,11 +180,14 @@ class ArteTVBaseIE(InfoExtractor):
 
 class ArteTVPlus7IE(ArteTVBaseIE):
     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>[^/?#&]+)'
+    _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
@@ -240,10 +243,10 @@ class ArteTVPlus7IE(ArteTVBaseIE):
             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)
+        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
@@ -252,22 +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,
@@ -349,14 +347,13 @@ class ArteTVCinemaIE(ArteTVPlus7IE):
     _VALID_URL = r'https?://cinema\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>.+)'
 
     _TESTS = [{
-        'url': 'http://cinema.arte.tv/de/node/38291',
-        'md5': '6b275511a5107c60bacbeeda368c3aa1',
+        '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',
         },
     }]
 
@@ -422,6 +419,7 @@ class ArteTVPlaylistIE(ArteTVBaseIE):
         'info_dict': {
             'id': 'PL-013263',
             'title': 'Areva & Uramin',
+            'description': 'md5:a1dc0312ce357c262259139cfd48c9bf',
         },
         'playlist_mincount': 6,
     }, {

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/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.
@@ -746,7 +759,7 @@ class BBCIE(BBCCoUkIE):
 
         webpage = self._download_webpage(url, playlist_id)
 
-        json_ld_info = self._search_json_ld(webpage, playlist_id, default=None)
+        json_ld_info = self._search_json_ld(webpage, playlist_id, default={})
         timestamp = json_ld_info.get('timestamp')
 
         playlist_title = json_ld_info.get('title')
@@ -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

@@ -11,22 +11,13 @@ from ..compat import compat_urllib_parse_unquote
 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',
-        'info_dict': {
-            'id': '16537',
-            'ext': 'mp4',
-            'title': 'Singham Returns',
-            'description': 'md5:3d2ba5815f14911d5cc6a501ae0cf65d',
-        }
-    }, {
         # 2 formats
         'url': 'http://www.bigflix.com/Tamil-movies/Drama-movies/Madarasapatinam/16070',
         'info_dict': {
             'id': '16070',
             'ext': 'mp4',
             'title': 'Madarasapatinam',
-            'description': 'md5:63b9b8ed79189c6f0418c26d9a3452ca',
+            'description': 'md5:9f0470b26a4ba8e824c823b5d95c2f6b',
             'formats': 'mincount:2',
         },
         'params': {

youtube_dl/extractor/bilibili.py

@@ -25,13 +25,13 @@ class BiliBiliIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.bilibili.tv/video/av1074402/',
-        'md5': '5f7d29e1a2872f3df0cf76b1f87d3788',
+        'md5': '9fa226fe2b8a9a4d5a69b4c6a183417e',
         'info_dict': {
             'id': '1554319',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': '【金坷垃】金泡沫',
             'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923',
-            'duration': 308.067,
+            'duration': 308.315,
             'timestamp': 1398012660,
             'upload_date': '20140420',
             'thumbnail': 're:^https?://.+\.jpg',
@@ -41,23 +41,20 @@ class BiliBiliIE(InfoExtractor):
     }, {
         'url': 'http://www.bilibili.com/video/av1041170/',
         'info_dict': {
-            'id': '1041170',
+            'id': '1507019',
+            'ext': 'mp4',
             '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！~',
+            'timestamp': 1396530060,
+            'upload_date': '20140403',
+            'uploader': '枫叶逝去',
+            'uploader_id': '520116',
         },
-        '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&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&illust_id=56912929',
-        },
-        'playlist': [{
-            'md5': '55cdadedf3254caaa0d5d27cf20a8f9c',
-            'info_dict': {
-                'id': '4808130_part1',
-                'ext': 'flv',
+            'id': '7802182',
+            'ext': 'mp4',
             'title': '【长篇】哆啦A梦443【钉铛】',
             'description': '(2016.05.27)来组合客人的脸吧&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&illust_id=56912929',
             'timestamp': 1464564180,
@@ -65,49 +62,12 @@ class BiliBiliIE(InfoExtractor):
             '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&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&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&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&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&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&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',
+            'ext': 'mp4',
             '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',

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

@@ -24,7 +24,8 @@ class BIQLEIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Ребенок в шоке от автоматической мойки',
             'uploader': 'Dmitry Kotov',
-        }
+        },
+        'skip': ' This video was marked as adult.  Embedding adult videos on external sites is prohibited.',
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/bloomberg.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -20,6 +21,18 @@ class BloombergIE(InfoExtractor):
         '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,
@@ -33,7 +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',
@@ -298,13 +310,19 @@ class BrightcoveLegacyIE(InfoExtractor):
             info_url, player_key, 'Downloading playlist information')
 
         json_data = json.loads(playlist_info)
-        if 'videoList' not in json_data:
-            raise ExtractorError('Empty playlist')
+        if 'videoList' in json_data:
             playlist_info = json_data['videoList']
-        videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
+            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')
+
+        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'])
@@ -528,14 +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_native', m3u8_id='hls', fatal=False))
-            elif source_type == 'application/dash+xml':
+            elif ext == 'mpd':
                 if not src:
                     continue
                 formats.extend(self._extract_mpd_formats(src, video_id, 'dash', fatal=False))
@@ -551,7 +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({
@@ -585,6 +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 = {}
@@ -597,7 +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/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/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

@@ -4,9 +4,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     js_to_json,
     smuggle_url,
+    try_get,
 )
 
 
@@ -25,8 +27,22 @@ class CBCIE(InfoExtractor):
             'upload_date': '20160203',
             'uploader': 'CBCC-NEW',
         },
+        'skip': 'Geo-restricted to Canada',
     }, {
-        # with clipId
+        # 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': {
@@ -64,6 +80,7 @@ class CBCIE(InfoExtractor):
                 'uploader': 'CBCC-NEW',
             },
         }],
+        'skip': 'Geo-restricted to Canada',
     }]
 
     @classmethod
@@ -81,6 +98,12 @@ class CBCIE(InfoExtractor):
             media_id = player_info.get('mediaId')
             if not media_id:
                 clip_id = player_info['clipId']
+                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]
@@ -104,6 +127,7 @@ class CBCPlayerIE(InfoExtractor):
             'upload_date': '20160210',
             'uploader': 'CBCC-NEW',
         },
+        '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',

youtube_dl/extractor/cbs.py

@@ -1,17 +1,13 @@
 from __future__ import unicode_literals
 
-import re
-
-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 {
@@ -21,9 +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'(?:cbs:(?P<content_id>\w+)|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/(?:video|artist)|colbertlateshow\.com/(?:video|podcasts))/[^/]+/(?P<display_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/',
@@ -38,25 +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/',
@@ -68,44 +59,5 @@ class CBSIE(CBSBaseIE):
     TP_RELEASE_URL_TEMPLATE = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true'
 
     def _real_extract(self, url):
-        content_id, display_id = re.match(self._VALID_URL, url).groups()
-        if not content_id:
-            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

@@ -1,12 +1,10 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import calendar
-import datetime
-
 from .anvato import AnvatoIE
 from .sendtonews import SendtoNewsIE
 from ..compat import compat_urlparse
+from ..utils import unified_timestamp
 
 
 class CBSLocalIE(AnvatoIE):
@@ -71,10 +69,7 @@ class CBSLocalIE(AnvatoIE):
 
         time_str = self._html_search_regex(
             r'class="entry-date">([^<]+)<', webpage, 'released date', fatal=False)
-        timestamp = None
-        if time_str:
-            timestamp = calendar.timegm(datetime.datetime.strptime(
-                time_str, '%b %d, %Y %I:%M %p').timetuple())
+        timestamp = unified_timestamp(time_str)
 
         info_dict.update({
             'display_id': display_id,

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/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/chaturbate.py

@@ -17,7 +17,8 @@ class ChaturbateIE(InfoExtractor):
         },
         'params': {
             'skip_download': True,
-        }
+        },
+        'skip': 'Room is offline',
     }, {
         'url': 'https://en.chaturbate.com/siswet19/',
         'only_matching': True,

youtube_dl/extractor/chirbit.py

@@ -1,30 +1,33 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import base64
+
 from .common import InfoExtractor
-from ..utils import (
-    parse_duration,
-    int_or_none,
-)
+from ..utils import parse_duration
 
 
 class ChirbitIE(InfoExtractor):
     IE_NAME = 'chirbit'
     _VALID_URL = r'https?://(?:www\.)?chirb\.it/(?:(?:wp|pl)/|fb_chirbit_player\.swf\?key=)?(?P<id>[\da-zA-Z]+)'
     _TESTS = [{
-        'url': 'http://chirb.it/PrIPv5',
-        'md5': '9847b0dad6ac3e074568bf2cfb197de8',
+        'url': 'http://chirb.it/be2abG',
         'info_dict': {
-            'id': 'PrIPv5',
+            'id': 'be2abG',
             'ext': 'mp3',
-            'title': 'Фасадстрой',
-            'duration': 52,
-            'view_count': int,
-            'comment_count': int,
+            'title': 'md5:f542ea253f5255240be4da375c6a5d7e',
+            'description': 'md5:f24a4e22a71763e32da5fed59e47c770',
+            'duration': 306,
+        },
+        'params': {
+            'skip_download': True,
         }
     }, {
         'url': 'https://chirb.it/fb_chirbit_player.swf?key=PrIPv5',
         'only_matching': True,
+    }, {
+        'url': 'https://chirb.it/wp/MN58c2',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -33,27 +36,30 @@ class ChirbitIE(InfoExtractor):
         webpage = self._download_webpage(
             'http://chirb.it/%s' % audio_id, audio_id)
 
-        audio_url = self._search_regex(
-            r'"setFile"\s*,\s*"([^"]+)"', webpage, 'audio url')
+        data_fd = self._search_regex(
+            r'data-fd=(["\'])(?P<url>(?:(?!\1).)+)\1',
+            webpage, 'data fd', group='url')
+
+        # Reverse engineered from https://chirb.it/js/chirbit.player.js (look
+        # for soundURL)
+        audio_url = base64.b64decode(
+            data_fd[::-1].encode('ascii')).decode('utf-8')
 
         title = self._search_regex(
-            r'itemprop="name">([^<]+)', webpage, 'title')
-        duration = parse_duration(self._html_search_meta(
-            'duration', webpage, 'duration', fatal=False))
-        view_count = int_or_none(self._search_regex(
-            r'itemprop="playCount"\s*>(\d+)', webpage,
-            'listen count', fatal=False))
-        comment_count = int_or_none(self._search_regex(
-            r'>(\d+) Comments?:', webpage,
-            'comment count', fatal=False))
+            r'class=["\']chirbit-title["\'][^>]*>([^<]+)', webpage, 'title')
+        description = self._search_regex(
+            r'<h3>Description</h3>\s*<pre[^>]*>([^<]+)</pre>',
+            webpage, 'description', default=None)
+        duration = parse_duration(self._search_regex(
+            r'class=["\']c-length["\'][^>]*>([^<]+)',
+            webpage, 'duration', fatal=False))
 
         return {
             'id': audio_id,
             'url': audio_url,
             'title': title,
+            'description': description,
             'duration': duration,
-            'view_count': view_count,
-            'comment_count': comment_count,
         }
 
 

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 (
@@ -17,17 +16,16 @@ 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|to))/
+        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 = [
-        {
+    _TEST = {
         'url': 'https://www.cloudy.ec/v/af511e2527aac',
         'md5': '5cb253ace826a42f35b4740539bedf07',
         'info_dict': {
@@ -35,19 +33,9 @@ class CloudyIE(InfoExtractor):
             'ext': 'flv',
             'title': 'Funny Cats and Animals Compilation june 2013',
         }
-        },
-        {
-            'url': 'http://www.videoraj.to/v/47f399fd8bb60',
-            'md5': '7d0f8799d91efd4eda26587421c3c3b0',
-            'info_dict': {
-                'id': '47f399fd8bb60',
-                'ext': 'flv',
-                'title': 'Burning a New iPhone 5 with Gasoline - Will it Survive?',
     }
-        }
-    ]
 
-    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/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',
@@ -35,241 +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|tosh)\.(?:cc\.)?com/
-                         ((?:full-)?episodes/(?:[0-9a-z]{6}/)?(?P<episode>.*)|
-                          (?P<clip>
-                              (?:(?:guests/[^/]+|videos|video-(?:clips|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',
-        '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',
-        },
-        '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',
-                },
-            },
-            {
-                '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',
-        'only_matching': True,
-    }, {
         'url': 'http://tosh.cc.com/video-clips/68g93d/twitter-users-share-summer-plans',
+        'info_dict': {
+            'description': 'Tosh asked fans to share their summer plans.',
+            'title': 'Twitter Users Share Summer Plans',
+        },
+        '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',
+                },
+            },
+        }]
+    }, {
+        '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

@@ -44,6 +44,7 @@ from ..utils import (
     sanitized_Request,
     unescapeHTML,
     unified_strdate,
+    unified_timestamp,
     url_basename,
     xpath_element,
     xpath_text,
@@ -53,6 +54,9 @@ from ..utils import (
     mimetype2ext,
     update_Request,
     update_url_query,
+    parse_m3u8_attributes,
+    extract_attributes,
+    parse_codecs,
 )
 
 
@@ -160,6 +164,7 @@ 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.
@@ -722,9 +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)
@@ -748,10 +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):
@@ -800,40 +812,66 @@ 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)
+        default = kwargs.get('default', NO_DEFAULT)
         if not json_ld:
-            return {}
-        return self._json_ld(json_ld, video_id, fatal=kwargs.get('fatal', True))
+            return default if default is not NO_DEFAULT else {}
+        # JSON-LD may be malformed and thus `fatal` should be respected.
+        # At the same time `default` may be passed that assumes `fatal=False`
+        # for _search_regex. Let's simulate the same behavior here as well.
+        fatal = kwargs.get('fatal', True) if default == NO_DEFAULT else False
+        return self._json_ld(json_ld, video_id, fatal=fatal, 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 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(json_ld.get('name')),
-                    'episode_number': int_or_none(json_ld.get('episodeNumber')),
-                    'description': unescapeHTML(json_ld.get('description')),
+                        'episode': unescapeHTML(e.get('name')),
+                        'episode_number': int_or_none(e.get('episodeNumber')),
+                        'description': unescapeHTML(e.get('description')),
                     })
-                part_of_season = json_ld.get('partOfSeason')
+                    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 = json_ld.get('partOfSeries')
+                    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(json_ld.get('datePublished')),
-                    'title': unescapeHTML(json_ld.get('headline')),
-                    'description': unescapeHTML(json_ld.get('articleBody')),
+                        '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
@@ -875,7 +913,11 @@ 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:
@@ -883,7 +925,8 @@ 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
@@ -1100,7 +1143,7 @@ class InfoExtractor(object):
             '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',
         }
@@ -1150,23 +1193,11 @@ 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:
@@ -1191,6 +1222,7 @@ class InfoExtractor(object):
                     'url': format_url(line.strip()),
                     'tbr': tbr,
                     'ext': ext,
+                    'fps': float_or_none(last_info.get('FRAME-RATE')),
                     'protocol': entry_protocol,
                     'preference': preference,
                 }
@@ -1199,24 +1231,17 @@ 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
@@ -1471,6 +1496,13 @@ 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 []
 
@@ -1503,8 +1535,16 @@ 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:
@@ -1541,7 +1581,7 @@ 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':
@@ -1585,16 +1625,40 @@ 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('$$', '$')
+
+                            # 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')}
+                                        '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'],
@@ -1621,6 +1685,62 @@ 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()
@@ -1681,7 +1801,7 @@ 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)
@@ -1734,6 +1854,13 @@ 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 +74,93 @@ 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)
-        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,
+        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'),
             })
-        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)
+        else:
+            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, fatal=False) 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/crunchyroll.py

@@ -114,6 +114,21 @@ class CrunchyrollIE(CrunchyrollBaseIE):
             # rtmp
             'skip_download': True,
         },
+    }, {
+        'url': 'http://www.crunchyroll.com/rezero-starting-life-in-another-world-/episode-5-the-morning-of-our-promise-is-still-distant-702409',
+        'info_dict': {
+            'id': '702409',
+            'ext': 'mp4',
+            'title': 'Re:ZERO -Starting Life in Another World- Episode 5 – The Morning of Our Promise Is Still Distant',
+            'description': 'md5:97664de1ab24bbf77a9c01918cb7dca9',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'TV TOKYO',
+            'upload_date': '20160508',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
     }, {
         'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
         'only_matching': True,
@@ -336,9 +351,18 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             if video_encode_id in video_encode_ids:
                 continue
             video_encode_ids.append(video_encode_id)
+
+            video_file = xpath_text(stream_info, './file')
+            if not video_file:
+                continue
+            if video_file.startswith('http'):
+                formats.extend(self._extract_m3u8_formats(
+                    video_file, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+                continue
+
             video_url = xpath_text(stream_info, './host')
-            video_play_path = xpath_text(stream_info, './file')
-            if not video_url or not video_play_path:
+            if not video_url:
                 continue
             metadata = stream_info.find('./metadata')
             format_info = {
@@ -353,7 +377,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
                 parsed_video_url = compat_urlparse.urlparse(video_url)
                 direct_video_url = compat_urlparse.urlunparse(parsed_video_url._replace(
                     netloc='v.lvlt.crcdn.net',
-                    path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_play_path.split(':')[-1])))
+                    path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_file.split(':')[-1])))
                 if self._is_valid_url(direct_video_url, video_id, video_format):
                     format_info.update({
                         'url': direct_video_url,
@@ -363,7 +387,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 
             format_info.update({
                 'url': video_url,
-                'play_path': video_play_path,
+                'play_path': video_file,
                 'ext': 'flv',
             })
             formats.append(format_info)

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/ctsnews.py

@@ -1,13 +1,12 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import parse_iso8601, ExtractorError
+from ..utils import unified_timestamp
 
 
 class CtsNewsIE(InfoExtractor):
     IE_DESC = '華視新聞'
-    # https connection failed (Connection reset)
     _VALID_URL = r'https?://news\.cts\.com\.tw/[a-z]+/[a-z]+/\d+/(?P<id>\d+)\.html'
     _TESTS = [{
         'url': 'http://news.cts.com.tw/cts/international/201501/201501291578109.html',
@@ -16,7 +15,7 @@ class CtsNewsIE(InfoExtractor):
             'id': '201501291578109',
             'ext': 'mp4',
             'title': '以色列.真主黨交火 3人死亡',
-            'description': 'md5:95e9b295c898b7ff294f09d450178d7d',
+            'description': '以色列和黎巴嫩真主黨，爆發五年最嚴重衝突，雙方砲轟交火，兩名以軍死亡，還有一名西班牙籍的聯合國維和人...',
             'timestamp': 1422528540,
             'upload_date': '20150129',
         }
@@ -28,7 +27,7 @@ class CtsNewsIE(InfoExtractor):
             'id': '201309031304098',
             'ext': 'mp4',
             'title': '韓國31歲童顏男 貌如十多歲小孩',
-            'description': 'md5:f183feeba3752b683827aab71adad584',
+            'description': '越有年紀的人，越希望看起來年輕一點，而南韓卻有一位31歲的男子，看起來像是11、12歲的小孩，身...',
             'thumbnail': 're:^https?://.*\.jpg$',
             'timestamp': 1378205880,
             'upload_date': '20130903',
@@ -36,8 +35,7 @@ class CtsNewsIE(InfoExtractor):
     }, {
         # With Youtube embedded video
         'url': 'http://news.cts.com.tw/cts/money/201501/201501291578003.html',
-        'md5': '1d842c771dc94c8c3bca5af2cc1db9c5',
-        'add_ie': ['Youtube'],
+        'md5': 'e4726b2ccd70ba2c319865e28f0a91d1',
         'info_dict': {
             'id': 'OVbfO7d0_hQ',
             'ext': 'mp4',
@@ -47,42 +45,37 @@ class CtsNewsIE(InfoExtractor):
             'upload_date': '20150128',
             'uploader_id': 'TBSCTS',
             'uploader': '中華電視公司',
-        }
+        },
+        'add_ie': ['Youtube'],
     }]
 
     def _real_extract(self, url):
         news_id = self._match_id(url)
         page = self._download_webpage(url, news_id)
 
-        if self._search_regex(r'(CTSPlayer2)', page, 'CTSPlayer2 identifier', default=None):
-            feed_url = self._html_search_regex(
-                r'(http://news\.cts\.com\.tw/action/mp4feed\.php\?news_id=\d+)',
-                page, 'feed url')
-            video_url = self._download_webpage(
-                feed_url, news_id, note='Fetching feed')
+        news_id = self._hidden_inputs(page).get('get_id')
+
+        if news_id:
+            mp4_feed = self._download_json(
+                'http://news.cts.com.tw/action/test_mp4feed.php',
+                news_id, note='Fetching feed', query={'news_id': news_id})
+            video_url = mp4_feed['source_url']
         else:
             self.to_screen('Not CTSPlayer video, trying Youtube...')
             youtube_url = self._search_regex(
-                r'src="(//www\.youtube\.com/embed/[^"]+)"', page, 'youtube url',
-                default=None)
-            if not youtube_url:
-                raise ExtractorError('The news includes no videos!', expected=True)
+                r'src="(//www\.youtube\.com/embed/[^"]+)"', page, 'youtube url')
 
-            return {
-                '_type': 'url',
-                'url': youtube_url,
-                'ie_key': 'Youtube',
-            }
+            return self.url_result(youtube_url, ie='Youtube')
 
         description = self._html_search_meta('description', page)
-        title = self._html_search_meta('title', page)
+        title = self._html_search_meta('title', page, fatal=True)
         thumbnail = self._html_search_meta('image', page)
 
         datetime_str = self._html_search_regex(
-            r'(\d{4}/\d{2}/\d{2} \d{2}:\d{2})', page, 'date and time')
-        # Transform into ISO 8601 format with timezone info
-        datetime_str = datetime_str.replace('/', '-') + ':00+0800'
-        timestamp = parse_iso8601(datetime_str, delimiter=' ')
+            r'(\d{4}/\d{2}/\d{2} \d{2}:\d{2})', page, 'date and time', fatal=False)
+        timestamp = None
+        if datetime_str:
+            timestamp = unified_timestamp(datetime_str) - 8 * 3600
 
         return {
             'id': news_id,

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}\?.*\bplay=(?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': {
@@ -28,7 +28,8 @@ class CWTVIE(InfoExtractor):
         'params': {
             # m3u8 download
             'skip_download': True,
-        }
+        },
+        'skip': 'redirect to http://cwtv.com/shows/arrow/',
     }, {
         'url': 'http://www.cwseed.com/shows/whose-line-is-it-anyway/jeff-davis-4/?play=24282b12-ead2-42f2-95ad-26770c2c6088',
         'info_dict': {
@@ -44,22 +45,43 @@ class CWTVIE(InfoExtractor):
             'upload_date': '20151006',
             'timestamp': 1444107300,
         },
-        'params': {
-            # 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):
         video_id = self._match_id(url)
-        video_data = self._download_json(
-            'http://metaframe.digitalsmiths.tv/v2/CWtv/assets/%s/partner/132?format=json' % video_id, video_id)
-
-        formats = self._extract_m3u8_formats(
-            video_data['videos']['variantplaylist']['uri'], video_id, 'mp4')
+        video_data = None
+        formats = []
+        for partner in (154, 213):
+            vdata = self._download_json(
+                'http://metaframe.digitalsmiths.tv/v2/CWtv/assets/%s/partner/%d?format=json' % (video_id, partner), video_id, fatal=False)
+            if not vdata:
+                continue
+            video_data = vdata
+            for quality, quality_data in vdata.get('videos', {}).items():
+                quality_url = quality_data.get('uri')
+                if not quality_url:
+                    continue
+                if quality == 'variantplaylist':
+                    formats.extend(self._extract_m3u8_formats(
+                        quality_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
+                else:
+                    tbr = int_or_none(quality_data.get('bitrate'))
+                    format_id = 'http' + ('-%d' % tbr if tbr else '')
+                    if self._is_valid_url(quality_url, video_id, format_id):
+                        formats.append({
+                            'format_id': format_id,
+                            'url': quality_url,
+                            'tbr': tbr,
+                        })
         self._sort_formats(formats)
 
         thumbnails = [{

youtube_dl/extractor/dailymail.py

@@ -5,19 +5,20 @@ 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/sciencetech/video-1288527/Turn-video-impressionist-masterpiece.html',
-        'md5': '2f639d446394f53f3a33658b518b6615',
+        'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html',
+        'md5': 'f6129624562251f628296c3a9ffde124',
         'info_dict': {
-            'id': '1288527',
+            'id': '1295863',
             'ext': 'mp4',
-            'title': 'Turn any video into an impressionist masterpiece',
-            'description': 'md5:88ddbcb504367987b2708bb38677c9d2',
+            'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'',
+            'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84',
         }
     }
 
@@ -26,7 +27,7 @@ class DailyMailIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
         video_data = self._parse_json(self._search_regex(
             r"data-opts='({.+?})'", webpage, 'video data'), video_id)
-        title = video_data['title']
+        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)
 
@@ -55,7 +56,7 @@ class DailyMailIE(InfoExtractor):
         return {
             'id': video_id,
             'title': title,
-            'description': video_data.get('descr'),
+            '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:
+        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(
-                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 + '/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/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,
+        }

youtube_dl/extractor/dreisat.py

@@ -17,8 +17,12 @@ class DreiSatIE(ZDFIE):
                 'ext': 'mp4',
                 'title': 'Waidmannsheil',
                 'description': 'md5:cce00ca1d70e21425e72c86a98a56817',
-                'uploader': '3sat',
+                'uploader': 'SCHWEIZWEIT',
+                'uploader_id': '100000210',
                 'upload_date': '20140913'
+            },
+            'params': {
+                'skip_download': True,  # m3u8 downloads
             }
         },
         {

youtube_dl/extractor/drtuber.py

@@ -3,7 +3,10 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import str_to_int
+from ..utils import (
+    NO_DEFAULT,
+    str_to_int,
+)
 
 
 class DrTuberIE(InfoExtractor):
@@ -17,7 +20,6 @@ class DrTuberIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'hot perky blonde naked golf',
             'like_count': int,
-            'dislike_count': int,
             'comment_count': int,
             'categories': ['Babe', 'Blonde', 'Erotic', 'Outdoor', 'Softcore', 'Solo'],
             'thumbnail': 're:https?://.*\.jpg$',
@@ -36,25 +38,29 @@ class DrTuberIE(InfoExtractor):
             r'<source src="([^"]+)"', webpage, 'video URL')
 
         title = self._html_search_regex(
-            [r'<p[^>]+class="title_substrate">([^<]+)</p>', r'<title>([^<]+) - \d+'],
+            (r'class="title_watch"[^>]*><p>([^<]+)<',
+             r'<p[^>]+class="title_substrate">([^<]+)</p>',
+             r'<title>([^<]+) - \d+'),
             webpage, 'title')
 
         thumbnail = self._html_search_regex(
             r'poster="([^"]+)"',
             webpage, 'thumbnail', fatal=False)
 
-        def extract_count(id_, name):
+        def extract_count(id_, name, default=NO_DEFAULT):
             return str_to_int(self._html_search_regex(
                 r'<span[^>]+(?:class|id)="%s"[^>]*>([\d,\.]+)</span>' % id_,
-                webpage, '%s count' % name, fatal=False))
+                webpage, '%s count' % name, default=default, fatal=False))
 
         like_count = extract_count('rate_likes', 'like')
-        dislike_count = extract_count('rate_dislikes', 'dislike')
+        dislike_count = extract_count('rate_dislikes', 'dislike', default=None)
         comment_count = extract_count('comments_count', 'comment')
 
         cats_str = self._search_regex(
-            r'<div[^>]+class="categories_list">(.+?)</div>', webpage, 'categories', fatal=False)
-        categories = [] if not cats_str else re.findall(r'<a title="([^"]+)"', cats_str)
+            r'<div[^>]+class="categories_list">(.+?)</div>',
+            webpage, 'categories', fatal=False)
+        categories = [] if not cats_str else re.findall(
+            r'<a title="([^"]+)"', cats_str)
 
         return {
             'id': video_id,

youtube_dl/extractor/eagleplatform.py

@@ -50,6 +50,14 @@ class EaglePlatformIE(InfoExtractor):
         'skip': 'Georestricted',
     }]
 
+    @staticmethod
+    def _extract_url(webpage):
+        mobj = re.search(
+            r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//.+?\.media\.eagleplatform\.com/index/player\?.+?)\1',
+            webpage)
+        if mobj is not None:
+            return mobj.group('url')
+
     @staticmethod
     def _handle_error(response):
         status = int_or_none(response.get('status', 200))

youtube_dl/extractor/ellentv.py

@@ -6,12 +6,13 @@ import json
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    NO_DEFAULT,
 )
 
 
 class EllenTVIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?(?:ellentv|ellentube)\.com/videos/(?P<id>[a-z0-9_-]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.ellentv.com/videos/0-ipq1gsai/',
         'md5': '4294cf98bc165f218aaa0b89e0fd8042',
         'info_dict': {
@@ -22,24 +23,47 @@ class EllenTVIE(InfoExtractor):
             'timestamp': 1428035648,
             'upload_date': '20150403',
             'uploader_id': 'batchUser',
-        }
-    }
+        },
+    }, {
+        # not available via http://widgets.ellentube.com/
+        'url': 'http://www.ellentv.com/videos/1-szkgu2m2/',
+        'info_dict': {
+            'id': '1_szkgu2m2',
+            'ext': 'flv',
+            'title': "Ellen's Amazingly Talented Audience",
+            'description': 'md5:86ff1e376ff0d717d7171590e273f0a5',
+            'timestamp': 1255140900,
+            'upload_date': '20091010',
+            'uploader_id': 'ellenkaltura@gmail.com',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
+        URLS = ('http://widgets.ellentube.com/videos/%s' % video_id, url)
+
+        for num, url_ in enumerate(URLS, 1):
             webpage = self._download_webpage(
-            'http://widgets.ellentube.com/videos/%s' % video_id,
-            video_id)
+                url_, video_id, fatal=num == len(URLS))
+
+            default = NO_DEFAULT if num == len(URLS) else None
 
             partner_id = self._search_regex(
-            r"var\s+partnerId\s*=\s*'([^']+)", webpage, 'partner id')
+                r"var\s+partnerId\s*=\s*'([^']+)", webpage, 'partner id',
+                default=default)
 
             kaltura_id = self._search_regex(
                 [r'id="kaltura_player_([^"]+)"',
                  r"_wb_entry_id\s*:\s*'([^']+)",
                  r'data-kaltura-entry-id="([^"]+)'],
-            webpage, 'kaltura id')
+                webpage, 'kaltura id', default=default)
+
+            if partner_id and kaltura_id:
+                break
 
         return self.url_result('kaltura:%s:%s' % (partner_id, kaltura_id), 'Kaltura')
 

youtube_dl/extractor/engadget.py

@@ -4,9 +4,10 @@ from .common import InfoExtractor
 
 
 class EngadgetIE(InfoExtractor):
-    _VALID_URL = r'https?://www.engadget.com/video/(?P<id>\d+)'
+    _VALID_URL = r'https?://www.engadget.com/video/(?P<id>[^/?#]+)'
 
-    _TEST = {
+    _TESTS = [{
+        # video with 5min ID
         'url': 'http://www.engadget.com/video/518153925/',
         'md5': 'c6820d4828a5064447a4d9fc73f312c9',
         'info_dict': {
@@ -15,8 +16,12 @@ class EngadgetIE(InfoExtractor):
             'title': 'Samsung Galaxy Tab Pro 8.4 Review',
         },
         'add_ie': ['FiveMin'],
-    }
+    }, {
+        # video with vidible ID
+        'url': 'https://www.engadget.com/video/57a28462134aa15a39f0421a/',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        return self.url_result('5min:%s' % video_id)
+        return self.url_result('aol-video:%s' % video_id)

youtube_dl/extractor/eporner.py

@@ -4,19 +4,23 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
+    encode_base_n,
+    ExtractorError,
+    int_or_none,
     parse_duration,
     str_to_int,
 )
 
 
 class EpornerIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?eporner\.com/hd-porn/(?P<id>\w+)/(?P<display_id>[\w-]+)'
+    _VALID_URL = r'https?://(?:www\.)?eporner\.com/hd-porn/(?P<id>\w+)(?:/(?P<display_id>[\w-]+))?'
     _TESTS = [{
         'url': 'http://www.eporner.com/hd-porn/95008/Infamous-Tiffany-Teen-Strip-Tease-Video/',
         'md5': '39d486f046212d8e1b911c52ab4691f8',
         'info_dict': {
-            'id': '95008',
+            'id': 'qlDUmNsj6VS',
             'display_id': 'Infamous-Tiffany-Teen-Strip-Tease-Video',
             'ext': 'mp4',
             'title': 'Infamous Tiffany Teen Strip Tease Video',
@@ -28,34 +32,72 @@ class EpornerIE(InfoExtractor):
         # New (May 2016) URL layout
         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0/Star-Wars-XXX-Parody/',
         'only_matching': True,
+    }, {
+        'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0',
+        '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')
+        display_id = mobj.group('display_id') or video_id
 
-        webpage = self._download_webpage(url, display_id)
-        title = self._html_search_regex(
-            r'<title>(.*?) - EPORNER', webpage, 'title')
+        webpage, urlh = self._download_webpage_handle(url, display_id)
 
-        redirect_url = 'http://www.eporner.com/config5/%s' % video_id
-        player_code = self._download_webpage(
-            redirect_url, display_id, note='Downloading player config')
+        video_id = self._match_id(compat_str(urlh.geturl()))
 
-        sources = self._search_regex(
-            r'(?s)sources\s*:\s*\[\s*({.+?})\s*\]', player_code, 'sources')
+        hash = self._search_regex(
+            r'hash\s*:\s*["\']([\da-f]{32})', webpage, 'hash')
+
+        title = self._og_search_title(webpage, default=None) or self._html_search_regex(
+            r'<title>(.+?) - EPORNER', webpage, 'title')
+
+        # Reverse engineered from vjs.js
+        def calc_hash(s):
+            return ''.join((encode_base_n(int(s[lb:lb + 8], 16), 36) for lb in range(0, 32, 8)))
+
+        video = self._download_json(
+            'http://www.eporner.com/xhr/video/%s' % video_id,
+            display_id, note='Downloading video JSON',
+            query={
+                'hash': calc_hash(hash),
+                'device': 'generic',
+                'domain': 'www.eporner.com',
+                'fallback': 'false',
+            })
+
+        if video.get('available') is False:
+            raise ExtractorError(
+                '%s said: %s' % (self.IE_NAME, video['message']), expected=True)
+
+        sources = video['sources']
 
         formats = []
-        for video_url, format_id in re.findall(r'file\s*:\s*"([^"]+)",\s*label\s*:\s*"([^"]+)"', sources):
-            fmt = {
-                'url': video_url,
+        for kind, formats_dict in sources.items():
+            if not isinstance(formats_dict, dict):
+                continue
+            for format_id, format_dict in formats_dict.items():
+                if not isinstance(format_dict, dict):
+                    continue
+                src = format_dict.get('src')
+                if not isinstance(src, compat_str) or not src.startswith('http'):
+                    continue
+                if kind == 'hls':
+                    formats.extend(self._extract_m3u8_formats(
+                        src, display_id, 'mp4', entry_protocol='m3u8_native',
+                        m3u8_id=kind, fatal=False))
+                else:
+                    height = int_or_none(self._search_regex(
+                        r'(\d+)[pP]', format_id, 'height', default=None))
+                    fps = int_or_none(self._search_regex(
+                        r'(\d+)fps', format_id, 'fps', default=None))
+
+                    formats.append({
+                        'url': src,
                         'format_id': format_id,
-            }
-            m = re.search(r'^(\d+)', format_id)
-            if m:
-                fmt['height'] = int(m.group(1))
-            formats.append(fmt)
+                        'height': height,
+                        'fps': fps,
+                    })
         self._sort_formats(formats)
 
         duration = parse_duration(self._html_search_meta('duration', webpage))

youtube_dl/extractor/expotv.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
@@ -12,23 +10,22 @@ from ..utils import (
 class ExpoTVIE(InfoExtractor):
     _VALID_URL = r'https?://www\.expotv\.com/videos/[^?#]*/(?P<id>[0-9]+)($|[?#])'
     _TEST = {
-        'url': 'http://www.expotv.com/videos/reviews/1/24/LinneCardscom/17561',
-        'md5': '2985e6d7a392b2f7a05e0ca350fe41d0',
+        'url': 'http://www.expotv.com/videos/reviews/3/40/NYX-Butter-lipstick/667916',
+        'md5': 'fe1d728c3a813ff78f595bc8b7a707a8',
         'info_dict': {
-            'id': '17561',
+            'id': '667916',
             'ext': 'mp4',
-            'upload_date': '20060212',
-            'title': 'My Favorite Online Scrapbook Store',
-            'view_count': int,
-            'description': 'You\'ll find most everything you need at this virtual store front.',
-            'uploader': 'Anna T.',
+            'title': 'NYX Butter Lipstick Little Susie',
+            'description': 'Goes on like butter, but looks better!',
             'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'Stephanie S.',
+            'upload_date': '20150520',
+            'view_count': int,
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
         player_key = self._search_regex(
@@ -66,7 +63,7 @@ class ExpoTVIE(InfoExtractor):
             fatal=False)
         upload_date = unified_strdate(self._search_regex(
             r'<h5>Reviewed on ([0-9/.]+)</h5>', webpage, 'upload date',
-            fatal=False))
+            fatal=False), day_first=False)
 
         return {
             'id': video_id,

youtube_dl/extractor/extractors.py

@@ -20,7 +20,10 @@ from .adobetv import (
     AdobeTVVideoIE,
 )
 from .adultswim import AdultSwimIE
-from .aenetworks import AENetworksIE
+from .aenetworks import (
+    AENetworksIE,
+    HistoryTopicIE,
+)
 from .afreecatv import AfreecaTVIE
 from .aftonbladet import AftonbladetIE
 from .airmozilla import AirMozillaIE
@@ -41,10 +44,10 @@ from .appletrailers import (
     AppleTrailersSectionIE,
 )
 from .archiveorg import ArchiveOrgIE
+from .arkena import ArkenaIE
 from .ard import (
     ARDIE,
     ARDMediathekIE,
-    SportschauIE,
 )
 from .arte import (
     ArteTvIE,
@@ -71,6 +74,8 @@ from .bandcamp import BandcampIE, BandcampAlbumIE
 from .bbc import (
     BBCCoUkIE,
     BBCCoUkArticleIE,
+    BBCCoUkIPlayerPlaylistIE,
+    BBCCoUkPlaylistIE,
     BBCIE,
 )
 from .beeg import BeegIE
@@ -108,6 +113,10 @@ from .camwithher import CamWithHerIE
 from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
 from .canvas import CanvasIE
+from .carambatv import (
+    CarambaTVIE,
+    CarambaTVPageIE,
+)
 from .cbc import (
     CBCIE,
     CBCPlayerIE,
@@ -131,10 +140,11 @@ from .chirbit import (
     ChirbitProfileIE,
 )
 from .cinchcast import CinchcastIE
-from .cliprs import ClipRsIE
 from .clipfish import ClipfishIE
 from .cliphunter import CliphunterIE
+from .cliprs import ClipRsIE
 from .clipsyndicate import ClipsyndicateIE
+from .closertotruth import CloserToTruthIE
 from .cloudy import CloudyIE
 from .clubic import ClubicIE
 from .clyp import ClypIE
@@ -147,7 +157,12 @@ from .cnn import (
 )
 from .coub import CoubIE
 from .collegerama import CollegeRamaIE
-from .comedycentral import ComedyCentralIE, ComedyCentralShowsIE
+from .comedycentral import (
+    ComedyCentralIE,
+    ComedyCentralShortnameIE,
+    ComedyCentralTVIE,
+    ToshIE,
+)
 from .comcarcoff import ComCarCoffIE
 from .commonmistakes import CommonMistakesIE, UnicodeBOMIE
 from .commonprotocols import RtmpIE
@@ -162,6 +177,8 @@ from .crunchyroll import (
 )
 from .cspan import CSpanIE
 from .ctsnews import CtsNewsIE
+from .ctv import CTVIE
+from .ctvnews import CTVNewsIE
 from .cultureunplugged import CultureUnpluggedIE
 from .cwtv import CWTVIE
 from .dailymail import DailyMailIE
@@ -204,6 +221,7 @@ from .dvtv import DVTVIE
 from .dumpert import DumpertIE
 from .defense import DefenseGouvFrIE
 from .discovery import DiscoveryIE
+from .discoverygo import DiscoveryGoIE
 from .dispeak import DigitallySpeakingIE
 from .dropbox import DropboxIE
 from .dw import (
@@ -245,6 +263,7 @@ from .fivemin import FiveMinIE
 from .fivetv import FiveTVIE
 from .fktv import FKTVIE
 from .flickr import FlickrIE
+from .flipagram import FlipagramIE
 from .folketinget import FolketingetIE
 from .footyroom import FootyRoomIE
 from .formula1 import Formula1IE
@@ -253,10 +272,7 @@ from .fox import FOXIE
 from .foxgay import FoxgayIE
 from .foxnews import FoxNewsIE
 from .foxsports import FoxSportsIE
-from .franceculture import (
-    FranceCultureIE,
-    FranceCultureEmissionIE,
-)
+from .franceculture import FranceCultureIE
 from .franceinter import FranceInterIE
 from .francetv import (
     PluzzIE,
@@ -270,8 +286,8 @@ from .freespeech import FreespeechIE
 from .freevideo import FreeVideoIE
 from .funimation import FunimationIE
 from .funnyordie import FunnyOrDieIE
+from .fusion import FusionIE
 from .gameinformer import GameInformerIE
-from .gamekings import GamekingsIE
 from .gameone import (
     GameOneIE,
     GameOnePlaylistIE,
@@ -279,7 +295,6 @@ from .gameone import (
 from .gamersyde import GamersydeIE
 from .gamespot import GameSpotIE
 from .gamestar import GameStarIE
-from .gametrailers import GametrailersIE
 from .gazeta import GazetaIE
 from .gdcvault import GDCVaultIE
 from .generic import GenericIE
@@ -293,7 +308,6 @@ from .globo import (
 )
 from .godtube import GodTubeIE
 from .godtv import GodTVIE
-from .goldenmoustache import GoldenMoustacheIE
 from .golem import GolemIE
 from .googledrive import GoogleDriveIE
 from .googleplus import GooglePlusIE
@@ -308,6 +322,7 @@ from .heise import HeiseIE
 from .hellporno import HellPornoIE
 from .helsinki import HelsinkiIE
 from .hentaistigma import HentaiStigmaIE
+from .hgtv import HGTVIE
 from .historicfilms import HistoricFilmsIE
 from .hitbox import HitboxIE, HitboxLiveIE
 from .hornbunny import HornBunnyIE
@@ -315,6 +330,10 @@ from .hotnewhiphop import HotNewHipHopIE
 from .hotstar import HotStarIE
 from .howcast import HowcastIE
 from .howstuffworks import HowStuffWorksIE
+from .hrti import (
+    HRTiIE,
+    HRTiPlaylistIE,
+)
 from .huffpost import HuffPostIE
 from .hypem import HypemIE
 from .iconosquare import IconosquareIE
@@ -353,6 +372,7 @@ from .jove import JoveIE
 from .jwplatform import JWPlatformIE
 from .jpopsukitv import JpopsukiIE
 from .kaltura import KalturaIE
+from .kamcord import KamcordIE
 from .kanalplay import KanalPlayIE
 from .kankan import KankanIE
 from .karaoketv import KaraoketvIE
@@ -376,6 +396,10 @@ from .kuwo import (
 )
 from .la7 import LA7IE
 from .laola1tv import Laola1TvIE
+from .lcp import (
+    LcpPlayIE,
+    LcpIE,
+)
 from .learnr import LearnrIE
 from .lecture2go import Lecture2GoIE
 from .lemonde import LemondeIE
@@ -417,6 +441,7 @@ from .makerschannel import MakersChannelIE
 from .makertv import MakerTVIE
 from .matchtv import MatchTVIE
 from .mdr import MDRIE
+from .meta import METAIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
 from .mgoon import MgoonIE
@@ -449,10 +474,10 @@ from .motherless import MotherlessIE
 from .motorsport import MotorsportIE
 from .movieclips import MovieClipsIE
 from .moviezine import MoviezineIE
+from .msn import MSNIE
 from .mtv import (
     MTVIE,
     MTVServicesEmbeddedIE,
-    MTVIggyIE,
     MTVDEIE,
 )
 from .muenchentv import MuenchenTVIE
@@ -464,8 +489,9 @@ from .myvi import MyviIE
 from .myvideo import MyVideoIE
 from .myvidster import MyVidsterIE
 from .nationalgeographic import (
+    NationalGeographicVideoIE,
     NationalGeographicIE,
-    NationalGeographicChannelIE,
+    NationalGeographicEpisodeGuideIE,
 )
 from .naver import NaverIE
 from .nba import NBAIE
@@ -475,7 +501,6 @@ from .nbc import (
     NBCNewsIE,
     NBCSportsIE,
     NBCSportsVPlayerIE,
-    MSNBCIE,
 )
 from .ndr import (
     NDRIE,
@@ -503,7 +528,6 @@ from .nextmedia import (
     NextMediaActionNewsIE,
     AppleDailyIE,
 )
-from .nextmovie import NextMovieIE
 from .nfb import NFBIE
 from .nfl import NFLIE
 from .nhl import (
@@ -512,9 +536,15 @@ from .nhl import (
     NHLVideocenterCategoryIE,
     NHLIE,
 )
-from .nick import NickIE
+from .nick import (
+    NickIE,
+    NickDeIE,
+)
 from .niconico import NiconicoIE, NiconicoPlaylistIE
+from .ninecninemedia import NineCNineMediaIE
 from .ninegag import NineGagIE
+from .ninenow import NineNowIE
+from .nintendo import NintendoIE
 from .noco import NocoIE
 from .normalboots import NormalbootsIE
 from .nosvideo import NosVideoIE
@@ -559,8 +589,13 @@ from .nytimes import (
     NYTimesArticleIE,
 )
 from .nuvid import NuvidIE
+from .odatv import OdaTVIE
 from .odnoklassniki import OdnoklassnikiIE
 from .oktoberfesttv import OktoberfestTVIE
+from .onet import (
+    OnetIE,
+    OnetChannelIE,
+)
 from .onionstudios import OnionStudiosIE
 from .ooyala import (
     OoyalaIE,
@@ -599,6 +634,8 @@ from .pluralsight import (
     PluralsightCourseIE,
 )
 from .podomatic import PodomaticIE
+from .pokemon import PokemonIE
+from .polskieradio import PolskieRadioIE
 from .porn91 import Porn91IE
 from .pornhd import PornHdIE
 from .pornhub import (
@@ -622,7 +659,10 @@ from .qqmusic import (
     QQMusicToplistIE,
     QQMusicPlaylistIE,
 )
-from .r7 import R7IE
+from .r7 import (
+    R7IE,
+    R7ArticleIE,
+)
 from .radiocanada import (
     RadioCanadaIE,
     RadioCanadaAudioVideoIE,
@@ -649,16 +689,20 @@ from .revision3 import (
 from .rice import RICEIE
 from .ringtv import RingTVIE
 from .ro220 import Ro220IE
+from .rockstargames import RockstarGamesIE
+from .roosterteeth import RoosterTeethIE
 from .rottentomatoes import RottenTomatoesIE
 from .roxwel import RoxwelIE
+from .rozhlas import RozhlasIE
 from .rtbf import RTBFIE
 from .rte import RteIE, RteRadioIE
 from .rtlnl import RtlNlIE
 from .rtl2 import RTL2IE
 from .rtp import RTPIE
 from .rts import RTSIE
-from .rtve import RTVEALaCartaIE, RTVELiveIE, RTVEInfantilIE
+from .rtve import RTVEALaCartaIE, RTVELiveIE, RTVEInfantilIE, RTVELiveIE, RTVETelevisionIE
 from .rtvnh import RTVNHIE
+from .rudo import RudoIE
 from .ruhd import RUHDIE
 from .ruleporn import RulePornIE
 from .rutube import (
@@ -693,10 +737,12 @@ from .shahid import ShahidIE
 from .shared import SharedIE
 from .sharesix import ShareSixIE
 from .sina import SinaIE
+from .sixplay import SixPlayIE
 from .skynewsarabia import (
     SkyNewsArabiaIE,
     SkyNewsArabiaArticleIE,
 )
+from .skysports import SkySportsIE
 from .slideshare import SlideshareIE
 from .slutload import SlutloadIE
 from .smotri import (
@@ -707,6 +753,7 @@ from .smotri import (
 )
 from .snotr import SnotrIE
 from .sohu import SohuIE
+from .sonyliv import SonyLIVIE
 from .soundcloud import (
     SoundcloudIE,
     SoundcloudSetIE,
@@ -737,6 +784,7 @@ from .sportbox import (
     SportBoxEmbedIE,
 )
 from .sportdeutschland import SportDeutschlandIE
+from .sportschau import SportschauIE
 from .srgssr import (
     SRGSSRIE,
     SRGSSRPlayIE,
@@ -745,6 +793,7 @@ from .srmediathek import SRMediathekIE
 from .ssa import SSAIE
 from .stanfordoc import StanfordOpenClassroomIE
 from .steam import SteamIE
+from .streamable import StreamableIE
 from .streamcloud import StreamcloudIE
 from .streamcz import StreamCZIE
 from .streetvoice import StreetVoiceIE
@@ -760,7 +809,6 @@ from .tagesschau import (
     TagesschauPlayerIE,
     TagesschauIE,
 )
-from .tapely import TapelyIE
 from .tass import TassIE
 from .tdslifeway import TDSLifewayIE
 from .teachertube import (
@@ -777,6 +825,7 @@ from .telecinco import TelecincoIE
 from .telegraaf import TelegraafIE
 from .telemb import TeleMBIE
 from .teletask import TeleTaskIE
+from .telewebion import TelewebionIE
 from .testurl import TestURLIE
 from .tf1 import TF1IE
 from .theintercept import TheInterceptIE
@@ -843,6 +892,7 @@ from .tvc import (
 from .tvigle import TvigleIE
 from .tvland import TVLandIE
 from .tvp import (
+    TVPEmbedIE,
     TVPIE,
     TVPSeriesIE,
 )
@@ -861,6 +911,7 @@ from .twitch import (
     TwitchProfileIE,
     TwitchPastBroadcastsIE,
     TwitchStreamIE,
+    TwitchClipsIE,
 )
 from .twitter import (
     TwitterCardIE,
@@ -874,7 +925,9 @@ from .udemy import (
 from .udn import UDNEmbedIE
 from .digiteka import DigitekaIE
 from .unistra import UnistraIE
+from .uol import UOLIE
 from .urort import UrortIE
+from .urplay import URPlayIE
 from .usatoday import USATodayIE
 from .ustream import UstreamIE, UstreamChannelIE
 from .ustudio import (
@@ -901,6 +954,7 @@ from .vice import (
     ViceIE,
     ViceShowIE,
 )
+from .vidbit import VidbitIE
 from .viddler import ViddlerIE
 from .videodetective import VideoDetectiveIE
 from .videofyme import VideofyMeIE
@@ -949,9 +1003,11 @@ from .viki import (
 from .vk import (
     VKIE,
     VKUserVideosIE,
+    VKWallPostIE,
 )
 from .vlive import VLiveIE
 from .vodlocker import VodlockerIE
+from .vodplatform import VODPlatformIE
 from .voicerepublic import VoiceRepublicIE
 from .voxmedia import VoxMediaIE
 from .vporn import VpornIE
@@ -977,7 +1033,10 @@ from .weiqitv import WeiqiTVIE
 from .wimp import WimpIE
 from .wistia import WistiaIE
 from .worldstarhiphop import WorldStarHipHopIE
-from .wrzuta import WrzutaIE
+from .wrzuta import (
+    WrzutaIE,
+    WrzutaPlaylistIE,
+)
 from .wsj import WSJIE
 from .xbef import XBefIE
 from .xboxclips import XboxClipsIE
@@ -1031,6 +1090,7 @@ from .youtube import (
     YoutubeSearchDateIE,
     YoutubeSearchIE,
     YoutubeSearchURLIE,
+    YoutubeSharedVideoIE,
     YoutubeShowIE,
     YoutubeSubscriptionsIE,
     YoutubeTruncatedIDIE,

youtube_dl/extractor/facebook.py

@@ -27,7 +27,7 @@ class FacebookIE(InfoExtractor):
     _VALID_URL = r'''(?x)
                 (?:
                     https?://
-                        (?:\w+\.)?facebook\.com/
+                        (?:[\w-]+\.)?facebook\.com/
                         (?:[^#]*?\#!/)?
                         (?:
                             (?:
@@ -127,8 +127,26 @@ class FacebookIE(InfoExtractor):
     }, {
         'url': 'https://www.facebook.com/groups/164828000315060/permalink/764967300301124/',
         'only_matching': True,
+    }, {
+        'url': 'https://zh-hk.facebook.com/peoplespower/videos/1135894589806027/',
+        'only_matching': True,
     }]
 
+    @staticmethod
+    def _extract_url(webpage):
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
+        if mobj is not None:
+            return mobj.group('url')
+
+        # Facebook API embed
+        # see https://developers.facebook.com/docs/plugins/embedded-video-player
+        mobj = re.search(r'''(?x)<div[^>]+
+                class=(?P<q1>[\'"])[^\'"]*\bfb-(?:video|post)\b[^\'"]*(?P=q1)[^>]+
+                data-href=(?P<q2>[\'"])(?P<url>(?:https?:)?//(?:www\.)?facebook.com/.+?)(?P=q2)''', webpage)
+        if mobj is not None:
+            return mobj.group('url')
+
     def _login(self):
         (useremail, password) = self._get_login_info()
         if useremail is None:
@@ -204,12 +222,25 @@ class FacebookIE(InfoExtractor):
 
         BEFORE = '{swf.addParam(param[0], param[1]);});'
         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
-        m = re.search(re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER), webpage)
-        if m:
-            swf_params = m.group(1).replace('\\\\', '\\').replace('\\"', '"')
+        PATTERN = re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER)
+
+        for m in re.findall(PATTERN, webpage):
+            swf_params = m.replace('\\\\', '\\').replace('\\"', '"')
             data = dict(json.loads(swf_params))
             params_raw = compat_urllib_parse_unquote(data['params'])
-            video_data = json.loads(params_raw)['video_data']
+            video_data_candidate = json.loads(params_raw)['video_data']
+            for _, f in video_data_candidate.items():
+                if not f:
+                    continue
+                if isinstance(f, dict):
+                    f = [f]
+                if not isinstance(f, list):
+                    continue
+                if f[0].get('video_id') == video_id:
+                    video_data = video_data_candidate
+                    break
+            if video_data:
+                break
 
         def video_data_list2dict(video_data):
             ret = {}
@@ -239,6 +270,8 @@ class FacebookIE(InfoExtractor):
 
         formats = []
         for format_id, f in video_data.items():
+            if f and isinstance(f, dict):
+                f = [f]
             if not f or not isinstance(f, list):
                 continue
             for quality in ('sd', 'hd'):

youtube_dl/extractor/fivemin.py

@@ -1,24 +1,11 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..compat import (
-    compat_parse_qs,
-    compat_urllib_parse_urlencode,
-    compat_urllib_parse_urlparse,
-    compat_urlparse,
-)
-from ..utils import (
-    ExtractorError,
-    parse_duration,
-    replace_extension,
-)
 
 
 class FiveMinIE(InfoExtractor):
     IE_NAME = '5min'
-    _VALID_URL = r'(?:5min:(?P<id>\d+)(?::(?P<sid>\d+))?|https?://[^/]*?5min\.com/Scripts/PlayerSeed\.js\?(?P<query>.*))'
+    _VALID_URL = r'(?:5min:|https?://(?:[^/]*?5min\.com/|delivery\.vidible\.tv/aol)(?:(?:Scripts/PlayerSeed\.js|playerseed/?)?\?.*?playList=)?)(?P<id>\d+)'
 
     _TESTS = [
         {
@@ -29,8 +16,16 @@ class FiveMinIE(InfoExtractor):
                 'id': '518013791',
                 'ext': 'mp4',
                 'title': 'iPad Mini with Retina Display Review',
+                'description': 'iPad mini with Retina Display review',
                 'duration': 177,
+                'uploader': 'engadget',
+                'upload_date': '20131115',
+                'timestamp': 1384515288,
             },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            }
         },
         {
             # From http://on.aol.com/video/how-to-make-a-next-level-fruit-salad-518086247
@@ -44,108 +39,16 @@ class FiveMinIE(InfoExtractor):
             },
             'skip': 'no longer available',
         },
+        {
+            'url': 'http://embed.5min.com/518726732/',
+            'only_matching': True,
+        },
+        {
+            'url': 'http://delivery.vidible.tv/aol?playList=518013791',
+            'only_matching': True,
+        }
     ]
-    _ERRORS = {
-        'ErrorVideoNotExist': 'We\'re sorry, but the video you are trying to watch does not exist.',
-        'ErrorVideoNoLongerAvailable': 'We\'re sorry, but the video you are trying to watch is no longer available.',
-        'ErrorVideoRejected': 'We\'re sorry, but the video you are trying to watch has been removed.',
-        'ErrorVideoUserNotGeo': 'We\'re sorry, but the video you are trying to watch cannot be viewed from your current location.',
-        'ErrorVideoLibraryRestriction': 'We\'re sorry, but the video you are trying to watch is currently unavailable for viewing at this domain.',
-        'ErrorExposurePermission': 'We\'re sorry, but the video you are trying to watch is currently unavailable for viewing at this domain.',
-    }
-    _QUALITIES = {
-        1: {
-            'width': 640,
-            'height': 360,
-        },
-        2: {
-            'width': 854,
-            'height': 480,
-        },
-        4: {
-            'width': 1280,
-            'height': 720,
-        },
-        8: {
-            'width': 1920,
-            'height': 1080,
-        },
-        16: {
-            'width': 640,
-            'height': 360,
-        },
-        32: {
-            'width': 854,
-            'height': 480,
-        },
-        64: {
-            'width': 1280,
-            'height': 720,
-        },
-        128: {
-            'width': 640,
-            'height': 360,
-        },
-    }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        sid = mobj.group('sid')
-
-        if mobj.group('query'):
-            qs = compat_parse_qs(mobj.group('query'))
-            if not qs.get('playList'):
-                raise ExtractorError('Invalid URL', expected=True)
-            video_id = qs['playList'][0]
-            if qs.get('sid'):
-                sid = qs['sid'][0]
-
-        embed_url = 'https://embed.5min.com/playerseed/?playList=%s' % video_id
-        if not sid:
-            embed_page = self._download_webpage(embed_url, video_id,
-                                                'Downloading embed page')
-            sid = self._search_regex(r'sid=(\d+)', embed_page, 'sid')
-
-        response = self._download_json(
-            'https://syn.5min.com/handlers/SenseHandler.ashx?' +
-            compat_urllib_parse_urlencode({
-                'func': 'GetResults',
-                'playlist': video_id,
-                'sid': sid,
-                'isPlayerSeed': 'true',
-                'url': embed_url,
-            }),
-            video_id)
-        if not response['success']:
-            raise ExtractorError(
-                '%s said: %s' % (
-                    self.IE_NAME,
-                    self._ERRORS.get(response['errorMessage'], response['errorMessage'])),
-                expected=True)
-        info = response['binding'][0]
-
-        formats = []
-        parsed_video_url = compat_urllib_parse_urlparse(compat_parse_qs(
-            compat_urllib_parse_urlparse(info['EmbededURL']).query)['videoUrl'][0])
-        for rendition in info['Renditions']:
-            if rendition['RenditionType'] == 'aac' or rendition['RenditionType'] == 'm3u8':
-                continue
-            else:
-                rendition_url = compat_urlparse.urlunparse(parsed_video_url._replace(path=replace_extension(parsed_video_url.path.replace('//', '/%s/' % rendition['ID']), rendition['RenditionType'])))
-                quality = self._QUALITIES.get(rendition['ID'], {})
-                formats.append({
-                    'format_id': '%s-%d' % (rendition['RenditionType'], rendition['ID']),
-                    'url': rendition_url,
-                    'width': quality.get('width'),
-                    'height': quality.get('height'),
-                })
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': info['Title'],
-            'thumbnail': info.get('ThumbURL'),
-            'duration': parse_duration(info.get('Duration')),
-            'formats': formats,
-        }
+        video_id = self._match_id(url)
+        return self.url_result('aol-video:%s' % video_id)

youtube_dl/extractor/flipagram.py

@@ -0,0 +1,115 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    int_or_none,
+    float_or_none,
+    try_get,
+    unified_timestamp,
+)
+
+
+class FlipagramIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?flipagram\.com/f/(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'https://flipagram.com/f/nyvTSJMKId',
+        'md5': '888dcf08b7ea671381f00fab74692755',
+        'info_dict': {
+            'id': 'nyvTSJMKId',
+            'ext': 'mp4',
+            'title': 'Flipagram by sjuria101 featuring Midnight Memories by One Direction',
+            'description': 'md5:d55e32edc55261cae96a41fa85ff630e',
+            'duration': 35.571,
+            'timestamp': 1461244995,
+            'upload_date': '20160421',
+            'uploader': 'kitty juria',
+            'uploader_id': 'sjuria101',
+            'creator': 'kitty juria',
+            'view_count': int,
+            'like_count': int,
+            'repost_count': int,
+            'comment_count': int,
+            'comments': list,
+            'formats': 'mincount:2',
+        },
+    }
+
+    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'window\.reactH2O\s*=\s*({.+});', webpage, 'video data'),
+            video_id)
+
+        flipagram = video_data['flipagram']
+        video = flipagram['video']
+
+        json_ld = self._search_json_ld(webpage, video_id, default={})
+        title = json_ld.get('title') or flipagram['captionText']
+        description = json_ld.get('description') or flipagram.get('captionText')
+
+        formats = [{
+            'url': video['url'],
+            'width': int_or_none(video.get('width')),
+            'height': int_or_none(video.get('height')),
+            'filesize': int_or_none(video_data.get('size')),
+        }]
+
+        preview_url = try_get(
+            flipagram, lambda x: x['music']['track']['previewUrl'], compat_str)
+        if preview_url:
+            formats.append({
+                'url': preview_url,
+                'ext': 'm4a',
+                'vcodec': 'none',
+            })
+
+        self._sort_formats(formats)
+
+        counts = flipagram.get('counts', {})
+        user = flipagram.get('user', {})
+        video_data = flipagram.get('video', {})
+
+        thumbnails = [{
+            'url': self._proto_relative_url(cover['url']),
+            'width': int_or_none(cover.get('width')),
+            'height': int_or_none(cover.get('height')),
+            'filesize': int_or_none(cover.get('size')),
+        } for cover in flipagram.get('covers', []) if cover.get('url')]
+
+        # Note that this only retrieves comments that are initally loaded.
+        # For videos with large amounts of comments, most won't be retrieved.
+        comments = []
+        for comment in video_data.get('comments', {}).get(video_id, {}).get('items', []):
+            text = comment.get('comment')
+            if not text or not isinstance(text, list):
+                continue
+            comments.append({
+                'author': comment.get('user', {}).get('name'),
+                'author_id': comment.get('user', {}).get('username'),
+                'id': comment.get('id'),
+                'text': text[0],
+                'timestamp': unified_timestamp(comment.get('created')),
+            })
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'duration': float_or_none(flipagram.get('duration'), 1000),
+            'thumbnails': thumbnails,
+            'timestamp': unified_timestamp(flipagram.get('iso8601Created')),
+            'uploader': user.get('name'),
+            'uploader_id': user.get('username'),
+            'creator': user.get('name'),
+            'view_count': int_or_none(counts.get('plays')),
+            'like_count': int_or_none(counts.get('likes')),
+            'repost_count': int_or_none(counts.get('reflips')),
+            'comment_count': int_or_none(counts.get('comments')),
+            'comments': comments,
+            'formats': formats,
+        }

youtube_dl/extractor/formula1.py

@@ -5,8 +5,8 @@ from .common import InfoExtractor
 
 
 class Formula1IE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?formula1\.com/content/fom-website/en/video/\d{4}/\d{1,2}/(?P<id>.+?)\.html'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?formula1\.com/(?:content/fom-website/)?en/video/\d{4}/\d{1,2}/(?P<id>.+?)\.html'
+    _TESTS = [{
         'url': 'http://www.formula1.com/content/fom-website/en/video/2016/5/Race_highlights_-_Spain_2016.html',
         'md5': '8c79e54be72078b26b89e0e111c0502b',
         'info_dict': {
@@ -15,7 +15,10 @@ class Formula1IE(InfoExtractor):
             'title': 'Race highlights - Spain 2016',
         },
         'add_ie': ['Ooyala'],
-    }
+    }, {
+        'url': 'http://www.formula1.com/en/video/2016/5/Race_highlights_-_Spain_2016.html',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         display_id = self._match_id(url)

youtube_dl/extractor/fourtube.py

@@ -43,14 +43,14 @@ class FourTubeIE(InfoExtractor):
             'uploadDate', webpage))
         thumbnail = self._html_search_meta('thumbnailUrl', webpage)
         uploader_id = self._html_search_regex(
-            r'<a class="img-avatar" href="[^"]+/channels/([^/"]+)" title="Go to [^"]+ page">',
+            r'<a class="item-to-subscribe" href="[^"]+/channels/([^/"]+)" title="Go to [^"]+ page">',
             webpage, 'uploader id', fatal=False)
         uploader = self._html_search_regex(
-            r'<a class="img-avatar" href="[^"]+/channels/[^/"]+" title="Go to ([^"]+) page">',
+            r'<a class="item-to-subscribe" href="[^"]+/channels/[^/"]+" title="Go to ([^"]+) page">',
             webpage, 'uploader', fatal=False)
 
         categories_html = self._search_regex(
-            r'(?s)><i class="icon icon-tag"></i>\s*Categories / Tags\s*.*?<ul class="list">(.*?)</ul>',
+            r'(?s)><i class="icon icon-tag"></i>\s*Categories / Tags\s*.*?<ul class="[^"]*?list[^"]*?">(.*?)</ul>',
             webpage, 'categories', fatal=False)
         categories = None
         if categories_html:
@@ -59,10 +59,10 @@ class FourTubeIE(InfoExtractor):
                     r'(?s)<li><a.*?>(.*?)</a>', categories_html)]
 
         view_count = str_to_int(self._search_regex(
-            r'<meta itemprop="interactionCount" content="UserPlays:([0-9,]+)">',
+            r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:([0-9,]+)">',
             webpage, 'view count', fatal=False))
         like_count = str_to_int(self._search_regex(
-            r'<meta itemprop="interactionCount" content="UserLikes:([0-9,]+)">',
+            r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserLikes:([0-9,]+)">',
             webpage, 'like count', fatal=False))
         duration = parse_duration(self._html_search_meta('duration', webpage))
 

youtube_dl/extractor/fox.py

@@ -2,7 +2,10 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import smuggle_url
+from ..utils import (
+    smuggle_url,
+    update_url_query,
+)
 
 
 class FOXIE(InfoExtractor):
@@ -29,11 +32,12 @@ class FOXIE(InfoExtractor):
 
         release_url = self._parse_json(self._search_regex(
             r'"fox_pdk_player"\s*:\s*({[^}]+?})', webpage, 'fox_pdk_player'),
-            video_id)['release_url'] + '&switch=http'
+            video_id)['release_url']
 
         return {
             '_type': 'url_transparent',
             'ie_key': 'ThePlatform',
-            'url': smuggle_url(release_url, {'force_smil_url': True}),
+            'url': smuggle_url(update_url_query(
+                release_url, {'switch': 'http'}), {'force_smil_url': True}),
             'id': video_id,
         }

youtube_dl/extractor/foxsports.py

@@ -1,7 +1,10 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import smuggle_url
+from ..utils import (
+    smuggle_url,
+    update_url_query,
+)
 
 
 class FoxSportsIE(InfoExtractor):
@@ -9,11 +12,15 @@ class FoxSportsIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.foxsports.com/video?vid=432609859715',
+        'md5': 'b49050e955bebe32c301972e4012ac17',
         'info_dict': {
-            'id': 'gA0bHB3Ladz3',
-            'ext': 'flv',
+            'id': 'i0qKWsk3qJaM',
+            'ext': 'mp4',
             'title': 'Courtney Lee on going up 2-0 in series vs. Blazers',
             'description': 'Courtney Lee talks about Memphis being focused.',
+            'upload_date': '20150423',
+            'timestamp': 1429761109,
+            'uploader': 'NEWA-FNG-FOXSPORTS',
         },
         'add_ie': ['ThePlatform'],
     }
@@ -28,5 +35,8 @@ class FoxSportsIE(InfoExtractor):
                 r"data-player-config='([^']+)'", webpage, 'data player config'),
             video_id)
 
-        return self.url_result(smuggle_url(
-            config['releaseURL'] + '&manifest=f4m', {'force_smil_url': True}))
+        return self.url_result(smuggle_url(update_url_query(
+            config['releaseURL'], {
+                'mbr': 'true',
+                'switch': 'http',
+            }), {'force_smil_url': True}))