release 2016.02.10

More robust with getting keys and ids from website.

.gitignore

@@ -31,3 +31,5 @@ updates_key.pem
 test/testdata
 .tox
 youtube-dl.zsh
+.idea
+.idea/*

.travis.yml

@@ -2,14 +2,16 @@ language: python
 python:
   - "2.6"
   - "2.7"
+  - "3.2"
   - "3.3"
   - "3.4"
+  - "3.5"
+sudo: false
 script: nosetests test --verbose
 notifications:
   email:
     - filippo.valsorda@gmail.com
     - phihag@phihag.de
-    - jaime.marquinez.ferrandiz+travis@gmail.com
     - yasoob.khld@gmail.com
 #  irc:
 #    channels:

AUTHORS

@@ -92,3 +92,68 @@ Tithen-Firion
 Zack Fernandes
 cryptonaut
 Adrian Kretz
+Mathias Rav
+Petr Kutalek
+Will Glynn
+Max Reimann
+Cédric Luthi
+Thijs Vermeir
+Joel Leclerc
+Christopher Krooss
+Ondřej Caletka
+Dinesh S
+Johan K. Jensen
+Yen Chi Hsuan
+Enam Mijbah Noor
+David Luhmer
+Shaya Goldberg
+Paul Hartmann
+Frans de Jonge
+Robin de Rooij
+Ryan Schmidt
+Leslie P. Polzer
+Duncan Keall
+Alexander Mamay
+Devin J. Pohly
+Eduardo Ferro Aldama
+Jeff Buchbinder
+Amish Bhadeshia
+Joram Schrijver
+Will W.
+Mohammad Teimori Pabandi
+Roman Le Négrate
+Matthias Küch
+Julian Richen
+Ping O.
+Mister Hat
+Peter Ding
+jackyzy823
+George Brighton
+Remita Amine
+Aurélio A. Heckert
+Bernhard Minks
+sceext
+Zach Bruggeman
+Tjark Saul
+slangangular
+Behrouz Abbasi
+ngld
+nyuszika7h
+Shaun Walbridge
+Lee Jenkins
+Anssi Hannula
+Lukáš Lalinský
+Qijiang Fan
+Rémy Léone
+Marco Ferragina
+reiv
+Muratcan Simsek
+Evan Lu
+flatgreen
+Brian Foley
+Vignesh Venkat
+Tom Gijselinck
+Founder Fang
+Andrew Alexeyew
+Saso Bezlaj
+Erwin de Haan

CONTRIBUTING.md

@@ -0,0 +1,155 @@
+**Please include the full output of youtube-dl when run with `-v`**, i.e. add `-v` flag to your command line, copy the **whole** output and post it in the issue body wrapped in \`\`\` for better formatting. It should look similar to this:
+```
+$ youtube-dl -v http://www.youtube.com/watch?v=BaW_jenozKcj
+[debug] System config: []
+[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 2015.12.06
+[debug] Git HEAD: 135392e
+[debug] Python version 2.6.6 - Windows-2003Server-5.2.3790-SP2
+[debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4
+[debug] Proxy map: {}
+...
+```
+**Do not post screenshots of verbose log only plain text is acceptable.**
+
+The output (including the first lines) contains important debugging information. Issues without the full output are often not reproducible and therefore do not get solved in short order, if ever.
+
+Please re-read your issue once again to avoid a couple of common mistakes (you can and should use this as a checklist):
+
+### Is the description of the issue itself sufficient?
+
+We often get issue reports that we cannot really decipher. While in most cases we eventually get the required information after asking back multiple times, this poses an unnecessary drain on our resources. Many contributors, including myself, are also not native speakers, so we may misread some parts.
+
+So please elaborate on what feature you are requesting, or what bug you want to be fixed. Make sure that it's obvious
+
+- What the problem is
+- How it could be fixed
+- How your proposed solution would look like
+
+If your report is shorter than two lines, it is almost certainly missing some of these, which makes it hard for us to respond to it. We're often too polite to close the issue outright, but the missing info makes misinterpretation likely. As a committer myself, I often get frustrated by these issues, since the only possible way for me to move forward on them is to ask for clarification over and over.
+
+For bug reports, this means that your report should contain the *complete* output of youtube-dl when called with the `-v` flag. The error message you get for (most) bugs even says so, but you would not believe how many of our bug reports do not contain this information.
+
+If your server has multiple IPs or you suspect censorship, adding `--call-home` may be a good idea to get more diagnostics. If the error is `ERROR: Unable to extract ...` and you cannot reproduce it from multiple countries, add `--dump-pages` (warning: this will yield a rather large output, redirect it to the file `log.txt` by adding `>log.txt 2>&1` to your command-line) or upload the `.dump` files you get when you add `--write-pages` [somewhere](https://gist.github.com/).
+
+**Site support requests must contain an example URL**. An example URL is a URL you might want to download, like `http://www.youtube.com/watch?v=BaW_jenozKc`. There should be an obvious video present. Except under very special circumstances, the main page of a video service (e.g. `http://www.youtube.com/`) is *not* an example URL.
+
+###  Are you using the latest version?
+
+Before reporting any issue, type `youtube-dl -U`. This should report that you're up-to-date. About 20% of the reports we receive are already fixed, but people are using outdated versions. This goes for feature requests as well.
+
+###  Is the issue already documented?
+
+Make sure that someone has not already opened the issue you're trying to open. Search at the top of the window or browse the [GitHub Issues](https://github.com/rg3/youtube-dl/search?type=Issues) of this repository. If there is an issue, feel free to write something along the lines of "This affects me as well, with version 2015.01.01. Here is some more information on the issue: ...". While some issues may be old, a new post into them often spurs rapid activity.
+
+###  Why are existing options not enough?
+
+Before requesting a new feature, please have a quick peek at [the list of supported options](https://github.com/rg3/youtube-dl/blob/master/README.md#synopsis). Many feature requests are for features that actually exist already! Please, absolutely do show off your work in the issue report and detail how the existing similar options do *not* solve your problem.
+
+###  Is there enough context in your bug report?
+
+People want to solve problems, and often think they do us a favor by breaking down their larger problems (e.g. wanting to skip already downloaded files) to a specific request (e.g. requesting us to look whether the file exists before downloading the info page). However, what often happens is that they break down the problem into two steps: One simple, and one impossible (or extremely complicated one).
+
+We are then presented with a very complicated request when the original problem could be solved far easier, e.g. by recording the downloaded video IDs in a separate file. To avoid this, you must include the greater context where it is non-obvious. In particular, every feature request that does not consist of adding support for a new site should contain a use case scenario that explains in what situation the missing feature would be useful.
+
+###  Does the issue involve one problem, and one problem only?
+
+Some of our users seem to think there is a limit of issues they can or should open. There is no limit of issues they can or should open. While it may seem appealing to be able to dump all your issues into one ticket, that means that someone who solves one of your issues cannot mark the issue as closed. Typically, reporting a bunch of issues leads to the ticket lingering since nobody wants to attack that behemoth, until someone mercifully splits the issue into multiple ones.
+
+In particular, every site support request issue should only pertain to services at one site (generally under a common domain, but always using the same backend technology). Do not request support for vimeo user videos, Whitehouse podcasts, and Google Plus pages in the same issue. Also, make sure that you don't post bug reports alongside feature requests. As a rule of thumb, a feature request does not include outputs of youtube-dl that are not immediately related to the feature at hand. Do not post reports of a network error alongside the request for a new video service.
+
+###  Is anyone going to need the feature?
+
+Only post features that you (or an incapacitated friend you can personally talk to) require. Do not post features because they seem like a good idea. If they are really useful, they will be requested by someone who requires them.
+
+###  Is your question about youtube-dl?
+
+It may sound strange, but some bug reports we receive are completely unrelated to youtube-dl and relate to a different or even the reporter's own application. Please make sure that you are actually using youtube-dl. If you are using a UI for youtube-dl, report the bug to the maintainer of the actual application providing the UI. On the other hand, if your UI for youtube-dl fails in some way you believe is related to youtube-dl, by all means, go ahead and report the bug.
+
+# DEVELOPER INSTRUCTIONS
+
+Most users do not need to build youtube-dl and can [download the builds](http://rg3.github.io/youtube-dl/download.html) or get them from their distribution.
+
+To run youtube-dl as a developer, you don't need to build anything either. Simply execute
+
+    python -m youtube_dl
+
+To run the test, simply invoke your favorite test runner, or execute a test file directly; any of the following work:
+
+    python -m unittest discover
+    python test/test_download.py
+    nosetests
+
+If you want to create a build of youtube-dl yourself, you'll need
+
+* python
+* make
+* pandoc
+* zip
+* nosetests
+
+### Adding support for a new site
+
+If you want to add support for a new site, you can follow this quick list (assuming your service is called `yourextractor`):
+
+1. [Fork this repository](https://github.com/rg3/youtube-dl/fork)
+2. Check out the source code with `git clone git@github.com:YOUR_GITHUB_USERNAME/youtube-dl.git`
+3. Start a new git branch with `cd youtube-dl; git checkout -b yourextractor`
+4. Start with this simple template and save it to `youtube_dl/extractor/yourextractor.py`:
+    ```python
+    # coding: utf-8
+    from __future__ import unicode_literals
+
+    from .common import InfoExtractor
+
+
+    class YourExtractorIE(InfoExtractor):
+        _VALID_URL = r'https?://(?:www\.)?yourextractor\.com/watch/(?P<id>[0-9]+)'
+        _TEST = {
+            'url': 'http://yourextractor.com/watch/42',
+            'md5': 'TODO: md5 sum of the first 10241 bytes of the video file (use --test)',
+            'info_dict': {
+                'id': '42',
+                'ext': 'mp4',
+                'title': 'Video title goes here',
+                'thumbnail': 're:^https?://.*\.jpg$',
+                # TODO more properties, either as:
+                # * A value
+                # * MD5 checksum; start the string with md5:
+                # * A regular expression; start the string with re:
+                # * Any Python type (for example int or float)
+            }
+        }
+
+        def _real_extract(self, url):
+            video_id = self._match_id(url)
+            webpage = self._download_webpage(url, video_id)
+
+            # TODO more code goes here, for example ...
+            title = self._html_search_regex(r'<h1>(.+?)</h1>', webpage, 'title')
+
+            return {
+                'id': video_id,
+                'title': title,
+                'description': self._og_search_description(webpage),
+                'uploader': self._search_regex(r'<div[^>]+id="uploader"[^>]*>([^<]+)<', webpage, 'uploader', fatal=False),
+                # TODO more properties (see youtube_dl/extractor/common.py)
+            }
+    ```
+5. Add an import in [`youtube_dl/extractor/__init__.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/__init__.py).
+6. Run `python test/test_download.py TestDownload.test_YourExtractor`. This *should fail* at first, but you can continually re-run it until you're done. 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/master/youtube_dl/extractor/common.py#L62-L200). Add tests and code for as many as you want.
+8. If you can, check the code with [flake8](https://pypi.python.org/pypi/flake8).
+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/__init__.py
+        $ git add youtube_dl/extractor/yourextractor.py
+        $ git commit -m '[yourextractor] Add new extractor'
+        $ git push origin yourextractor
+
+10. Finally, [create a pull request](https://help.github.com/articles/creating-a-pull-request). We'll then review and merge it.
+
+In any case, thank you very much for your contributions!
+

Makefile

@@ -1,10 +1,8 @@
-all: youtube-dl README.md README.txt youtube-dl.1 youtube-dl.bash-completion youtube-dl.zsh youtube-dl.fish
+all: youtube-dl README.md CONTRIBUTING.md README.txt youtube-dl.1 youtube-dl.bash-completion youtube-dl.zsh youtube-dl.fish supportedsites
 
 clean:
-	rm -rf youtube-dl.1.temp.md youtube-dl.1 youtube-dl.bash-completion README.txt MANIFEST build/ dist/ .coverage cover/ youtube-dl.tar.gz youtube-dl.zsh youtube-dl.fish *.dump *.part
-
-cleanall: clean
-	rm -f youtube-dl youtube-dl.exe
+	rm -rf youtube-dl.1.temp.md youtube-dl.1 youtube-dl.bash-completion README.txt MANIFEST build/ dist/ .coverage cover/ youtube-dl.tar.gz youtube-dl.zsh youtube-dl.fish *.dump *.part *.info.json *.mp4 *.flv *.mp3 *.avi CONTRIBUTING.md.tmp youtube-dl youtube-dl.exe
+	find . -name "*.pyc" -delete
 
 PREFIX ?= /usr/local
 BINDIR ?= $(PREFIX)/bin
@@ -35,13 +33,22 @@ install: youtube-dl youtube-dl.1 youtube-dl.bash-completion youtube-dl.zsh youtu
 	install -d $(DESTDIR)$(SYSCONFDIR)/fish/completions
 	install -m 644 youtube-dl.fish $(DESTDIR)$(SYSCONFDIR)/fish/completions/youtube-dl.fish
 
+codetest:
+	flake8 .
+
 test:
 	#nosetests --with-coverage --cover-package=youtube_dl --cover-html --verbose --processes 4 test
 	nosetests --verbose test
+	$(MAKE) codetest
+
+ot: offlinetest
+
+offlinetest: codetest
+	nosetests --verbose test --exclude test_download.py --exclude test_age_restriction.py --exclude test_subtitles.py --exclude test_write_annotations.py --exclude test_youtube_lists.py
 
 tar: youtube-dl.tar.gz
 
-.PHONY: all clean install test tar bash-completion pypi-files zsh-completion fish-completion
+.PHONY: all clean install test tar bash-completion pypi-files zsh-completion fish-completion ot offlinetest codetest supportedsites
 
 pypi-files: youtube-dl.bash-completion README.txt youtube-dl.1 youtube-dl.fish
 
@@ -54,28 +61,34 @@ youtube-dl: youtube_dl/*.py youtube_dl/*/*.py
 	chmod a+x youtube-dl
 
 README.md: youtube_dl/*.py youtube_dl/*/*.py
-	COLUMNS=80 python -m youtube_dl --help | python devscripts/make_readme.py
+	COLUMNS=80 $(PYTHON) youtube_dl/__main__.py --help | $(PYTHON) devscripts/make_readme.py
+
+CONTRIBUTING.md: README.md
+	$(PYTHON) devscripts/make_contributing.py README.md CONTRIBUTING.md
+
+supportedsites:
+	$(PYTHON) devscripts/make_supportedsites.py docs/supportedsites.md
 
 README.txt: README.md
 	pandoc -f markdown -t plain README.md -o README.txt
 
 youtube-dl.1: README.md
-	python devscripts/prepare_manpage.py >youtube-dl.1.temp.md
+	$(PYTHON) devscripts/prepare_manpage.py >youtube-dl.1.temp.md
 	pandoc -s -f markdown -t man youtube-dl.1.temp.md -o youtube-dl.1
 	rm -f youtube-dl.1.temp.md
 
 youtube-dl.bash-completion: youtube_dl/*.py youtube_dl/*/*.py devscripts/bash-completion.in
-	python devscripts/bash-completion.py
+	$(PYTHON) devscripts/bash-completion.py
 
 bash-completion: youtube-dl.bash-completion
 
 youtube-dl.zsh: youtube_dl/*.py youtube_dl/*/*.py devscripts/zsh-completion.in
-	python devscripts/zsh-completion.py
+	$(PYTHON) devscripts/zsh-completion.py
 
 zsh-completion: youtube-dl.zsh
 
 youtube-dl.fish: youtube_dl/*.py youtube_dl/*/*.py devscripts/fish-completion.in
-	python devscripts/fish-completion.py
+	$(PYTHON) devscripts/fish-completion.py
 
 fish-completion: youtube-dl.fish
 

devscripts/bash-completion.py

@@ -5,7 +5,7 @@ import os
 from os.path import dirname as dirn
 import sys
 
-sys.path.append(dirn(dirn((os.path.abspath(__file__)))))
+sys.path.insert(0, dirn(dirn((os.path.abspath(__file__)))))
 import youtube_dl
 
 BASH_COMPLETION_FILE = "youtube-dl.bash-completion"

devscripts/check-porn.py

@@ -28,7 +28,7 @@ for test in get_testcases():
     if METHOD == 'EURISTIC':
         try:
             webpage = compat_urllib_request.urlopen(test['url'], timeout=10).read()
-        except:
+        except Exception:
             print('\nFail: {0}'.format(test['name']))
             continue
 
@@ -45,12 +45,12 @@ for test in get_testcases():
 
         RESULT = ('.' + domain + '\n' in LIST or '\n' + domain + '\n' in LIST)
 
-    if RESULT and ('info_dict' not in test or 'age_limit' not in test['info_dict']
-                   or test['info_dict']['age_limit'] != 18):
+    if RESULT and ('info_dict' not in test or 'age_limit' not in test['info_dict'] or
+                   test['info_dict']['age_limit'] != 18):
         print('\nPotential missing age_limit check: {0}'.format(test['name']))
 
-    elif not RESULT and ('info_dict' in test and 'age_limit' in test['info_dict']
-                         and test['info_dict']['age_limit'] == 18):
+    elif not RESULT and ('info_dict' in test and 'age_limit' in test['info_dict'] and
+                         test['info_dict']['age_limit'] == 18):
         print('\nPotential false negative: {0}'.format(test['name']))
 
     else:

devscripts/fish-completion.py

@@ -6,7 +6,7 @@ import os
 from os.path import dirname as dirn
 import sys
 
-sys.path.append(dirn(dirn((os.path.abspath(__file__)))))
+sys.path.insert(0, dirn(dirn((os.path.abspath(__file__)))))
 import youtube_dl
 from youtube_dl.utils import shell_quote
 

devscripts/generate_aes_testdata.py

@@ -0,0 +1,42 @@
+from __future__ import unicode_literals
+
+import codecs
+import subprocess
+
+import os
+import sys
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl.utils import intlist_to_bytes
+from youtube_dl.aes import aes_encrypt, key_expansion
+
+secret_msg = b'Secret message goes here'
+
+
+def hex_str(int_list):
+    return codecs.encode(intlist_to_bytes(int_list), 'hex')
+
+
+def openssl_encode(algo, key, iv):
+    cmd = ['openssl', 'enc', '-e', '-' + algo, '-K', hex_str(key), '-iv', hex_str(iv)]
+    prog = subprocess.Popen(cmd, stdin=subprocess.PIPE, stdout=subprocess.PIPE)
+    out, _ = prog.communicate(secret_msg)
+    return out
+
+iv = key = [0x20, 0x15] + 14 * [0]
+
+r = openssl_encode('aes-128-cbc', key, iv)
+print('aes_cbc_decrypt')
+print(repr(r))
+
+password = key
+new_key = aes_encrypt(password, key_expansion(password))
+r = openssl_encode('aes-128-ctr', new_key, iv)
+print('aes_decrypt_text 16')
+print(repr(r))
+
+password = key + 16 * [0]
+new_key = aes_encrypt(password, key_expansion(password)) * (32 // 16)
+r = openssl_encode('aes-256-ctr', new_key, iv)
+print('aes_decrypt_text 32')
+print(repr(r))

devscripts/gh-pages/update-copyright.py

@@ -5,7 +5,7 @@ from __future__ import with_statement, unicode_literals
 
 import datetime
 import glob
-import io  # For Python 2 compatibilty
+import io  # For Python 2 compatibility
 import os
 import re
 

devscripts/gh-pages/update-sites.py

@@ -6,7 +6,7 @@ import os
 import textwrap
 
 # We must be able to import youtube_dl
-sys.path.append(os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))))
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))))
 
 import youtube_dl
 
@@ -16,7 +16,7 @@ def main():
         template = tmplf.read()
 
     ie_htmls = []
-    for ie in sorted(youtube_dl.gen_extractors(), key=lambda i: i.IE_NAME.lower()):
+    for ie in youtube_dl.list_extractors(age_limit=None):
         ie_html = '<b>{}</b>'.format(ie.IE_NAME)
         ie_desc = getattr(ie, 'IE_DESC', None)
         if ie_desc is False:

devscripts/make_contributing.py

@@ -0,0 +1,32 @@
+#!/usr/bin/env python
+from __future__ import unicode_literals
+
+import io
+import optparse
+import re
+
+
+def main():
+    parser = optparse.OptionParser(usage='%prog INFILE OUTFILE')
+    options, args = parser.parse_args()
+    if len(args) != 2:
+        parser.error('Expected an input and an output filename')
+
+    infile, outfile = args
+
+    with io.open(infile, encoding='utf-8') as inf:
+        readme = inf.read()
+
+    bug_text = re.search(
+        r'(?s)#\s*BUGS\s*[^\n]*\s*(.*?)#\s*COPYRIGHT', readme).group(1)
+    dev_text = re.search(
+        r'(?s)(#\s*DEVELOPER INSTRUCTIONS.*?)#\s*EMBEDDING YOUTUBE-DL',
+        readme).group(1)
+
+    out = bug_text + dev_text
+
+    with io.open(outfile, 'w', encoding='utf-8') as outf:
+        outf.write(out)
+
+if __name__ == '__main__':
+    main()

devscripts/make_supportedsites.py

@@ -0,0 +1,45 @@
+#!/usr/bin/env python
+from __future__ import unicode_literals
+
+import io
+import optparse
+import os
+import sys
+
+
+# Import youtube_dl
+ROOT_DIR = os.path.join(os.path.dirname(__file__), '..')
+sys.path.insert(0, ROOT_DIR)
+import youtube_dl
+
+
+def main():
+    parser = optparse.OptionParser(usage='%prog OUTFILE.md')
+    options, args = parser.parse_args()
+    if len(args) != 1:
+        parser.error('Expected an output filename')
+
+    outfile, = args
+
+    def gen_ies_md(ies):
+        for ie in ies:
+            ie_md = '**{0}**'.format(ie.IE_NAME)
+            ie_desc = getattr(ie, 'IE_DESC', None)
+            if ie_desc is False:
+                continue
+            if ie_desc is not None:
+                ie_md += ': {0}'.format(ie.IE_DESC)
+            if not ie.working():
+                ie_md += ' (Currently broken)'
+            yield ie_md
+
+    ies = sorted(youtube_dl.gen_extractors(), key=lambda i: i.IE_NAME.lower())
+    out = '# Supported sites\n' + ''.join(
+        ' - ' + md + '\n'
+        for md in gen_ies_md(ies))
+
+    with io.open(outfile, 'w', encoding='utf-8') as outf:
+        outf.write(out)
+
+if __name__ == '__main__':
+    main()

devscripts/prepare_manpage.py

@@ -8,13 +8,55 @@ import re
 ROOT_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 README_FILE = os.path.join(ROOT_DIR, 'README.md')
 
+
+def filter_options(readme):
+    ret = ''
+    in_options = False
+    for line in readme.split('\n'):
+        if line.startswith('# '):
+            if line[2:].startswith('OPTIONS'):
+                in_options = True
+            else:
+                in_options = False
+
+        if in_options:
+            if line.lstrip().startswith('-'):
+                option, description = re.split(r'\s{2,}', line.lstrip())
+                split_option = option.split(' ')
+
+                if not split_option[-1].startswith('-'):  # metavar
+                    option = ' '.join(split_option[:-1] + ['*%s*' % split_option[-1]])
+
+                # Pandoc's definition_lists. See http://pandoc.org/README.html
+                # for more information.
+                ret += '\n%s\n:   %s\n' % (option, description)
+            else:
+                ret += line.lstrip() + '\n'
+        else:
+            ret += line + '\n'
+
+    return ret
+
 with io.open(README_FILE, encoding='utf-8') as f:
     readme = f.read()
 
-PREFIX = '%YOUTUBE-DL(1)\n\n# NAME\n'
-readme = re.sub(r'(?s)# INSTALLATION.*?(?=# DESCRIPTION)', '', readme)
+PREFIX = '''%YOUTUBE-DL(1)
+
+# NAME
+
+youtube\-dl \- download videos from youtube.com or other video platforms
+
+# SYNOPSIS
+
+**youtube-dl** \[OPTIONS\] URL [URL...]
+
+'''
+readme = re.sub(r'(?s)^.*?(?=# DESCRIPTION)', '', readme)
+readme = re.sub(r'\s+youtube-dl \[OPTIONS\] URL \[URL\.\.\.\]', '', readme)
 readme = PREFIX + readme
 
+readme = filter_options(readme)
+
 if sys.version_info < (3, 0):
     print(readme.encode('utf-8'))
 else:

devscripts/release.sh

@@ -35,7 +35,7 @@ if [ ! -z "$useless_files" ]; then echo "ERROR: Non-.py files in youtube_dl: $us
 if [ ! -f "updates_key.pem" ]; then echo 'ERROR: updates_key.pem missing'; exit 1; fi
 
 /bin/echo -e "\n### First of all, testing..."
-make cleanall
+make clean
 if $skip_tests ; then
     echo 'SKIPPING TESTS'
 else
@@ -45,9 +45,9 @@ fi
 /bin/echo -e "\n### Changing version in version.py..."
 sed -i "s/__version__ = '.*'/__version__ = '$version'/" youtube_dl/version.py
 
-/bin/echo -e "\n### Committing README.md and youtube_dl/version.py..."
-make README.md
-git add README.md youtube_dl/version.py
+/bin/echo -e "\n### Committing documentation and youtube_dl/version.py..."
+make README.md CONTRIBUTING.md supportedsites
+git add README.md CONTRIBUTING.md docs/supportedsites.md youtube_dl/version.py
 git commit -m "release $version"
 
 /bin/echo -e "\n### Now tagging, signing and pushing..."

devscripts/zsh-completion.py

@@ -5,7 +5,7 @@ import os
 from os.path import dirname as dirn
 import sys
 
-sys.path.append(dirn(dirn((os.path.abspath(__file__)))))
+sys.path.insert(0, dirn(dirn((os.path.abspath(__file__)))))
 import youtube_dl
 
 ZSH_COMPLETION_FILE = "youtube-dl.zsh"

docs/supportedsites.md

@@ -0,0 +1,779 @@
+# Supported sites
+ - **1tv**: Первый канал
+ - **1up.com**
+ - **20min**
+ - **220.ro**
+ - **22tracks:genre**
+ - **22tracks:track**
+ - **24video**
+ - **3sat**
+ - **4tube**
+ - **56.com**
+ - **5min**
+ - **8tracks**
+ - **91porn**
+ - **9gag**
+ - **abc.net.au**
+ - **Abc7News**
+ - **AcademicEarth:Course**
+ - **acast**
+ - **acast:channel**
+ - **AddAnime**
+ - **AdobeTV**
+ - **AdobeTVChannel**
+ - **AdobeTVShow**
+ - **AdobeTVVideo**
+ - **AdultSwim**
+ - **aenetworks**: A+E Networks: A&E, Lifetime, History.com, FYI Network
+ - **Aftonbladet**
+ - **AirMozilla**
+ - **AlJazeera**
+ - **Allocine**
+ - **AlphaPorno**
+ - **anitube.se**
+ - **AnySex**
+ - **Aparat**
+ - **AppleConnect**
+ - **AppleDaily**: 臺灣蘋果日報
+ - **appletrailers**
+ - **appletrailers:section**
+ - **archive.org**: archive.org videos
+ - **ARD**
+ - **ARD:mediathek**: Saarländischer Rundfunk
+ - **ARD:mediathek**
+ - **arte.tv**
+ - **arte.tv:+7**
+ - **arte.tv:cinema**
+ - **arte.tv:concert**
+ - **arte.tv:creative**
+ - **arte.tv:ddc**
+ - **arte.tv:embed**
+ - **arte.tv:future**
+ - **AtresPlayer**
+ - **ATTTechChannel**
+ - **AudiMedia**
+ - **audiomack**
+ - **audiomack:album**
+ - **Azubu**
+ - **AzubuLive**
+ - **BaiduVideo**: 百度视频
+ - **bambuser**
+ - **bambuser:channel**
+ - **Bandcamp**
+ - **Bandcamp:album**
+ - **bbc**: BBC
+ - **bbc.co.uk**: BBC iPlayer
+ - **bbc.co.uk:article**: BBC articles
+ - **BeatportPro**
+ - **Beeg**
+ - **BehindKink**
+ - **Bet**
+ - **Bigflix**
+ - **Bild**: Bild.de
+ - **BiliBili**
+ - **BleacherReport**
+ - **BleacherReportCMS**
+ - **blinkx**
+ - **Bloomberg**
+ - **Bpb**: Bundeszentrale für politische Bildung
+ - **BR**: Bayerischer Rundfunk Mediathek
+ - **Break**
+ - **brightcove:legacy**
+ - **brightcove:new**
+ - **bt:article**: Bergens Tidende Articles
+ - **bt:vestlendingen**: Bergens Tidende - Vestlendingen
+ - **BuzzFeed**
+ - **BYUtv**
+ - **Camdemy**
+ - **CamdemyFolder**
+ - **canalc2.tv**
+ - **Canalplus**: canalplus.fr, piwiplus.fr and d8.tv
+ - **Canvas**
+ - **CBS**
+ - **CBSNews**: CBS News
+ - **CBSNewsLiveVideo**: CBS News Live Videos
+ - **CBSSports**
+ - **CeskaTelevize**
+ - **channel9**: Channel 9
+ - **Chaturbate**
+ - **Chilloutzone**
+ - **chirbit**
+ - **chirbit:profile**
+ - **Cinchcast**
+ - **Cinemassacre**
+ - **Clipfish**
+ - **cliphunter**
+ - **Clipsyndicate**
+ - **cloudtime**: CloudTime
+ - **Cloudy**
+ - **Clubic**
+ - **Clyp**
+ - **cmt.com**
+ - **CNET**
+ - **CNN**
+ - **CNNArticle**
+ - **CNNBlogs**
+ - **CollegeHumor**
+ - **CollegeRama**
+ - **ComCarCoff**
+ - **ComedyCentral**
+ - **ComedyCentralShows**: The Daily Show / The Colbert Report
+ - **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
+ - **Cracked**
+ - **Criterion**
+ - **CrooksAndLiars**
+ - **Crunchyroll**
+ - **crunchyroll:playlist**
+ - **CSpan**: C-SPAN
+ - **CtsNews**: 華視新聞
+ - **culturebox.francetvinfo.fr**
+ - **CultureUnplugged**
+ - **CWTV**
+ - **dailymotion**
+ - **dailymotion:playlist**
+ - **dailymotion:user**
+ - **DailymotionCloud**
+ - **daum.net**
+ - **daum.net:clip**
+ - **daum.net:playlist**
+ - **daum.net:user**
+ - **DBTV**
+ - **DCN**
+ - **dcn:live**
+ - **dcn:season**
+ - **dcn:video**
+ - **DctpTv**
+ - **DeezerPlaylist**
+ - **defense.gouv.fr**
+ - **democracynow**
+ - **DHM**: Filmarchiv - Deutsches Historisches Museum
+ - **Digiteka**
+ - **Discovery**
+ - **Dotsub**
+ - **DouyuTV**: 斗鱼
+ - **DPlay**
+ - **dramafever**
+ - **dramafever:series**
+ - **DRBonanza**
+ - **Dropbox**
+ - **DrTuber**
+ - **DRTV**
+ - **Dump**
+ - **Dumpert**
+ - **dvtv**: http://video.aktualne.cz/
+ - **EaglePlatform**
+ - **EbaumsWorld**
+ - **EchoMsk**
+ - **eHow**
+ - **Einthusan**
+ - **eitb.tv**
+ - **EllenTV**
+ - **EllenTV:clips**
+ - **ElPais**: El País
+ - **Embedly**
+ - **EMPFlix**
+ - **Engadget**
+ - **Eporner**
+ - **EroProfile**
+ - **Escapist**
+ - **ESPN**
+ - **EsriVideo**
+ - **Europa**
+ - **EveryonesMixtape**
+ - **exfm**: ex.fm
+ - **ExpoTV**
+ - **ExtremeTube**
+ - **facebook**
+ - **facebook:post**
+ - **faz.net**
+ - **fc2**
+ - **Fczenit**
+ - **fernsehkritik.tv**
+ - **Firstpost**
+ - **FiveTV**
+ - **Flickr**
+ - **Folketinget**: Folketinget (ft.dk; Danish parliament)
+ - **FootyRoom**
+ - **FOX**
+ - **Foxgay**
+ - **FoxNews**: Fox News and Fox Business Video
+ - **FoxSports**
+ - **france2.fr:generation-quoi**
+ - **FranceCulture**
+ - **FranceCultureEmission**
+ - **FranceInter**
+ - **francetv**: France 2, 3, 4, 5 and Ô
+ - **francetvinfo.fr**
+ - **Freesound**
+ - **freespeech.org**
+ - **FreeVideo**
+ - **Funimation**
+ - **FunnyOrDie**
+ - **GameInformer**
+ - **Gamekings**
+ - **GameOne**
+ - **gameone:playlist**
+ - **Gamersyde**
+ - **GameSpot**
+ - **GameStar**
+ - **Gametrailers**
+ - **Gazeta**
+ - **GDCVault**
+ - **generic**: Generic downloader that works on some sites
+ - **Gfycat**
+ - **GiantBomb**
+ - **Giga**
+ - **Glide**: Glide mobile video messages (glide.me)
+ - **Globo**
+ - **GloboArticle**
+ - **GodTube**
+ - **GoldenMoustache**
+ - **Golem**
+ - **GoogleDrive**
+ - **Goshgay**
+ - **GPUTechConf**
+ - **Groupon**
+ - **Hark**
+ - **HearThisAt**
+ - **Heise**
+ - **HellPorno**
+ - **Helsinki**: helsinki.fi
+ - **HentaiStigma**
+ - **HistoricFilms**
+ - **hitbox**
+ - **hitbox:live**
+ - **HornBunny**
+ - **HotNewHipHop**
+ - **HotStar**
+ - **Howcast**
+ - **HowStuffWorks**
+ - **HuffPost**: Huffington Post
+ - **Hypem**
+ - **Iconosquare**
+ - **ign.com**
+ - **imdb**: Internet Movie Database trailers
+ - **imdb:list**: Internet Movie Database lists
+ - **Imgur**
+ - **ImgurAlbum**
+ - **Ina**
+ - **Indavideo**
+ - **IndavideoEmbed**
+ - **InfoQ**
+ - **Instagram**
+ - **instagram:user**: Instagram user profile
+ - **InternetVideoArchive**
+ - **IPrima**
+ - **iqiyi**: 爱奇艺
+ - **Ir90Tv**
+ - **ivi**: ivi.ru
+ - **ivi:compilation**: ivi.ru compilations
+ - **ivideon**: Ivideon TV
+ - **Izlesene**
+ - **JadoreCettePub**
+ - **JeuxVideo**
+ - **Jove**
+ - **jpopsuki.tv**
+ - **JWPlatform**
+ - **Kaltura**
+ - **KanalPlay**: Kanal 5/9/11 Play
+ - **Kankan**
+ - **Karaoketv**
+ - **KarriereVideos**
+ - **keek**
+ - **KeezMovies**
+ - **KhanAcademy**
+ - **KickStarter**
+ - **KonserthusetPlay**
+ - **kontrtube**: KontrTube.ru - Труба зовёт
+ - **KrasView**: Красвью
+ - **Ku6**
+ - **kuwo:album**: 酷我音乐 - 专辑
+ - **kuwo:category**: 酷我音乐 - 分类
+ - **kuwo:chart**: 酷我音乐 - 排行榜
+ - **kuwo:mv**: 酷我音乐 - MV
+ - **kuwo:singer**: 酷我音乐 - 歌手
+ - **kuwo:song**: 酷我音乐
+ - **la7.tv**
+ - **Laola1Tv**
+ - **Lecture2Go**
+ - **Lemonde**
+ - **Letv**: 乐视网
+ - **LetvCloud**: 乐视云
+ - **LetvPlaylist**
+ - **LetvTv**
+ - **Libsyn**
+ - **life:embed**
+ - **lifenews**: LIFE | NEWS
+ - **limelight**
+ - **limelight:channel**
+ - **limelight:channel_list**
+ - **LiveLeak**
+ - **livestream**
+ - **livestream:original**
+ - **LnkGo**
+ - **LoveHomePorn**
+ - **lrt.lt**
+ - **lynda**: lynda.com videos
+ - **lynda:course**: lynda.com online courses
+ - **m6**
+ - **macgamestore**: MacGameStore trailers
+ - **mailru**: Видео@Mail.Ru
+ - **MakerTV**
+ - **Malemotion**
+ - **MatchTV**
+ - **MDR**: MDR.DE and KiKA
+ - **media.ccc.de**
+ - **metacafe**
+ - **Metacritic**
+ - **Mgoon**
+ - **Minhateca**
+ - **MinistryGrid**
+ - **miomio.tv**
+ - **MiTele**: mitele.es
+ - **mixcloud**
+ - **MLB**
+ - **MoeVideo**: LetitBit video services: moevideo.net, playreplay.net and videochart.net
+ - **Mofosex**
+ - **Mojvideo**
+ - **Moniker**: allmyvideos.net and vidspot.net
+ - **mooshare**: Mooshare.biz
+ - **Morningstar**: morningstar.com
+ - **Motherless**
+ - **Motorsport**: motorsport.com
+ - **MovieClips**
+ - **MovieFap**
+ - **Moviezine**
+ - **MPORA**
+ - **MSNBC**
+ - **MTV**
+ - **mtv.de**
+ - **mtviggy.com**
+ - **mtvservices:embedded**
+ - **MuenchenTV**: münchen.tv
+ - **MusicPlayOn**
+ - **muzu.tv**
+ - **Mwave**
+ - **MySpace**
+ - **MySpace:album**
+ - **MySpass**
+ - **Myvi**
+ - **myvideo**
+ - **MyVidster**
+ - **n-tv.de**
+ - **NationalGeographic**
+ - **Naver**
+ - **NBA**
+ - **NBC**
+ - **NBCNews**
+ - **NBCSports**
+ - **NBCSportsVPlayer**
+ - **ndr**: NDR.de - Norddeutscher Rundfunk
+ - **ndr:embed**
+ - **ndr:embed:base**
+ - **NDTV**
+ - **NerdCubedFeed**
+ - **Nerdist**
+ - **netease:album**: 网易云音乐 - 专辑
+ - **netease:djradio**: 网易云音乐 - 电台
+ - **netease:mv**: 网易云音乐 - MV
+ - **netease:playlist**: 网易云音乐 - 歌单
+ - **netease:program**: 网易云音乐 - 电台节目
+ - **netease:singer**: 网易云音乐 - 歌手
+ - **netease:song**: 网易云音乐
+ - **Netzkino**
+ - **Newgrounds**
+ - **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**
+ - **nhl.com:news**: NHL news
+ - **nhl.com:videocenter**: NHL videocenter category
+ - **nick.com**
+ - **niconico**: ニコニコ動画
+ - **NiconicoPlaylist**
+ - **njoy**: N-JOY
+ - **njoy:embed**
+ - **Noco**
+ - **Normalboots**
+ - **NosVideo**
+ - **Nova**: TN.cz, Prásk.tv, Nova.cz, Novaplus.cz, FANDA.tv, Krásná.cz and Doma.cz
+ - **novamov**: NovaMov
+ - **nowness**
+ - **nowness:playlist**
+ - **nowness:series**
+ - **NowTV** (Currently broken)
+ - **NowTVList**
+ - **nowvideo**: NowVideo
+ - **npo**: npo.nl and ntr.nl
+ - **npo.nl:live**
+ - **npo.nl:radio**
+ - **npo.nl:radio:fragment**
+ - **Npr**
+ - **NRK**
+ - **NRKPlaylist**
+ - **NRKTV**: NRK TV and NRK Radio
+ - **ntv.ru**
+ - **Nuvid**
+ - **NYTimes**
+ - **NYTimesArticle**
+ - **ocw.mit.edu**
+ - **Odnoklassniki**
+ - **OktoberfestTV**
+ - **on.aol.com**
+ - **OnionStudios**
+ - **Ooyala**
+ - **OoyalaExternal**
+ - **OraTV**
+ - **orf:fm4**: radio FM4
+ - **orf:iptv**: iptv.ORF.at
+ - **orf:oe1**: Radio Österreich 1
+ - **orf:tvthek**: ORF TVthek
+ - **pandora.tv**: 판도라TV
+ - **parliamentlive.tv**: UK parliament videos
+ - **Patreon**
+ - **pbs**: Public Broadcasting Service (PBS) and member stations: PBS: Public Broadcasting Service, APT - Alabama Public Television (WBIQ), GPB/Georgia Public Broadcasting (WGTV), Mississippi Public Broadcasting (WMPN), Nashville Public Television (WNPT), WFSU-TV (WFSU), WSRE (WSRE), WTCI (WTCI), WPBA/Channel 30 (WPBA), Alaska Public Media (KAKM), Arizona PBS (KAET), KNME-TV/Channel 5 (KNME), Vegas PBS (KLVX), AETN/ARKANSAS ETV NETWORK (KETS), KET (WKLE), WKNO/Channel 10 (WKNO), LPB/LOUISIANA PUBLIC BROADCASTING (WLPB), OETA (KETA), Ozarks Public Television (KOZK), WSIU Public Broadcasting (WSIU), KEET TV (KEET), KIXE/Channel 9 (KIXE), KPBS San Diego (KPBS), KQED (KQED), KVIE Public Television (KVIE), PBS SoCal/KOCE (KOCE), ValleyPBS (KVPT), CONNECTICUT PUBLIC TELEVISION (WEDH), KNPB Channel 5 (KNPB), SOPTV (KSYS), Rocky Mountain PBS (KRMA), KENW-TV3 (KENW), KUED Channel 7 (KUED), Wyoming PBS (KCWC), Colorado Public Television / KBDI 12 (KBDI), KBYU-TV (KBYU), Thirteen/WNET New York (WNET), WGBH/Channel 2 (WGBH), WGBY (WGBY), NJTV Public Media NJ (WNJT), WLIW21 (WLIW), mpt/Maryland Public Television (WMPB), WETA Television and Radio (WETA), WHYY (WHYY), PBS 39 (WLVT), WVPT - Your Source for PBS and More! (WVPT), Howard University Television (WHUT), WEDU PBS (WEDU), WGCU Public Media (WGCU), WPBT2 (WPBT), WUCF TV (WUCF), WUFT/Channel 5 (WUFT), WXEL/Channel 42 (WXEL), WLRN/Channel 17 (WLRN), WUSF Public Broadcasting (WUSF), ETV (WRLK), UNC-TV (WUNC), PBS Hawaii - Oceanic Cable Channel 10 (KHET), Idaho Public Television (KAID), KSPS (KSPS), OPB (KOPB), KWSU/Channel 10 & KTNW/Channel 31 (KWSU), WILL-TV (WILL), Network Knowledge - WSEC/Springfield (WSEC), WTTW11 (WTTW), Iowa Public Television/IPTV (KDIN), Nine Network (KETC), PBS39 Fort Wayne (WFWA), WFYI Indianapolis (WFYI), Milwaukee Public Television (WMVS), WNIN (WNIN), WNIT Public Television (WNIT), WPT (WPNE), WVUT/Channel 22 (WVUT), WEIU/Channel 51 (WEIU), WQPT-TV (WQPT), WYCC PBS Chicago (WYCC), WIPB-TV (WIPB), WTIU (WTIU), CET  (WCET), ThinkTVNetwork (WPTD), WBGU-TV (WBGU), WGVU TV (WGVU), NET1 (KUON), Pioneer Public Television (KWCM), SDPB Television (KUSD), TPT (KTCA), KSMQ (KSMQ), KPTS/Channel 8 (KPTS), KTWU/Channel 11 (KTWU), East Tennessee PBS (WSJK), WCTE-TV (WCTE), WLJT, Channel 11 (WLJT), WOSU TV (WOSU), WOUB/WOUC (WOUB), WVPB (WVPB), WKYU-PBS (WKYU), KERA 13 (KERA), MPBN (WCBB), Mountain Lake PBS (WCFE), NHPTV (WENH), Vermont PBS (WETK), witf (WITF), WQED Multimedia (WQED), WMHT Educational Telecommunications (WMHT), Q-TV (WDCQ), WTVS Detroit Public TV (WTVS), CMU Public Television (WCMU), WKAR-TV (WKAR), WNMU-TV Public TV 13 (WNMU), WDSE - WRPT (WDSE), WGTE TV (WGTE), Lakeland Public Television (KAWE), KMOS-TV - Channels 6.1, 6.2 and 6.3 (KMOS), MontanaPBS (KUSM), KRWG/Channel 22 (KRWG), KACV (KACV), KCOS/Channel 13 (KCOS), WCNY/Channel 24 (WCNY), WNED (WNED), WPBS (WPBS), WSKG Public TV (WSKG), WXXI (WXXI), WPSU (WPSU), WVIA Public Media Studios (WVIA), WTVI (WTVI), Western Reserve PBS (WNEO), WVIZ/PBS ideastream (WVIZ), KCTS 9 (KCTS), Basin PBS (KPBT), KUHT / Channel 8 (KUHT), KLRN (KLRN), KLRU (KLRU), WTJX Channel 12 (WTJX), WCVE PBS (WCVE), KBTC Public Television (KBTC)
+ - **pcmag**
+ - **Periscope**: Periscope
+ - **PhilharmonieDeParis**: Philharmonie de Paris
+ - **phoenix.de**
+ - **Photobucket**
+ - **Pinkbike**
+ - **Pladform**
+ - **PlanetaPlay**
+ - **play.fm**
+ - **played.to**
+ - **Playtvak**: Playtvak.cz, iDNES.cz and Lidovky.cz
+ - **Playvid**
+ - **Playwire**
+ - **pluralsight**
+ - **pluralsight:course**
+ - **plus.google**: Google Plus
+ - **pluzz.francetv.fr**
+ - **podomatic**
+ - **PornHd**
+ - **PornHub**
+ - **PornHubPlaylist**
+ - **Pornotube**
+ - **PornoVoisines**
+ - **PornoXO**
+ - **PrimeShareTV**
+ - **PromptFile**
+ - **prosiebensat1**: ProSiebenSat.1 Digital
+ - **Puls4**
+ - **Pyvideo**
+ - **qqmusic**: QQ音乐
+ - **qqmusic:album**: QQ音乐 - 专辑
+ - **qqmusic:playlist**: QQ音乐 - 歌单
+ - **qqmusic:singer**: QQ音乐 - 歌手
+ - **qqmusic:toplist**: QQ音乐 - 排行榜
+ - **QuickVid**
+ - **R7**
+ - **radio.de**
+ - **radiobremen**
+ - **radiofrance**
+ - **RadioJavan**
+ - **Rai**
+ - **RaiTV**
+ - **RBMARadio**
+ - **RDS**: RDS.ca
+ - **RedTube**
+ - **RegioTV**
+ - **Restudy**
+ - **ReverbNation**
+ - **Revision3**
+ - **RingTV**
+ - **RottenTomatoes**
+ - **Roxwel**
+ - **RTBF**
+ - **rte**: Raidió Teilifís Éireann TV
+ - **rte:radio**: Raidió Teilifís Éireann radio
+ - **rtl.nl**: rtl.nl and rtlxl.nl
+ - **RTL2**
+ - **RTP**
+ - **RTS**: RTS.ch
+ - **rtve.es:alacarta**: RTVE a la carta
+ - **rtve.es:infantil**: RTVE infantil
+ - **rtve.es:live**: RTVE.es live streams
+ - **RTVNH**
+ - **RUHD**
+ - **RulePorn**
+ - **rutube**: Rutube videos
+ - **rutube:channel**: Rutube channels
+ - **rutube:embed**: Rutube embedded videos
+ - **rutube:movie**: Rutube movies
+ - **rutube:person**: Rutube person videos
+ - **RUTV**: RUTV.RU
+ - **Ruutu**
+ - **safari**: safaribooksonline.com online video
+ - **safari:course**: safaribooksonline.com online courses
+ - **Sandia**: Sandia National Laboratories
+ - **Sapo**: SAPO Vídeos
+ - **savefrom.net**
+ - **SBS**: sbs.com.au
+ - **schooltv**
+ - **SciVee**
+ - **screen.yahoo:search**: Yahoo screen search
+ - **Screencast**
+ - **ScreencastOMatic**
+ - **ScreenwaveMedia**
+ - **SenateISVP**
+ - **ServingSys**
+ - **Sexu**
+ - **SexyKarma**: Sexy Karma and Watch Indian Porn
+ - **Shahid**
+ - **Shared**: shared.sx and vivo.sx
+ - **ShareSix**
+ - **Sina**
+ - **skynewsarabia:video**
+ - **skynewsarabia:video**
+ - **Slideshare**
+ - **Slutload**
+ - **smotri**: Smotri.com
+ - **smotri:broadcast**: Smotri.com broadcasts
+ - **smotri:community**: Smotri.com community videos
+ - **smotri:user**: Smotri.com user videos
+ - **SnagFilms**
+ - **SnagFilmsEmbed**
+ - **Snotr**
+ - **Sohu**
+ - **soundcloud**
+ - **soundcloud:playlist**
+ - **soundcloud:search**: Soundcloud search
+ - **soundcloud:set**
+ - **soundcloud:user**
+ - **soundgasm**
+ - **soundgasm:profile**
+ - **southpark.cc.com**
+ - **southpark.cc.com:español**
+ - **southpark.de**
+ - **southpark.nl**
+ - **southparkstudios.dk**
+ - **Space**
+ - **SpankBang**
+ - **Spankwire**
+ - **Spiegel**
+ - **Spiegel:Article**: Articles on spiegel.de
+ - **Spiegeltv**
+ - **Spike**
+ - **Sport5**
+ - **SportBox**
+ - **SportBoxEmbed**
+ - **SportDeutschland**
+ - **Sportschau**
+ - **SRGSSR**
+ - **SRGSSRPlay**: srf.ch, rts.ch, rsi.ch, rtr.ch and swissinfo.ch play sites
+ - **SSA**
+ - **stanfordoc**: Stanford Open ClassRoom
+ - **Steam**
+ - **Stitcher**
+ - **streamcloud.eu**
+ - **StreamCZ**
+ - **StreetVoice**
+ - **SunPorno**
+ - **SVT**
+ - **SVTPlay**: SVT Play and Öppet arkiv
+ - **SWRMediathek**
+ - **Syfy**
+ - **SztvHu**
+ - **Tagesschau**
+ - **Tapely**
+ - **Tass**
+ - **teachertube**: teachertube.com videos
+ - **teachertube:user:collection**: teachertube.com user and collection videos
+ - **TeachingChannel**
+ - **Teamcoco**
+ - **TeamFour**
+ - **TechTalks**
+ - **techtv.mit.edu**
+ - **ted**
+ - **Tele13**
+ - **TeleBruxelles**
+ - **Telecinco**: telecinco.es, cuatro.com and mediaset.es
+ - **Telegraaf**
+ - **TeleMB**
+ - **TeleTask**
+ - **TenPlay**
+ - **TF1**
+ - **TheIntercept**
+ - **TheOnion**
+ - **ThePlatform**
+ - **ThePlatformFeed**
+ - **TheSixtyOne**
+ - **ThisAmericanLife**
+ - **ThisAV**
+ - **THVideo**
+ - **THVideoPlaylist**
+ - **tinypic**: tinypic.com videos
+ - **tlc.de**
+ - **TMZ**
+ - **TMZArticle**
+ - **TNAFlix**
+ - **toggle**
+ - **tou.tv**
+ - **Toypics**: Toypics user profile
+ - **ToypicsUser**: Toypics user profile
+ - **TrailerAddict** (Currently broken)
+ - **Trilulilu**
+ - **trollvids**
+ - **TruTube**
+ - **Tube8**
+ - **TubiTv**
+ - **tudou**
+ - **tudou:album**
+ - **tudou:playlist**
+ - **Tumblr**
+ - **tunein:clip**
+ - **tunein:program**
+ - **tunein:station**
+ - **tunein:topic**
+ - **Turbo**
+ - **Tutv**
+ - **tv.dfb.de**
+ - **TV2**
+ - **TV2Article**
+ - **TV4**: tv4.se and tv4play.se
+ - **TVC**
+ - **TVCArticle**
+ - **tvigle**: Интернет-телевидение Tvigle.ru
+ - **tvland.com**
+ - **tvp.pl**
+ - **tvp.pl:Series**
+ - **TVPlay**: TV3Play and related services
+ - **Tweakers**
+ - **twitch:bookmarks**
+ - **twitch:chapter**
+ - **twitch:past_broadcasts**
+ - **twitch:profile**
+ - **twitch:stream**
+ - **twitch:video**
+ - **twitch:vod**
+ - **twitter**
+ - **twitter:card**
+ - **Ubu**
+ - **udemy**
+ - **udemy:course**
+ - **UDNEmbed**: 聯合影音
+ - **Unistra**
+ - **Urort**: NRK P3 Urørt
+ - **ustream**
+ - **ustream:channel**
+ - **Varzesh3**
+ - **Vbox7**
+ - **VeeHD**
+ - **Veoh**
+ - **Vessel**
+ - **Vesti**: Вести.Ru
+ - **Vevo**
+ - **VGTV**: VGTV, BTTV, FTV, Aftenposten and Aftonbladet
+ - **vh1.com**
+ - **Vice**
+ - **Viddler**
+ - **video.google:search**: Google Video search
+ - **video.mit.edu**
+ - **VideoDetective**
+ - **videofy.me**
+ - **VideoMega** (Currently broken)
+ - **videomore**
+ - **videomore:season**
+ - **videomore:video**
+ - **VideoPremium**
+ - **VideoTt**: video.tt - Your True Tube (Currently broken)
+ - **videoweed**: VideoWeed
+ - **vidme**
+ - **vidme:user**
+ - **vidme:user:likes**
+ - **Vidzi**
+ - **vier**
+ - **vier:videos**
+ - **Viewster**
+ - **Viidea**
+ - **viki**
+ - **viki:channel**
+ - **vimeo**
+ - **vimeo:album**
+ - **vimeo:channel**
+ - **vimeo:group**
+ - **vimeo:likes**: Vimeo user likes
+ - **vimeo:review**: Review pages on vimeo
+ - **vimeo:user**
+ - **vimeo:watchlater**: Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)
+ - **Vimple**: Vimple - one-click video hosting
+ - **Vine**
+ - **vine:user**
+ - **vk**: VK
+ - **vk:uservideos**: VK - User's Videos
+ - **vlive**
+ - **Vodlocker**
+ - **VoiceRepublic**
+ - **Vporn**
+ - **vpro**: npo.nl and ntr.nl
+ - **VRT**
+ - **vube**: Vube.com
+ - **VuClip**
+ - **vulture.com**
+ - **Walla**
+ - **WashingtonPost**
+ - **wat.tv**
+ - **WayOfTheMaster**
+ - **WDR**
+ - **wdr:mobile**
+ - **WDRMaus**: Sendung mit der Maus
+ - **WebOfStories**
+ - **WebOfStoriesPlaylist**
+ - **Weibo**
+ - **WeiqiTV**: WQTV
+ - **wholecloud**: WholeCloud
+ - **Wimp**
+ - **Wistia**
+ - **WNL**
+ - **WorldStarHipHop**
+ - **wrzuta.pl**
+ - **WSJ**: Wall Street Journal
+ - **XBef**
+ - **XboxClips**
+ - **XFileShare**: XFileShare based sites: GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net, filehoot.com and vidto.me
+ - **XHamster**
+ - **XHamsterEmbed**
+ - **XMinus**
+ - **XNXX**
+ - **Xstream**
+ - **XTube**
+ - **XTubeUser**: XTube user profile
+ - **Xuite**: 隨意窩Xuite影音
+ - **XVideos**
+ - **XXXYMovies**
+ - **Yahoo**: Yahoo screen and movies
+ - **Yam**: 蕃薯藤yam天空部落
+ - **yandexmusic:album**: Яндекс.Музыка - Альбом
+ - **yandexmusic:playlist**: Яндекс.Музыка - Плейлист
+ - **yandexmusic:track**: Яндекс.Музыка - Трек
+ - **YesJapan**
+ - **yinyuetai:video**: 音悦Tai
+ - **Ynet**
+ - **YouJizz**
+ - **youku**: 优酷
+ - **YouPorn**
+ - **YourUpload**
+ - **youtube**: YouTube.com
+ - **youtube:channel**: YouTube.com channels
+ - **youtube:favorites**: YouTube.com favourite videos, ":ytfav" for short (requires authentication)
+ - **youtube:history**: Youtube watch history, ":ythistory" for short (requires authentication)
+ - **youtube:playlist**: YouTube.com playlists
+ - **youtube:playlists**: YouTube.com user/channel playlists
+ - **youtube:recommended**: YouTube.com recommended videos, ":ytrec" for short (requires authentication)
+ - **youtube:search**: YouTube.com searches
+ - **youtube:search:date**: YouTube.com searches, newest videos first
+ - **youtube:search_url**: YouTube.com search URLs
+ - **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)
+ - **youtube:watchlater**: Youtube watch later list, ":ytwatchlater" for short (requires authentication)
+ - **Zapiks**
+ - **ZDF**
+ - **ZDFChannel**
+ - **zingmp3:album**: mp3.zing.vn albums
+ - **zingmp3:song**: mp3.zing.vn songs
+ - **ZippCast**

setup.cfg

@@ -1,2 +1,6 @@
 [wheel]
 universal = True
+
+[flake8]
+exclude = youtube_dl/extractor/__init__.py,devscripts/buildserver.py,setup.py,build,.git
+ignore = E402,E501,E731

setup.py

@@ -28,7 +28,7 @@ py2exe_options = {
     "compressed": 1,
     "optimize": 2,
     "dist_dir": '.',
-    "dll_excludes": ['w9xpopen.exe'],
+    "dll_excludes": ['w9xpopen.exe', 'crypt32.dll'],
 }
 
 py2exe_console = [{

test/helper.py

@@ -82,51 +82,90 @@ class FakeYDL(YoutubeDL):
 
 def gettestcases(include_onlymatching=False):
     for ie in youtube_dl.extractor.gen_extractors():
-        t = getattr(ie, '_TEST', None)
-        if t:
-            assert not hasattr(ie, '_TESTS'), \
-                '%s has _TEST and _TESTS' % type(ie).__name__
-            tests = [t]
-        else:
-            tests = getattr(ie, '_TESTS', [])
-        for t in tests:
-            if not include_onlymatching and t.get('only_matching', False):
-                continue
-            t['name'] = type(ie).__name__[:-len('IE')]
-            yield t
+        for tc in ie.get_testcases(include_onlymatching):
+            yield tc
 
 
 md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
 
 
-def expect_info_dict(self, expected_dict, got_dict):
+def expect_value(self, got, expected, field):
+    if isinstance(expected, compat_str) and expected.startswith('re:'):
+        match_str = expected[len('re:'):]
+        match_rex = re.compile(match_str)
+
+        self.assertTrue(
+            isinstance(got, compat_str),
+            'Expected a %s object, but got %s for field %s' % (
+                compat_str.__name__, type(got).__name__, field))
+        self.assertTrue(
+            match_rex.match(got),
+            'field %s (value: %r) should match %r' % (field, got, match_str))
+    elif isinstance(expected, compat_str) and expected.startswith('startswith:'):
+        start_str = expected[len('startswith:'):]
+        self.assertTrue(
+            isinstance(got, compat_str),
+            'Expected a %s object, but got %s for field %s' % (
+                compat_str.__name__, type(got).__name__, field))
+        self.assertTrue(
+            got.startswith(start_str),
+            'field %s (value: %r) should start with %r' % (field, got, start_str))
+    elif isinstance(expected, compat_str) and expected.startswith('contains:'):
+        contains_str = expected[len('contains:'):]
+        self.assertTrue(
+            isinstance(got, compat_str),
+            'Expected a %s object, but got %s for field %s' % (
+                compat_str.__name__, type(got).__name__, field))
+        self.assertTrue(
+            contains_str in got,
+            'field %s (value: %r) should contain %r' % (field, got, contains_str))
+    elif isinstance(expected, type):
+        self.assertTrue(
+            isinstance(got, expected),
+            'Expected type %r for field %s, but got value %r of type %r' % (expected, field, got, type(got)))
+    elif isinstance(expected, dict) and isinstance(got, dict):
+        expect_dict(self, got, expected)
+    elif isinstance(expected, list) and isinstance(got, list):
+        self.assertEqual(
+            len(expected), len(got),
+            'Expect a list of length %d, but got a list of length %d for field %s' % (
+                len(expected), len(got), field))
+        for index, (item_got, item_expected) in enumerate(zip(got, expected)):
+            type_got = type(item_got)
+            type_expected = type(item_expected)
+            self.assertEqual(
+                type_expected, type_got,
+                'Type mismatch for list item at index %d for field %s, expected %r, got %r' % (
+                    index, field, type_expected, type_got))
+            expect_value(self, item_got, item_expected, field)
+    else:
+        if isinstance(expected, compat_str) and expected.startswith('md5:'):
+            got = 'md5:' + md5(got)
+        elif isinstance(expected, compat_str) and expected.startswith('mincount:'):
+            self.assertTrue(
+                isinstance(got, (list, dict)),
+                'Expected field %s to be a list or a dict, but it is of type %s' % (
+                    field, type(got).__name__))
+            expected_num = int(expected.partition(':')[2])
+            assertGreaterEqual(
+                self, len(got), expected_num,
+                'Expected %d items in field %s, but only got %d' % (expected_num, field, len(got)))
+            return
+        self.assertEqual(
+            expected, got,
+            'Invalid value for field %s, expected %r, got %r' % (field, expected, got))
+
+
+def expect_dict(self, got_dict, expected_dict):
     for info_field, expected in expected_dict.items():
-        if isinstance(expected, compat_str) and expected.startswith('re:'):
-            got = got_dict.get(info_field)
-            match_str = expected[len('re:'):]
-            match_rex = re.compile(match_str)
+        got = got_dict.get(info_field)
+        expect_value(self, got, expected, info_field)
 
-            self.assertTrue(
-                isinstance(got, compat_str),
-                'Expected a %s object, but got %s for field %s' % (
-                    compat_str.__name__, type(got).__name__, info_field))
-            self.assertTrue(
-                match_rex.match(got),
-                'field %s (value: %r) should match %r' % (info_field, got, match_str))
-        elif isinstance(expected, type):
-            got = got_dict.get(info_field)
-            self.assertTrue(isinstance(got, expected),
-                            'Expected type %r for field %s, but got value %r of type %r' % (expected, info_field, got, type(got)))
-        else:
-            if isinstance(expected, compat_str) and expected.startswith('md5:'):
-                got = 'md5:' + md5(got_dict.get(info_field))
-            else:
-                got = got_dict.get(info_field)
-            self.assertEqual(expected, got,
-                             'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
 
+def expect_info_dict(self, got_dict, expected_dict):
+    expect_dict(self, got_dict, expected_dict)
     # Check for the presence of mandatory fields
-    if got_dict.get('_type') != 'playlist':
+    if got_dict.get('_type') not in ('playlist', 'multi_video'):
         for key in ('id', 'url', 'title', 'ext'):
             self.assertTrue(got_dict.get(key), 'Missing mandatory field %s' % key)
     # Check for mandatory fields that are automatically set by YoutubeDL
@@ -136,7 +175,7 @@ def expect_info_dict(self, expected_dict, got_dict):
     # Are checkable fields missing from the test case definition?
     test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))
                           for key, value in got_dict.items()
-                          if value and key in ('title', 'description', 'uploader', 'upload_date', 'timestamp', 'uploader_id', 'location'))
+                          if value and key in ('id', 'title', 'description', 'uploader', 'upload_date', 'timestamp', 'uploader_id', 'location', 'age_limit'))
     missing_keys = set(test_info_dict.keys()) - set(expected_dict.keys())
     if missing_keys:
         def _repr(v):
@@ -144,11 +183,19 @@ def expect_info_dict(self, expected_dict, got_dict):
                 return "'%s'" % v.replace('\\', '\\\\').replace("'", "\\'").replace('\n', '\\n')
             else:
                 return repr(v)
-        info_dict_str = ''.join(
-            '    %s: %s,\n' % (_repr(k), _repr(v))
-            for k, v in test_info_dict.items())
+        info_dict_str = ''
+        if len(missing_keys) != len(expected_dict):
+            info_dict_str += ''.join(
+                '    %s: %s,\n' % (_repr(k), _repr(v))
+                for k, v in test_info_dict.items() if k not in missing_keys)
+
+            if info_dict_str:
+                info_dict_str += '\n'
+        info_dict_str += ''.join(
+            '    %s: %s,\n' % (_repr(k), _repr(test_info_dict[k]))
+            for k in missing_keys)
         write_string(
-            '\n\'info_dict\': {\n' + info_dict_str + '}\n', out=sys.stderr)
+            '\n\'info_dict\': {\n' + info_dict_str + '},\n', out=sys.stderr)
         self.assertFalse(
             missing_keys,
             'Missing keys in test definition: %s' % (
@@ -161,7 +208,9 @@ def assertRegexpMatches(self, text, regexp, msg=None):
     else:
         m = re.match(regexp, text)
         if not m:
-            note = 'Regexp didn\'t match: %r not found in %r' % (regexp, text)
+            note = 'Regexp didn\'t match: %r not found' % (regexp)
+            if len(text) < 1000:
+                note += ' in %r' % text
             if msg is None:
                 msg = note
             else:

test/parameters.json

@@ -7,8 +7,7 @@
     "forcethumbnail": false, 
     "forcetitle": false, 
     "forceurl": false, 
-    "format": null, 
-    "format_limit": null, 
+    "format": "best",
     "ignoreerrors": false, 
     "listformats": null, 
     "logtostderr": false, 
@@ -28,7 +27,7 @@
     "retries": 10, 
     "simulate": false, 
     "subtitleslang": null, 
-    "subtitlesformat": "srt",
+    "subtitlesformat": "best",
     "test": true, 
     "updatetime": true, 
     "usenetrc": false, 
@@ -39,5 +38,6 @@
     "writesubtitles": false,
     "allsubtitles": false,
     "listssubtitles": false,
-    "socket_timeout": 20
+    "socket_timeout": 20,
+    "fixup": "never"
 }

test/test_InfoExtractor.py

@@ -35,10 +35,36 @@ class TestInfoExtractor(unittest.TestCase):
             <meta name="og:title" content='Foo'/>
             <meta content="Some video's description " name="og:description"/>
             <meta property='og:image' content='http://domain.com/pic.jpg?key1=val1&amp;key2=val2'/>
+            <meta content='application/x-shockwave-flash' property='og:video:type'>
+            <meta content='Foo' property=og:foobar>
+            <meta name="og:test1" content='foo > < bar'/>
+            <meta name="og:test2" content="foo >//< bar"/>
             '''
         self.assertEqual(ie._og_search_title(html), 'Foo')
         self.assertEqual(ie._og_search_description(html), 'Some video\'s description ')
         self.assertEqual(ie._og_search_thumbnail(html), 'http://domain.com/pic.jpg?key1=val1&key2=val2')
+        self.assertEqual(ie._og_search_video_url(html, default=None), None)
+        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')
+
+    def test_html_search_meta(self):
+        ie = self.ie
+        html = '''
+            <meta name="a" content="1" />
+            <meta name='b' content='2'>
+            <meta name="c" content='3'>
+            <meta name=d content='4'>
+            <meta property="e" content='5' >
+            <meta content="6" name="f">
+        '''
+
+        self.assertEqual(ie._html_search_meta('a', html), '1')
+        self.assertEqual(ie._html_search_meta('b', html), '2')
+        self.assertEqual(ie._html_search_meta('c', html), '3')
+        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')
 
 if __name__ == '__main__':
     unittest.main()

test/test_YoutubeDL.py

@@ -8,9 +8,17 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+import copy
+
 from test.helper import FakeYDL, assertRegexpMatches
 from youtube_dl import YoutubeDL
+from youtube_dl.compat import compat_str, compat_urllib_error
 from youtube_dl.extractor import YoutubeIE
+from youtube_dl.extractor.common import InfoExtractor
+from youtube_dl.postprocessor.common import PostProcessor
+from youtube_dl.utils import ExtractorError, match_filter_func
+
+TEST_URL = 'http://localhost/sample.mp4'
 
 
 class YDL(FakeYDL):
@@ -43,8 +51,8 @@ class TestFormatSelection(unittest.TestCase):
         ydl = YDL()
         ydl.params['prefer_free_formats'] = True
         formats = [
-            {'ext': 'webm', 'height': 460, 'url': 'x'},
-            {'ext': 'mp4', 'height': 460, 'url': 'y'},
+            {'ext': 'webm', 'height': 460, 'url': TEST_URL},
+            {'ext': 'mp4', 'height': 460, 'url': TEST_URL},
         ]
         info_dict = _make_result(formats)
         yie = YoutubeIE(ydl)
@@ -57,8 +65,8 @@ class TestFormatSelection(unittest.TestCase):
         ydl = YDL()
         ydl.params['prefer_free_formats'] = True
         formats = [
-            {'ext': 'webm', 'height': 720, 'url': 'a'},
-            {'ext': 'mp4', 'height': 1080, 'url': 'b'},
+            {'ext': 'webm', 'height': 720, 'url': TEST_URL},
+            {'ext': 'mp4', 'height': 1080, 'url': TEST_URL},
         ]
         info_dict['formats'] = formats
         yie = YoutubeIE(ydl)
@@ -71,9 +79,9 @@ class TestFormatSelection(unittest.TestCase):
         ydl = YDL()
         ydl.params['prefer_free_formats'] = False
         formats = [
-            {'ext': 'webm', 'height': 720, 'url': '_'},
-            {'ext': 'mp4', 'height': 720, 'url': '_'},
-            {'ext': 'flv', 'height': 720, 'url': '_'},
+            {'ext': 'webm', 'height': 720, 'url': TEST_URL},
+            {'ext': 'mp4', 'height': 720, 'url': TEST_URL},
+            {'ext': 'flv', 'height': 720, 'url': TEST_URL},
         ]
         info_dict['formats'] = formats
         yie = YoutubeIE(ydl)
@@ -85,8 +93,8 @@ class TestFormatSelection(unittest.TestCase):
         ydl = YDL()
         ydl.params['prefer_free_formats'] = False
         formats = [
-            {'ext': 'flv', 'height': 720, 'url': '_'},
-            {'ext': 'webm', 'height': 720, 'url': '_'},
+            {'ext': 'flv', 'height': 720, 'url': TEST_URL},
+            {'ext': 'webm', 'height': 720, 'url': TEST_URL},
         ]
         info_dict['formats'] = formats
         yie = YoutubeIE(ydl)
@@ -95,45 +103,13 @@ class TestFormatSelection(unittest.TestCase):
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['ext'], 'flv')
 
-    def test_format_limit(self):
-        formats = [
-            {'format_id': 'meh', 'url': 'http://example.com/meh', 'preference': 1},
-            {'format_id': 'good', 'url': 'http://example.com/good', 'preference': 2},
-            {'format_id': 'great', 'url': 'http://example.com/great', 'preference': 3},
-            {'format_id': 'excellent', 'url': 'http://example.com/exc', 'preference': 4},
-        ]
-        info_dict = _make_result(formats)
-
-        ydl = YDL()
-        ydl.process_ie_result(info_dict)
-        downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], 'excellent')
-
-        ydl = YDL({'format_limit': 'good'})
-        assert ydl.params['format_limit'] == 'good'
-        ydl.process_ie_result(info_dict.copy())
-        downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], 'good')
-
-        ydl = YDL({'format_limit': 'great', 'format': 'all'})
-        ydl.process_ie_result(info_dict.copy())
-        self.assertEqual(ydl.downloaded_info_dicts[0]['format_id'], 'meh')
-        self.assertEqual(ydl.downloaded_info_dicts[1]['format_id'], 'good')
-        self.assertEqual(ydl.downloaded_info_dicts[2]['format_id'], 'great')
-        self.assertTrue('3' in ydl.msgs[0])
-
-        ydl = YDL()
-        ydl.params['format_limit'] = 'excellent'
-        ydl.process_ie_result(info_dict.copy())
-        downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], 'excellent')
-
     def test_format_selection(self):
         formats = [
-            {'format_id': '35', 'ext': 'mp4', 'preference': 1, 'url': '_'},
-            {'format_id': '45', 'ext': 'webm', 'preference': 2, 'url': '_'},
-            {'format_id': '47', 'ext': 'webm', 'preference': 3, 'url': '_'},
-            {'format_id': '2', 'ext': 'flv', 'preference': 4, 'url': '_'},
+            {'format_id': '35', 'ext': 'mp4', 'preference': 1, 'url': TEST_URL},
+            {'format_id': 'example-with-dashes', 'ext': 'webm', 'preference': 1, 'url': TEST_URL},
+            {'format_id': '45', 'ext': 'webm', 'preference': 2, 'url': TEST_URL},
+            {'format_id': '47', 'ext': 'webm', 'preference': 3, 'url': TEST_URL},
+            {'format_id': '2', 'ext': 'flv', 'preference': 4, 'url': TEST_URL},
         ]
         info_dict = _make_result(formats)
 
@@ -162,12 +138,17 @@ class TestFormatSelection(unittest.TestCase):
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['format_id'], '35')
 
+        ydl = YDL({'format': 'example-with-dashes'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'example-with-dashes')
+
     def test_format_selection_audio(self):
         formats = [
-            {'format_id': 'audio-low', 'ext': 'webm', 'preference': 1, 'vcodec': 'none', 'url': '_'},
-            {'format_id': 'audio-mid', 'ext': 'webm', 'preference': 2, 'vcodec': 'none', 'url': '_'},
-            {'format_id': 'audio-high', 'ext': 'flv', 'preference': 3, 'vcodec': 'none', 'url': '_'},
-            {'format_id': 'vid', 'ext': 'mp4', 'preference': 4, 'url': '_'},
+            {'format_id': 'audio-low', 'ext': 'webm', 'preference': 1, 'vcodec': 'none', 'url': TEST_URL},
+            {'format_id': 'audio-mid', 'ext': 'webm', 'preference': 2, 'vcodec': 'none', 'url': TEST_URL},
+            {'format_id': 'audio-high', 'ext': 'flv', 'preference': 3, 'vcodec': 'none', 'url': TEST_URL},
+            {'format_id': 'vid', 'ext': 'mp4', 'preference': 4, 'url': TEST_URL},
         ]
         info_dict = _make_result(formats)
 
@@ -182,8 +163,8 @@ class TestFormatSelection(unittest.TestCase):
         self.assertEqual(downloaded['format_id'], 'audio-low')
 
         formats = [
-            {'format_id': 'vid-low', 'ext': 'mp4', 'preference': 1, 'url': '_'},
-            {'format_id': 'vid-high', 'ext': 'mp4', 'preference': 2, 'url': '_'},
+            {'format_id': 'vid-low', 'ext': 'mp4', 'preference': 1, 'url': TEST_URL},
+            {'format_id': 'vid-high', 'ext': 'mp4', 'preference': 2, 'url': TEST_URL},
         ]
         info_dict = _make_result(formats)
 
@@ -192,11 +173,42 @@ class TestFormatSelection(unittest.TestCase):
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['format_id'], 'vid-high')
 
+    def test_format_selection_audio_exts(self):
+        formats = [
+            {'format_id': 'mp3-64', 'ext': 'mp3', 'abr': 64, 'url': 'http://_', 'vcodec': 'none'},
+            {'format_id': 'ogg-64', 'ext': 'ogg', 'abr': 64, 'url': 'http://_', 'vcodec': 'none'},
+            {'format_id': 'aac-64', 'ext': 'aac', 'abr': 64, 'url': 'http://_', 'vcodec': 'none'},
+            {'format_id': 'mp3-32', 'ext': 'mp3', 'abr': 32, 'url': 'http://_', 'vcodec': 'none'},
+            {'format_id': 'aac-32', 'ext': 'aac', 'abr': 32, 'url': 'http://_', 'vcodec': 'none'},
+        ]
+
+        info_dict = _make_result(formats)
+        ydl = YDL({'format': 'best'})
+        ie = YoutubeIE(ydl)
+        ie._sort_formats(info_dict['formats'])
+        ydl.process_ie_result(copy.deepcopy(info_dict))
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'aac-64')
+
+        ydl = YDL({'format': 'mp3'})
+        ie = YoutubeIE(ydl)
+        ie._sort_formats(info_dict['formats'])
+        ydl.process_ie_result(copy.deepcopy(info_dict))
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'mp3-64')
+
+        ydl = YDL({'prefer_free_formats': True})
+        ie = YoutubeIE(ydl)
+        ie._sort_formats(info_dict['formats'])
+        ydl.process_ie_result(copy.deepcopy(info_dict))
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'ogg-64')
+
     def test_format_selection_video(self):
         formats = [
-            {'format_id': 'dash-video-low', 'ext': 'mp4', 'preference': 1, 'acodec': 'none', 'url': '_'},
-            {'format_id': 'dash-video-high', 'ext': 'mp4', 'preference': 2, 'acodec': 'none', 'url': '_'},
-            {'format_id': 'vid', 'ext': 'mp4', 'preference': 3, 'url': '_'},
+            {'format_id': 'dash-video-low', 'ext': 'mp4', 'preference': 1, 'acodec': 'none', 'url': TEST_URL},
+            {'format_id': 'dash-video-high', 'ext': 'mp4', 'preference': 2, 'acodec': 'none', 'url': TEST_URL},
+            {'format_id': 'vid', 'ext': 'mp4', 'preference': 3, 'url': TEST_URL},
         ]
         info_dict = _make_result(formats)
 
@@ -210,6 +222,16 @@ class TestFormatSelection(unittest.TestCase):
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['format_id'], 'dash-video-low')
 
+        formats = [
+            {'format_id': 'vid-vcodec-dot', 'ext': 'mp4', 'preference': 1, 'vcodec': 'avc1.123456', 'acodec': 'none', 'url': TEST_URL},
+        ]
+        info_dict = _make_result(formats)
+
+        ydl = YDL({'format': 'bestvideo[vcodec=avc1.123456]'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'vid-vcodec-dot')
+
     def test_youtube_format_selection(self):
         order = [
             '38', '37', '46', '22', '45', '35', '44', '18', '34', '43', '6', '5', '36', '17', '13',
@@ -218,35 +240,234 @@ class TestFormatSelection(unittest.TestCase):
             # 3D
             '85', '84', '102', '83', '101', '82', '100',
             # Dash video
-            '138', '137', '248', '136', '247', '135', '246',
+            '137', '248', '136', '247', '135', '246',
             '245', '244', '134', '243', '133', '242', '160',
             # Dash audio
             '141', '172', '140', '171', '139',
         ]
 
-        for f1id, f2id in zip(order, order[1:]):
-            f1 = YoutubeIE._formats[f1id].copy()
-            f1['format_id'] = f1id
-            f1['url'] = 'url:' + f1id
-            f2 = YoutubeIE._formats[f2id].copy()
-            f2['format_id'] = f2id
-            f2['url'] = 'url:' + f2id
+        def format_info(f_id):
+            info = YoutubeIE._formats[f_id].copy()
 
+            # XXX: In real cases InfoExtractor._parse_mpd_formats() fills up 'acodec'
+            # and 'vcodec', while in tests such information is incomplete since
+            # commit a6c2c24479e5f4827ceb06f64d855329c0a6f593
+            # test_YoutubeDL.test_youtube_format_selection is broken without
+            # this fix
+            if 'acodec' in info and 'vcodec' not in info:
+                info['vcodec'] = 'none'
+            elif 'vcodec' in info and 'acodec' not in info:
+                info['acodec'] = 'none'
+
+            info['format_id'] = f_id
+            info['url'] = 'url:' + f_id
+            return info
+        formats_order = [format_info(f_id) for f_id in order]
+
+        info_dict = _make_result(list(formats_order), extractor='youtube')
+        ydl = YDL({'format': 'bestvideo+bestaudio'})
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], '137+141')
+        self.assertEqual(downloaded['ext'], 'mp4')
+
+        info_dict = _make_result(list(formats_order), extractor='youtube')
+        ydl = YDL({'format': 'bestvideo[height>=999999]+bestaudio/best'})
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], '38')
+
+        info_dict = _make_result(list(formats_order), extractor='youtube')
+        ydl = YDL({'format': 'bestvideo/best,bestaudio'})
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
+        ydl.process_ie_result(info_dict)
+        downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
+        self.assertEqual(downloaded_ids, ['137', '141'])
+
+        info_dict = _make_result(list(formats_order), extractor='youtube')
+        ydl = YDL({'format': '(bestvideo[ext=mp4],bestvideo[ext=webm])+bestaudio'})
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
+        ydl.process_ie_result(info_dict)
+        downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
+        self.assertEqual(downloaded_ids, ['137+141', '248+141'])
+
+        info_dict = _make_result(list(formats_order), extractor='youtube')
+        ydl = YDL({'format': '(bestvideo[ext=mp4],bestvideo[ext=webm])[height<=720]+bestaudio'})
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
+        ydl.process_ie_result(info_dict)
+        downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
+        self.assertEqual(downloaded_ids, ['136+141', '247+141'])
+
+        info_dict = _make_result(list(formats_order), extractor='youtube')
+        ydl = YDL({'format': '(bestvideo[ext=none]/bestvideo[ext=webm])+bestaudio'})
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
+        ydl.process_ie_result(info_dict)
+        downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
+        self.assertEqual(downloaded_ids, ['248+141'])
+
+        for f1, f2 in zip(formats_order, formats_order[1:]):
             info_dict = _make_result([f1, f2], extractor='youtube')
-            ydl = YDL()
+            ydl = YDL({'format': 'best/bestvideo'})
             yie = YoutubeIE(ydl)
             yie._sort_formats(info_dict['formats'])
             ydl.process_ie_result(info_dict)
             downloaded = ydl.downloaded_info_dicts[0]
-            self.assertEqual(downloaded['format_id'], f1id)
+            self.assertEqual(downloaded['format_id'], f1['format_id'])
 
             info_dict = _make_result([f2, f1], extractor='youtube')
-            ydl = YDL()
+            ydl = YDL({'format': 'best/bestvideo'})
             yie = YoutubeIE(ydl)
             yie._sort_formats(info_dict['formats'])
             ydl.process_ie_result(info_dict)
             downloaded = ydl.downloaded_info_dicts[0]
-            self.assertEqual(downloaded['format_id'], f1id)
+            self.assertEqual(downloaded['format_id'], f1['format_id'])
+
+    def test_invalid_format_specs(self):
+        def assert_syntax_error(format_spec):
+            ydl = YDL({'format': format_spec})
+            info_dict = _make_result([{'format_id': 'foo', 'url': TEST_URL}])
+            self.assertRaises(SyntaxError, ydl.process_ie_result, info_dict)
+
+        assert_syntax_error('bestvideo,,best')
+        assert_syntax_error('+bestaudio')
+        assert_syntax_error('bestvideo+')
+        assert_syntax_error('/')
+
+    def test_format_filtering(self):
+        formats = [
+            {'format_id': 'A', 'filesize': 500, 'width': 1000},
+            {'format_id': 'B', 'filesize': 1000, 'width': 500},
+            {'format_id': 'C', 'filesize': 1000, 'width': 400},
+            {'format_id': 'D', 'filesize': 2000, 'width': 600},
+            {'format_id': 'E', 'filesize': 3000},
+            {'format_id': 'F'},
+            {'format_id': 'G', 'filesize': 1000000},
+        ]
+        for f in formats:
+            f['url'] = 'http://_/'
+            f['ext'] = 'unknown'
+        info_dict = _make_result(formats)
+
+        ydl = YDL({'format': 'best[filesize<3000]'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'D')
+
+        ydl = YDL({'format': 'best[filesize<=3000]'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'E')
+
+        ydl = YDL({'format': 'best[filesize <= ? 3000]'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'F')
+
+        ydl = YDL({'format': 'best [filesize = 1000] [width>450]'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'B')
+
+        ydl = YDL({'format': 'best [filesize = 1000] [width!=450]'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'C')
+
+        ydl = YDL({'format': '[filesize>?1]'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'G')
+
+        ydl = YDL({'format': '[filesize<1M]'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'E')
+
+        ydl = YDL({'format': '[filesize<1MiB]'})
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'G')
+
+        ydl = YDL({'format': 'all[width>=400][width<=600]'})
+        ydl.process_ie_result(info_dict)
+        downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
+        self.assertEqual(downloaded_ids, ['B', 'C', 'D'])
+
+        ydl = YDL({'format': 'best[height<40]'})
+        try:
+            ydl.process_ie_result(info_dict)
+        except ExtractorError:
+            pass
+        self.assertEqual(ydl.downloaded_info_dicts, [])
+
+
+class TestYoutubeDL(unittest.TestCase):
+    def test_subtitles(self):
+        def s_formats(lang, autocaption=False):
+            return [{
+                'ext': ext,
+                'url': 'http://localhost/video.%s.%s' % (lang, ext),
+                '_auto': autocaption,
+            } for ext in ['vtt', 'srt', 'ass']]
+        subtitles = dict((l, s_formats(l)) for l in ['en', 'fr', 'es'])
+        auto_captions = dict((l, s_formats(l, True)) for l in ['it', 'pt', 'es'])
+        info_dict = {
+            'id': 'test',
+            'title': 'Test',
+            'url': 'http://localhost/video.mp4',
+            'subtitles': subtitles,
+            'automatic_captions': auto_captions,
+            'extractor': 'TEST',
+        }
+
+        def get_info(params={}):
+            params.setdefault('simulate', True)
+            ydl = YDL(params)
+            ydl.report_warning = lambda *args, **kargs: None
+            return ydl.process_video_result(info_dict, download=False)
+
+        result = get_info()
+        self.assertFalse(result.get('requested_subtitles'))
+        self.assertEqual(result['subtitles'], subtitles)
+        self.assertEqual(result['automatic_captions'], auto_captions)
+
+        result = get_info({'writesubtitles': True})
+        subs = result['requested_subtitles']
+        self.assertTrue(subs)
+        self.assertEqual(set(subs.keys()), set(['en']))
+        self.assertTrue(subs['en'].get('data') is None)
+        self.assertEqual(subs['en']['ext'], 'ass')
+
+        result = get_info({'writesubtitles': True, 'subtitlesformat': 'foo/srt'})
+        subs = result['requested_subtitles']
+        self.assertEqual(subs['en']['ext'], 'srt')
+
+        result = get_info({'writesubtitles': True, 'subtitleslangs': ['es', 'fr', 'it']})
+        subs = result['requested_subtitles']
+        self.assertTrue(subs)
+        self.assertEqual(set(subs.keys()), set(['es', 'fr']))
+
+        result = get_info({'writesubtitles': True, 'writeautomaticsub': True, 'subtitleslangs': ['es', 'pt']})
+        subs = result['requested_subtitles']
+        self.assertTrue(subs)
+        self.assertEqual(set(subs.keys()), set(['es', 'pt']))
+        self.assertFalse(subs['es']['_auto'])
+        self.assertTrue(subs['pt']['_auto'])
+
+        result = get_info({'writeautomaticsub': True, 'subtitleslangs': ['es', 'pt']})
+        subs = result['requested_subtitles']
+        self.assertTrue(subs)
+        self.assertEqual(set(subs.keys()), set(['es', 'pt']))
+        self.assertTrue(subs['es']['_auto'])
+        self.assertTrue(subs['pt']['_auto'])
 
     def test_add_extra_info(self):
         test_dict = {
@@ -282,5 +503,197 @@ class TestFormatSelection(unittest.TestCase):
             'vbr': 10,
         }), '^\s*10k$')
 
+    def test_postprocessors(self):
+        filename = 'post-processor-testfile.mp4'
+        audiofile = filename + '.mp3'
+
+        class SimplePP(PostProcessor):
+            def run(self, info):
+                with open(audiofile, 'wt') as f:
+                    f.write('EXAMPLE')
+                return [info['filepath']], info
+
+        def run_pp(params, PP):
+            with open(filename, 'wt') as f:
+                f.write('EXAMPLE')
+            ydl = YoutubeDL(params)
+            ydl.add_post_processor(PP())
+            ydl.post_process(filename, {'filepath': filename})
+
+        run_pp({'keepvideo': True}, SimplePP)
+        self.assertTrue(os.path.exists(filename), '%s doesn\'t exist' % filename)
+        self.assertTrue(os.path.exists(audiofile), '%s doesn\'t exist' % audiofile)
+        os.unlink(filename)
+        os.unlink(audiofile)
+
+        run_pp({'keepvideo': False}, SimplePP)
+        self.assertFalse(os.path.exists(filename), '%s exists' % filename)
+        self.assertTrue(os.path.exists(audiofile), '%s doesn\'t exist' % audiofile)
+        os.unlink(audiofile)
+
+        class ModifierPP(PostProcessor):
+            def run(self, info):
+                with open(info['filepath'], 'wt') as f:
+                    f.write('MODIFIED')
+                return [], info
+
+        run_pp({'keepvideo': False}, ModifierPP)
+        self.assertTrue(os.path.exists(filename), '%s doesn\'t exist' % filename)
+        os.unlink(filename)
+
+    def test_match_filter(self):
+        class FilterYDL(YDL):
+            def __init__(self, *args, **kwargs):
+                super(FilterYDL, self).__init__(*args, **kwargs)
+                self.params['simulate'] = True
+
+            def process_info(self, info_dict):
+                super(YDL, self).process_info(info_dict)
+
+            def _match_entry(self, info_dict, incomplete):
+                res = super(FilterYDL, self)._match_entry(info_dict, incomplete)
+                if res is None:
+                    self.downloaded_info_dicts.append(info_dict)
+                return res
+
+        first = {
+            'id': '1',
+            'url': TEST_URL,
+            'title': 'one',
+            'extractor': 'TEST',
+            'duration': 30,
+            'filesize': 10 * 1024,
+        }
+        second = {
+            'id': '2',
+            'url': TEST_URL,
+            'title': 'two',
+            'extractor': 'TEST',
+            'duration': 10,
+            'description': 'foo',
+            'filesize': 5 * 1024,
+        }
+        videos = [first, second]
+
+        def get_videos(filter_=None):
+            ydl = FilterYDL({'match_filter': filter_})
+            for v in videos:
+                ydl.process_ie_result(v, download=True)
+            return [v['id'] for v in ydl.downloaded_info_dicts]
+
+        res = get_videos()
+        self.assertEqual(res, ['1', '2'])
+
+        def f(v):
+            if v['id'] == '1':
+                return None
+            else:
+                return 'Video id is not 1'
+        res = get_videos(f)
+        self.assertEqual(res, ['1'])
+
+        f = match_filter_func('duration < 30')
+        res = get_videos(f)
+        self.assertEqual(res, ['2'])
+
+        f = match_filter_func('description = foo')
+        res = get_videos(f)
+        self.assertEqual(res, ['2'])
+
+        f = match_filter_func('description =? foo')
+        res = get_videos(f)
+        self.assertEqual(res, ['1', '2'])
+
+        f = match_filter_func('filesize > 5KiB')
+        res = get_videos(f)
+        self.assertEqual(res, ['1'])
+
+    def test_playlist_items_selection(self):
+        entries = [{
+            'id': compat_str(i),
+            'title': compat_str(i),
+            'url': TEST_URL,
+        } for i in range(1, 5)]
+        playlist = {
+            '_type': 'playlist',
+            'id': 'test',
+            'entries': entries,
+            'extractor': 'test:playlist',
+            'extractor_key': 'test:playlist',
+            'webpage_url': 'http://example.com',
+        }
+
+        def get_ids(params):
+            ydl = YDL(params)
+            # make a copy because the dictionary can be modified
+            ydl.process_ie_result(playlist.copy())
+            return [int(v['id']) for v in ydl.downloaded_info_dicts]
+
+        result = get_ids({})
+        self.assertEqual(result, [1, 2, 3, 4])
+
+        result = get_ids({'playlistend': 10})
+        self.assertEqual(result, [1, 2, 3, 4])
+
+        result = get_ids({'playlistend': 2})
+        self.assertEqual(result, [1, 2])
+
+        result = get_ids({'playliststart': 10})
+        self.assertEqual(result, [])
+
+        result = get_ids({'playliststart': 2})
+        self.assertEqual(result, [2, 3, 4])
+
+        result = get_ids({'playlist_items': '2-4'})
+        self.assertEqual(result, [2, 3, 4])
+
+        result = get_ids({'playlist_items': '2,4'})
+        self.assertEqual(result, [2, 4])
+
+        result = get_ids({'playlist_items': '10'})
+        self.assertEqual(result, [])
+
+    def test_urlopen_no_file_protocol(self):
+        # see https://github.com/rg3/youtube-dl/issues/8227
+        ydl = YDL()
+        self.assertRaises(compat_urllib_error.URLError, ydl.urlopen, 'file:///etc/passwd')
+
+    def test_do_not_override_ie_key_in_url_transparent(self):
+        ydl = YDL()
+
+        class Foo1IE(InfoExtractor):
+            _VALID_URL = r'foo1:'
+
+            def _real_extract(self, url):
+                return {
+                    '_type': 'url_transparent',
+                    'url': 'foo2:',
+                    'ie_key': 'Foo2',
+                }
+
+        class Foo2IE(InfoExtractor):
+            _VALID_URL = r'foo2:'
+
+            def _real_extract(self, url):
+                return {
+                    '_type': 'url',
+                    'url': 'foo3:',
+                    'ie_key': 'Foo3',
+                }
+
+        class Foo3IE(InfoExtractor):
+            _VALID_URL = r'foo3:'
+
+            def _real_extract(self, url):
+                return _make_result([{'url': TEST_URL}])
+
+        ydl.add_info_extractor(Foo1IE(ydl))
+        ydl.add_info_extractor(Foo2IE(ydl))
+        ydl.add_info_extractor(Foo3IE(ydl))
+        ydl.extract_info('foo1:')
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['url'], TEST_URL)
+
+
 if __name__ == '__main__':
     unittest.main()

test/test_aes.py

@@ -0,0 +1,55 @@
+#!/usr/bin/env python
+
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl.aes import aes_decrypt, aes_encrypt, aes_cbc_decrypt, aes_decrypt_text
+from youtube_dl.utils import bytes_to_intlist, intlist_to_bytes
+import base64
+
+# the encrypted data can be generate with 'devscripts/generate_aes_testdata.py'
+
+
+class TestAES(unittest.TestCase):
+    def setUp(self):
+        self.key = self.iv = [0x20, 0x15] + 14 * [0]
+        self.secret_msg = b'Secret message goes here'
+
+    def test_encrypt(self):
+        msg = b'message'
+        key = list(range(16))
+        encrypted = aes_encrypt(bytes_to_intlist(msg), key)
+        decrypted = intlist_to_bytes(aes_decrypt(encrypted, key))
+        self.assertEqual(decrypted, msg)
+
+    def test_cbc_decrypt(self):
+        data = bytes_to_intlist(
+            b"\x97\x92+\xe5\x0b\xc3\x18\x91ky9m&\xb3\xb5@\xe6'\xc2\x96.\xc8u\x88\xab9-[\x9e|\xf1\xcd"
+        )
+        decrypted = intlist_to_bytes(aes_cbc_decrypt(data, self.key, self.iv))
+        self.assertEqual(decrypted.rstrip(b'\x08'), self.secret_msg)
+
+    def test_decrypt_text(self):
+        password = intlist_to_bytes(self.key).decode('utf-8')
+        encrypted = base64.b64encode(
+            intlist_to_bytes(self.iv[:8]) +
+            b'\x17\x15\x93\xab\x8d\x80V\xcdV\xe0\t\xcdo\xc2\xa5\xd8ksM\r\xe27N\xae'
+        ).decode('utf-8')
+        decrypted = (aes_decrypt_text(encrypted, password, 16))
+        self.assertEqual(decrypted, self.secret_msg)
+
+        password = intlist_to_bytes(self.key).decode('utf-8')
+        encrypted = base64.b64encode(
+            intlist_to_bytes(self.iv[:8]) +
+            b'\x0b\xe6\xa4\xd9z\x0e\xb8\xb9\xd0\xd4i_\x85\x1d\x99\x98_\xe5\x80\xe7.\xbf\xa5\x83'
+        ).decode('utf-8')
+        decrypted = (aes_decrypt_text(encrypted, password, 32))
+        self.assertEqual(decrypted, self.secret_msg)
+
+if __name__ == '__main__':
+    unittest.main()

test/test_age_restriction.py

@@ -45,11 +45,6 @@ class TestAgeRestriction(unittest.TestCase):
             'http://www.youporn.com/watch/505835/sex-ed-is-it-safe-to-masturbate-daily/',
             '505835.mp4', 2, old_age=25)
 
-    def test_pornotube(self):
-        self._assert_restricted(
-            'http://pornotube.com/c/173/m/1689755/Marilyn-Monroe-Bathing',
-            '1689755.flv', 13)
-
 
 if __name__ == '__main__':
     unittest.main()

test/test_all_urls.py

@@ -14,7 +14,6 @@ from test.helper import gettestcases
 from youtube_dl.extractor import (
     FacebookIE,
     gen_extractors,
-    TwitchIE,
     YoutubeIE,
 )
 
@@ -57,10 +56,10 @@ class TestAllURLsMatching(unittest.TestCase):
         assertChannel('https://www.youtube.com/channel/HCtnHdj3df7iM/videos')
 
     def test_youtube_user_matching(self):
-        self.assertMatch('www.youtube.com/NASAgovVideo/videos', ['youtube:user'])
+        self.assertMatch('http://www.youtube.com/NASAgovVideo/videos', ['youtube:user'])
 
     def test_youtube_feeds(self):
-        self.assertMatch('https://www.youtube.com/feed/watch_later', ['youtube:watch_later'])
+        self.assertMatch('https://www.youtube.com/feed/watch_later', ['youtube:watchlater'])
         self.assertMatch('https://www.youtube.com/feed/subscriptions', ['youtube:subscriptions'])
         self.assertMatch('https://www.youtube.com/feed/recommended', ['youtube:recommended'])
         self.assertMatch('https://www.youtube.com/my_favorites', ['youtube:favorites'])
@@ -72,18 +71,6 @@ class TestAllURLsMatching(unittest.TestCase):
         self.assertMatch('http://www.youtube.com/results?search_query=making+mustard', ['youtube:search_url'])
         self.assertMatch('https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video', ['youtube:search_url'])
 
-    def test_twitch_channelid_matching(self):
-        self.assertTrue(TwitchIE.suitable('twitch.tv/vanillatv'))
-        self.assertTrue(TwitchIE.suitable('www.twitch.tv/vanillatv'))
-        self.assertTrue(TwitchIE.suitable('http://www.twitch.tv/vanillatv'))
-        self.assertTrue(TwitchIE.suitable('http://www.twitch.tv/vanillatv/'))
-
-    def test_twitch_videoid_matching(self):
-        self.assertTrue(TwitchIE.suitable('http://www.twitch.tv/vanillatv/b/328087483'))
-
-    def test_twitch_chapterid_matching(self):
-        self.assertTrue(TwitchIE.suitable('http://www.twitch.tv/tsm_theoddone/c/2349361'))
-
     def test_youtube_extract(self):
         assertExtractId = lambda url, id: self.assertEqual(YoutubeIE.extract_id(url), id)
         assertExtractId('http://www.youtube.com/watch?&v=BaW_jenozKc', 'BaW_jenozKc')
@@ -115,15 +102,13 @@ class TestAllURLsMatching(unittest.TestCase):
         self.assertMatch(':ythistory', ['youtube:history'])
         self.assertMatch(':thedailyshow', ['ComedyCentralShows'])
         self.assertMatch(':tds', ['ComedyCentralShows'])
-        self.assertMatch(':colbertreport', ['ComedyCentralShows'])
-        self.assertMatch(':cr', ['ComedyCentralShows'])
 
     def test_vimeo_matching(self):
-        self.assertMatch('http://vimeo.com/channels/tributes', ['vimeo:channel'])
-        self.assertMatch('http://vimeo.com/channels/31259', ['vimeo:channel'])
-        self.assertMatch('http://vimeo.com/channels/31259/53576664', ['vimeo'])
-        self.assertMatch('http://vimeo.com/user7108434', ['vimeo:user'])
-        self.assertMatch('http://vimeo.com/user7108434/videos', ['vimeo:user'])
+        self.assertMatch('https://vimeo.com/channels/tributes', ['vimeo:channel'])
+        self.assertMatch('https://vimeo.com/channels/31259', ['vimeo:channel'])
+        self.assertMatch('https://vimeo.com/channels/31259/53576664', ['vimeo'])
+        self.assertMatch('https://vimeo.com/user7108434', ['vimeo:user'])
+        self.assertMatch('https://vimeo.com/user7108434/videos', ['vimeo:user'])
         self.assertMatch('https://vimeo.com/user21297594/review/75524534/3c257a1b5d', ['vimeo:review'])
 
     # https://github.com/rg3/youtube-dl/issues/1930
@@ -136,8 +121,8 @@ class TestAllURLsMatching(unittest.TestCase):
 
     def test_pbs(self):
         # https://github.com/rg3/youtube-dl/issues/2350
-        self.assertMatch('http://video.pbs.org/viralplayer/2365173446/', ['PBS'])
-        self.assertMatch('http://video.pbs.org/widget/partnerplayer/980042464/', ['PBS'])
+        self.assertMatch('http://video.pbs.org/viralplayer/2365173446/', ['pbs'])
+        self.assertMatch('http://video.pbs.org/widget/partnerplayer/980042464/', ['pbs'])
 
     def test_yahoo_https(self):
         # https://github.com/rg3/youtube-dl/issues/2701

test/test_compat.py

@@ -13,7 +13,12 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 from youtube_dl.utils import get_filesystem_encoding
 from youtube_dl.compat import (
     compat_getenv,
+    compat_etree_fromstring,
     compat_expanduser,
+    compat_shlex_split,
+    compat_str,
+    compat_urllib_parse_unquote,
+    compat_urllib_parse_unquote_plus,
 )
 
 
@@ -42,5 +47,46 @@ class TestCompat(unittest.TestCase):
             dir(youtube_dl.compat))) - set(['unicode_literals'])
         self.assertEqual(all_names, sorted(present_names))
 
+    def test_compat_urllib_parse_unquote(self):
+        self.assertEqual(compat_urllib_parse_unquote('abc%20def'), 'abc def')
+        self.assertEqual(compat_urllib_parse_unquote('%7e/abc+def'), '~/abc+def')
+        self.assertEqual(compat_urllib_parse_unquote(''), '')
+        self.assertEqual(compat_urllib_parse_unquote('%'), '%')
+        self.assertEqual(compat_urllib_parse_unquote('%%'), '%%')
+        self.assertEqual(compat_urllib_parse_unquote('%%%'), '%%%')
+        self.assertEqual(compat_urllib_parse_unquote('%2F'), '/')
+        self.assertEqual(compat_urllib_parse_unquote('%2f'), '/')
+        self.assertEqual(compat_urllib_parse_unquote('%E6%B4%A5%E6%B3%A2'), '津波')
+        self.assertEqual(
+            compat_urllib_parse_unquote('''<meta property="og:description" content="%E2%96%81%E2%96%82%E2%96%83%E2%96%84%25%E2%96%85%E2%96%86%E2%96%87%E2%96%88" />
+%<a href="https://ar.wikipedia.org/wiki/%D8%AA%D8%B3%D9%88%D9%86%D8%A7%D9%85%D9%8A">%a'''),
+            '''<meta property="og:description" content="▁▂▃▄%▅▆▇█" />
+%<a href="https://ar.wikipedia.org/wiki/تسونامي">%a''')
+        self.assertEqual(
+            compat_urllib_parse_unquote('''%28%5E%E2%97%A3_%E2%97%A2%5E%29%E3%81%A3%EF%B8%BB%E3%83%87%E2%95%90%E4%B8%80    %E2%87%80    %E2%87%80    %E2%87%80    %E2%87%80    %E2%87%80    %E2%86%B6%I%Break%25Things%'''),
+            '''(^◣_◢^)っ︻デ═一    ⇀    ⇀    ⇀    ⇀    ⇀    ↶%I%Break%Things%''')
+
+    def test_compat_urllib_parse_unquote_plus(self):
+        self.assertEqual(compat_urllib_parse_unquote_plus('abc%20def'), 'abc def')
+        self.assertEqual(compat_urllib_parse_unquote_plus('%7e/abc+def'), '~/abc def')
+
+    def test_compat_shlex_split(self):
+        self.assertEqual(compat_shlex_split('-option "one two"'), ['-option', 'one two'])
+
+    def test_compat_etree_fromstring(self):
+        xml = '''
+            <root foo="bar" spam="中文">
+                <normal>foo</normal>
+                <chinese>中文</chinese>
+                <foo><bar>spam</bar></foo>
+            </root>
+        '''
+        doc = compat_etree_fromstring(xml.encode('utf-8'))
+        self.assertTrue(isinstance(doc.attrib['foo'], compat_str))
+        self.assertTrue(isinstance(doc.attrib['spam'], compat_str))
+        self.assertTrue(isinstance(doc.find('normal').text, compat_str))
+        self.assertTrue(isinstance(doc.find('chinese').text, compat_str))
+        self.assertTrue(isinstance(doc.find('foo/bar').text, compat_str))
+
 if __name__ == '__main__':
     unittest.main()

test/test_download.py

@@ -89,7 +89,7 @@ def generator(test_case):
 
         for tc in test_cases:
             info_dict = tc.get('info_dict', {})
-            if not tc.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
+            if not (info_dict.get('id') and info_dict.get('ext')):
                 raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?')
 
         if 'skip' in test_case:
@@ -102,7 +102,7 @@ def generator(test_case):
 
         params = get_params(test_case.get('params', {}))
         if is_playlist and 'playlist' not in test_case:
-            params.setdefault('extract_flat', True)
+            params.setdefault('extract_flat', 'in_playlist')
             params.setdefault('skip_download', True)
 
         ydl = YoutubeDL(params, auto_init=False)
@@ -116,7 +116,7 @@ def generator(test_case):
         expect_warnings(ydl, test_case.get('expected_warnings', []))
 
         def get_tc_filename(tc):
-            return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
+            return ydl.prepare_filename(tc.get('info_dict', {}))
 
         res_dict = None
 
@@ -136,7 +136,9 @@ def generator(test_case):
                     # We're not using .download here sine that is just a shim
                     # for outside error handling, and returns the exit code
                     # instead of the result dict.
-                    res_dict = ydl.extract_info(test_case['url'])
+                    res_dict = ydl.extract_info(
+                        test_case['url'],
+                        force_generic_extractor=params.get('force_generic_extractor', False))
                 except (DownloadError, ExtractorError) as err:
                     # Check if the exception is not a network related one
                     if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503):
@@ -153,9 +155,9 @@ def generator(test_case):
                     break
 
             if is_playlist:
-                self.assertEqual(res_dict['_type'], 'playlist')
+                self.assertTrue(res_dict['_type'] in ['playlist', 'multi_video'])
                 self.assertTrue('entries' in res_dict)
-                expect_info_dict(self, test_case.get('info_dict', {}), res_dict)
+                expect_info_dict(self, res_dict, test_case.get('info_dict', {}))
 
             if 'playlist_mincount' in test_case:
                 assertGreaterEqual(
@@ -204,7 +206,7 @@ def generator(test_case):
                 with io.open(info_json_fn, encoding='utf-8') as infof:
                     info_dict = json.load(infof)
 
-                expect_info_dict(self, tc.get('info_dict', {}), info_dict)
+                expect_info_dict(self, info_dict, tc.get('info_dict', {}))
         finally:
             try_rm_tcs_files()
             if is_playlist and res_dict is not None and res_dict.get('entries'):

test/test_execution.py

@@ -1,4 +1,6 @@
 #!/usr/bin/env python
+# coding: utf-8
+
 from __future__ import unicode_literals
 
 import unittest
@@ -6,6 +8,9 @@ import unittest
 import sys
 import os
 import subprocess
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl.utils import encodeArgument
 
 rootDir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 
@@ -27,5 +32,12 @@ class TestExecution(unittest.TestCase):
     def test_main_exec(self):
         subprocess.check_call([sys.executable, 'youtube_dl/__main__.py', '--version'], cwd=rootDir, stdout=_DEV_NULL)
 
+    def test_cmdline_umlauts(self):
+        p = subprocess.Popen(
+            [sys.executable, 'youtube_dl/__main__.py', encodeArgument('ä'), '--version'],
+            cwd=rootDir, stdout=_DEV_NULL, stderr=subprocess.PIPE)
+        _, stderr = p.communicate()
+        self.assertFalse(stderr)
+
 if __name__ == '__main__':
     unittest.main()

test/test_http.py

@@ -0,0 +1,119 @@
+#!/usr/bin/env python
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl import YoutubeDL
+from youtube_dl.compat import compat_http_server, compat_urllib_request
+import ssl
+import threading
+
+TEST_DIR = os.path.dirname(os.path.abspath(__file__))
+
+
+class HTTPTestRequestHandler(compat_http_server.BaseHTTPRequestHandler):
+    def log_message(self, format, *args):
+        pass
+
+    def do_GET(self):
+        if self.path == '/video.html':
+            self.send_response(200)
+            self.send_header('Content-Type', 'text/html; charset=utf-8')
+            self.end_headers()
+            self.wfile.write(b'<html><video src="/vid.mp4" /></html>')
+        elif self.path == '/vid.mp4':
+            self.send_response(200)
+            self.send_header('Content-Type', 'video/mp4')
+            self.end_headers()
+            self.wfile.write(b'\x00\x00\x00\x00\x20\x66\x74[video]')
+        else:
+            assert False
+
+
+class FakeLogger(object):
+    def debug(self, msg):
+        pass
+
+    def warning(self, msg):
+        pass
+
+    def error(self, msg):
+        pass
+
+
+class TestHTTP(unittest.TestCase):
+    def setUp(self):
+        certfn = os.path.join(TEST_DIR, 'testcert.pem')
+        self.httpd = compat_http_server.HTTPServer(
+            ('localhost', 0), HTTPTestRequestHandler)
+        self.httpd.socket = ssl.wrap_socket(
+            self.httpd.socket, certfile=certfn, server_side=True)
+        self.port = self.httpd.socket.getsockname()[1]
+        self.server_thread = threading.Thread(target=self.httpd.serve_forever)
+        self.server_thread.daemon = True
+        self.server_thread.start()
+
+    def test_nocheckcertificate(self):
+        if sys.version_info >= (2, 7, 9):  # No certificate checking anyways
+            ydl = YoutubeDL({'logger': FakeLogger()})
+            self.assertRaises(
+                Exception,
+                ydl.extract_info, 'https://localhost:%d/video.html' % self.port)
+
+        ydl = YoutubeDL({'logger': FakeLogger(), 'nocheckcertificate': True})
+        r = ydl.extract_info('https://localhost:%d/video.html' % self.port)
+        self.assertEqual(r['url'], 'https://localhost:%d/vid.mp4' % self.port)
+
+
+def _build_proxy_handler(name):
+    class HTTPTestRequestHandler(compat_http_server.BaseHTTPRequestHandler):
+        proxy_name = name
+
+        def log_message(self, format, *args):
+            pass
+
+        def do_GET(self):
+            self.send_response(200)
+            self.send_header('Content-Type', 'text/plain; charset=utf-8')
+            self.end_headers()
+            self.wfile.write('{self.proxy_name}: {self.path}'.format(self=self).encode('utf-8'))
+    return HTTPTestRequestHandler
+
+
+class TestProxy(unittest.TestCase):
+    def setUp(self):
+        self.proxy = compat_http_server.HTTPServer(
+            ('localhost', 0), _build_proxy_handler('normal'))
+        self.port = self.proxy.socket.getsockname()[1]
+        self.proxy_thread = threading.Thread(target=self.proxy.serve_forever)
+        self.proxy_thread.daemon = True
+        self.proxy_thread.start()
+
+        self.cn_proxy = compat_http_server.HTTPServer(
+            ('localhost', 0), _build_proxy_handler('cn'))
+        self.cn_port = self.cn_proxy.socket.getsockname()[1]
+        self.cn_proxy_thread = threading.Thread(target=self.cn_proxy.serve_forever)
+        self.cn_proxy_thread.daemon = True
+        self.cn_proxy_thread.start()
+
+    def test_proxy(self):
+        cn_proxy = 'localhost:{0}'.format(self.cn_port)
+        ydl = YoutubeDL({
+            'proxy': 'localhost:{0}'.format(self.port),
+            'cn_verification_proxy': cn_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)
+        response = ydl.urlopen(req).read().decode('utf-8')
+        self.assertEqual(response, 'cn: {0}'.format(url))
+
+if __name__ == '__main__':
+    unittest.main()

test/test_jsinterp.py

@@ -0,0 +1,109 @@
+#!/usr/bin/env python
+
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl.jsinterp import JSInterpreter
+
+
+class TestJSInterpreter(unittest.TestCase):
+    def test_basic(self):
+        jsi = JSInterpreter('function x(){;}')
+        self.assertEqual(jsi.call_function('x'), None)
+
+        jsi = JSInterpreter('function x3(){return 42;}')
+        self.assertEqual(jsi.call_function('x3'), 42)
+
+        jsi = JSInterpreter('var x5 = function(){return 42;}')
+        self.assertEqual(jsi.call_function('x5'), 42)
+
+    def test_calc(self):
+        jsi = JSInterpreter('function x4(a){return 2*a+1;}')
+        self.assertEqual(jsi.call_function('x4', 3), 7)
+
+    def test_empty_return(self):
+        jsi = JSInterpreter('function f(){return; y()}')
+        self.assertEqual(jsi.call_function('f'), None)
+
+    def test_morespace(self):
+        jsi = JSInterpreter('function x (a) { return 2 * a + 1 ; }')
+        self.assertEqual(jsi.call_function('x', 3), 7)
+
+        jsi = JSInterpreter('function f () { x =  2  ; return x; }')
+        self.assertEqual(jsi.call_function('f'), 2)
+
+    def test_strange_chars(self):
+        jsi = JSInterpreter('function $_xY1 ($_axY1) { var $_axY2 = $_axY1 + 1; return $_axY2; }')
+        self.assertEqual(jsi.call_function('$_xY1', 20), 21)
+
+    def test_operators(self):
+        jsi = JSInterpreter('function f(){return 1 << 5;}')
+        self.assertEqual(jsi.call_function('f'), 32)
+
+        jsi = JSInterpreter('function f(){return 19 & 21;}')
+        self.assertEqual(jsi.call_function('f'), 17)
+
+        jsi = JSInterpreter('function f(){return 11 >> 2;}')
+        self.assertEqual(jsi.call_function('f'), 2)
+
+    def test_array_access(self):
+        jsi = JSInterpreter('function f(){var x = [1,2,3]; x[0] = 4; x[0] = 5; x[2] = 7; return x;}')
+        self.assertEqual(jsi.call_function('f'), [5, 2, 7])
+
+    def test_parens(self):
+        jsi = JSInterpreter('function f(){return (1) + (2) * ((( (( (((((3)))))) )) ));}')
+        self.assertEqual(jsi.call_function('f'), 7)
+
+        jsi = JSInterpreter('function f(){return (1 + 2) * 3;}')
+        self.assertEqual(jsi.call_function('f'), 9)
+
+    def test_assignments(self):
+        jsi = JSInterpreter('function f(){var x = 20; x = 30 + 1; return x;}')
+        self.assertEqual(jsi.call_function('f'), 31)
+
+        jsi = JSInterpreter('function f(){var x = 20; x += 30 + 1; return x;}')
+        self.assertEqual(jsi.call_function('f'), 51)
+
+        jsi = JSInterpreter('function f(){var x = 20; x -= 30 + 1; return x;}')
+        self.assertEqual(jsi.call_function('f'), -11)
+
+    def test_comments(self):
+        'Skipping: Not yet fully implemented'
+        return
+        jsi = JSInterpreter('''
+        function x() {
+            var x = /* 1 + */ 2;
+            var y = /* 30
+            * 40 */ 50;
+            return x + y;
+        }
+        ''')
+        self.assertEqual(jsi.call_function('x'), 52)
+
+        jsi = JSInterpreter('''
+        function f() {
+            var x = "/*";
+            var y = 1 /* comment */ + 2;
+            return y;
+        }
+        ''')
+        self.assertEqual(jsi.call_function('f'), 3)
+
+    def test_precedence(self):
+        jsi = JSInterpreter('''
+        function x() {
+            var a = [10, 20, 30, 40, 50];
+            var b = 6;
+            a[0]=a[b%a.length];
+            return a;
+        }''')
+        self.assertEqual(jsi.call_function('x'), [20, 20, 30, 40, 50])
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_netrc.py

@@ -0,0 +1,26 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+
+from youtube_dl.extractor import (
+    gen_extractors,
+)
+
+
+class TestNetRc(unittest.TestCase):
+    def test_netrc_present(self):
+        for ie in gen_extractors():
+            if not hasattr(ie, '_login'):
+                continue
+            self.assertTrue(
+                hasattr(ie, '_NETRC_MACHINE'),
+                'Extractor %s supports login, but is missing a _NETRC_MACHINE property' % ie.IE_NAME)
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_postprocessors.py

@@ -0,0 +1,17 @@
+#!/usr/bin/env python
+
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl.postprocessor import MetadataFromTitlePP
+
+
+class TestMetadataFromTitle(unittest.TestCase):
+    def test_format_to_regex(self):
+        pp = MetadataFromTitlePP(None, '%(title)s - %(artist)s')
+        self.assertEqual(pp._titleregex, '(?P<title>.+)\ \-\ (?P<artist>.+)')

test/test_subtitles.py

@@ -11,12 +11,23 @@ from test.helper import FakeYDL, md5
 
 
 from youtube_dl.extractor import (
-    BlipTVIE,
     YoutubeIE,
     DailymotionIE,
     TEDIE,
     VimeoIE,
     WallaIE,
+    CeskaTelevizeIE,
+    LyndaIE,
+    NPOIE,
+    ComedyCentralIE,
+    NRKTVIE,
+    RaiTVIE,
+    VikiIE,
+    ThePlatformIE,
+    ThePlatformFeedIE,
+    RTVEALaCartaIE,
+    FunnyOrDieIE,
+    DemocracynowIE,
 )
 
 
@@ -26,48 +37,44 @@ class BaseTestSubtitles(unittest.TestCase):
 
     def setUp(self):
         self.DL = FakeYDL()
-        self.ie = self.IE(self.DL)
+        self.ie = self.IE()
+        self.DL.add_info_extractor(self.ie)
 
     def getInfoDict(self):
-        info_dict = self.ie.extract(self.url)
+        info_dict = self.DL.extract_info(self.url, download=False)
         return info_dict
 
     def getSubtitles(self):
         info_dict = self.getInfoDict()
-        return info_dict['subtitles']
+        subtitles = info_dict['requested_subtitles']
+        if not subtitles:
+            return subtitles
+        for sub_info in subtitles.values():
+            if sub_info.get('data') is None:
+                uf = self.DL.urlopen(sub_info['url'])
+                sub_info['data'] = uf.read().decode('utf-8')
+        return dict((l, sub_info['data']) for l, sub_info in subtitles.items())
 
 
 class TestYoutubeSubtitles(BaseTestSubtitles):
     url = 'QRS8MkLhQmM'
     IE = YoutubeIE
 
-    def test_youtube_no_writesubtitles(self):
-        self.DL.params['writesubtitles'] = False
-        subtitles = self.getSubtitles()
-        self.assertEqual(subtitles, None)
-
-    def test_youtube_subtitles(self):
-        self.DL.params['writesubtitles'] = True
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '4cd9278a35ba2305f47354ee13472260')
-
-    def test_youtube_subtitles_lang(self):
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['subtitleslangs'] = ['it']
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['it']), '164a51f16f260476a05b50fe4c2f161d')
-
     def test_youtube_allsubtitles(self):
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
         self.assertEqual(len(subtitles.keys()), 13)
+        self.assertEqual(md5(subtitles['en']), '3cb210999d3e021bd6c7f0ea751eab06')
+        self.assertEqual(md5(subtitles['it']), '6d752b98c31f1cf8d597050c7a2cb4b5')
+        for lang in ['fr', 'de']:
+            self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
 
-    def test_youtube_subtitles_sbv_format(self):
+    def test_youtube_subtitles_ttml_format(self):
         self.DL.params['writesubtitles'] = True
-        self.DL.params['subtitlesformat'] = 'sbv'
+        self.DL.params['subtitlesformat'] = 'ttml'
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '13aeaa0c245a8bed9a451cb643e3ad8b')
+        self.assertEqual(md5(subtitles['en']), 'e306f8c42842f723447d9f63ad65df54')
 
     def test_youtube_subtitles_vtt_format(self):
         self.DL.params['writesubtitles'] = True
@@ -75,12 +82,6 @@ class TestYoutubeSubtitles(BaseTestSubtitles):
         subtitles = self.getSubtitles()
         self.assertEqual(md5(subtitles['en']), '3cb210999d3e021bd6c7f0ea751eab06')
 
-    def test_youtube_list_subtitles(self):
-        self.DL.expect_warning('Video doesn\'t have automatic captions')
-        self.DL.params['listsubtitles'] = True
-        info_dict = self.getInfoDict()
-        self.assertEqual(info_dict, None)
-
     def test_youtube_automatic_captions(self):
         self.url = '8YoUxe5ncPo'
         self.DL.params['writeautomaticsub'] = True
@@ -88,61 +89,36 @@ class TestYoutubeSubtitles(BaseTestSubtitles):
         subtitles = self.getSubtitles()
         self.assertTrue(subtitles['it'] is not None)
 
+    def test_youtube_translated_subtitles(self):
+        # This video has a subtitles track, which can be translated
+        self.url = 'Ky9eprVWzlI'
+        self.DL.params['writeautomaticsub'] = True
+        self.DL.params['subtitleslangs'] = ['it']
+        subtitles = self.getSubtitles()
+        self.assertTrue(subtitles['it'] is not None)
+
     def test_youtube_nosubtitles(self):
         self.DL.expect_warning('video doesn\'t have subtitles')
         self.url = 'n5BB19UTcdA'
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
-        self.assertEqual(len(subtitles), 0)
-
-    def test_youtube_multiple_langs(self):
-        self.url = 'QRS8MkLhQmM'
-        self.DL.params['writesubtitles'] = True
-        langs = ['it', 'fr', 'de']
-        self.DL.params['subtitleslangs'] = langs
-        subtitles = self.getSubtitles()
-        for lang in langs:
-            self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
+        self.assertFalse(subtitles)
 
 
 class TestDailymotionSubtitles(BaseTestSubtitles):
     url = 'http://www.dailymotion.com/video/xczg00'
     IE = DailymotionIE
 
-    def test_no_writesubtitles(self):
-        subtitles = self.getSubtitles()
-        self.assertEqual(subtitles, None)
-
-    def test_subtitles(self):
-        self.DL.params['writesubtitles'] = True
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '976553874490cba125086bbfea3ff76f')
-
-    def test_subtitles_lang(self):
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['subtitleslangs'] = ['fr']
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['fr']), '594564ec7d588942e384e920e5341792')
-
     def test_allsubtitles(self):
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
-        self.assertEqual(len(subtitles.keys()), 5)
-
-    def test_list_subtitles(self):
-        self.DL.expect_warning('Automatic Captions not supported by this server')
-        self.DL.params['listsubtitles'] = True
-        info_dict = self.getInfoDict()
-        self.assertEqual(info_dict, None)
-
-    def test_automatic_captions(self):
-        self.DL.expect_warning('Automatic Captions not supported by this server')
-        self.DL.params['writeautomaticsub'] = True
-        self.DL.params['subtitleslang'] = ['en']
-        subtitles = self.getSubtitles()
-        self.assertTrue(len(subtitles.keys()) == 0)
+        self.assertTrue(len(subtitles.keys()) >= 6)
+        self.assertEqual(md5(subtitles['en']), '976553874490cba125086bbfea3ff76f')
+        self.assertEqual(md5(subtitles['fr']), '594564ec7d588942e384e920e5341792')
+        for lang in ['es', 'fr', 'de']:
+            self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
 
     def test_nosubtitles(self):
         self.DL.expect_warning('video doesn\'t have subtitles')
@@ -150,120 +126,35 @@ class TestDailymotionSubtitles(BaseTestSubtitles):
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
-        self.assertEqual(len(subtitles), 0)
-
-    def test_multiple_langs(self):
-        self.DL.params['writesubtitles'] = True
-        langs = ['es', 'fr', 'de']
-        self.DL.params['subtitleslangs'] = langs
-        subtitles = self.getSubtitles()
-        for lang in langs:
-            self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
+        self.assertFalse(subtitles)
 
 
 class TestTedSubtitles(BaseTestSubtitles):
     url = 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html'
     IE = TEDIE
 
-    def test_no_writesubtitles(self):
-        subtitles = self.getSubtitles()
-        self.assertEqual(subtitles, None)
-
-    def test_subtitles(self):
-        self.DL.params['writesubtitles'] = True
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '4262c1665ff928a2dada178f62cb8d14')
-
-    def test_subtitles_lang(self):
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['subtitleslangs'] = ['fr']
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['fr']), '66a63f7f42c97a50f8c0e90bc7797bb5')
-
     def test_allsubtitles(self):
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
         self.assertTrue(len(subtitles.keys()) >= 28)
-
-    def test_list_subtitles(self):
-        self.DL.expect_warning('Automatic Captions not supported by this server')
-        self.DL.params['listsubtitles'] = True
-        info_dict = self.getInfoDict()
-        self.assertEqual(info_dict, None)
-
-    def test_automatic_captions(self):
-        self.DL.expect_warning('Automatic Captions not supported by this server')
-        self.DL.params['writeautomaticsub'] = True
-        self.DL.params['subtitleslang'] = ['en']
-        subtitles = self.getSubtitles()
-        self.assertTrue(len(subtitles.keys()) == 0)
-
-    def test_multiple_langs(self):
-        self.DL.params['writesubtitles'] = True
-        langs = ['es', 'fr', 'de']
-        self.DL.params['subtitleslangs'] = langs
-        subtitles = self.getSubtitles()
-        for lang in langs:
+        self.assertEqual(md5(subtitles['en']), '4262c1665ff928a2dada178f62cb8d14')
+        self.assertEqual(md5(subtitles['fr']), '66a63f7f42c97a50f8c0e90bc7797bb5')
+        for lang in ['es', 'fr', 'de']:
             self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
 
 
-class TestBlipTVSubtitles(BaseTestSubtitles):
-    url = 'http://blip.tv/a/a-6603250'
-    IE = BlipTVIE
-
-    def test_list_subtitles(self):
-        self.DL.expect_warning('Automatic Captions not supported by this server')
-        self.DL.params['listsubtitles'] = True
-        info_dict = self.getInfoDict()
-        self.assertEqual(info_dict, None)
-
-    def test_allsubtitles(self):
-        self.DL.expect_warning('Automatic Captions not supported by this server')
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['allsubtitles'] = True
-        subtitles = self.getSubtitles()
-        self.assertEqual(set(subtitles.keys()), set(['en']))
-        self.assertEqual(md5(subtitles['en']), '5b75c300af65fe4476dff79478bb93e4')
-
-
 class TestVimeoSubtitles(BaseTestSubtitles):
     url = 'http://vimeo.com/76979871'
     IE = VimeoIE
 
-    def test_no_writesubtitles(self):
-        subtitles = self.getSubtitles()
-        self.assertEqual(subtitles, None)
-
-    def test_subtitles(self):
-        self.DL.params['writesubtitles'] = True
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '26399116d23ae3cf2c087cea94bc43b4')
-
-    def test_subtitles_lang(self):
-        self.DL.params['writesubtitles'] = True
-        self.DL.params['subtitleslangs'] = ['fr']
-        subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['fr']), 'b6191146a6c5d3a452244d853fde6dc8')
-
     def test_allsubtitles(self):
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
         self.assertEqual(set(subtitles.keys()), set(['de', 'en', 'es', 'fr']))
-
-    def test_list_subtitles(self):
-        self.DL.expect_warning('Automatic Captions not supported by this server')
-        self.DL.params['listsubtitles'] = True
-        info_dict = self.getInfoDict()
-        self.assertEqual(info_dict, None)
-
-    def test_automatic_captions(self):
-        self.DL.expect_warning('Automatic Captions not supported by this server')
-        self.DL.params['writeautomaticsub'] = True
-        self.DL.params['subtitleslang'] = ['en']
-        subtitles = self.getSubtitles()
-        self.assertTrue(len(subtitles.keys()) == 0)
+        self.assertEqual(md5(subtitles['en']), '8062383cf4dec168fc40a088aa6d5888')
+        self.assertEqual(md5(subtitles['fr']), 'b6191146a6c5d3a452244d853fde6dc8')
 
     def test_nosubtitles(self):
         self.DL.expect_warning('video doesn\'t have subtitles')
@@ -271,27 +162,13 @@ class TestVimeoSubtitles(BaseTestSubtitles):
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
-        self.assertEqual(len(subtitles), 0)
-
-    def test_multiple_langs(self):
-        self.DL.params['writesubtitles'] = True
-        langs = ['es', 'fr', 'de']
-        self.DL.params['subtitleslangs'] = langs
-        subtitles = self.getSubtitles()
-        for lang in langs:
-            self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
+        self.assertFalse(subtitles)
 
 
 class TestWallaSubtitles(BaseTestSubtitles):
     url = 'http://vod.walla.co.il/movie/2705958/the-yes-men'
     IE = WallaIE
 
-    def test_list_subtitles(self):
-        self.DL.expect_warning('Automatic Captions not supported by this server')
-        self.DL.params['listsubtitles'] = True
-        info_dict = self.getInfoDict()
-        self.assertEqual(info_dict, None)
-
     def test_allsubtitles(self):
         self.DL.expect_warning('Automatic Captions not supported by this server')
         self.DL.params['writesubtitles'] = True
@@ -306,7 +183,175 @@ class TestWallaSubtitles(BaseTestSubtitles):
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
-        self.assertEqual(len(subtitles), 0)
+        self.assertFalse(subtitles)
+
+
+class TestCeskaTelevizeSubtitles(BaseTestSubtitles):
+    url = 'http://www.ceskatelevize.cz/ivysilani/10600540290-u6-uzasny-svet-techniky'
+    IE = CeskaTelevizeIE
+
+    def test_allsubtitles(self):
+        self.DL.expect_warning('Automatic Captions not supported by this server')
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['cs']))
+        self.assertTrue(len(subtitles['cs']) > 20000)
+
+    def test_nosubtitles(self):
+        self.DL.expect_warning('video doesn\'t have subtitles')
+        self.url = 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220'
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertFalse(subtitles)
+
+
+class TestLyndaSubtitles(BaseTestSubtitles):
+    url = 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html'
+    IE = LyndaIE
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['en']))
+        self.assertEqual(md5(subtitles['en']), '09bbe67222259bed60deaa26997d73a7')
+
+
+class TestNPOSubtitles(BaseTestSubtitles):
+    url = 'http://www.npo.nl/nos-journaal/28-08-2014/POW_00722860'
+    IE = NPOIE
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['nl']))
+        self.assertEqual(md5(subtitles['nl']), 'fc6435027572b63fb4ab143abd5ad3f4')
+
+
+class TestMTVSubtitles(BaseTestSubtitles):
+    url = 'http://www.cc.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother'
+    IE = ComedyCentralIE
+
+    def getInfoDict(self):
+        return super(TestMTVSubtitles, self).getInfoDict()['entries'][0]
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['en']))
+        self.assertEqual(md5(subtitles['en']), 'b9f6ca22a6acf597ec76f61749765e65')
+
+
+class TestNRKSubtitles(BaseTestSubtitles):
+    url = 'http://tv.nrk.no/serie/ikke-gjoer-dette-hjemme/DMPV73000411/sesong-2/episode-1'
+    IE = NRKTVIE
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['no']))
+        self.assertEqual(md5(subtitles['no']), '544fa917d3197fcbee64634559221cc2')
+
+
+class TestRaiSubtitles(BaseTestSubtitles):
+    url = 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html'
+    IE = RaiTVIE
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['it']))
+        self.assertEqual(md5(subtitles['it']), 'b1d90a98755126b61e667567a1f6680a')
+
+
+class TestVikiSubtitles(BaseTestSubtitles):
+    url = 'http://www.viki.com/videos/1060846v-punch-episode-18'
+    IE = VikiIE
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['en']))
+        self.assertEqual(md5(subtitles['en']), '53cb083a5914b2d84ef1ab67b880d18a')
+
+
+class TestThePlatformSubtitles(BaseTestSubtitles):
+    # from http://www.3playmedia.com/services-features/tools/integrations/theplatform/
+    # (see http://theplatform.com/about/partners/type/subtitles-closed-captioning/)
+    url = 'theplatform:JFUjUE1_ehvq'
+    IE = ThePlatformIE
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['en']))
+        self.assertEqual(md5(subtitles['en']), '97e7670cbae3c4d26ae8bcc7fdd78d4b')
+
+
+class TestThePlatformFeedSubtitles(BaseTestSubtitles):
+    url = 'http://feed.theplatform.com/f/7wvmTC/msnbc_video-p-test?form=json&pretty=true&range=-40&byGuid=n_hardball_5biden_140207'
+    IE = ThePlatformFeedIE
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['en']))
+        self.assertEqual(md5(subtitles['en']), '48649a22e82b2da21c9a67a395eedade')
+
+
+class TestRtveSubtitles(BaseTestSubtitles):
+    url = 'http://www.rtve.es/alacarta/videos/los-misterios-de-laura/misterios-laura-capitulo-32-misterio-del-numero-17-2-parte/2428621/'
+    IE = RTVEALaCartaIE
+
+    def test_allsubtitles(self):
+        print('Skipping, only available from Spain')
+        return
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['es']))
+        self.assertEqual(md5(subtitles['es']), '69e70cae2d40574fb7316f31d6eb7fca')
+
+
+class TestFunnyOrDieSubtitles(BaseTestSubtitles):
+    url = 'http://www.funnyordie.com/videos/224829ff6d/judd-apatow-will-direct-your-vine'
+    IE = FunnyOrDieIE
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['en']))
+        self.assertEqual(md5(subtitles['en']), 'c5593c193eacd353596c11c2d4f9ecc4')
+
+
+class TestDemocracynowSubtitles(BaseTestSubtitles):
+    url = 'http://www.democracynow.org/shows/2015/7/3'
+    IE = DemocracynowIE
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['en']))
+        self.assertEqual(md5(subtitles['en']), 'acaca989e24a9e45a6719c9b3d60815c')
+
+    def test_subtitles_in_page(self):
+        self.url = 'http://www.democracynow.org/2015/7/3/this_flag_comes_down_today_bree'
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['en']))
+        self.assertEqual(md5(subtitles['en']), 'acaca989e24a9e45a6719c9b3d60815c')
 
 
 if __name__ == '__main__':

test/test_swfinterp.py

@@ -34,8 +34,8 @@ def _make_testfunc(testfile):
     def test_func(self):
         as_file = os.path.join(TEST_DIR, testfile)
         swf_file = os.path.join(TEST_DIR, test_id + '.swf')
-        if ((not os.path.exists(swf_file))
-                or os.path.getmtime(swf_file) < os.path.getmtime(as_file)):
+        if ((not os.path.exists(swf_file)) or
+                os.path.getmtime(swf_file) < os.path.getmtime(as_file)):
             # Recompile
             try:
                 subprocess.check_call([

test/test_unicode_literals.py

@@ -1,9 +1,13 @@
 from __future__ import unicode_literals
 
-import io
+# Allow direct execution
 import os
-import re
+import sys
 import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+import io
+import re
 
 rootDir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 
@@ -13,10 +17,22 @@ IGNORED_FILES = [
     'buildserver.py',
 ]
 
+IGNORED_DIRS = [
+    '.git',
+    '.tox',
+]
+
+from test.helper import assertRegexpMatches
+
 
 class TestUnicodeLiterals(unittest.TestCase):
     def test_all_files(self):
-        for dirpath, _, filenames in os.walk(rootDir):
+        for dirpath, dirnames, filenames in os.walk(rootDir):
+            for ignore_dir in IGNORED_DIRS:
+                if ignore_dir in dirnames:
+                    # If we remove the directory from dirnames os.walk won't
+                    # recurse into it
+                    dirnames.remove(ignore_dir)
             for basename in filenames:
                 if not basename.endswith('.py'):
                     continue
@@ -29,9 +45,10 @@ class TestUnicodeLiterals(unittest.TestCase):
 
                 if "'" not in code and '"' not in code:
                     continue
-                self.assertRegexpMatches(
+                assertRegexpMatches(
+                    self,
                     code,
-                    r'(?:#.*\n*)?from __future__ import (?:[a-z_]+,\s*)*unicode_literals',
+                    r'(?:(?:#.*?|\s*)\n)*from __future__ import (?:[a-z_]+,\s*)*unicode_literals',
                     'unicode_literals import  missing in %s' % fn)
 
                 m = re.search(r'(?<=\s)u[\'"](?!\)|,|$)', code)

test/test_update.py

@@ -0,0 +1,30 @@
+#!/usr/bin/env python
+
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+
+import json
+from youtube_dl.update import rsa_verify
+
+
+class TestUpdate(unittest.TestCase):
+    def test_rsa_verify(self):
+        UPDATES_RSA_KEY = (0x9d60ee4d8f805312fdb15a62f87b95bd66177b91df176765d13514a0f1754bcd2057295c5b6f1d35daa6742c3ffc9a82d3e118861c207995a8031e151d863c9927e304576bc80692bc8e094896fcf11b66f3e29e04e3a71e9a11558558acea1840aec37fc396fb6b65dc81a1c4144e03bd1c011de62e3f1357b327d08426fe93, 65537)
+        with open(os.path.join(os.path.dirname(os.path.abspath(__file__)), 'versions.json'), 'rb') as f:
+            versions_info = f.read().decode()
+        versions_info = json.loads(versions_info)
+        signature = versions_info['signature']
+        del versions_info['signature']
+        self.assertTrue(rsa_verify(
+            json.dumps(versions_info, sort_keys=True).encode('utf-8'),
+            signature, UPDATES_RSA_KEY))
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_utils.py

@@ -16,39 +16,64 @@ import json
 import xml.etree.ElementTree
 
 from youtube_dl.utils import (
+    age_restricted,
+    args_to_str,
     clean_html,
     DateRange,
+    detect_exe_version,
+    determine_ext,
+    dict_get,
+    encode_compat_str,
     encodeFilename,
+    escape_rfc3986,
+    escape_url,
+    ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,
-    orderedSet,
-    OnDemandPagedList,
     InAdvancePagedList,
+    intlist_to_bytes,
+    is_html,
+    js_to_json,
+    limit_length,
+    OnDemandPagedList,
+    orderedSet,
     parse_duration,
+    parse_filesize,
+    parse_iso8601,
     read_batch_urls,
     sanitize_filename,
+    sanitize_path,
+    prepend_extension,
+    replace_extension,
+    remove_quotes,
     shell_quote,
     smuggle_url,
     str_to_int,
+    strip_jsonp,
     struct_unpack,
     timeconvert,
     unescapeHTML,
     unified_strdate,
     unsmuggle_url,
+    uppercase_escape,
+    lowercase_escape,
     url_basename,
     urlencode_postdata,
-    xpath_with_ns,
-    parse_iso8601,
-    strip_jsonp,
-    uppercase_escape,
-    limit_length,
-    escape_rfc3986,
-    escape_url,
-    js_to_json,
-    intlist_to_bytes,
-    args_to_str,
-    parse_filesize,
     version_tuple,
+    xpath_with_ns,
+    xpath_element,
+    xpath_text,
+    xpath_attr,
+    render_table,
+    match_str,
+    parse_dfxp_time_expr,
+    dfxp2srt,
+    cli_option,
+    cli_valueless_option,
+    cli_bool_option,
+)
+from youtube_dl.compat import (
+    compat_etree_fromstring,
 )
 
 
@@ -77,6 +102,15 @@ class TestUtil(unittest.TestCase):
         tests = '\u043a\u0438\u0440\u0438\u043b\u043b\u0438\u0446\u0430'
         self.assertEqual(sanitize_filename(tests), tests)
 
+        self.assertEqual(
+            sanitize_filename('New World record at 0:12:34'),
+            'New World record at 0_12_34')
+
+        self.assertEqual(sanitize_filename('--gasdgf'), '_-gasdgf')
+        self.assertEqual(sanitize_filename('--gasdgf', is_id=True), '--gasdgf')
+        self.assertEqual(sanitize_filename('.gasdgf'), 'gasdgf')
+        self.assertEqual(sanitize_filename('.gasdgf', is_id=True), '.gasdgf')
+
         forbidden = '"\0\\/'
         for fc in forbidden:
             for fbc in forbidden:
@@ -117,6 +151,67 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(sanitize_filename('_BD_eEpuzXw', is_id=True), '_BD_eEpuzXw')
         self.assertEqual(sanitize_filename('N0Y__7-UOdI', is_id=True), 'N0Y__7-UOdI')
 
+    def test_sanitize_path(self):
+        if sys.platform != 'win32':
+            return
+
+        self.assertEqual(sanitize_path('abc'), 'abc')
+        self.assertEqual(sanitize_path('abc/def'), 'abc\\def')
+        self.assertEqual(sanitize_path('abc\\def'), 'abc\\def')
+        self.assertEqual(sanitize_path('abc|def'), 'abc#def')
+        self.assertEqual(sanitize_path('<>:"|?*'), '#######')
+        self.assertEqual(sanitize_path('C:/abc/def'), 'C:\\abc\\def')
+        self.assertEqual(sanitize_path('C?:/abc/def'), 'C##\\abc\\def')
+
+        self.assertEqual(sanitize_path('\\\\?\\UNC\\ComputerName\\abc'), '\\\\?\\UNC\\ComputerName\\abc')
+        self.assertEqual(sanitize_path('\\\\?\\UNC/ComputerName/abc'), '\\\\?\\UNC\\ComputerName\\abc')
+
+        self.assertEqual(sanitize_path('\\\\?\\C:\\abc'), '\\\\?\\C:\\abc')
+        self.assertEqual(sanitize_path('\\\\?\\C:/abc'), '\\\\?\\C:\\abc')
+        self.assertEqual(sanitize_path('\\\\?\\C:\\ab?c\\de:f'), '\\\\?\\C:\\ab#c\\de#f')
+        self.assertEqual(sanitize_path('\\\\?\\C:\\abc'), '\\\\?\\C:\\abc')
+
+        self.assertEqual(
+            sanitize_path('youtube/%(uploader)s/%(autonumber)s-%(title)s-%(upload_date)s.%(ext)s'),
+            'youtube\\%(uploader)s\\%(autonumber)s-%(title)s-%(upload_date)s.%(ext)s')
+
+        self.assertEqual(
+            sanitize_path('youtube/TheWreckingYard ./00001-Not bad, Especially for Free! (1987 Yamaha 700)-20141116.mp4.part'),
+            'youtube\\TheWreckingYard #\\00001-Not bad, Especially for Free! (1987 Yamaha 700)-20141116.mp4.part')
+        self.assertEqual(sanitize_path('abc/def...'), 'abc\\def..#')
+        self.assertEqual(sanitize_path('abc.../def'), 'abc..#\\def')
+        self.assertEqual(sanitize_path('abc.../def...'), 'abc..#\\def..#')
+
+        self.assertEqual(sanitize_path('../abc'), '..\\abc')
+        self.assertEqual(sanitize_path('../../abc'), '..\\..\\abc')
+        self.assertEqual(sanitize_path('./abc'), 'abc')
+        self.assertEqual(sanitize_path('./../abc'), '..\\abc')
+
+    def test_prepend_extension(self):
+        self.assertEqual(prepend_extension('abc.ext', 'temp'), 'abc.temp.ext')
+        self.assertEqual(prepend_extension('abc.ext', 'temp', 'ext'), 'abc.temp.ext')
+        self.assertEqual(prepend_extension('abc.unexpected_ext', 'temp', 'ext'), 'abc.unexpected_ext.temp')
+        self.assertEqual(prepend_extension('abc', 'temp'), 'abc.temp')
+        self.assertEqual(prepend_extension('.abc', 'temp'), '.abc.temp')
+        self.assertEqual(prepend_extension('.abc.ext', 'temp'), '.abc.temp.ext')
+
+    def test_replace_extension(self):
+        self.assertEqual(replace_extension('abc.ext', 'temp'), 'abc.temp')
+        self.assertEqual(replace_extension('abc.ext', 'temp', 'ext'), 'abc.temp')
+        self.assertEqual(replace_extension('abc.unexpected_ext', 'temp', 'ext'), 'abc.unexpected_ext.temp')
+        self.assertEqual(replace_extension('abc', 'temp'), 'abc.temp')
+        self.assertEqual(replace_extension('.abc', 'temp'), '.abc.temp')
+        self.assertEqual(replace_extension('.abc.ext', 'temp'), '.abc.temp')
+
+    def test_remove_quotes(self):
+        self.assertEqual(remove_quotes(None), None)
+        self.assertEqual(remove_quotes('"'), '"')
+        self.assertEqual(remove_quotes("'"), "'")
+        self.assertEqual(remove_quotes(';'), ';')
+        self.assertEqual(remove_quotes('";'), '";')
+        self.assertEqual(remove_quotes('""'), '')
+        self.assertEqual(remove_quotes('";"'), ';')
+
     def test_ordered_set(self):
         self.assertEqual(orderedSet([1, 1, 2, 3, 4, 4, 5, 6, 7, 3, 5]), [1, 2, 3, 4, 5, 6, 7])
         self.assertEqual(orderedSet([]), [])
@@ -126,8 +221,10 @@ class TestUtil(unittest.TestCase):
 
     def test_unescape_html(self):
         self.assertEqual(unescapeHTML('%20;'), '%20;')
-        self.assertEqual(
-            unescapeHTML('&eacute;'), 'é')
+        self.assertEqual(unescapeHTML('&#x2F;'), '/')
+        self.assertEqual(unescapeHTML('&#47;'), '/')
+        self.assertEqual(unescapeHTML('&eacute;'), 'é')
+        self.assertEqual(unescapeHTML('&#2013266066;'), '&#2013266066;')
 
     def test_daterange(self):
         _20century = DateRange("19000101", "20000101")
@@ -142,8 +239,24 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(unified_strdate('8/7/2009'), '20090708')
         self.assertEqual(unified_strdate('Dec 14, 2012'), '20121214')
         self.assertEqual(unified_strdate('2012/10/11 01:56:38 +0000'), '20121011')
+        self.assertEqual(unified_strdate('1968 12 10'), '19681210')
         self.assertEqual(unified_strdate('1968-12-10'), '19681210')
         self.assertEqual(unified_strdate('28/01/2014 21:00:00 +0100'), '20140128')
+        self.assertEqual(
+            unified_strdate('11/26/2014 11:30:00 AM PST', day_first=False),
+            '20141126')
+        self.assertEqual(
+            unified_strdate('2/2/2015 6:47:40 PM', day_first=False),
+            '20150202')
+        self.assertEqual(unified_strdate('25-09-2014'), '20140925')
+        self.assertEqual(unified_strdate('UNKNOWN DATE FORMAT'), None)
+
+    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)
+        self.assertEqual(determine_ext('http://example.com/foo/bar.nonext/?download', None), None)
+        self.assertEqual(determine_ext('http://example.com/foo/bar/mp4?download', None), None)
+        self.assertEqual(determine_ext('http://example.com/foo/bar.m3u8//?download'), 'm3u8')
 
     def test_find_xpath_attr(self):
         testxml = '''<root>
@@ -151,12 +264,21 @@ class TestUtil(unittest.TestCase):
             <node x="a"/>
             <node x="a" y="c" />
             <node x="b" y="d" />
+            <node x="" />
         </root>'''
-        doc = xml.etree.ElementTree.fromstring(testxml)
+        doc = compat_etree_fromstring(testxml)
 
+        self.assertEqual(find_xpath_attr(doc, './/fourohfour', 'n'), None)
         self.assertEqual(find_xpath_attr(doc, './/fourohfour', 'n', 'v'), None)
+        self.assertEqual(find_xpath_attr(doc, './/node', 'n'), None)
+        self.assertEqual(find_xpath_attr(doc, './/node', 'n', 'v'), None)
+        self.assertEqual(find_xpath_attr(doc, './/node', 'x'), doc[1])
         self.assertEqual(find_xpath_attr(doc, './/node', 'x', 'a'), doc[1])
+        self.assertEqual(find_xpath_attr(doc, './/node', 'x', 'b'), doc[3])
+        self.assertEqual(find_xpath_attr(doc, './/node', 'y'), doc[2])
         self.assertEqual(find_xpath_attr(doc, './/node', 'y', 'c'), doc[2])
+        self.assertEqual(find_xpath_attr(doc, './/node', 'y', 'd'), doc[3])
+        self.assertEqual(find_xpath_attr(doc, './/node', 'x', ''), doc[4])
 
     def test_xpath_with_ns(self):
         testxml = '''<root xmlns:media="http://example.com/">
@@ -165,12 +287,56 @@ class TestUtil(unittest.TestCase):
                 <url>http://server.com/download.mp3</url>
             </media:song>
         </root>'''
-        doc = xml.etree.ElementTree.fromstring(testxml)
+        doc = compat_etree_fromstring(testxml)
         find = lambda p: doc.find(xpath_with_ns(p, {'media': 'http://example.com/'}))
         self.assertTrue(find('media:song') is not None)
         self.assertEqual(find('media:song/media:author').text, 'The Author')
         self.assertEqual(find('media:song/url').text, 'http://server.com/download.mp3')
 
+    def test_xpath_element(self):
+        doc = xml.etree.ElementTree.Element('root')
+        div = xml.etree.ElementTree.SubElement(doc, 'div')
+        p = xml.etree.ElementTree.SubElement(div, 'p')
+        p.text = 'Foo'
+        self.assertEqual(xpath_element(doc, 'div/p'), p)
+        self.assertEqual(xpath_element(doc, ['div/p']), p)
+        self.assertEqual(xpath_element(doc, ['div/bar', 'div/p']), p)
+        self.assertEqual(xpath_element(doc, 'div/bar', default='default'), 'default')
+        self.assertEqual(xpath_element(doc, ['div/bar'], default='default'), 'default')
+        self.assertTrue(xpath_element(doc, 'div/bar') is None)
+        self.assertTrue(xpath_element(doc, ['div/bar']) is None)
+        self.assertTrue(xpath_element(doc, ['div/bar'], 'div/baz') is None)
+        self.assertRaises(ExtractorError, xpath_element, doc, 'div/bar', fatal=True)
+        self.assertRaises(ExtractorError, xpath_element, doc, ['div/bar'], fatal=True)
+        self.assertRaises(ExtractorError, xpath_element, doc, ['div/bar', 'div/baz'], fatal=True)
+
+    def test_xpath_text(self):
+        testxml = '''<root>
+            <div>
+                <p>Foo</p>
+            </div>
+        </root>'''
+        doc = compat_etree_fromstring(testxml)
+        self.assertEqual(xpath_text(doc, 'div/p'), 'Foo')
+        self.assertEqual(xpath_text(doc, 'div/bar', default='default'), 'default')
+        self.assertTrue(xpath_text(doc, 'div/bar') is None)
+        self.assertRaises(ExtractorError, xpath_text, doc, 'div/bar', fatal=True)
+
+    def test_xpath_attr(self):
+        testxml = '''<root>
+            <div>
+                <p x="a">Foo</p>
+            </div>
+        </root>'''
+        doc = compat_etree_fromstring(testxml)
+        self.assertEqual(xpath_attr(doc, 'div/p', 'x'), 'a')
+        self.assertEqual(xpath_attr(doc, 'div/bar', 'x'), None)
+        self.assertEqual(xpath_attr(doc, 'div/p', 'y'), None)
+        self.assertEqual(xpath_attr(doc, 'div/bar', 'x', default='default'), 'default')
+        self.assertEqual(xpath_attr(doc, 'div/p', 'y', default='default'), 'default')
+        self.assertRaises(ExtractorError, xpath_attr, doc, 'div/bar', 'x', fatal=True)
+        self.assertRaises(ExtractorError, xpath_attr, doc, 'div/p', 'y', fatal=True)
+
     def test_smuggle_url(self):
         data = {"ö": "ö", "abc": [3]}
         url = 'https://foo.bar/baz?x=y#a'
@@ -203,6 +369,8 @@ class TestUtil(unittest.TestCase):
 
     def test_parse_duration(self):
         self.assertEqual(parse_duration(None), None)
+        self.assertEqual(parse_duration(False), None)
+        self.assertEqual(parse_duration('invalid'), None)
         self.assertEqual(parse_duration('1'), 1)
         self.assertEqual(parse_duration('1337:12'), 80232)
         self.assertEqual(parse_duration('9:12:43'), 33163)
@@ -224,6 +392,10 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_duration('5 s'), 5)
         self.assertEqual(parse_duration('3 min'), 180)
         self.assertEqual(parse_duration('2.5 hours'), 9000)
+        self.assertEqual(parse_duration('02:03:04'), 7384)
+        self.assertEqual(parse_duration('01:02:03:04'), 93784)
+        self.assertEqual(parse_duration('1 hour 3 minutes'), 3780)
+        self.assertEqual(parse_duration('87 Min.'), 5220)
 
     def test_fix_xml_ampersands(self):
         self.assertEqual(
@@ -279,11 +451,39 @@ class TestUtil(unittest.TestCase):
         data = urlencode_postdata({'username': 'foo@bar.com', 'password': '1234'})
         self.assertTrue(isinstance(data, bytes))
 
+    def test_dict_get(self):
+        FALSE_VALUES = {
+            'none': None,
+            'false': False,
+            'zero': 0,
+            'empty_string': '',
+            'empty_list': [],
+        }
+        d = FALSE_VALUES.copy()
+        d['a'] = 42
+        self.assertEqual(dict_get(d, 'a'), 42)
+        self.assertEqual(dict_get(d, 'b'), None)
+        self.assertEqual(dict_get(d, 'b', 42), 42)
+        self.assertEqual(dict_get(d, ('a', )), 42)
+        self.assertEqual(dict_get(d, ('b', 'a', )), 42)
+        self.assertEqual(dict_get(d, ('b', 'c', 'a', 'd', )), 42)
+        self.assertEqual(dict_get(d, ('b', 'c', )), None)
+        self.assertEqual(dict_get(d, ('b', 'c', ), 42), 42)
+        for key, false_value in FALSE_VALUES.items():
+            self.assertEqual(dict_get(d, ('b', 'c', key, )), None)
+            self.assertEqual(dict_get(d, ('b', 'c', key, ), skip_false_values=False), false_value)
+
+    def test_encode_compat_str(self):
+        self.assertEqual(encode_compat_str(b'\xd1\x82\xd0\xb5\xd1\x81\xd1\x82', 'utf-8'), 'тест')
+        self.assertEqual(encode_compat_str('тест', 'utf-8'), 'тест')
+
     def test_parse_iso8601(self):
         self.assertEqual(parse_iso8601('2014-03-23T23:04:26+0100'), 1395612266)
         self.assertEqual(parse_iso8601('2014-03-23T22:04:26+0000'), 1395612266)
         self.assertEqual(parse_iso8601('2014-03-23T22:04:26Z'), 1395612266)
         self.assertEqual(parse_iso8601('2014-03-23T22:04:26.1234Z'), 1395612266)
+        self.assertEqual(parse_iso8601('2015-09-29T08:27:31.727'), 1443515251)
+        self.assertEqual(parse_iso8601('2015-09-29T08-27-31.727'), None)
 
     def test_strip_jsonp(self):
         stripped = strip_jsonp('cb ([ {"id":"532cb",\n\n\n"x":\n3}\n]\n);')
@@ -294,10 +494,18 @@ class TestUtil(unittest.TestCase):
         d = json.loads(stripped)
         self.assertEqual(d, {'STATUS': 'OK'})
 
+        stripped = strip_jsonp('ps.embedHandler({"status": "success"});')
+        d = json.loads(stripped)
+        self.assertEqual(d, {'status': 'success'})
+
     def test_uppercase_escape(self):
         self.assertEqual(uppercase_escape('aä'), 'aä')
         self.assertEqual(uppercase_escape('\\U0001d550'), '𝕐')
 
+    def test_lowercase_escape(self):
+        self.assertEqual(lowercase_escape('aä'), 'aä')
+        self.assertEqual(lowercase_escape('\\u0026'), '&')
+
     def test_limit_length(self):
         self.assertEqual(limit_length(None, 12), None)
         self.assertEqual(limit_length('foo', 12), 'foo')
@@ -350,6 +558,13 @@ class TestUtil(unittest.TestCase):
             "playlist":[{"controls":{"all":null}}]
         }''')
 
+        inp = '''"The CW\\'s \\'Crazy Ex-Girlfriend\\'"'''
+        self.assertEqual(js_to_json(inp), '''"The CW's 'Crazy Ex-Girlfriend'"''')
+
+        inp = '"SAND Number: SAND 2013-7800P\\nPresenter: Tom Russo\\nHabanero Software Training - Xyce Software\\nXyce, Sandia\\u0027s"'
+        json_code = js_to_json(inp)
+        self.assertEqual(json.loads(json_code), json.loads(inp))
+
     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"})
@@ -357,6 +572,22 @@ class TestUtil(unittest.TestCase):
         on = js_to_json('{"abc": true}')
         self.assertEqual(json.loads(on), {'abc': True})
 
+        # Ignore JavaScript code as well
+        on = js_to_json('''{
+            "x": 1,
+            y: "a",
+            z: some.code
+        }''')
+        d = json.loads(on)
+        self.assertEqual(d['x'], 1)
+        self.assertEqual(d['y'], 'a')
+
+        on = js_to_json('["abc", "def",]')
+        self.assertEqual(json.loads(on), ['abc', 'def'])
+
+        on = js_to_json('{"abc": "def",}')
+        self.assertEqual(json.loads(on), {'abc': 'def'})
+
     def test_clean_html(self):
         self.assertEqual(clean_html('a:\nb'), 'a: b')
         self.assertEqual(clean_html('a:\n   "b"'), 'a:    "b"')
@@ -387,5 +618,180 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(version_tuple('10.23.344'), (10, 23, 344))
         self.assertEqual(version_tuple('10.1-6'), (10, 1, 6))  # avconv style
 
+    def test_detect_exe_version(self):
+        self.assertEqual(detect_exe_version('''ffmpeg version 1.2.1
+built on May 27 2013 08:37:26 with gcc 4.7 (Debian 4.7.3-4)
+configuration: --prefix=/usr --extra-'''), '1.2.1')
+        self.assertEqual(detect_exe_version('''ffmpeg version N-63176-g1fb4685
+built on May 15 2014 22:09:06 with gcc 4.8.2 (GCC)'''), 'N-63176-g1fb4685')
+        self.assertEqual(detect_exe_version('''X server found. dri2 connection failed!
+Trying to open render node...
+Success at /dev/dri/renderD128.
+ffmpeg version 2.4.4 Copyright (c) 2000-2014 the FFmpeg ...'''), '2.4.4')
+
+    def test_age_restricted(self):
+        self.assertFalse(age_restricted(None, 10))  # unrestricted content
+        self.assertFalse(age_restricted(1, None))  # unrestricted policy
+        self.assertFalse(age_restricted(8, 10))
+        self.assertTrue(age_restricted(18, 14))
+        self.assertFalse(age_restricted(18, 18))
+
+    def test_is_html(self):
+        self.assertFalse(is_html(b'\x49\x44\x43<html'))
+        self.assertTrue(is_html(b'<!DOCTYPE foo>\xaaa'))
+        self.assertTrue(is_html(  # UTF-8 with BOM
+            b'\xef\xbb\xbf<!DOCTYPE foo>\xaaa'))
+        self.assertTrue(is_html(  # UTF-16-LE
+            b'\xff\xfe<\x00h\x00t\x00m\x00l\x00>\x00\xe4\x00'
+        ))
+        self.assertTrue(is_html(  # UTF-16-BE
+            b'\xfe\xff\x00<\x00h\x00t\x00m\x00l\x00>\x00\xe4'
+        ))
+        self.assertTrue(is_html(  # UTF-32-BE
+            b'\x00\x00\xFE\xFF\x00\x00\x00<\x00\x00\x00h\x00\x00\x00t\x00\x00\x00m\x00\x00\x00l\x00\x00\x00>\x00\x00\x00\xe4'))
+        self.assertTrue(is_html(  # UTF-32-LE
+            b'\xFF\xFE\x00\x00<\x00\x00\x00h\x00\x00\x00t\x00\x00\x00m\x00\x00\x00l\x00\x00\x00>\x00\x00\x00\xe4\x00\x00\x00'))
+
+    def test_render_table(self):
+        self.assertEqual(
+            render_table(
+                ['a', 'bcd'],
+                [[123, 4], [9999, 51]]),
+            'a    bcd\n'
+            '123  4\n'
+            '9999 51')
+
+    def test_match_str(self):
+        self.assertRaises(ValueError, match_str, 'xy>foobar', {})
+        self.assertFalse(match_str('xy', {'x': 1200}))
+        self.assertTrue(match_str('!xy', {'x': 1200}))
+        self.assertTrue(match_str('x', {'x': 1200}))
+        self.assertFalse(match_str('!x', {'x': 1200}))
+        self.assertTrue(match_str('x', {'x': 0}))
+        self.assertFalse(match_str('x>0', {'x': 0}))
+        self.assertFalse(match_str('x>0', {}))
+        self.assertTrue(match_str('x>?0', {}))
+        self.assertTrue(match_str('x>1K', {'x': 1200}))
+        self.assertFalse(match_str('x>2K', {'x': 1200}))
+        self.assertTrue(match_str('x>=1200 & x < 1300', {'x': 1200}))
+        self.assertFalse(match_str('x>=1100 & x < 1200', {'x': 1200}))
+        self.assertFalse(match_str('y=a212', {'y': 'foobar42'}))
+        self.assertTrue(match_str('y=foobar42', {'y': 'foobar42'}))
+        self.assertFalse(match_str('y!=foobar42', {'y': 'foobar42'}))
+        self.assertTrue(match_str('y!=foobar2', {'y': 'foobar42'}))
+        self.assertFalse(match_str(
+            'like_count > 100 & dislike_count <? 50 & description',
+            {'like_count': 90, 'description': 'foo'}))
+        self.assertTrue(match_str(
+            'like_count > 100 & dislike_count <? 50 & description',
+            {'like_count': 190, 'description': 'foo'}))
+        self.assertFalse(match_str(
+            'like_count > 100 & dislike_count <? 50 & description',
+            {'like_count': 190, 'dislike_count': 60, 'description': 'foo'}))
+        self.assertFalse(match_str(
+            'like_count > 100 & dislike_count <? 50 & description',
+            {'like_count': 190, 'dislike_count': 10}))
+
+    def test_parse_dfxp_time_expr(self):
+        self.assertEqual(parse_dfxp_time_expr(None), None)
+        self.assertEqual(parse_dfxp_time_expr(''), None)
+        self.assertEqual(parse_dfxp_time_expr('0.1'), 0.1)
+        self.assertEqual(parse_dfxp_time_expr('0.1s'), 0.1)
+        self.assertEqual(parse_dfxp_time_expr('00:00:01'), 1.0)
+        self.assertEqual(parse_dfxp_time_expr('00:00:01.100'), 1.1)
+        self.assertEqual(parse_dfxp_time_expr('00:00:01:100'), 1.1)
+
+    def test_dfxp2srt(self):
+        dfxp_data = '''<?xml version="1.0" encoding="UTF-8"?>
+            <tt xmlns="http://www.w3.org/ns/ttml" xml:lang="en" xmlns:tts="http://www.w3.org/ns/ttml#parameter">
+            <body>
+                <div xml:lang="en">
+                    <p begin="0" end="1">The following line contains Chinese characters and special symbols</p>
+                    <p begin="1" end="2">第二行<br/>♪♪</p>
+                    <p begin="2" dur="1"><span>Third<br/>Line</span></p>
+                    <p begin="3" end="-1">Lines with invalid timestamps are ignored</p>
+                    <p begin="-1" end="-1">Ignore, two</p>
+                    <p begin="3" dur="-1">Ignored, three</p>
+                </div>
+            </body>
+            </tt>'''
+        srt_data = '''1
+00:00:00,000 --> 00:00:01,000
+The following line contains Chinese characters and special symbols
+
+2
+00:00:01,000 --> 00:00:02,000
+第二行
+♪♪
+
+3
+00:00:02,000 --> 00:00:03,000
+Third
+Line
+
+'''
+        self.assertEqual(dfxp2srt(dfxp_data), srt_data)
+
+        dfxp_data_no_default_namespace = '''<?xml version="1.0" encoding="UTF-8"?>
+            <tt xml:lang="en" xmlns:tts="http://www.w3.org/ns/ttml#parameter">
+            <body>
+                <div xml:lang="en">
+                    <p begin="0" end="1">The first line</p>
+                </div>
+            </body>
+            </tt>'''
+        srt_data = '''1
+00:00:00,000 --> 00:00:01,000
+The first line
+
+'''
+        self.assertEqual(dfxp2srt(dfxp_data_no_default_namespace), srt_data)
+
+    def test_cli_option(self):
+        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'), [])
+
+    def test_cli_valueless_option(self):
+        self.assertEqual(cli_valueless_option(
+            {'downloader': 'external'}, '--external-downloader', 'downloader', 'external'), ['--external-downloader'])
+        self.assertEqual(cli_valueless_option(
+            {'downloader': 'internal'}, '--external-downloader', 'downloader', 'external'), [])
+        self.assertEqual(cli_valueless_option(
+            {'nocheckcertificate': True}, '--no-check-certificate', 'nocheckcertificate'), ['--no-check-certificate'])
+        self.assertEqual(cli_valueless_option(
+            {'nocheckcertificate': False}, '--no-check-certificate', 'nocheckcertificate'), [])
+        self.assertEqual(cli_valueless_option(
+            {'checkcertificate': True}, '--no-check-certificate', 'checkcertificate', False), [])
+        self.assertEqual(cli_valueless_option(
+            {'checkcertificate': False}, '--no-check-certificate', 'checkcertificate', False), ['--no-check-certificate'])
+
+    def test_cli_bool_option(self):
+        self.assertEqual(
+            cli_bool_option(
+                {'nocheckcertificate': True}, '--no-check-certificate', 'nocheckcertificate'),
+            ['--no-check-certificate', 'true'])
+        self.assertEqual(
+            cli_bool_option(
+                {'nocheckcertificate': True}, '--no-check-certificate', 'nocheckcertificate', separator='='),
+            ['--no-check-certificate=true'])
+        self.assertEqual(
+            cli_bool_option(
+                {'nocheckcertificate': True}, '--check-certificate', 'nocheckcertificate', 'false', 'true'),
+            ['--check-certificate', 'false'])
+        self.assertEqual(
+            cli_bool_option(
+                {'nocheckcertificate': True}, '--check-certificate', 'nocheckcertificate', 'false', 'true', '='),
+            ['--check-certificate=false'])
+        self.assertEqual(
+            cli_bool_option(
+                {'nocheckcertificate': False}, '--check-certificate', 'nocheckcertificate', 'false', 'true'),
+            ['--check-certificate', 'true'])
+        self.assertEqual(
+            cli_bool_option(
+                {'nocheckcertificate': False}, '--check-certificate', 'nocheckcertificate', 'false', 'true', '='),
+            ['--check-certificate=true'])
+
+
 if __name__ == '__main__':
     unittest.main()

test/test_write_annotations.py

@@ -33,7 +33,7 @@ params = get_params({
 
 
 TEST_ID = 'gr51aVj-mLg'
-ANNOTATIONS_FILE = TEST_ID + '.flv.annotations.xml'
+ANNOTATIONS_FILE = TEST_ID + '.annotations.xml'
 EXPECTED_ANNOTATIONS = ['Speech bubble', 'Note', 'Title', 'Spotlight', 'Label']
 
 
@@ -66,7 +66,7 @@ class TestAnnotations(unittest.TestCase):
                 textTag = a.find('TEXT')
                 text = textTag.text
                 self.assertTrue(text in expected)  # assertIn only added in python 2.7
-                # remove the first occurance, there could be more than one annotation with the same text
+                # remove the first occurrence, there could be more than one annotation with the same text
                 expected.remove(text)
         # We should have seen (and removed) all the expected annotation texts.
         self.assertEqual(len(expected), 0, 'Not all expected annotations were found.')

test/test_write_info_json.py

@@ -1,76 +0,0 @@
-#!/usr/bin/env python
-# coding: utf-8
-from __future__ import unicode_literals
-
-# Allow direct execution
-import os
-import sys
-import unittest
-sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
-
-from test.helper import get_params
-
-
-import io
-import json
-
-import youtube_dl.YoutubeDL
-import youtube_dl.extractor
-
-
-class YoutubeDL(youtube_dl.YoutubeDL):
-    def __init__(self, *args, **kwargs):
-        super(YoutubeDL, self).__init__(*args, **kwargs)
-        self.to_stderr = self.to_screen
-
-params = get_params({
-    'writeinfojson': True,
-    'skip_download': True,
-    'writedescription': True,
-})
-
-
-TEST_ID = 'BaW_jenozKc'
-INFO_JSON_FILE = TEST_ID + '.info.json'
-DESCRIPTION_FILE = TEST_ID + '.mp4.description'
-EXPECTED_DESCRIPTION = '''test chars:  "'/\ä↭𝕐
-test URL: https://github.com/rg3/youtube-dl/issues/1892
-
-This is a test video for youtube-dl.
-
-For more information, contact phihag@phihag.de .'''
-
-
-class TestInfoJSON(unittest.TestCase):
-    def setUp(self):
-        # Clear old files
-        self.tearDown()
-
-    def test_info_json(self):
-        ie = youtube_dl.extractor.YoutubeIE()
-        ydl = YoutubeDL(params)
-        ydl.add_info_extractor(ie)
-        ydl.download([TEST_ID])
-        self.assertTrue(os.path.exists(INFO_JSON_FILE))
-        with io.open(INFO_JSON_FILE, 'r', encoding='utf-8') as jsonf:
-            jd = json.load(jsonf)
-        self.assertEqual(jd['upload_date'], '20121002')
-        self.assertEqual(jd['description'], EXPECTED_DESCRIPTION)
-        self.assertEqual(jd['id'], TEST_ID)
-        self.assertEqual(jd['extractor'], 'youtube')
-        self.assertEqual(jd['title'], '''youtube-dl test video "'/\ä↭𝕐''')
-        self.assertEqual(jd['uploader'], 'Philipp Hagemeister')
-
-        self.assertTrue(os.path.exists(DESCRIPTION_FILE))
-        with io.open(DESCRIPTION_FILE, 'r', encoding='utf-8') as descf:
-            descr = descf.read()
-        self.assertEqual(descr, EXPECTED_DESCRIPTION)
-
-    def tearDown(self):
-        if os.path.exists(INFO_JSON_FILE):
-            os.remove(INFO_JSON_FILE)
-        if os.path.exists(DESCRIPTION_FILE):
-            os.remove(DESCRIPTION_FILE)
-
-if __name__ == '__main__':
-    unittest.main()

test/test_youtube_lists.py

@@ -34,7 +34,7 @@ class TestYoutubeLists(unittest.TestCase):
         ie = YoutubePlaylistIE(dl)
         # TODO find a > 100 (paginating?) videos course
         result = ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
-        entries = result['entries']
+        entries = list(result['entries'])
         self.assertEqual(YoutubeIE().extract_id(entries[0]['url']), 'j9WZyLZCBzs')
         self.assertEqual(len(entries), 25)
         self.assertEqual(YoutubeIE().extract_id(entries[-1]['url']), 'rYefUsYuEp0')
@@ -57,5 +57,14 @@ class TestYoutubeLists(unittest.TestCase):
         entries = result['entries']
         self.assertEqual(len(entries), 100)
 
+    def test_youtube_flat_playlist_titles(self):
+        dl = FakeYDL()
+        dl.params['extract_flat'] = True
+        ie = YoutubePlaylistIE(dl)
+        result = ie.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
+        self.assertIsPlaylist(result)
+        for entry in result['entries']:
+            self.assertTrue(entry.get('title'))
+
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_signature.py

@@ -8,11 +8,11 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-
 import io
 import re
 import string
 
+from test.helper import FakeYDL
 from youtube_dl.extractor import YoutubeIE
 from youtube_dl.compat import compat_str, compat_urlretrieve
 
@@ -64,6 +64,12 @@ _TESTS = [
         'js',
         '4646B5181C6C3020DF1D9C7FCFEA.AD80ABF70C39BD369CCCAE780AFBB98FA6B6CB42766249D9488C288',
         '82C8849D94266724DC6B6AF89BBFA087EACCD963.B93C07FBA084ACAEFCF7C9D1FD0203C6C1815B6B'
+    ),
+    (
+        'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflKjOTVq/html5player.js',
+        'js',
+        '312AA52209E3623129A412D56A40F11CB0AF14AE.3EE09501CB14E3BCDC3B2AE808BF3F1D14E7FBF12',
+        '112AA5220913623229A412D56A40F11CB0AF14AE.3EE0950FCB14EEBCDC3B2AE808BF331D14E7FBF3',
     )
 ]
 
@@ -88,7 +94,8 @@ def make_tfunc(url, stype, sig_input, expected_sig):
         if not os.path.exists(fn):
             compat_urlretrieve(url, fn)
 
-        ie = YoutubeIE()
+        ydl = FakeYDL()
+        ie = YoutubeIE(ydl)
         if stype == 'js':
             with io.open(fn, encoding='utf-8') as testf:
                 jscode = testf.read()

test/testcert.pem

@@ -0,0 +1,52 @@
+-----BEGIN PRIVATE KEY-----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+-----END PRIVATE KEY-----
+-----BEGIN CERTIFICATE-----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+-----END CERTIFICATE-----

test/versions.json

@@ -0,0 +1,34 @@
+{
+    "latest": "2013.01.06", 
+    "signature": "72158cdba391628569ffdbea259afbcf279bbe3d8aeb7492690735dc1cfa6afa754f55c61196f3871d429599ab22f2667f1fec98865527b32632e7f4b3675a7ef0f0fbe084d359256ae4bba68f0d33854e531a70754712f244be71d4b92e664302aa99653ee4df19800d955b6c4149cd2b3f24288d6e4b40b16126e01f4c8ce6", 
+    "versions": {
+        "2013.01.02": {
+            "bin": [
+                "http://youtube-dl.org/downloads/2013.01.02/youtube-dl", 
+                "f5b502f8aaa77675c4884938b1e4871ebca2611813a0c0e74f60c0fbd6dcca6b"
+            ], 
+            "exe": [
+                "http://youtube-dl.org/downloads/2013.01.02/youtube-dl.exe", 
+                "75fa89d2ce297d102ff27675aa9d92545bbc91013f52ec52868c069f4f9f0422"
+            ], 
+            "tar": [
+                "http://youtube-dl.org/downloads/2013.01.02/youtube-dl-2013.01.02.tar.gz", 
+                "6a66d022ac8e1c13da284036288a133ec8dba003b7bd3a5179d0c0daca8c8196"
+            ]
+        }, 
+        "2013.01.06": {
+            "bin": [
+                "http://youtube-dl.org/downloads/2013.01.06/youtube-dl", 
+                "64b6ed8865735c6302e836d4d832577321b4519aa02640dc508580c1ee824049"
+            ], 
+            "exe": [
+                "http://youtube-dl.org/downloads/2013.01.06/youtube-dl.exe", 
+                "58609baf91e4389d36e3ba586e21dab882daaaee537e4448b1265392ae86ff84"
+            ], 
+            "tar": [
+                "http://youtube-dl.org/downloads/2013.01.06/youtube-dl-2013.01.06.tar.gz", 
+                "fe77ab20a95d980ed17a659aa67e371fdd4d656d19c4c7950e7b720b0c2f1a86"
+            ]
+        }
+    }
+}

tox.ini

@@ -1,8 +1,13 @@
 [tox]
-envlist = py26,py27,py33
+envlist = py26,py27,py33,py34,py35
 [testenv]
 deps =
    nose
    coverage
-commands = nosetests --verbose {posargs:test}  # --with-coverage --cover-package=youtube_dl --cover-html
+# We need a valid $HOME for test_compat_expanduser
+passenv = HOME
+defaultargs = test --exclude test_download.py --exclude test_age_restriction.py
+    --exclude test_subtitles.py --exclude test_write_annotations.py
+    --exclude test_youtube_lists.py
+commands = nosetests --verbose {posargs:{[testenv]defaultargs}}  # --with-coverage --cover-package=youtube_dl --cover-html
                                                # test.test_download:TestDownload.test_NowVideo

youtube_dl/__init__.py

@@ -19,13 +19,15 @@ from .compat import (
     compat_expanduser,
     compat_getpass,
     compat_print,
+    compat_shlex_split,
     workaround_optparse_bug9161,
 )
 from .utils import (
     DateRange,
-    DEFAULT_OUTTMPL,
     decodeOption,
+    DEFAULT_OUTTMPL,
     DownloadError,
+    match_filter_func,
     MaxDownloadsReached,
     preferredencoding,
     read_batch_urls,
@@ -38,18 +40,8 @@ from .update import update_self
 from .downloader import (
     FileDownloader,
 )
-from .extractor import gen_extractors
+from .extractor import gen_extractors, list_extractors
 from .YoutubeDL import YoutubeDL
-from .postprocessor import (
-    AtomicParsleyPP,
-    FFmpegAudioFixPP,
-    FFmpegMetadataPP,
-    FFmpegVideoConvertor,
-    FFmpegExtractAudioPP,
-    FFmpegEmbedSubtitlePP,
-    XAttrMetadataPP,
-    ExecAfterDownloadPP,
-)
 
 
 def _real_main(argv=None):
@@ -105,24 +97,22 @@ def _real_main(argv=None):
     _enc = preferredencoding()
     all_urls = [url.decode(_enc, 'ignore') if isinstance(url, bytes) else url for url in all_urls]
 
-    extractors = gen_extractors()
-
     if opts.list_extractors:
-        for ie in sorted(extractors, key=lambda ie: ie.IE_NAME.lower()):
+        for ie in list_extractors(opts.age_limit):
             compat_print(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else ''))
             matchedUrls = [url for url in all_urls if ie.suitable(url)]
             for mu in matchedUrls:
                 compat_print('  ' + mu)
         sys.exit(0)
     if opts.list_extractor_descriptions:
-        for ie in sorted(extractors, key=lambda ie: ie.IE_NAME.lower()):
+        for ie in list_extractors(opts.age_limit):
             if not ie._WORKING:
                 continue
             desc = getattr(ie, 'IE_DESC', ie.IE_NAME)
             if desc is False:
                 continue
             if hasattr(ie, 'SEARCH_KEY'):
-                _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny')
+                _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
                 _COUNTS = ('', '5', '10', 'all')
                 desc += ' (Example: "%s%s:%s" )' % (ie.SEARCH_KEY, random.choice(_COUNTS), random.choice(_SEARCHES))
             compat_print(desc)
@@ -155,10 +145,13 @@ def _real_main(argv=None):
             parser.error('invalid max_filesize specified')
         opts.max_filesize = numeric_limit
     if opts.retries is not None:
-        try:
-            opts.retries = int(opts.retries)
-        except (TypeError, ValueError):
-            parser.error('invalid retry count specified')
+        if opts.retries in ('inf', 'infinite'):
+            opts_retries = float('inf')
+        else:
+            try:
+                opts_retries = int(opts.retries)
+            except (TypeError, ValueError):
+                parser.error('invalid retry count specified')
     if opts.buffersize is not None:
         numeric_buffersize = FileDownloader.parse_bytes(opts.buffersize)
         if numeric_buffersize is None:
@@ -176,8 +169,12 @@ def _real_main(argv=None):
         if not opts.audioquality.isdigit():
             parser.error('invalid audio quality specified')
     if opts.recodevideo is not None:
-        if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg', 'mkv']:
+        if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg', 'mkv', 'avi']:
             parser.error('invalid video recode format specified')
+    if opts.convertsubtitles is not None:
+        if opts.convertsubtitles not in ['srt', 'vtt', 'ass']:
+            parser.error('invalid subtitle format specified')
+
     if opts.date is not None:
         date = DateRange.day(opts.date)
     else:
@@ -192,33 +189,94 @@ def _real_main(argv=None):
     if opts.allsubtitles and not opts.writeautomaticsub:
         opts.writesubtitles = True
 
-    if sys.version_info < (3,):
-        # In Python 2, sys.argv is a bytestring (also note http://bugs.python.org/issue2128 for Windows systems)
-        if opts.outtmpl is not None:
-            opts.outtmpl = opts.outtmpl.decode(preferredencoding())
-    outtmpl = ((opts.outtmpl is not None and opts.outtmpl)
-               or (opts.format == '-1' and opts.usetitle and '%(title)s-%(id)s-%(format)s.%(ext)s')
-               or (opts.format == '-1' and '%(id)s-%(format)s.%(ext)s')
-               or (opts.usetitle and opts.autonumber and '%(autonumber)s-%(title)s-%(id)s.%(ext)s')
-               or (opts.usetitle and '%(title)s-%(id)s.%(ext)s')
-               or (opts.useid and '%(id)s.%(ext)s')
-               or (opts.autonumber and '%(autonumber)s-%(id)s.%(ext)s')
-               or DEFAULT_OUTTMPL)
+    outtmpl = ((opts.outtmpl is not None and opts.outtmpl) or
+               (opts.format == '-1' and opts.usetitle and '%(title)s-%(id)s-%(format)s.%(ext)s') or
+               (opts.format == '-1' and '%(id)s-%(format)s.%(ext)s') or
+               (opts.usetitle and opts.autonumber and '%(autonumber)s-%(title)s-%(id)s.%(ext)s') or
+               (opts.usetitle and '%(title)s-%(id)s.%(ext)s') or
+               (opts.useid and '%(id)s.%(ext)s') or
+               (opts.autonumber and '%(autonumber)s-%(id)s.%(ext)s') or
+               DEFAULT_OUTTMPL)
     if not os.path.splitext(outtmpl)[1] and opts.extractaudio:
         parser.error('Cannot download a video and extract audio into the same'
                      ' file! Use "{0}.%(ext)s" instead of "{0}" as the output'
                      ' template'.format(outtmpl))
 
-    any_printing = opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson or opts.dump_single_json
+    any_getting = opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson or opts.dump_single_json
+    any_printing = opts.print_json
     download_archive_fn = compat_expanduser(opts.download_archive) if opts.download_archive is not None else opts.download_archive
 
+    # PostProcessors
+    postprocessors = []
+    # Add the metadata pp first, the other pps will copy it
+    if opts.metafromtitle:
+        postprocessors.append({
+            'key': 'MetadataFromTitle',
+            'titleformat': opts.metafromtitle
+        })
+    if opts.addmetadata:
+        postprocessors.append({'key': 'FFmpegMetadata'})
+    if opts.extractaudio:
+        postprocessors.append({
+            'key': 'FFmpegExtractAudio',
+            'preferredcodec': opts.audioformat,
+            'preferredquality': opts.audioquality,
+            'nopostoverwrites': opts.nopostoverwrites,
+        })
+    if opts.recodevideo:
+        postprocessors.append({
+            'key': 'FFmpegVideoConvertor',
+            'preferedformat': opts.recodevideo,
+        })
+    if opts.convertsubtitles:
+        postprocessors.append({
+            'key': 'FFmpegSubtitlesConvertor',
+            'format': opts.convertsubtitles,
+        })
+    if opts.embedsubtitles:
+        postprocessors.append({
+            'key': 'FFmpegEmbedSubtitle',
+        })
+    if opts.xattrs:
+        postprocessors.append({'key': 'XAttrMetadata'})
+    if opts.embedthumbnail:
+        already_have_thumbnail = opts.writethumbnail or opts.write_all_thumbnails
+        postprocessors.append({
+            'key': 'EmbedThumbnail',
+            'already_have_thumbnail': already_have_thumbnail
+        })
+        if not already_have_thumbnail:
+            opts.writethumbnail = True
+    # Please keep ExecAfterDownload towards the bottom as it allows the user to modify the final file in any way.
+    # So if the user is able to remove the file before your postprocessor runs it might cause a few problems.
+    if opts.exec_cmd:
+        postprocessors.append({
+            'key': 'ExecAfterDownload',
+            'exec_cmd': opts.exec_cmd,
+        })
+    if opts.xattr_set_filesize:
+        try:
+            import xattr
+            xattr  # Confuse flake8
+        except ImportError:
+            parser.error('setting filesize xattr requested but python-xattr is not available')
+    external_downloader_args = None
+    if opts.external_downloader_args:
+        external_downloader_args = compat_shlex_split(opts.external_downloader_args)
+    postprocessor_args = None
+    if opts.postprocessor_args:
+        postprocessor_args = compat_shlex_split(opts.postprocessor_args)
+    match_filter = (
+        None if opts.match_filter is None
+        else match_filter_func(opts.match_filter))
+
     ydl_opts = {
         'usenetrc': opts.usenetrc,
         'username': opts.username,
         'password': opts.password,
         'twofactor': opts.twofactor,
         'videopassword': opts.videopassword,
-        'quiet': (opts.quiet or any_printing),
+        'quiet': (opts.quiet or any_getting or any_printing),
         'no_warnings': opts.no_warnings,
         'forceurl': opts.geturl,
         'forcetitle': opts.gettitle,
@@ -228,20 +286,20 @@ def _real_main(argv=None):
         'forceduration': opts.getduration,
         'forcefilename': opts.getfilename,
         'forceformat': opts.getformat,
-        'forcejson': opts.dumpjson,
+        'forcejson': opts.dumpjson or opts.print_json,
         'dump_single_json': opts.dump_single_json,
-        'simulate': opts.simulate or any_printing,
+        'simulate': opts.simulate or any_getting,
         'skip_download': opts.skip_download,
         'format': opts.format,
-        'format_limit': opts.format_limit,
         'listformats': opts.listformats,
         'outtmpl': outtmpl,
         'autonumber_size': opts.autonumber_size,
         'restrictfilenames': opts.restrictfilenames,
         'ignoreerrors': opts.ignoreerrors,
+        'force_generic_extractor': opts.force_generic_extractor,
         'ratelimit': opts.ratelimit,
         'nooverwrites': opts.nooverwrites,
-        'retries': opts.retries,
+        'retries': opts_retries,
         'buffersize': opts.buffersize,
         'noresizebuffer': opts.noresizebuffer,
         'continuedl': opts.continue_dl,
@@ -249,6 +307,7 @@ def _real_main(argv=None):
         'progress_with_newline': opts.progress_with_newline,
         'playliststart': opts.playliststart,
         'playlistend': opts.playlistend,
+        'playlistreverse': opts.playlist_reverse,
         'noplaylist': opts.noplaylist,
         'logtostderr': opts.outtmpl == '-',
         'consoletitle': opts.consoletitle,
@@ -258,6 +317,7 @@ def _real_main(argv=None):
         'writeannotations': opts.writeannotations,
         'writeinfojson': opts.writeinfojson,
         'writethumbnail': opts.writethumbnail,
+        'write_all_thumbnails': opts.write_all_thumbnails,
         'writesubtitles': opts.writesubtitles,
         'writeautomaticsub': opts.writeautomaticsub,
         'allsubtitles': opts.allsubtitles,
@@ -294,37 +354,31 @@ def _real_main(argv=None):
         'default_search': opts.default_search,
         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
         'encoding': opts.encoding,
-        'exec_cmd': opts.exec_cmd,
         'extract_flat': opts.extract_flat,
+        'merge_output_format': opts.merge_output_format,
+        'postprocessors': postprocessors,
+        'fixup': opts.fixup,
+        'source_address': opts.source_address,
+        'call_home': opts.call_home,
+        'sleep_interval': opts.sleep_interval,
+        'external_downloader': opts.external_downloader,
+        'list_thumbnails': opts.list_thumbnails,
+        'playlist_items': opts.playlist_items,
+        'xattr_set_filesize': opts.xattr_set_filesize,
+        'match_filter': match_filter,
+        'no_color': opts.no_color,
+        'ffmpeg_location': opts.ffmpeg_location,
+        'hls_prefer_native': opts.hls_prefer_native,
+        'hls_use_mpegts': opts.hls_use_mpegts,
+        'external_downloader_args': external_downloader_args,
+        'postprocessor_args': postprocessor_args,
+        'cn_verification_proxy': opts.cn_verification_proxy,
     }
 
     with YoutubeDL(ydl_opts) as ydl:
-        # PostProcessors
-        # Add the metadata pp first, the other pps will copy it
-        if opts.addmetadata:
-            ydl.add_post_processor(FFmpegMetadataPP())
-        if opts.extractaudio:
-            ydl.add_post_processor(FFmpegExtractAudioPP(preferredcodec=opts.audioformat, preferredquality=opts.audioquality, nopostoverwrites=opts.nopostoverwrites))
-        if opts.recodevideo:
-            ydl.add_post_processor(FFmpegVideoConvertor(preferedformat=opts.recodevideo))
-        if opts.embedsubtitles:
-            ydl.add_post_processor(FFmpegEmbedSubtitlePP(subtitlesformat=opts.subtitlesformat))
-        if opts.xattrs:
-            ydl.add_post_processor(XAttrMetadataPP())
-        if opts.embedthumbnail:
-            if not opts.addmetadata:
-                ydl.add_post_processor(FFmpegAudioFixPP())
-            ydl.add_post_processor(AtomicParsleyPP())
-
-        # Please keep ExecAfterDownload towards the bottom as it allows the user to modify the final file in any way.
-        # So if the user is able to remove the file before your postprocessor runs it might cause a few problems.
-        if opts.exec_cmd:
-            ydl.add_post_processor(ExecAfterDownloadPP(
-                verboseOutput=opts.verbose, exec_cmd=opts.exec_cmd))
-
         # Update version
         if opts.update_self:
-            update_self(ydl.to_screen, opts.verbose)
+            update_self(ydl.to_screen, opts.verbose, ydl._opener)
 
         # Remove cache dir
         if opts.rm_cachedir:
@@ -336,7 +390,9 @@ def _real_main(argv=None):
                 sys.exit()
 
             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
-            parser.error('you must provide at least one URL')
+            parser.error(
+                'You must provide at least one URL.\n'
+                'Type youtube-dl --help to see a list of all options.')
 
         try:
             if opts.load_info_filename is not None:
@@ -359,3 +415,5 @@ def main(argv=None):
         sys.exit('ERROR: fixed output name but more than one file to download')
     except KeyboardInterrupt:
         sys.exit('\nERROR: Interrupted by user')
+
+__all__ = ['main', 'YoutubeDL', 'gen_extractors', 'list_extractors']

youtube_dl/__main__.py

@@ -11,7 +11,7 @@ if __package__ is None and not hasattr(sys, "frozen"):
     # direct call of __main__.py
     import os.path
     path = os.path.realpath(os.path.abspath(__file__))
-    sys.path.append(os.path.dirname(os.path.dirname(path)))
+    sys.path.insert(0, os.path.dirname(os.path.dirname(path)))
 
 import youtube_dl
 

youtube_dl/aes.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-__all__ = ['aes_encrypt', 'key_expansion', 'aes_ctr_decrypt', 'aes_cbc_decrypt', 'aes_decrypt_text']
-
 import base64
 from math import ceil
 
@@ -154,7 +152,7 @@ def aes_decrypt_text(data, password, key_size_bytes):
     """
     NONCE_LENGTH_BYTES = 8
 
-    data = bytes_to_intlist(base64.b64decode(data))
+    data = bytes_to_intlist(base64.b64decode(data.encode('utf-8')))
     password = bytes_to_intlist(password.encode('utf-8'))
 
     key = password[:key_size_bytes] + [0] * (key_size_bytes - len(password))
@@ -329,3 +327,5 @@ def inc(data):
             data[i] = data[i] + 1
             break
     return data
+
+__all__ = ['aes_encrypt', 'key_expansion', 'aes_ctr_decrypt', 'aes_cbc_decrypt', 'aes_decrypt_text']

youtube_dl/compat.py

@@ -1,11 +1,20 @@
 from __future__ import unicode_literals
 
+import binascii
+import collections
+import email
 import getpass
+import io
 import optparse
 import os
 import re
+import shlex
+import shutil
+import socket
 import subprocess
 import sys
+import itertools
+import xml.etree.ElementTree
 
 
 try:
@@ -33,21 +42,26 @@ try:
 except ImportError:  # Python 2
     import urlparse as compat_urlparse
 
+try:
+    import urllib.response as compat_urllib_response
+except ImportError:  # Python 2
+    import urllib as compat_urllib_response
+
 try:
     import http.cookiejar as compat_cookiejar
 except ImportError:  # Python 2
     import cookielib as compat_cookiejar
 
+try:
+    import http.cookies as compat_cookies
+except ImportError:  # Python 2
+    import Cookie as compat_cookies
+
 try:
     import html.entities as compat_html_entities
 except ImportError:  # Python 2
     import htmlentitydefs as compat_html_entities
 
-try:
-    import html.parser as compat_html_parser
-except ImportError:  # Python 2
-    import HTMLParser as compat_html_parser
-
 try:
     import http.client as compat_http_client
 except ImportError:  # Python 2
@@ -71,43 +85,171 @@ except ImportError:
     compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w')
 
 try:
-    from urllib.parse import unquote as compat_urllib_parse_unquote
+    import http.server as compat_http_server
 except ImportError:
-    def compat_urllib_parse_unquote(string, encoding='utf-8', errors='replace'):
-        if string == '':
+    import BaseHTTPServer as compat_http_server
+
+try:
+    compat_str = unicode  # Python 2
+except NameError:
+    compat_str = str
+
+try:
+    from urllib.parse import unquote_to_bytes as compat_urllib_parse_unquote_to_bytes
+    from urllib.parse import unquote as compat_urllib_parse_unquote
+    from urllib.parse import unquote_plus as compat_urllib_parse_unquote_plus
+except ImportError:  # Python 2
+    _asciire = (compat_urllib_parse._asciire if hasattr(compat_urllib_parse, '_asciire')
+                else re.compile('([\x00-\x7f]+)'))
+
+    # HACK: The following are the correct unquote_to_bytes, unquote and unquote_plus
+    # implementations from cpython 3.4.3's stdlib. Python 2's version
+    # is apparently broken (see https://github.com/rg3/youtube-dl/pull/6244)
+
+    def compat_urllib_parse_unquote_to_bytes(string):
+        """unquote_to_bytes('abc%20def') -> b'abc def'."""
+        # Note: strings are encoded as UTF-8. This is only an issue if it contains
+        # unescaped non-ASCII characters, which URIs should not.
+        if not string:
+            # Is it a string-like object?
+            string.split
+            return b''
+        if isinstance(string, compat_str):
+            string = string.encode('utf-8')
+        bits = string.split(b'%')
+        if len(bits) == 1:
             return string
-        res = string.split('%')
-        if len(res) == 1:
+        res = [bits[0]]
+        append = res.append
+        for item in bits[1:]:
+            try:
+                append(compat_urllib_parse._hextochr[item[:2]])
+                append(item[2:])
+            except KeyError:
+                append(b'%')
+                append(item)
+        return b''.join(res)
+
+    def compat_urllib_parse_unquote(string, encoding='utf-8', errors='replace'):
+        """Replace %xx escapes by their single-character equivalent. The optional
+        encoding and errors parameters specify how to decode percent-encoded
+        sequences into Unicode characters, as accepted by the bytes.decode()
+        method.
+        By default, percent-encoded sequences are decoded with UTF-8, and invalid
+        sequences are replaced by a placeholder character.
+
+        unquote('abc%20def') -> 'abc def'.
+        """
+        if '%' not in string:
+            string.split
             return string
         if encoding is None:
             encoding = 'utf-8'
         if errors is None:
             errors = 'replace'
-        # pct_sequence: contiguous sequence of percent-encoded bytes, decoded
-        pct_sequence = b''
-        string = res[0]
-        for item in res[1:]:
-            try:
-                if not item:
-                    raise ValueError
-                pct_sequence += item[:2].decode('hex')
-                rest = item[2:]
-                if not rest:
-                    # This segment was just a single percent-encoded character.
-                    # May be part of a sequence of code units, so delay decoding.
-                    # (Stored in pct_sequence).
-                    continue
-            except ValueError:
-                rest = '%' + item
-            # Encountered non-percent-encoded characters. Flush the current
-            # pct_sequence.
-            string += pct_sequence.decode(encoding, errors) + rest
-            pct_sequence = b''
-        if pct_sequence:
-            # Flush the final pct_sequence
-            string += pct_sequence.decode(encoding, errors)
-        return string
+        bits = _asciire.split(string)
+        res = [bits[0]]
+        append = res.append
+        for i in range(1, len(bits), 2):
+            append(compat_urllib_parse_unquote_to_bytes(bits[i]).decode(encoding, errors))
+            append(bits[i + 1])
+        return ''.join(res)
 
+    def compat_urllib_parse_unquote_plus(string, encoding='utf-8', errors='replace'):
+        """Like unquote(), but also replace plus signs by spaces, as required for
+        unquoting HTML form values.
+
+        unquote_plus('%7e/abc+def') -> '~/abc def'
+        """
+        string = string.replace('+', ' ')
+        return compat_urllib_parse_unquote(string, encoding, errors)
+
+try:
+    from urllib.request import DataHandler as compat_urllib_request_DataHandler
+except ImportError:  # Python < 3.4
+    # Ported from CPython 98774:1733b3bd46db, Lib/urllib/request.py
+    class compat_urllib_request_DataHandler(compat_urllib_request.BaseHandler):
+        def data_open(self, req):
+            # data URLs as specified in RFC 2397.
+            #
+            # ignores POSTed data
+            #
+            # syntax:
+            # dataurl   := "data:" [ mediatype ] [ ";base64" ] "," data
+            # mediatype := [ type "/" subtype ] *( ";" parameter )
+            # data      := *urlchar
+            # parameter := attribute "=" value
+            url = req.get_full_url()
+
+            scheme, data = url.split(":", 1)
+            mediatype, data = data.split(",", 1)
+
+            # even base64 encoded data URLs might be quoted so unquote in any case:
+            data = compat_urllib_parse_unquote_to_bytes(data)
+            if mediatype.endswith(";base64"):
+                data = binascii.a2b_base64(data)
+                mediatype = mediatype[:-7]
+
+            if not mediatype:
+                mediatype = "text/plain;charset=US-ASCII"
+
+            headers = email.message_from_string(
+                "Content-type: %s\nContent-length: %d\n" % (mediatype, len(data)))
+
+            return compat_urllib_response.addinfourl(io.BytesIO(data), headers, url)
+
+try:
+    compat_basestring = basestring  # Python 2
+except NameError:
+    compat_basestring = str
+
+try:
+    compat_chr = unichr  # Python 2
+except NameError:
+    compat_chr = chr
+
+try:
+    from xml.etree.ElementTree import ParseError as compat_xml_parse_error
+except ImportError:  # Python 2.6
+    from xml.parsers.expat import ExpatError as compat_xml_parse_error
+
+if sys.version_info[0] >= 3:
+    compat_etree_fromstring = xml.etree.ElementTree.fromstring
+else:
+    # python 2.x tries to encode unicode strings with ascii (see the
+    # XMLParser._fixtext method)
+    etree = xml.etree.ElementTree
+
+    try:
+        _etree_iter = etree.Element.iter
+    except AttributeError:  # Python <=2.6
+        def _etree_iter(root):
+            for el in root.findall('*'):
+                yield el
+                for sub in _etree_iter(el):
+                    yield sub
+
+    # on 2.6 XML doesn't have a parser argument, function copied from CPython
+    # 2.7 source
+    def _XML(text, parser=None):
+        if not parser:
+            parser = etree.XMLParser(target=etree.TreeBuilder())
+        parser.feed(text)
+        return parser.close()
+
+    def _element_factory(*args, **kwargs):
+        el = etree.Element(*args, **kwargs)
+        for k, v in el.items():
+            if isinstance(v, bytes):
+                el.set(k, v.decode('utf-8'))
+        return el
+
+    def compat_etree_fromstring(text):
+        doc = _XML(text, parser=etree.XMLParser(target=etree.TreeBuilder(element_factory=_element_factory)))
+        for el in _etree_iter(doc):
+            if el.text is not None and isinstance(el.text, bytes):
+                el.text = el.text.decode('utf-8')
+        return doc
 
 try:
     from urllib.parse import parse_qs as compat_parse_qs
@@ -117,7 +259,7 @@ except ImportError:  # Python 2
 
     def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False,
                    encoding='utf-8', errors='replace'):
-        qs, _coerce_result = qs, unicode
+        qs, _coerce_result = qs, compat_str
         pairs = [s2 for s1 in qs.split('&') for s2 in s1.split(';')]
         r = []
         for name_value in pairs:
@@ -156,21 +298,6 @@ except ImportError:  # Python 2
                 parsed_result[name] = [value]
         return parsed_result
 
-try:
-    compat_str = unicode  # Python 2
-except NameError:
-    compat_str = str
-
-try:
-    compat_chr = unichr  # Python 2
-except NameError:
-    compat_chr = chr
-
-try:
-    from xml.etree.ElementTree import ParseError as compat_xml_parse_error
-except ImportError:  # Python 2.6
-    from xml.parsers.expat import ExpatError as compat_xml_parse_error
-
 try:
     from shlex import quote as shlex_quote
 except ImportError:  # Python < 3.3
@@ -181,6 +308,17 @@ except ImportError:  # Python < 3.3
             return "'" + s.replace("'", "'\"'\"'") + "'"
 
 
+if sys.version_info >= (2, 7, 3):
+    compat_shlex_split = shlex.split
+else:
+    # 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)
+
+
 def compat_ord(c):
     if type(c) is int:
         return c
@@ -295,9 +433,11 @@ if sys.version_info < (3, 0) and sys.platform == 'win32':
 else:
     compat_getpass = getpass.getpass
 
-# Old 2.6 and 2.7 releases require kwargs to be bytes
+# Python < 2.6.5 require kwargs to be bytes
 try:
-    (lambda x: x)(**{'x': 0})
+    def _testfunc(x):
+        pass
+    _testfunc(**{'x': 0})
 except TypeError:
     def compat_kwargs(kwargs):
         return dict((bytes(k), v) for k, v in kwargs.items())
@@ -305,6 +445,32 @@ else:
     compat_kwargs = lambda kwargs: kwargs
 
 
+if sys.version_info < (2, 7):
+    def compat_socket_create_connection(address, timeout, source_address=None):
+        host, port = address
+        err = None
+        for res in socket.getaddrinfo(host, port, 0, socket.SOCK_STREAM):
+            af, socktype, proto, canonname, sa = res
+            sock = None
+            try:
+                sock = socket.socket(af, socktype, proto)
+                sock.settimeout(timeout)
+                if source_address:
+                    sock.bind(source_address)
+                sock.connect(sa)
+                return sock
+            except socket.error as _:
+                err = _
+                if sock is not None:
+                    sock.close()
+        if err is not None:
+            raise err
+        else:
+            raise socket.error("getaddrinfo returns an empty list")
+else:
+    compat_socket_create_connection = socket.create_connection
+
+
 # Fix https://github.com/rg3/youtube-dl/issues/4223
 # See http://bugs.python.org/issue9161 for what is broken
 def workaround_optparse_bug9161():
@@ -325,28 +491,87 @@ def workaround_optparse_bug9161():
             return real_add_option(self, *bargs, **bkwargs)
         optparse.OptionGroup.add_option = _compat_add_option
 
+if hasattr(shutil, 'get_terminal_size'):  # Python >= 3.3
+    compat_get_terminal_size = shutil.get_terminal_size
+else:
+    _terminal_size = collections.namedtuple('terminal_size', ['columns', 'lines'])
+
+    def compat_get_terminal_size(fallback=(80, 24)):
+        columns = compat_getenv('COLUMNS')
+        if columns:
+            columns = int(columns)
+        else:
+            columns = None
+        lines = compat_getenv('LINES')
+        if lines:
+            lines = int(lines)
+        else:
+            lines = None
+
+        if columns is None or lines is None or columns <= 0 or lines <= 0:
+            try:
+                sp = subprocess.Popen(
+                    ['stty', 'size'],
+                    stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+                out, err = sp.communicate()
+                _lines, _columns = map(int, out.split())
+            except Exception:
+                _columns, _lines = _terminal_size(*fallback)
+
+            if columns is None or columns <= 0:
+                columns = _columns
+            if lines is None or lines <= 0:
+                lines = _lines
+        return _terminal_size(columns, lines)
+
+try:
+    itertools.count(start=0, step=1)
+    compat_itertools_count = itertools.count
+except TypeError:  # Python 2.6
+    def compat_itertools_count(start=0, step=1):
+        n = start
+        while True:
+            yield n
+            n += step
+
+if sys.version_info >= (3, 0):
+    from tokenize import tokenize as compat_tokenize_tokenize
+else:
+    from tokenize import generate_tokens as compat_tokenize_tokenize
 
 __all__ = [
     'compat_HTTPError',
+    'compat_basestring',
     'compat_chr',
     'compat_cookiejar',
+    'compat_cookies',
+    'compat_etree_fromstring',
     'compat_expanduser',
+    'compat_get_terminal_size',
     'compat_getenv',
     'compat_getpass',
     'compat_html_entities',
-    'compat_html_parser',
     'compat_http_client',
+    'compat_http_server',
+    'compat_itertools_count',
     'compat_kwargs',
     'compat_ord',
     'compat_parse_qs',
     'compat_print',
+    'compat_shlex_split',
+    'compat_socket_create_connection',
     'compat_str',
     'compat_subprocess_get_DEVNULL',
+    'compat_tokenize_tokenize',
     'compat_urllib_error',
     'compat_urllib_parse',
     'compat_urllib_parse_unquote',
+    'compat_urllib_parse_unquote_plus',
+    'compat_urllib_parse_unquote_to_bytes',
     'compat_urllib_parse_urlparse',
     'compat_urllib_request',
+    'compat_urllib_request_DataHandler',
+    'compat_urllib_response',
     'compat_urlparse',
     'compat_urlretrieve',
     'compat_xml_parse_error',

youtube_dl/downloader/__init__.py

@@ -1,35 +1,46 @@
 from __future__ import unicode_literals
 
 from .common import FileDownloader
+from .external import get_external_downloader
+from .f4m import F4mFD
 from .hls import HlsFD
 from .hls import NativeHlsFD
 from .http import HttpFD
-from .mplayer import MplayerFD
+from .rtsp import RtspFD
 from .rtmp import RtmpFD
-from .f4m import F4mFD
+from .dash import DashSegmentsFD
 
 from ..utils import (
-    determine_ext,
+    determine_protocol,
 )
 
+PROTOCOL_MAP = {
+    'rtmp': RtmpFD,
+    'm3u8_native': NativeHlsFD,
+    'm3u8': HlsFD,
+    'mms': RtspFD,
+    'rtsp': RtspFD,
+    'f4m': F4mFD,
+    'http_dash_segments': DashSegmentsFD,
+}
 
-def get_suitable_downloader(info_dict):
+
+def get_suitable_downloader(info_dict, params={}):
     """Get the downloader class that can handle the info dict."""
-    url = info_dict['url']
-    protocol = info_dict.get('protocol')
+    protocol = determine_protocol(info_dict)
+    info_dict['protocol'] = protocol
 
-    if url.startswith('rtmp'):
-        return RtmpFD
-    if protocol == 'm3u8_native':
+    external_downloader = params.get('external_downloader')
+    if external_downloader is not None:
+        ed = get_external_downloader(external_downloader)
+        if ed.supports(info_dict):
+            return ed
+
+    if protocol == 'm3u8' and params.get('hls_prefer_native'):
         return NativeHlsFD
-    if (protocol == 'm3u8') or (protocol is None and determine_ext(url) == 'm3u8'):
-        return HlsFD
-    if url.startswith('mms') or url.startswith('rtsp'):
-        return MplayerFD
-    if determine_ext(url) == 'f4m':
-        return F4mFD
-    else:
-        return HttpFD
+
+    return PROTOCOL_MAP.get(protocol, HttpFD)
+
 
 __all__ = [
     'get_suitable_downloader',

youtube_dl/downloader/common.py

@@ -1,4 +1,4 @@
-from __future__ import unicode_literals
+from __future__ import division, unicode_literals
 
 import os
 import re
@@ -6,8 +6,9 @@ import sys
 import time
 
 from ..utils import (
-    compat_str,
     encodeFilename,
+    error_to_compat_str,
+    decodeArgument,
     format_bytes,
     timeconvert,
 )
@@ -25,21 +26,26 @@ class FileDownloader(object):
 
     Available options:
 
-    verbose:           Print additional info to stdout.
-    quiet:             Do not print messages to stdout.
-    ratelimit:         Download speed limit, in bytes/sec.
-    retries:           Number of times to retry for HTTP error 5xx
-    buffersize:        Size of download buffer in bytes.
-    noresizebuffer:    Do not automatically resize the download buffer.
-    continuedl:        Try to continue downloads if possible.
-    noprogress:        Do not print the progress bar.
-    logtostderr:       Log messages to stderr instead of stdout.
-    consoletitle:      Display progress in console window's titlebar.
-    nopart:            Do not use temporary .part files.
-    updatetime:        Use the Last-modified header to set output file timestamps.
-    test:              Download only first bytes to test the downloader.
-    min_filesize:      Skip files smaller than this size
-    max_filesize:      Skip files larger than this size
+    verbose:            Print additional info to stdout.
+    quiet:              Do not print messages to stdout.
+    ratelimit:          Download speed limit, in bytes/sec.
+    retries:            Number of times to retry for HTTP error 5xx
+    buffersize:         Size of download buffer in bytes.
+    noresizebuffer:     Do not automatically resize the download buffer.
+    continuedl:         Try to continue downloads if possible.
+    noprogress:         Do not print the progress bar.
+    logtostderr:        Log messages to stderr instead of stdout.
+    consoletitle:       Display progress in console window's titlebar.
+    nopart:             Do not use temporary .part files.
+    updatetime:         Use the Last-modified header to set output file timestamps.
+    test:               Download only first bytes to test the downloader.
+    min_filesize:       Skip files smaller than this size
+    max_filesize:       Skip files larger than this size
+    xattr_set_filesize: Set ytdl.filesize user xattribute with expected size.
+                        (experimental)
+    external_downloader_args:  A list of additional command-line arguments for the
+                        external downloader.
+    hls_use_mpegts:     Use the mpegts container for HLS videos.
 
     Subclasses of this one must re-define the real_download method.
     """
@@ -52,6 +58,7 @@ class FileDownloader(object):
         self.ydl = ydl
         self._progress_hooks = []
         self.params = params
+        self.add_progress_hook(self.report_progress)
 
     @staticmethod
     def format_seconds(seconds):
@@ -80,6 +87,8 @@ class FileDownloader(object):
     def calc_eta(start, now, total, current):
         if total is None:
             return None
+        if now is None:
+            now = time.time()
         dif = now - start
         if current == 0 or dif < 0.001:  # One millisecond
             return None
@@ -146,18 +155,19 @@ class FileDownloader(object):
     def report_error(self, *args, **kargs):
         self.ydl.report_error(*args, **kargs)
 
-    def slow_down(self, start_time, byte_counter):
+    def slow_down(self, start_time, now, byte_counter):
         """Sleep if the download speed is over the rate limit."""
         rate_limit = self.params.get('ratelimit', None)
         if rate_limit is None or byte_counter == 0:
             return
-        now = time.time()
+        if now is None:
+            now = time.time()
         elapsed = now - start_time
         if elapsed <= 0.0:
             return
         speed = float(byte_counter) / elapsed
         if speed > rate_limit:
-            time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit)
+            time.sleep(max((byte_counter // rate_limit) - elapsed, 0))
 
     def temp_name(self, filename):
         """Returns a temporary filename for the given filename."""
@@ -177,7 +187,7 @@ class FileDownloader(object):
                 return
             os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
         except (IOError, OSError) as err:
-            self.report_error('unable to rename file: %s' % compat_str(err))
+            self.report_error('unable to rename file: %s' % error_to_compat_str(err))
 
     def try_utime(self, filename, last_modified_hdr):
         """Try to set the last-modified time of the given file."""
@@ -196,7 +206,7 @@ class FileDownloader(object):
             return
         try:
             os.utime(filename, (time.time(), filetime))
-        except:
+        except Exception:
             pass
         return filetime
 
@@ -221,42 +231,64 @@ class FileDownloader(object):
             self.to_screen(clear_line + fullmsg, skip_eol=not is_last_line)
         self.to_console_title('youtube-dl ' + msg)
 
-    def report_progress(self, percent, data_len_str, speed, eta):
-        """Report download progress."""
-        if self.params.get('noprogress', False):
+    def report_progress(self, s):
+        if s['status'] == 'finished':
+            if self.params.get('noprogress', False):
+                self.to_screen('[download] Download completed')
+            else:
+                s['_total_bytes_str'] = format_bytes(s['total_bytes'])
+                if s.get('elapsed') is not None:
+                    s['_elapsed_str'] = self.format_seconds(s['elapsed'])
+                    msg_template = '100%% of %(_total_bytes_str)s in %(_elapsed_str)s'
+                else:
+                    msg_template = '100%% of %(_total_bytes_str)s'
+                self._report_progress_status(
+                    msg_template % s, is_last_line=True)
+
+        if self.params.get('noprogress'):
             return
-        if eta is not None:
-            eta_str = self.format_eta(eta)
-        else:
-            eta_str = 'Unknown ETA'
-        if percent is not None:
-            percent_str = self.format_percent(percent)
-        else:
-            percent_str = 'Unknown %'
-        speed_str = self.format_speed(speed)
 
-        msg = ('%s of %s at %s ETA %s' %
-               (percent_str, data_len_str, speed_str, eta_str))
-        self._report_progress_status(msg)
-
-    def report_progress_live_stream(self, downloaded_data_len, speed, elapsed):
-        if self.params.get('noprogress', False):
+        if s['status'] != 'downloading':
             return
-        downloaded_str = format_bytes(downloaded_data_len)
-        speed_str = self.format_speed(speed)
-        elapsed_str = FileDownloader.format_seconds(elapsed)
-        msg = '%s at %s (%s)' % (downloaded_str, speed_str, elapsed_str)
-        self._report_progress_status(msg)
 
-    def report_finish(self, data_len_str, tot_time):
-        """Report download finished."""
-        if self.params.get('noprogress', False):
-            self.to_screen('[download] Download completed')
+        if s.get('eta') is not None:
+            s['_eta_str'] = self.format_eta(s['eta'])
         else:
-            self._report_progress_status(
-                ('100%% of %s in %s' %
-                 (data_len_str, self.format_seconds(tot_time))),
-                is_last_line=True)
+            s['_eta_str'] = 'Unknown ETA'
+
+        if s.get('total_bytes') and s.get('downloaded_bytes') is not None:
+            s['_percent_str'] = self.format_percent(100 * s['downloaded_bytes'] / s['total_bytes'])
+        elif s.get('total_bytes_estimate') and s.get('downloaded_bytes') is not None:
+            s['_percent_str'] = self.format_percent(100 * s['downloaded_bytes'] / s['total_bytes_estimate'])
+        else:
+            if s.get('downloaded_bytes') == 0:
+                s['_percent_str'] = self.format_percent(0)
+            else:
+                s['_percent_str'] = 'Unknown %'
+
+        if s.get('speed') is not None:
+            s['_speed_str'] = self.format_speed(s['speed'])
+        else:
+            s['_speed_str'] = 'Unknown speed'
+
+        if s.get('total_bytes') is not None:
+            s['_total_bytes_str'] = format_bytes(s['total_bytes'])
+            msg_template = '%(_percent_str)s of %(_total_bytes_str)s at %(_speed_str)s ETA %(_eta_str)s'
+        elif s.get('total_bytes_estimate') is not None:
+            s['_total_bytes_estimate_str'] = format_bytes(s['total_bytes_estimate'])
+            msg_template = '%(_percent_str)s of ~%(_total_bytes_estimate_str)s at %(_speed_str)s ETA %(_eta_str)s'
+        else:
+            if s.get('downloaded_bytes') is not None:
+                s['_downloaded_bytes_str'] = format_bytes(s['downloaded_bytes'])
+                if s.get('elapsed'):
+                    s['_elapsed_str'] = self.format_seconds(s['elapsed'])
+                    msg_template = '%(_downloaded_bytes_str)s at %(_speed_str)s (%(_elapsed_str)s)'
+                else:
+                    msg_template = '%(_downloaded_bytes_str)s at %(_speed_str)s'
+            else:
+                msg_template = '%(_percent_str)s % at %(_speed_str)s ETA %(_eta_str)s'
+
+        self._report_progress_status(msg_template % s)
 
     def report_resuming_byte(self, resume_len):
         """Report attempt to resume at given byte."""
@@ -264,7 +296,7 @@ class FileDownloader(object):
 
     def report_retry(self, count, retries):
         """Report retry in case of HTTP error 5xx"""
-        self.to_screen('[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count, retries))
+        self.to_screen('[download] Got server HTTP error. Retrying (attempt %d of %.0f)...' % (count, retries))
 
     def report_file_already_downloaded(self, file_name):
         """Report file has already been fully downloaded."""
@@ -281,8 +313,20 @@ class FileDownloader(object):
         """Download to a filename using the info from info_dict
         Return True on success and False otherwise
         """
+
+        nooverwrites_and_exists = (
+            self.params.get('nooverwrites', False) and
+            os.path.exists(encodeFilename(filename))
+        )
+
+        continuedl_and_exists = (
+            self.params.get('continuedl', True) and
+            os.path.isfile(encodeFilename(filename)) and
+            not self.params.get('nopart', False)
+        )
+
         # Check file already present
-        if self.params.get('continuedl', False) and os.path.isfile(encodeFilename(filename)) and not self.params.get('nopart', False):
+        if filename != '-' and (nooverwrites_and_exists or continuedl_and_exists):
             self.report_file_already_downloaded(filename)
             self._hook_progress({
                 'filename': filename,
@@ -291,6 +335,11 @@ class FileDownloader(object):
             })
             return True
 
+        sleep_interval = self.params.get('sleep_interval')
+        if sleep_interval:
+            self.to_screen('[download] Sleeping %s seconds...' % sleep_interval)
+            time.sleep(sleep_interval)
+
         return self.real_download(filename, info_dict)
 
     def real_download(self, filename, info_dict):
@@ -302,19 +351,23 @@ class FileDownloader(object):
             ph(status)
 
     def add_progress_hook(self, ph):
-        """ ph gets called on download progress, with a dictionary with the entries
-        * filename: The final filename
-        * status: One of "downloading" and "finished"
-
-        It can also have some of the following entries:
-
-        * downloaded_bytes: Bytes on disks
-        * total_bytes: Total bytes, None if unknown
-        * tmpfilename: The filename we're currently writing to
-        * eta: The estimated time in seconds, None if unknown
-        * speed: The download speed in bytes/second, None if unknown
-
-        Hooks are guaranteed to be called at least once (with status "finished")
-        if the download is successful.
-        """
+        # See YoutubeDl.py (search for progress_hooks) for a description of
+        # this interface
         self._progress_hooks.append(ph)
+
+    def _debug_cmd(self, args, exe=None):
+        if not self.params.get('verbose', False):
+            return
+
+        str_args = [decodeArgument(a) for a in args]
+
+        if exe is None:
+            exe = os.path.basename(str_args[0])
+
+        try:
+            import pipes
+            shell_quote = lambda args: ' '.join(map(pipes.quote, str_args))
+        except ImportError:
+            shell_quote = repr
+        self.to_screen('[debug] %s command line: %s' % (
+            exe, shell_quote(str_args)))

youtube_dl/downloader/dash.py

@@ -0,0 +1,67 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import FileDownloader
+from ..utils import sanitized_Request
+
+
+class DashSegmentsFD(FileDownloader):
+    """
+    Download segments in a DASH manifest
+    """
+    def real_download(self, filename, info_dict):
+        self.report_destination(filename)
+        tmpfilename = self.temp_name(filename)
+        base_url = info_dict['url']
+        segment_urls = info_dict['segment_urls']
+
+        is_test = self.params.get('test', False)
+        remaining_bytes = self._TEST_FILE_SIZE if is_test else None
+        byte_counter = 0
+
+        def append_url_to_file(outf, target_url, target_name, remaining_bytes=None):
+            self.to_screen('[DashSegments] %s: Downloading %s' % (info_dict['id'], target_name))
+            req = sanitized_Request(target_url)
+            if remaining_bytes is not None:
+                req.add_header('Range', 'bytes=0-%d' % (remaining_bytes - 1))
+
+            data = self.ydl.urlopen(req).read()
+
+            if remaining_bytes is not None:
+                data = data[:remaining_bytes]
+
+            outf.write(data)
+            return len(data)
+
+        def combine_url(base_url, target_url):
+            if re.match(r'^https?://', target_url):
+                return target_url
+            return '%s%s%s' % (base_url, '' if base_url.endswith('/') else '/', target_url)
+
+        with open(tmpfilename, 'wb') as outf:
+            if info_dict.get('initialization_url'):
+                append_url_to_file(
+                    outf, combine_url(base_url, info_dict['initialization_url']),
+                    'initialization segment')
+            for i, segment_url in enumerate(segment_urls):
+                segment_len = append_url_to_file(
+                    outf, combine_url(base_url, segment_url),
+                    'segment %d / %d' % (i + 1, len(segment_urls)),
+                    remaining_bytes)
+                byte_counter += segment_len
+                if remaining_bytes is not None:
+                    remaining_bytes -= segment_len
+                    if remaining_bytes <= 0:
+                        break
+
+        self.try_rename(tmpfilename, filename)
+
+        self._hook_progress({
+            'downloaded_bytes': byte_counter,
+            'total_bytes': byte_counter,
+            'filename': filename,
+            'status': 'finished',
+        })
+
+        return True

youtube_dl/downloader/external.py

@@ -0,0 +1,155 @@
+from __future__ import unicode_literals
+
+import os.path
+import subprocess
+
+from .common import FileDownloader
+from ..utils import (
+    cli_option,
+    cli_valueless_option,
+    cli_bool_option,
+    cli_configuration_args,
+    encodeFilename,
+    encodeArgument,
+)
+
+
+class ExternalFD(FileDownloader):
+    def real_download(self, filename, info_dict):
+        self.report_destination(filename)
+        tmpfilename = self.temp_name(filename)
+
+        retval = self._call_downloader(tmpfilename, info_dict)
+        if retval == 0:
+            fsize = os.path.getsize(encodeFilename(tmpfilename))
+            self.to_screen('\r[%s] Downloaded %s bytes' % (self.get_basename(), fsize))
+            self.try_rename(tmpfilename, filename)
+            self._hook_progress({
+                'downloaded_bytes': fsize,
+                'total_bytes': fsize,
+                'filename': filename,
+                'status': 'finished',
+            })
+            return True
+        else:
+            self.to_stderr('\n')
+            self.report_error('%s exited with code %d' % (
+                self.get_basename(), retval))
+            return False
+
+    @classmethod
+    def get_basename(cls):
+        return cls.__name__[:-2].lower()
+
+    @property
+    def exe(self):
+        return self.params.get('external_downloader')
+
+    @classmethod
+    def supports(cls, info_dict):
+        return info_dict['protocol'] in ('http', 'https', 'ftp', 'ftps')
+
+    def _option(self, command_option, param):
+        return cli_option(self.params, command_option, param)
+
+    def _bool_option(self, command_option, param, true_value='true', false_value='false', separator=None):
+        return cli_bool_option(self.params, command_option, param, true_value, false_value, separator)
+
+    def _valueless_option(self, command_option, param, expected_value=True):
+        return cli_valueless_option(self.params, command_option, param, expected_value)
+
+    def _configuration_args(self, default=[]):
+        return cli_configuration_args(self.params, 'external_downloader_args', default)
+
+    def _call_downloader(self, tmpfilename, info_dict):
+        """ Either overwrite this or implement _make_cmd """
+        cmd = [encodeArgument(a) for a in self._make_cmd(tmpfilename, info_dict)]
+
+        self._debug_cmd(cmd)
+
+        p = subprocess.Popen(
+            cmd, stderr=subprocess.PIPE)
+        _, stderr = p.communicate()
+        if p.returncode != 0:
+            self.to_stderr(stderr)
+        return p.returncode
+
+
+class CurlFD(ExternalFD):
+    def _make_cmd(self, tmpfilename, info_dict):
+        cmd = [self.exe, '--location', '-o', tmpfilename]
+        for key, val in info_dict['http_headers'].items():
+            cmd += ['--header', '%s: %s' % (key, val)]
+        cmd += self._option('--interface', 'source_address')
+        cmd += self._option('--proxy', 'proxy')
+        cmd += self._valueless_option('--insecure', 'nocheckcertificate')
+        cmd += self._configuration_args()
+        cmd += ['--', info_dict['url']]
+        return cmd
+
+
+class AxelFD(ExternalFD):
+    def _make_cmd(self, tmpfilename, info_dict):
+        cmd = [self.exe, '-o', tmpfilename]
+        for key, val in info_dict['http_headers'].items():
+            cmd += ['-H', '%s: %s' % (key, val)]
+        cmd += self._configuration_args()
+        cmd += ['--', info_dict['url']]
+        return cmd
+
+
+class WgetFD(ExternalFD):
+    def _make_cmd(self, tmpfilename, info_dict):
+        cmd = [self.exe, '-O', tmpfilename, '-nv', '--no-cookies']
+        for key, val in info_dict['http_headers'].items():
+            cmd += ['--header', '%s: %s' % (key, val)]
+        cmd += self._option('--bind-address', 'source_address')
+        cmd += self._option('--proxy', 'proxy')
+        cmd += self._valueless_option('--no-check-certificate', 'nocheckcertificate')
+        cmd += self._configuration_args()
+        cmd += ['--', info_dict['url']]
+        return cmd
+
+
+class Aria2cFD(ExternalFD):
+    def _make_cmd(self, tmpfilename, info_dict):
+        cmd = [self.exe, '-c']
+        cmd += self._configuration_args([
+            '--min-split-size', '1M', '--max-connection-per-server', '4'])
+        dn = os.path.dirname(tmpfilename)
+        if dn:
+            cmd += ['--dir', dn]
+        cmd += ['--out', os.path.basename(tmpfilename)]
+        for key, val in info_dict['http_headers'].items():
+            cmd += ['--header', '%s: %s' % (key, val)]
+        cmd += self._option('--interface', 'source_address')
+        cmd += self._option('--all-proxy', 'proxy')
+        cmd += self._bool_option('--check-certificate', 'nocheckcertificate', 'false', 'true', '=')
+        cmd += ['--', info_dict['url']]
+        return cmd
+
+
+class HttpieFD(ExternalFD):
+    def _make_cmd(self, tmpfilename, info_dict):
+        cmd = ['http', '--download', '--output', tmpfilename, info_dict['url']]
+        for key, val in info_dict['http_headers'].items():
+            cmd += ['%s:%s' % (key, val)]
+        return cmd
+
+_BY_NAME = dict(
+    (klass.get_basename(), klass)
+    for name, klass in globals().items()
+    if name.endswith('FD') and name != 'ExternalFD'
+)
+
+
+def list_external_downloaders():
+    return sorted(_BY_NAME.keys())
+
+
+def get_external_downloader(external_downloader):
+    """ Given the name of the executable, see whether we support the given
+        downloader . """
+    # Drop .exe extension on Windows
+    bn = os.path.splitext(os.path.basename(external_downloader))[0]
+    return _BY_NAME[bn]

youtube_dl/downloader/f4m.py

@@ -1,21 +1,24 @@
-from __future__ import unicode_literals
+from __future__ import division, unicode_literals
 
 import base64
 import io
 import itertools
 import os
 import time
-import xml.etree.ElementTree as etree
 
-from .common import FileDownloader
-from .http import HttpFD
+from .fragment import FragmentFD
+from ..compat import (
+    compat_etree_fromstring,
+    compat_urlparse,
+    compat_urllib_error,
+    compat_urllib_parse_urlparse,
+)
 from ..utils import (
+    encodeFilename,
+    fix_xml_ampersands,
+    sanitize_open,
     struct_pack,
     struct_unpack,
-    compat_urlparse,
-    format_bytes,
-    encodeFilename,
-    sanitize_open,
     xpath_text,
 )
 
@@ -120,7 +123,8 @@ class FlvReader(io.BytesIO):
 
         self.read_unsigned_int()  # BootstrapinfoVersion
         # Profile,Live,Update,Reserved
-        self.read(1)
+        flags = self.read_unsigned_char()
+        live = flags & 0x20 != 0
         # time scale
         self.read_unsigned_int()
         # CurrentMediaTime
@@ -159,6 +163,7 @@ class FlvReader(io.BytesIO):
         return {
             'segments': segments,
             'fragments': fragments,
+            'live': live,
         }
 
     def read_bootstrap_info(self):
@@ -175,68 +180,129 @@ def build_fragments_list(boot_info):
     """ Return a list of (segment, fragment) for each fragment in the video """
     res = []
     segment_run_table = boot_info['segments'][0]
-    # I've only found videos with one segment
-    segment_run_entry = segment_run_table['segment_run'][0]
-    n_frags = segment_run_entry[1]
     fragment_run_entry_table = boot_info['fragments'][0]['fragments']
     first_frag_number = fragment_run_entry_table[0]['first']
-    for (i, frag_number) in zip(range(1, n_frags + 1), itertools.count(first_frag_number)):
-        res.append((1, frag_number))
+    fragments_counter = itertools.count(first_frag_number)
+    for segment, fragments_count in segment_run_table['segment_run']:
+        for _ in range(fragments_count):
+            res.append((segment, next(fragments_counter)))
+
+    if boot_info['live']:
+        res = res[-2:]
+
     return res
 
 
-def write_flv_header(stream, metadata):
-    """Writes the FLV header and the metadata to stream"""
+def write_unsigned_int(stream, val):
+    stream.write(struct_pack('!I', val))
+
+
+def write_unsigned_int_24(stream, val):
+    stream.write(struct_pack('!I', val)[1:])
+
+
+def write_flv_header(stream):
+    """Writes the FLV header to stream"""
     # FLV header
     stream.write(b'FLV\x01')
     stream.write(b'\x05')
     stream.write(b'\x00\x00\x00\x09')
-    # FLV File body
     stream.write(b'\x00\x00\x00\x00')
-    # FLVTAG
-    # Script data
-    stream.write(b'\x12')
-    # Size of the metadata with 3 bytes
-    stream.write(struct_pack('!L', len(metadata))[1:])
-    stream.write(b'\x00\x00\x00\x00\x00\x00\x00')
-    stream.write(metadata)
-    # Magic numbers extracted from the output files produced by AdobeHDS.php
-    #(https://github.com/K-S-V/Scripts)
-    stream.write(b'\x00\x00\x01\x73')
+
+
+def write_metadata_tag(stream, metadata):
+    """Writes optional metadata tag to stream"""
+    SCRIPT_TAG = b'\x12'
+    FLV_TAG_HEADER_LEN = 11
+
+    if metadata:
+        stream.write(SCRIPT_TAG)
+        write_unsigned_int_24(stream, len(metadata))
+        stream.write(b'\x00\x00\x00\x00\x00\x00\x00')
+        stream.write(metadata)
+        write_unsigned_int(stream, FLV_TAG_HEADER_LEN + len(metadata))
 
 
 def _add_ns(prop):
     return '{http://ns.adobe.com/f4m/1.0}%s' % prop
 
 
-class HttpQuietDownloader(HttpFD):
-    def to_screen(self, *args, **kargs):
-        pass
-
-
-class F4mFD(FileDownloader):
+class F4mFD(FragmentFD):
     """
     A downloader for f4m manifests or AdobeHDS.
     """
 
+    FD_NAME = 'f4m'
+
+    def _get_unencrypted_media(self, doc):
+        media = doc.findall(_add_ns('media'))
+        if not media:
+            self.report_error('No media found')
+        for e in (doc.findall(_add_ns('drmAdditionalHeader')) +
+                  doc.findall(_add_ns('drmAdditionalHeaderSet'))):
+            # If id attribute is missing it's valid for all media nodes
+            # without drmAdditionalHeaderId or drmAdditionalHeaderSetId attribute
+            if 'id' not in e.attrib:
+                self.report_error('Missing ID in f4m DRM')
+        media = list(filter(lambda e: 'drmAdditionalHeaderId' not in e.attrib and
+                                      'drmAdditionalHeaderSetId' not in e.attrib,
+                            media))
+        if not media:
+            self.report_error('Unsupported DRM')
+        return media
+
+    def _get_bootstrap_from_url(self, bootstrap_url):
+        bootstrap = self.ydl.urlopen(bootstrap_url).read()
+        return read_bootstrap_info(bootstrap)
+
+    def _update_live_fragments(self, bootstrap_url, latest_fragment):
+        fragments_list = []
+        retries = 30
+        while (not fragments_list) and (retries > 0):
+            boot_info = self._get_bootstrap_from_url(bootstrap_url)
+            fragments_list = build_fragments_list(boot_info)
+            fragments_list = [f for f in fragments_list if f[1] > latest_fragment]
+            if not fragments_list:
+                # Retry after a while
+                time.sleep(5.0)
+                retries -= 1
+
+        if not fragments_list:
+            self.report_error('Failed to update fragments')
+
+        return fragments_list
+
+    def _parse_bootstrap_node(self, node, base_url):
+        # Sometimes non empty inline bootstrap info can be specified along
+        # with bootstrap url attribute (e.g. dummy inline bootstrap info
+        # contains whitespace characters in [1]). We will prefer bootstrap
+        # url over inline bootstrap info when present.
+        # 1. http://live-1-1.rutube.ru/stream/1024/HDS/SD/C2NKsS85HQNckgn5HdEmOQ/1454167650/S-s604419906/move/four/dirs/upper/1024-576p.f4m
+        bootstrap_url = node.get('url')
+        if bootstrap_url:
+            bootstrap_url = compat_urlparse.urljoin(
+                base_url, bootstrap_url)
+            boot_info = self._get_bootstrap_from_url(bootstrap_url)
+        else:
+            bootstrap_url = None
+            bootstrap = base64.b64decode(node.text.encode('ascii'))
+            boot_info = read_bootstrap_info(bootstrap)
+        return boot_info, bootstrap_url
+
     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
         requested_bitrate = info_dict.get('tbr')
-        self.to_screen('[download] Downloading f4m manifest')
-        manifest = self.ydl.urlopen(man_url).read()
-        self.report_destination(filename)
-        http_dl = HttpQuietDownloader(
-            self.ydl,
-            {
-                'continuedl': True,
-                'quiet': True,
-                'noprogress': True,
-                'test': self.params.get('test', False),
-            }
-        )
+        self.to_screen('[%s] Downloading f4m manifest' % self.FD_NAME)
+        urlh = self.ydl.urlopen(man_url)
+        man_url = urlh.geturl()
+        # Some manifests may be malformed, e.g. prosiebensat1 generated manifests
+        # (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244
+        # and https://github.com/rg3/youtube-dl/issues/7823)
+        manifest = fix_xml_ampersands(urlh.read().decode('utf-8', 'ignore')).strip()
 
-        doc = etree.fromstring(manifest)
-        formats = [(int(f.attrib.get('bitrate', -1)), f) for f in doc.findall(_add_ns('media'))]
+        doc = compat_etree_fromstring(manifest)
+        formats = [(int(f.attrib.get('bitrate', -1)), f)
+                   for f in self._get_unencrypted_media(doc)]
         if requested_bitrate is None:
             # get the best format
             formats = sorted(formats, key=lambda f: f[0])
@@ -247,90 +313,91 @@ class F4mFD(FileDownloader):
 
         base_url = compat_urlparse.urljoin(man_url, media.attrib['url'])
         bootstrap_node = doc.find(_add_ns('bootstrapInfo'))
-        if bootstrap_node.text is None:
-            bootstrap_url = compat_urlparse.urljoin(
-                base_url, bootstrap_node.attrib['url'])
-            bootstrap = self.ydl.urlopen(bootstrap_url).read()
+        boot_info, bootstrap_url = self._parse_bootstrap_node(bootstrap_node, base_url)
+        live = boot_info['live']
+        metadata_node = media.find(_add_ns('metadata'))
+        if metadata_node is not None:
+            metadata = base64.b64decode(metadata_node.text.encode('ascii'))
         else:
-            bootstrap = base64.b64decode(bootstrap_node.text)
-        metadata = base64.b64decode(media.find(_add_ns('metadata')).text)
-        boot_info = read_bootstrap_info(bootstrap)
+            metadata = None
 
         fragments_list = build_fragments_list(boot_info)
-        if self.params.get('test', False):
+        test = self.params.get('test', False)
+        if test:
             # We only download the first fragment
             fragments_list = fragments_list[:1]
         total_frags = len(fragments_list)
         # For some akamai manifests we'll need to add a query to the fragment url
         akamai_pv = xpath_text(doc, _add_ns('pv-2.0'))
 
-        tmpfilename = self.temp_name(filename)
-        (dest_stream, tmpfilename) = sanitize_open(tmpfilename, 'wb')
-        write_flv_header(dest_stream, metadata)
-
-        # This dict stores the download progress, it's updated by the progress
-        # hook
-        state = {
-            'downloaded_bytes': 0,
-            'frag_counter': 0,
+        ctx = {
+            'filename': filename,
+            'total_frags': total_frags,
+            'live': live,
         }
-        start = time.time()
 
-        def frag_progress_hook(status):
-            frag_total_bytes = status.get('total_bytes', 0)
-            estimated_size = (state['downloaded_bytes'] +
-                              (total_frags - state['frag_counter']) * frag_total_bytes)
-            if status['status'] == 'finished':
-                state['downloaded_bytes'] += frag_total_bytes
-                state['frag_counter'] += 1
-                progress = self.calc_percent(state['frag_counter'], total_frags)
-                byte_counter = state['downloaded_bytes']
-            else:
-                frag_downloaded_bytes = status['downloaded_bytes']
-                byte_counter = state['downloaded_bytes'] + frag_downloaded_bytes
-                frag_progress = self.calc_percent(frag_downloaded_bytes,
-                                                  frag_total_bytes)
-                progress = self.calc_percent(state['frag_counter'], total_frags)
-                progress += frag_progress / float(total_frags)
+        self._prepare_frag_download(ctx)
 
-            eta = self.calc_eta(start, time.time(), estimated_size, byte_counter)
-            self.report_progress(progress, format_bytes(estimated_size),
-                                 status.get('speed'), eta)
-        http_dl.add_progress_hook(frag_progress_hook)
+        dest_stream = ctx['dest_stream']
+
+        write_flv_header(dest_stream)
+        if not live:
+            write_metadata_tag(dest_stream, metadata)
+
+        base_url_parsed = compat_urllib_parse_urlparse(base_url)
+
+        self._start_frag_download(ctx)
 
         frags_filenames = []
-        for (seg_i, frag_i) in fragments_list:
+        while fragments_list:
+            seg_i, frag_i = fragments_list.pop(0)
             name = 'Seg%d-Frag%d' % (seg_i, frag_i)
-            url = base_url + name
+            query = []
+            if base_url_parsed.query:
+                query.append(base_url_parsed.query)
             if akamai_pv:
-                url += '?' + akamai_pv.strip(';')
-            frag_filename = '%s-%s' % (tmpfilename, name)
-            success = http_dl.download(frag_filename, {'url': url})
-            if not success:
-                return False
-            with open(frag_filename, 'rb') as down:
+                query.append(akamai_pv.strip(';'))
+            if info_dict.get('extra_param_to_segment_url'):
+                query.append(info_dict['extra_param_to_segment_url'])
+            url_parsed = base_url_parsed._replace(path=base_url_parsed.path + name, query='&'.join(query))
+            frag_filename = '%s-%s' % (ctx['tmpfilename'], name)
+            try:
+                success = ctx['dl'].download(frag_filename, {'url': url_parsed.geturl()})
+                if not success:
+                    return False
+                (down, frag_sanitized) = sanitize_open(frag_filename, 'rb')
                 down_data = down.read()
+                down.close()
                 reader = FlvReader(down_data)
                 while True:
                     _, box_type, box_data = reader.read_box_info()
                     if box_type == b'mdat':
                         dest_stream.write(box_data)
                         break
-            frags_filenames.append(frag_filename)
+                if live:
+                    os.remove(encodeFilename(frag_sanitized))
+                else:
+                    frags_filenames.append(frag_sanitized)
+            except (compat_urllib_error.HTTPError, ) as err:
+                if live and (err.code == 404 or err.code == 410):
+                    # We didn't keep up with the live window. Continue
+                    # with the next available fragment.
+                    msg = 'Fragment %d unavailable' % frag_i
+                    self.report_warning(msg)
+                    fragments_list = []
+                else:
+                    raise
 
-        dest_stream.close()
-        self.report_finish(format_bytes(state['downloaded_bytes']), time.time() - start)
+            if not fragments_list and not test and live and bootstrap_url:
+                fragments_list = self._update_live_fragments(bootstrap_url, frag_i)
+                total_frags += len(fragments_list)
+                if fragments_list and (fragments_list[0][1] > frag_i + 1):
+                    msg = 'Missed %d fragments' % (fragments_list[0][1] - (frag_i + 1))
+                    self.report_warning(msg)
+
+        self._finish_frag_download(ctx)
 
-        self.try_rename(tmpfilename, filename)
         for frag_file in frags_filenames:
-            os.remove(frag_file)
-
-        fsize = os.path.getsize(encodeFilename(filename))
-        self._hook_progress({
-            'downloaded_bytes': fsize,
-            'total_bytes': fsize,
-            'filename': filename,
-            'status': 'finished',
-        })
+            os.remove(encodeFilename(frag_file))
 
         return True

youtube_dl/downloader/fragment.py

@@ -0,0 +1,122 @@
+from __future__ import division, unicode_literals
+
+import os
+import time
+
+from .common import FileDownloader
+from .http import HttpFD
+from ..utils import (
+    encodeFilename,
+    sanitize_open,
+)
+
+
+class HttpQuietDownloader(HttpFD):
+    def to_screen(self, *args, **kargs):
+        pass
+
+
+class FragmentFD(FileDownloader):
+    """
+    A base file downloader class for fragmented media (e.g. f4m/m3u8 manifests).
+    """
+
+    def _prepare_and_start_frag_download(self, ctx):
+        self._prepare_frag_download(ctx)
+        self._start_frag_download(ctx)
+
+    def _prepare_frag_download(self, ctx):
+        if 'live' not in ctx:
+            ctx['live'] = False
+        self.to_screen(
+            '[%s] Total fragments: %s'
+            % (self.FD_NAME, ctx['total_frags'] if not ctx['live'] else 'unknown (live)'))
+        self.report_destination(ctx['filename'])
+        dl = HttpQuietDownloader(
+            self.ydl,
+            {
+                'continuedl': True,
+                'quiet': True,
+                'noprogress': True,
+                'ratelimit': self.params.get('ratelimit', None),
+                'retries': self.params.get('retries', 0),
+                'test': self.params.get('test', False),
+            }
+        )
+        tmpfilename = self.temp_name(ctx['filename'])
+        dest_stream, tmpfilename = sanitize_open(tmpfilename, 'wb')
+        ctx.update({
+            'dl': dl,
+            'dest_stream': dest_stream,
+            'tmpfilename': tmpfilename,
+        })
+
+    def _start_frag_download(self, ctx):
+        total_frags = ctx['total_frags']
+        # This dict stores the download progress, it's updated by the progress
+        # hook
+        state = {
+            'status': 'downloading',
+            'downloaded_bytes': 0,
+            'frag_index': 0,
+            'frag_count': total_frags,
+            'filename': ctx['filename'],
+            'tmpfilename': ctx['tmpfilename'],
+        }
+
+        start = time.time()
+        ctx.update({
+            'started': start,
+            # Total complete fragments downloaded so far in bytes
+            'complete_frags_downloaded_bytes': 0,
+            # Amount of fragment's bytes downloaded by the time of the previous
+            # frag progress hook invocation
+            'prev_frag_downloaded_bytes': 0,
+        })
+
+        def frag_progress_hook(s):
+            if s['status'] not in ('downloading', 'finished'):
+                return
+
+            time_now = time.time()
+            state['elapsed'] = time_now - start
+            frag_total_bytes = s.get('total_bytes') or 0
+            if not ctx['live']:
+                estimated_size = (
+                    (ctx['complete_frags_downloaded_bytes'] + frag_total_bytes) /
+                    (state['frag_index'] + 1) * total_frags)
+                state['total_bytes_estimate'] = estimated_size
+
+            if s['status'] == 'finished':
+                state['frag_index'] += 1
+                state['downloaded_bytes'] += frag_total_bytes - ctx['prev_frag_downloaded_bytes']
+                ctx['complete_frags_downloaded_bytes'] = state['downloaded_bytes']
+                ctx['prev_frag_downloaded_bytes'] = 0
+            else:
+                frag_downloaded_bytes = s['downloaded_bytes']
+                state['downloaded_bytes'] += frag_downloaded_bytes - ctx['prev_frag_downloaded_bytes']
+                if not ctx['live']:
+                    state['eta'] = self.calc_eta(
+                        start, time_now, estimated_size,
+                        state['downloaded_bytes'])
+                state['speed'] = s.get('speed')
+                ctx['prev_frag_downloaded_bytes'] = frag_downloaded_bytes
+            self._hook_progress(state)
+
+        ctx['dl'].add_progress_hook(frag_progress_hook)
+
+        return start
+
+    def _finish_frag_download(self, ctx):
+        ctx['dest_stream'].close()
+        elapsed = time.time() - ctx['started']
+        self.try_rename(ctx['tmpfilename'], ctx['filename'])
+        fsize = os.path.getsize(encodeFilename(ctx['filename']))
+
+        self._hook_progress({
+            'downloaded_bytes': fsize,
+            'total_bytes': fsize,
+            'filename': ctx['filename'],
+            'status': 'finished',
+            'elapsed': elapsed,
+        })

youtube_dl/downloader/hls.py

@@ -3,14 +3,18 @@ from __future__ import unicode_literals
 import os
 import re
 import subprocess
+import sys
 
-from ..postprocessor.ffmpeg import FFmpegPostProcessor
 from .common import FileDownloader
+from .fragment import FragmentFD
+
+from ..compat import compat_urlparse
+from ..postprocessor.ffmpeg import FFmpegPostProcessor
 from ..utils import (
-    compat_urlparse,
-    compat_urllib_request,
-    check_executable,
+    encodeArgument,
     encodeFilename,
+    sanitize_open,
+    handle_youtubedl_headers,
 )
 
 
@@ -20,26 +24,48 @@ class HlsFD(FileDownloader):
         self.report_destination(filename)
         tmpfilename = self.temp_name(filename)
 
-        args = [
-            '-y', '-i', url, '-f', 'mp4', '-c', 'copy',
-            '-bsf:a', 'aac_adtstoasc',
-            encodeFilename(tmpfilename, for_subprocess=True)]
-
-        for program in ['avconv', 'ffmpeg']:
-            if check_executable(program, ['-version']):
-                break
-        else:
+        ffpp = FFmpegPostProcessor(downloader=self)
+        if not ffpp.available:
             self.report_error('m3u8 download detected but ffmpeg or avconv could not be found. Please install one.')
             return False
-        cmd = [program] + args
-
-        ffpp = FFmpegPostProcessor(downloader=self)
         ffpp.check_version()
 
-        retval = subprocess.call(cmd)
+        args = [ffpp.executable, '-y']
+
+        if info_dict['http_headers'] and re.match(r'^https?://', url):
+            # Trailing \r\n after each HTTP header is important to prevent warning from ffmpeg/avconv:
+            # [http @ 00000000003d2fa0] No trailing CRLF found in HTTP header.
+            headers = handle_youtubedl_headers(info_dict['http_headers'])
+            args += [
+                '-headers',
+                ''.join('%s: %s\r\n' % (key, val) for key, val in headers.items())]
+
+        args += ['-i', url, '-c', 'copy']
+        if self.params.get('hls_use_mpegts', False):
+            args += ['-f', 'mpegts']
+        else:
+            args += ['-f', 'mp4', '-bsf:a', 'aac_adtstoasc']
+
+        args = [encodeArgument(opt) for opt in args]
+        args.append(encodeFilename(ffpp._ffmpeg_filename_argument(tmpfilename), True))
+
+        self._debug_cmd(args)
+
+        proc = subprocess.Popen(args, stdin=subprocess.PIPE)
+        try:
+            retval = proc.wait()
+        except KeyboardInterrupt:
+            # subprocces.run would send the SIGKILL signal to ffmpeg and the
+            # mp4 file couldn't be played, but if we ask ffmpeg to quit it
+            # produces a file that is playable (this is mostly useful for live
+            # streams). Note that Windows is not affected and produces playable
+            # files (see https://github.com/rg3/youtube-dl/issues/8300).
+            if sys.platform != 'win32':
+                proc.communicate(b'q')
+            raise
         if retval == 0:
             fsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen('\r[%s] %s bytes' % (cmd[0], fsize))
+            self.to_screen('\r[%s] %s bytes' % (args[0], fsize))
             self.try_rename(tmpfilename, filename)
             self._hook_progress({
                 'downloaded_bytes': fsize,
@@ -50,58 +76,55 @@ class HlsFD(FileDownloader):
             return True
         else:
             self.to_stderr('\n')
-            self.report_error('%s exited with code %d' % (program, retval))
+            self.report_error('%s exited with code %d' % (ffpp.basename, retval))
             return False
 
 
-class NativeHlsFD(FileDownloader):
+class NativeHlsFD(FragmentFD):
     """ A more limited implementation that does not require ffmpeg """
 
-    def real_download(self, filename, info_dict):
-        url = info_dict['url']
-        self.report_destination(filename)
-        tmpfilename = self.temp_name(filename)
+    FD_NAME = 'hlsnative'
 
-        self.to_screen(
-            '[hlsnative] %s: Downloading m3u8 manifest' % info_dict['id'])
-        data = self.ydl.urlopen(url).read()
-        s = data.decode('utf-8', 'ignore')
-        segment_urls = []
+    def real_download(self, filename, info_dict):
+        man_url = info_dict['url']
+        self.to_screen('[%s] Downloading m3u8 manifest' % self.FD_NAME)
+        manifest = self.ydl.urlopen(man_url).read()
+
+        s = manifest.decode('utf-8', 'ignore')
+        fragment_urls = []
         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(url, line))
-                segment_urls.append(segment_url)
-
-        is_test = self.params.get('test', False)
-        remaining_bytes = self._TEST_FILE_SIZE if is_test else None
-        byte_counter = 0
-        with open(tmpfilename, 'wb') as outf:
-            for i, segurl in enumerate(segment_urls):
-                self.to_screen(
-                    '[hlsnative] %s: Downloading segment %d / %d' %
-                    (info_dict['id'], i + 1, len(segment_urls)))
-                seg_req = compat_urllib_request.Request(segurl)
-                if remaining_bytes is not None:
-                    seg_req.add_header('Range', 'bytes=0-%d' % (remaining_bytes - 1))
-
-                segment = self.ydl.urlopen(seg_req).read()
-                if remaining_bytes is not None:
-                    segment = segment[:remaining_bytes]
-                    remaining_bytes -= len(segment)
-                outf.write(segment)
-                byte_counter += len(segment)
-                if remaining_bytes is not None and remaining_bytes <= 0:
+                    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
 
-        self._hook_progress({
-            'downloaded_bytes': byte_counter,
-            'total_bytes': byte_counter,
+        ctx = {
             'filename': filename,
-            'status': 'finished',
-        })
-        self.try_rename(tmpfilename, filename)
+            'total_frags': len(fragment_urls),
+        }
+
+        self._prepare_and_start_frag_download(ctx)
+
+        frags_filenames = []
+        for i, frag_url in enumerate(fragment_urls):
+            frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i)
+            success = ctx['dl'].download(frag_filename, {'url': frag_url})
+            if not success:
+                return False
+            down, frag_sanitized = sanitize_open(frag_filename, 'rb')
+            ctx['dest_stream'].write(down.read())
+            down.close()
+            frags_filenames.append(frag_sanitized)
+
+        self._finish_frag_download(ctx)
+
+        for frag_file in frags_filenames:
+            os.remove(encodeFilename(frag_file))
+
         return True

youtube_dl/downloader/http.py

@@ -1,17 +1,18 @@
 from __future__ import unicode_literals
 
+import errno
 import os
+import socket
 import time
+import re
 
 from .common import FileDownloader
+from ..compat import compat_urllib_error
 from ..utils import (
-    compat_urllib_request,
-    compat_urllib_error,
     ContentTooShortError,
-
     encodeFilename,
     sanitize_open,
-    format_bytes,
+    sanitized_Request,
 )
 
 
@@ -23,20 +24,11 @@ class HttpFD(FileDownloader):
 
         # Do not include the Accept-Encoding header
         headers = {'Youtubedl-no-compression': 'True'}
-        if 'user_agent' in info_dict:
-            headers['Youtubedl-user-agent'] = info_dict['user_agent']
-        if 'http_referer' in info_dict:
-            headers['Referer'] = info_dict['http_referer']
         add_headers = info_dict.get('http_headers')
         if add_headers:
             headers.update(add_headers)
-        data = info_dict.get('http_post_data')
-        http_method = info_dict.get('http_method')
-        basic_request = compat_urllib_request.Request(url, data, headers)
-        request = compat_urllib_request.Request(url, data, headers)
-        if http_method is not None:
-            basic_request.get_method = lambda: http_method
-            request.get_method = lambda: http_method
+        basic_request = sanitized_Request(url, None, headers)
+        request = sanitized_Request(url, None, headers)
 
         is_test = self.params.get('test', False)
 
@@ -51,7 +43,7 @@ class HttpFD(FileDownloader):
 
         open_mode = 'wb'
         if resume_len != 0:
-            if self.params.get('continuedl', False):
+            if self.params.get('continuedl', True):
                 self.report_resuming_byte(resume_len)
                 request.add_header('Range', 'bytes=%d-' % resume_len)
                 open_mode = 'ab'
@@ -64,6 +56,24 @@ class HttpFD(FileDownloader):
             # Establish connection
             try:
                 data = self.ydl.urlopen(request)
+                # When trying to resume, Content-Range HTTP header of response has to be checked
+                # to match the value of requested Range HTTP header. This is due to a webservers
+                # that don't support resuming and serve a whole file with no Content-Range
+                # set in response despite of requested Range (see
+                # https://github.com/rg3/youtube-dl/issues/6057#issuecomment-126129799)
+                if resume_len > 0:
+                    content_range = data.headers.get('Content-Range')
+                    if content_range:
+                        content_range_m = re.search(r'bytes (\d+)-', content_range)
+                        # Content-Range is present and matches requested Range, resume is possible
+                        if content_range_m and resume_len == int(content_range_m.group(1)):
+                            break
+                    # Content-Range is either not present or invalid. Assuming remote webserver is
+                    # trying to send the whole file, resume is not possible, so wiping the local file
+                    # and performing entire redownload
+                    self.report_unable_to_resume()
+                    resume_len = 0
+                    open_mode = 'wb'
                 break
             except (compat_urllib_error.HTTPError, ) as err:
                 if (err.code < 500 or err.code >= 600) and err.code != 416:
@@ -94,6 +104,8 @@ class HttpFD(FileDownloader):
                             self._hook_progress({
                                 'filename': filename,
                                 'status': 'finished',
+                                'downloaded_bytes': resume_len,
+                                'total_bytes': resume_len,
                             })
                             return True
                         else:
@@ -102,6 +114,11 @@ class HttpFD(FileDownloader):
                             resume_len = 0
                             open_mode = 'wb'
                             break
+            except socket.error as e:
+                if e.errno != errno.ECONNRESET:
+                    # Connection reset is no problem, just retry
+                    raise
+
             # Retry
             count += 1
             if count <= retries:
@@ -132,20 +149,24 @@ class HttpFD(FileDownloader):
                 self.to_screen('\r[download] File is larger than max-filesize (%s bytes > %s bytes). Aborting.' % (data_len, max_data_len))
                 return False
 
-        data_len_str = format_bytes(data_len)
         byte_counter = 0 + resume_len
         block_size = self.params.get('buffersize', 1024)
         start = time.time()
+
+        # measure time over whole while-loop, so slow_down() and best_block_size() work together properly
+        now = None  # needed for slow_down() in the first loop run
+        before = start  # start measuring
         while True:
+
             # Download and write
-            before = time.time()
             data_block = data.read(block_size if not is_test else min(block_size, data_len - byte_counter))
-            after = time.time()
-            if len(data_block) == 0:
-                break
             byte_counter += len(data_block)
 
-            # Open file just in time
+            # exit loop when download is finished
+            if len(data_block) == 0:
+                break
+
+            # Open destination file just in time
             if stream is None:
                 try:
                     (stream, tmpfilename) = sanitize_open(tmpfilename, open_mode)
@@ -155,47 +176,62 @@ class HttpFD(FileDownloader):
                 except (OSError, IOError) as err:
                     self.report_error('unable to open for writing: %s' % str(err))
                     return False
+
+                if self.params.get('xattr_set_filesize', False) and data_len is not None:
+                    try:
+                        import xattr
+                        xattr.setxattr(tmpfilename, 'user.ytdl.filesize', str(data_len))
+                    except(OSError, IOError, ImportError) as err:
+                        self.report_error('unable to set filesize xattr: %s' % str(err))
+
             try:
                 stream.write(data_block)
             except (IOError, OSError) as err:
                 self.to_stderr('\n')
                 self.report_error('unable to write data: %s' % str(err))
                 return False
+
+            # Apply rate limit
+            self.slow_down(start, now, byte_counter - resume_len)
+
+            # end measuring of one loop run
+            now = time.time()
+            after = now
+
+            # Adjust block size
             if not self.params.get('noresizebuffer', False):
                 block_size = self.best_block_size(after - before, len(data_block))
 
+            before = after
+
             # Progress message
-            speed = self.calc_speed(start, time.time(), byte_counter - resume_len)
+            speed = self.calc_speed(start, now, byte_counter - resume_len)
             if data_len is None:
-                eta = percent = None
+                eta = None
             else:
-                percent = self.calc_percent(byte_counter, data_len)
                 eta = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len)
-            self.report_progress(percent, data_len_str, speed, eta)
 
             self._hook_progress({
+                'status': 'downloading',
                 'downloaded_bytes': byte_counter,
                 'total_bytes': data_len,
                 'tmpfilename': tmpfilename,
                 'filename': filename,
-                'status': 'downloading',
                 'eta': eta,
                 'speed': speed,
+                'elapsed': now - start,
             })
 
             if is_test and byte_counter == data_len:
                 break
 
-            # Apply rate limit
-            self.slow_down(start, byte_counter - resume_len)
-
         if stream is None:
             self.to_stderr('\n')
             self.report_error('Did not get any data blocks')
             return False
         if tmpfilename != '-':
             stream.close()
-        self.report_finish(data_len_str, (time.time() - start))
+
         if data_len is not None and byte_counter != data_len:
             raise ContentTooShortError(byte_counter, int(data_len))
         self.try_rename(tmpfilename, filename)
@@ -209,6 +245,7 @@ class HttpFD(FileDownloader):
             'total_bytes': byte_counter,
             'filename': filename,
             'status': 'finished',
+            'elapsed': time.time() - start,
         })
 
         return True

youtube_dl/downloader/rtmp.py

@@ -3,15 +3,14 @@ from __future__ import unicode_literals
 import os
 import re
 import subprocess
-import sys
 import time
 
 from .common import FileDownloader
+from ..compat import compat_str
 from ..utils import (
     check_executable,
-    compat_str,
     encodeFilename,
-    format_bytes,
+    encodeArgument,
     get_exe_version,
 )
 
@@ -51,23 +50,23 @@ class RtmpFD(FileDownloader):
                     if not resume_percent:
                         resume_percent = percent
                         resume_downloaded_data_len = downloaded_data_len
-                    eta = self.calc_eta(start, time.time(), 100 - resume_percent, percent - resume_percent)
-                    speed = self.calc_speed(start, time.time(), downloaded_data_len - resume_downloaded_data_len)
+                    time_now = time.time()
+                    eta = self.calc_eta(start, time_now, 100 - resume_percent, percent - resume_percent)
+                    speed = self.calc_speed(start, time_now, downloaded_data_len - resume_downloaded_data_len)
                     data_len = None
                     if percent > 0:
                         data_len = int(downloaded_data_len * 100 / percent)
-                    data_len_str = '~' + format_bytes(data_len)
-                    self.report_progress(percent, data_len_str, speed, eta)
-                    cursor_in_new_line = False
                     self._hook_progress({
+                        'status': 'downloading',
                         'downloaded_bytes': downloaded_data_len,
-                        'total_bytes': data_len,
+                        'total_bytes_estimate': data_len,
                         'tmpfilename': tmpfilename,
                         'filename': filename,
-                        'status': 'downloading',
                         'eta': eta,
+                        'elapsed': time_now - start,
                         'speed': speed,
                     })
+                    cursor_in_new_line = False
                 else:
                     # no percent for live streams
                     mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec', line)
@@ -75,15 +74,15 @@ class RtmpFD(FileDownloader):
                         downloaded_data_len = int(float(mobj.group(1)) * 1024)
                         time_now = time.time()
                         speed = self.calc_speed(start, time_now, downloaded_data_len)
-                        self.report_progress_live_stream(downloaded_data_len, speed, time_now - start)
-                        cursor_in_new_line = False
                         self._hook_progress({
                             'downloaded_bytes': downloaded_data_len,
                             'tmpfilename': tmpfilename,
                             'filename': filename,
                             'status': 'downloading',
+                            'elapsed': time_now - start,
                             'speed': speed,
                         })
+                        cursor_in_new_line = False
                     elif self.params.get('verbose', False):
                         if not cursor_in_new_line:
                             self.to_screen('')
@@ -104,6 +103,9 @@ class RtmpFD(FileDownloader):
         live = info_dict.get('rtmp_live', False)
         conn = info_dict.get('rtmp_conn', None)
         protocol = info_dict.get('rtmp_protocol', None)
+        real_time = info_dict.get('rtmp_real_time', False)
+        no_resume = info_dict.get('no_resume', False)
+        continue_dl = self.params.get('continuedl', True)
 
         self.report_destination(filename)
         tmpfilename = self.temp_name(filename)
@@ -115,9 +117,11 @@ class RtmpFD(FileDownloader):
             return False
 
         # Download using rtmpdump. rtmpdump returns exit code 2 when
-        # the connection was interrumpted and resuming appears to be
+        # the connection was interrupted and resuming appears to be
         # possible. This is part of rtmpdump's normal usage, AFAIK.
-        basic_args = ['rtmpdump', '--verbose', '-r', url, '-o', tmpfilename]
+        basic_args = [
+            'rtmpdump', '--verbose', '-r', url,
+            '-o', tmpfilename]
         if player_url is not None:
             basic_args += ['--swfVfy', player_url]
         if page_url is not None:
@@ -127,7 +131,7 @@ class RtmpFD(FileDownloader):
         if play_path is not None:
             basic_args += ['--playpath', play_path]
         if tc_url is not None:
-            basic_args += ['--tcUrl', url]
+            basic_args += ['--tcUrl', tc_url]
         if test:
             basic_args += ['--stop', '1']
         if flash_version is not None:
@@ -141,30 +145,18 @@ class RtmpFD(FileDownloader):
             basic_args += ['--conn', conn]
         if protocol is not None:
             basic_args += ['--protocol', protocol]
-        args = basic_args + [[], ['--resume', '--skip', '1']][not live and self.params.get('continuedl', False)]
+        if real_time:
+            basic_args += ['--realtime']
 
-        if sys.platform == 'win32' and sys.version_info < (3, 0):
-            # Windows subprocess module does not actually support Unicode
-            # on Python 2.x
-            # See http://stackoverflow.com/a/9951851/35070
-            subprocess_encoding = sys.getfilesystemencoding()
-            args = [a.encode(subprocess_encoding, 'ignore') for a in args]
-        else:
-            subprocess_encoding = None
+        args = basic_args
+        if not no_resume and continue_dl and not live:
+            args += ['--resume']
+        if not live and continue_dl:
+            args += ['--skip', '1']
 
-        if self.params.get('verbose', False):
-            if subprocess_encoding:
-                str_args = [
-                    a.decode(subprocess_encoding) if isinstance(a, bytes) else a
-                    for a in args]
-            else:
-                str_args = args
-            try:
-                import pipes
-                shell_quote = lambda args: ' '.join(map(pipes.quote, str_args))
-            except ImportError:
-                shell_quote = repr
-            self.to_screen('[debug] rtmpdump command line: ' + shell_quote(str_args))
+        args = [encodeArgument(a) for a in args]
+
+        self._debug_cmd(args, exe='rtmpdump')
 
         RD_SUCCESS = 0
         RD_FAILED = 1
@@ -181,11 +173,15 @@ class RtmpFD(FileDownloader):
             prevsize = os.path.getsize(encodeFilename(tmpfilename))
             self.to_screen('[rtmpdump] %s bytes' % prevsize)
             time.sleep(5.0)  # This seems to be needed
-            retval = run_rtmpdump(basic_args + ['-e'] + [[], ['-k', '1']][retval == RD_FAILED])
+            args = basic_args + ['--resume']
+            if retval == RD_FAILED:
+                args += ['--skip', '1']
+            args = [encodeArgument(a) for a in args]
+            retval = run_rtmpdump(args)
             cursize = os.path.getsize(encodeFilename(tmpfilename))
             if prevsize == cursize and retval == RD_FAILED:
                 break
-             # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those
+            # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those
             if prevsize == cursize and retval == RD_INCOMPLETE and cursize > 1024:
                 self.to_screen('[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
                 retval = RD_SUCCESS

youtube_dl/downloader/rtsp.py

@@ -4,31 +4,29 @@ import os
 import subprocess
 
 from .common import FileDownloader
-from ..compat import compat_subprocess_get_DEVNULL
 from ..utils import (
+    check_executable,
     encodeFilename,
 )
 
 
-class MplayerFD(FileDownloader):
+class RtspFD(FileDownloader):
     def real_download(self, filename, info_dict):
         url = info_dict['url']
         self.report_destination(filename)
         tmpfilename = self.temp_name(filename)
 
-        args = [
-            'mplayer', '-really-quiet', '-vo', 'null', '-vc', 'dummy',
-            '-dumpstream', '-dumpfile', tmpfilename, url]
-        # Check for mplayer first
-        try:
-            subprocess.call(
-                ['mplayer', '-h'],
-                stdout=compat_subprocess_get_DEVNULL(), stderr=subprocess.STDOUT)
-        except (OSError, IOError):
-            self.report_error('MMS or RTSP download detected but "%s" could not be run' % args[0])
+        if check_executable('mplayer', ['-h']):
+            args = [
+                'mplayer', '-really-quiet', '-vo', 'null', '-vc', 'dummy',
+                '-dumpstream', '-dumpfile', tmpfilename, url]
+        elif check_executable('mpv', ['-h']):
+            args = [
+                'mpv', '-really-quiet', '--vo=null', '--stream-dump=' + tmpfilename, url]
+        else:
+            self.report_error('MMS or RTSP download detected but neither "mplayer" nor "mpv" could be run. Please install any.')
             return False
 
-        # Download using mplayer.
         retval = subprocess.call(args)
         if retval == 0:
             fsize = os.path.getsize(encodeFilename(tmpfilename))
@@ -43,5 +41,5 @@ class MplayerFD(FileDownloader):
             return True
         else:
             self.to_stderr('\n')
-            self.report_error('mplayer exited with code %d' % retval)
+            self.report_error('%s exited with code %d' % (args[0], retval))
             return False

youtube_dl/extractor/__init__.py

@@ -1,97 +1,185 @@
 from __future__ import unicode_literals
 
 from .abc import ABCIE
+from .abc7news import Abc7NewsIE
 from .academicearth import AcademicEarthCourseIE
+from .acast import (
+    ACastIE,
+    ACastChannelIE,
+)
 from .addanime import AddAnimeIE
+from .adobetv import (
+    AdobeTVIE,
+    AdobeTVShowIE,
+    AdobeTVChannelIE,
+    AdobeTVVideoIE,
+)
 from .adultswim import AdultSwimIE
+from .aenetworks import AENetworksIE
 from .aftonbladet import AftonbladetIE
+from .airmozilla import AirMozillaIE
+from .aljazeera import AlJazeeraIE
+from .alphaporno import AlphaPornoIE
 from .anitube import AnitubeIE
 from .anysex import AnySexIE
 from .aol import AolIE
 from .allocine import AllocineIE
 from .aparat import AparatIE
-from .appletrailers import AppleTrailersIE
+from .appleconnect import AppleConnectIE
+from .appletrailers import (
+    AppleTrailersIE,
+    AppleTrailersSectionIE,
+)
 from .archiveorg import ArchiveOrgIE
-from .ard import ARDIE, ARDMediathekIE
+from .ard import (
+    ARDIE,
+    ARDMediathekIE,
+    SportschauIE,
+)
 from .arte import (
     ArteTvIE,
     ArteTVPlus7IE,
     ArteTVCreativeIE,
     ArteTVConcertIE,
     ArteTVFutureIE,
+    ArteTVCinemaIE,
     ArteTVDDCIE,
     ArteTVEmbedIE,
 )
-from .audiomack import AudiomackIE
-from .auengine import AUEngineIE
-from .azubu import AzubuIE
+from .atresplayer import AtresPlayerIE
+from .atttechchannel import ATTTechChannelIE
+from .audimedia import AudiMediaIE
+from .audiomack import AudiomackIE, AudiomackAlbumIE
+from .azubu import AzubuIE, AzubuLiveIE
+from .baidu import BaiduVideoIE
 from .bambuser import BambuserIE, BambuserChannelIE
 from .bandcamp import BandcampIE, BandcampAlbumIE
-from .bbccouk import BBCCoUkIE
+from .bbc import (
+    BBCCoUkIE,
+    BBCCoUkArticleIE,
+    BBCIE,
+)
 from .beeg import BeegIE
 from .behindkink import BehindKinkIE
+from .beatportpro import BeatportProIE
 from .bet import BetIE
+from .bigflix import BigflixIE
 from .bild import BildIE
 from .bilibili import BiliBiliIE
+from .bleacherreport import (
+    BleacherReportIE,
+    BleacherReportCMSIE,
+)
 from .blinkx import BlinkxIE
-from .bliptv import BlipTVIE, BlipTVUserIE
 from .bloomberg import BloombergIE
 from .bpb import BpbIE
 from .br import BRIE
 from .breakcom import BreakIE
-from .brightcove import BrightcoveIE
+from .brightcove import (
+    BrightcoveLegacyIE,
+    BrightcoveNewIE,
+)
 from .buzzfeed import BuzzFeedIE
 from .byutv import BYUtvIE
 from .c56 import C56IE
-from .canal13cl import Canal13clIE
+from .camdemy import (
+    CamdemyIE,
+    CamdemyFolderIE
+)
 from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
+from .canvas import CanvasIE
 from .cbs import CBSIE
-from .cbsnews import CBSNewsIE
+from .cbsnews import (
+    CBSNewsIE,
+    CBSNewsLiveVideoIE,
+)
+from .cbssports import CBSSportsIE
+from .ccc import CCCIE
 from .ceskatelevize import CeskaTelevizeIE
 from .channel9 import Channel9IE
+from .chaturbate import ChaturbateIE
 from .chilloutzone import ChilloutzoneIE
+from .chirbit import (
+    ChirbitIE,
+    ChirbitProfileIE,
+)
+from .cinchcast import CinchcastIE
 from .cinemassacre import CinemassacreIE
 from .clipfish import ClipfishIE
 from .cliphunter import CliphunterIE
 from .clipsyndicate import ClipsyndicateIE
 from .cloudy import CloudyIE
 from .clubic import ClubicIE
+from .clyp import ClypIE
 from .cmt import CMTIE
 from .cnet import CNETIE
 from .cnn import (
     CNNIE,
     CNNBlogsIE,
+    CNNArticleIE,
 )
 from .collegehumor import CollegeHumorIE
+from .collegerama import CollegeRamaIE
 from .comedycentral import ComedyCentralIE, ComedyCentralShowsIE
+from .comcarcoff import ComCarCoffIE
+from .commonmistakes import CommonMistakesIE, UnicodeBOMIE
 from .condenast import CondeNastIE
 from .cracked import CrackedIE
 from .criterion import CriterionIE
+from .crooksandliars import CrooksAndLiarsIE
 from .crunchyroll import (
     CrunchyrollIE,
     CrunchyrollShowPlaylistIE
 )
 from .cspan import CSpanIE
+from .ctsnews import CtsNewsIE
+from .cultureunplugged import CultureUnpluggedIE
+from .cwtv import CWTVIE
 from .dailymotion import (
     DailymotionIE,
     DailymotionPlaylistIE,
     DailymotionUserIE,
+    DailymotionCloudIE,
+)
+from .daum import (
+    DaumIE,
+    DaumClipIE,
+    DaumPlaylistIE,
+    DaumUserIE,
 )
-from .daum import DaumIE
 from .dbtv import DBTVIE
+from .dcn import (
+    DCNIE,
+    DCNVideoIE,
+    DCNLiveIE,
+    DCNSeasonIE,
+)
+from .dctp import DctpTvIE
 from .deezer import DeezerPlaylistIE
+from .democracynow import DemocracynowIE
 from .dfb import DFBIE
+from .dhm import DHMIE
 from .dotsub import DotsubIE
+from .douyutv import DouyuTVIE
+from .dplay import DPlayIE
+from .dramafever import (
+    DramaFeverIE,
+    DramaFeverSeriesIE,
+)
 from .dreisat import DreiSatIE
+from .drbonanza import DRBonanzaIE
 from .drtuber import DrTuberIE
 from .drtv import DRTVIE
+from .dvtv import DVTVIE
 from .dump import DumpIE
+from .dumpert import DumpertIE
 from .defense import DefenseGouvFrIE
 from .discovery import DiscoveryIE
-from .divxstage import DivxStageIE
 from .dropbox import DropboxIE
+from .eagleplatform import EaglePlatformIE
 from .ebaumsworld import EbaumsWorldIE
+from .echomsk import EchoMskIE
 from .ehow import EHowIE
 from .eighttracks import EightTracksIE
 from .einthusan import EinthusanIE
@@ -101,31 +189,42 @@ from .ellentv import (
     EllenTVClipsIE,
 )
 from .elpais import ElPaisIE
-from .empflix import EMPFlixIE
+from .embedly import EmbedlyIE
 from .engadget import EngadgetIE
 from .eporner import EpornerIE
+from .eroprofile import EroProfileIE
 from .escapist import EscapistIE
+from .espn import ESPNIE
+from .esri import EsriVideoIE
+from .europa import EuropaIE
 from .everyonesmixtape import EveryonesMixtapeIE
 from .exfm import ExfmIE
 from .expotv import ExpoTVIE
 from .extremetube import ExtremeTubeIE
-from .facebook import FacebookIE
+from .facebook import (
+    FacebookIE,
+    FacebookPostIE,
+)
 from .faz import FazIE
 from .fc2 import FC2IE
-from .firedrive import FiredriveIE
+from .fczenit import FczenitIE
 from .firstpost import FirstpostIE
 from .firsttv import FirstTVIE
 from .fivemin import FiveMinIE
-from .fktv import (
-    FKTVIE,
-    FKTVPosteckeIE,
-)
+from .fivetv import FiveTVIE
+from .fktv import FKTVIE
 from .flickr import FlickrIE
 from .folketinget import FolketingetIE
+from .footyroom import FootyRoomIE
 from .fourtube import FourTubeIE
+from .fox import FOXIE
 from .foxgay import FoxgayIE
 from .foxnews import FoxNewsIE
-from .franceculture import FranceCultureIE
+from .foxsports import FoxSportsIE
+from .franceculture import (
+    FranceCultureIE,
+    FranceCultureEmissionIE,
+)
 from .franceinter import FranceInterIE
 from .francetv import (
     PluzzIE,
@@ -137,76 +236,138 @@ from .francetv import (
 from .freesound import FreesoundIE
 from .freespeech import FreespeechIE
 from .freevideo import FreeVideoIE
+from .funimation import FunimationIE
 from .funnyordie import FunnyOrDieIE
+from .gameinformer import GameInformerIE
 from .gamekings import GamekingsIE
 from .gameone import (
     GameOneIE,
     GameOnePlaylistIE,
 )
+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
+from .gfycat import GfycatIE
+from .giantbomb import GiantBombIE
+from .giga import GigaIE
 from .glide import GlideIE
-from .globo import GloboIE
+from .globo import (
+    GloboIE,
+    GloboArticleIE,
+)
 from .godtube import GodTubeIE
 from .goldenmoustache import GoldenMoustacheIE
 from .golem import GolemIE
+from .googledrive import GoogleDriveIE
 from .googleplus import GooglePlusIE
 from .googlesearch import GoogleSearchIE
-from .gorillavid import GorillaVidIE
 from .goshgay import GoshgayIE
-from .grooveshark import GroovesharkIE
+from .gputechconf import GPUTechConfIE
+from .groupon import GrouponIE
 from .hark import HarkIE
+from .hearthisat import HearThisAtIE
 from .heise import HeiseIE
+from .hellporno import HellPornoIE
 from .helsinki import HelsinkiIE
 from .hentaistigma import HentaiStigmaIE
+from .historicfilms import HistoricFilmsIE
+from .hitbox import HitboxIE, HitboxLiveIE
 from .hornbunny import HornBunnyIE
-from .hostingbulk import HostingBulkIE
 from .hotnewhiphop import HotNewHipHopIE
+from .hotstar import HotStarIE
 from .howcast import HowcastIE
 from .howstuffworks import HowStuffWorksIE
 from .huffpost import HuffPostIE
 from .hypem import HypemIE
 from .iconosquare import IconosquareIE
-from .ign import IGNIE, OneUPIE
+from .ign import (
+    IGNIE,
+    OneUPIE,
+    PCMagIE,
+)
 from .imdb import (
     ImdbIE,
     ImdbListIE
 )
+from .imgur import (
+    ImgurIE,
+    ImgurAlbumIE,
+)
 from .ina import InaIE
+from .indavideo import (
+    IndavideoIE,
+    IndavideoEmbedIE,
+)
 from .infoq import InfoQIE
 from .instagram import InstagramIE, InstagramUserIE
 from .internetvideoarchive import InternetVideoArchiveIE
 from .iprima import IPrimaIE
+from .iqiyi import IqiyiIE
+from .ir90tv import Ir90TvIE
 from .ivi import (
     IviIE,
     IviCompilationIE
 )
+from .ivideon import IvideonIE
 from .izlesene import IzleseneIE
 from .jadorecettepub import JadoreCettePubIE
 from .jeuxvideo import JeuxVideoIE
 from .jove import JoveIE
-from .jukebox import JukeboxIE
+from .jwplatform import JWPlatformIE
 from .jpopsukitv import JpopsukiIE
+from .kaltura import KalturaIE
+from .kanalplay import KanalPlayIE
 from .kankan import KankanIE
+from .karaoketv import KaraoketvIE
+from .karrierevideos import KarriereVideosIE
 from .keezmovies import KeezMoviesIE
 from .khanacademy import KhanAcademyIE
 from .kickstarter import KickStarterIE
 from .keek import KeekIE
+from .konserthusetplay import KonserthusetPlayIE
 from .kontrtube import KontrTubeIE
 from .krasview import KrasViewIE
 from .ku6 import Ku6IE
+from .kuwo import (
+    KuwoIE,
+    KuwoAlbumIE,
+    KuwoChartIE,
+    KuwoSingerIE,
+    KuwoCategoryIE,
+    KuwoMvIE,
+)
 from .la7 import LA7IE
 from .laola1tv import Laola1TvIE
-from .lifenews import LifeNewsIE
+from .lecture2go import Lecture2GoIE
+from .lemonde import LemondeIE
+from .letv import (
+    LetvIE,
+    LetvTvIE,
+    LetvPlaylistIE,
+    LetvCloudIE,
+)
+from .libsyn import LibsynIE
+from .lifenews import (
+    LifeNewsIE,
+    LifeEmbedIE,
+)
+from .limelight import (
+    LimelightMediaIE,
+    LimelightChannelIE,
+    LimelightChannelListIE,
+)
 from .liveleak import LiveLeakIE
 from .livestream import (
     LivestreamIE,
     LivestreamOriginalIE,
     LivestreamShortenerIE,
 )
+from .lnkgo import LnkGoIE
+from .lovehomeporn import LoveHomePornIE
 from .lrt import LRTIE
 from .lynda import (
     LyndaIE,
@@ -215,13 +376,16 @@ from .lynda import (
 from .m6 import M6IE
 from .macgamestore import MacGameStoreIE
 from .mailru import MailRuIE
+from .makertv import MakerTVIE
 from .malemotion import MalemotionIE
+from .matchtv import MatchTVIE
 from .mdr import MDRIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
 from .mgoon import MgoonIE
 from .minhateca import MinhatecaIE
 from .ministrygrid import MinistryGridIE
+from .miomio import MioMioIE
 from .mit import TechTVMITIE, MITIE, OCWMITIE
 from .mitele import MiTeleIE
 from .mixcloud import MixcloudIE
@@ -237,111 +401,231 @@ from .motherless import MotherlessIE
 from .motorsport import MotorsportIE
 from .movieclips import MovieClipsIE
 from .moviezine import MoviezineIE
-from .movshare import MovShareIE
 from .mtv import (
     MTVIE,
     MTVServicesEmbeddedIE,
     MTVIggyIE,
+    MTVDEIE,
 )
 from .muenchentv import MuenchenTVIE
 from .musicplayon import MusicPlayOnIE
-from .musicvault import MusicVaultIE
 from .muzu import MuzuTVIE
+from .mwave import MwaveIE
 from .myspace import MySpaceIE, MySpaceAlbumIE
 from .myspass import MySpassIE
+from .myvi import MyviIE
 from .myvideo import MyVideoIE
 from .myvidster import MyVidsterIE
+from .nationalgeographic import NationalGeographicIE
 from .naver import NaverIE
 from .nba import NBAIE
 from .nbc import (
     NBCIE,
     NBCNewsIE,
+    NBCSportsIE,
+    NBCSportsVPlayerIE,
+    MSNBCIE,
+)
+from .ndr import (
+    NDRIE,
+    NJoyIE,
+    NDREmbedBaseIE,
+    NDREmbedIE,
+    NJoyEmbedIE,
 )
-from .ndr import NDRIE
 from .ndtv import NDTVIE
+from .netzkino import NetzkinoIE
+from .nerdcubed import NerdCubedFeedIE
+from .nerdist import NerdistIE
+from .neteasemusic import (
+    NetEaseMusicIE,
+    NetEaseMusicAlbumIE,
+    NetEaseMusicSingerIE,
+    NetEaseMusicListIE,
+    NetEaseMusicMvIE,
+    NetEaseMusicProgramIE,
+    NetEaseMusicDjRadioIE,
+)
 from .newgrounds import NewgroundsIE
 from .newstube import NewstubeIE
+from .nextmedia import (
+    NextMediaIE,
+    NextMediaActionNewsIE,
+    AppleDailyIE,
+)
+from .nextmovie import NextMovieIE
 from .nfb import NFBIE
 from .nfl import NFLIE
-from .nhl import NHLIE, NHLVideocenterIE
+from .nhl import (
+    NHLIE,
+    NHLNewsIE,
+    NHLVideocenterIE,
+)
+from .nick import NickIE
 from .niconico import NiconicoIE, NiconicoPlaylistIE
 from .ninegag import NineGagIE
 from .noco import NocoIE
 from .normalboots import NormalbootsIE
 from .nosvideo import NosVideoIE
-from .novamov import NovaMovIE
-from .nowness import NownessIE
-from .nowvideo import NowVideoIE
+from .nova import NovaIE
+from .novamov import (
+    NovaMovIE,
+    WholeCloudIE,
+    NowVideoIE,
+    VideoWeedIE,
+    CloudTimeIE,
+)
+from .nowness import (
+    NownessIE,
+    NownessPlaylistIE,
+    NownessSeriesIE,
+)
+from .nowtv import (
+    NowTVIE,
+    NowTVListIE,
+)
 from .npo import (
     NPOIE,
-    TegenlichtVproIE,
+    NPOLiveIE,
+    NPORadioIE,
+    NPORadioFragmentIE,
+    SchoolTVIE,
+    VPROIE,
+    WNLIE
 )
+from .npr import NprIE
 from .nrk import (
     NRKIE,
+    NRKPlaylistIE,
     NRKTVIE,
 )
-from .ntv import NTVIE
-from .nytimes import NYTimesIE
+from .ntvde import NTVDeIE
+from .ntvru import NTVRuIE
+from .nytimes import (
+    NYTimesIE,
+    NYTimesArticleIE,
+)
 from .nuvid import NuvidIE
+from .odnoklassniki import OdnoklassnikiIE
 from .oktoberfesttv import OktoberfestTVIE
-from .ooyala import OoyalaIE
+from .onionstudios import OnionStudiosIE
+from .ooyala import (
+    OoyalaIE,
+    OoyalaExternalIE,
+)
+from .ora import OraTVIE
 from .orf import (
     ORFTVthekIE,
     ORFOE1IE,
     ORFFM4IE,
+    ORFIPTVIE,
 )
+from .pandoratv import PandoraTVIE
 from .parliamentliveuk import ParliamentLiveUKIE
 from .patreon import PatreonIE
 from .pbs import PBSIE
+from .periscope import PeriscopeIE
+from .philharmoniedeparis import PhilharmonieDeParisIE
 from .phoenix import PhoenixIE
 from .photobucket import PhotobucketIE
+from .pinkbike import PinkbikeIE
 from .planetaplay import PlanetaPlayIE
+from .pladform import PladformIE
 from .played import PlayedIE
 from .playfm import PlayFMIE
+from .playtvak import PlaytvakIE
 from .playvid import PlayvidIE
+from .playwire import PlaywireIE
+from .pluralsight import (
+    PluralsightIE,
+    PluralsightCourseIE,
+)
 from .podomatic import PodomaticIE
+from .porn91 import Porn91IE
 from .pornhd import PornHdIE
-from .pornhub import PornHubIE
+from .pornhub import (
+    PornHubIE,
+    PornHubPlaylistIE,
+)
 from .pornotube import PornotubeIE
+from .pornovoisines import PornoVoisinesIE
 from .pornoxo import PornoXOIE
+from .primesharetv import PrimeShareTVIE
 from .promptfile import PromptFileIE
 from .prosiebensat1 import ProSiebenSat1IE
+from .puls4 import Puls4IE
 from .pyvideo import PyvideoIE
+from .qqmusic import (
+    QQMusicIE,
+    QQMusicSingerIE,
+    QQMusicAlbumIE,
+    QQMusicToplistIE,
+    QQMusicPlaylistIE,
+)
 from .quickvid import QuickVidIE
+from .r7 import R7IE
 from .radiode import RadioDeIE
+from .radiojavan import RadioJavanIE
+from .radiobremen import RadioBremenIE
 from .radiofrance import RadioFranceIE
-from .rai import RaiIE
+from .rai import (
+    RaiTVIE,
+    RaiIE,
+)
 from .rbmaradio import RBMARadioIE
+from .rds import RDSIE
 from .redtube import RedTubeIE
+from .regiotv import RegioTVIE
+from .restudy import RestudyIE
 from .reverbnation import ReverbNationIE
+from .revision3 import Revision3IE
 from .ringtv import RingTVIE
 from .ro220 import Ro220IE
 from .rottentomatoes import RottenTomatoesIE
 from .roxwel import RoxwelIE
 from .rtbf import RTBFIE
-from .rtlnl import RtlXlIE
-from .rtlnow import RTLnowIE
+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
+from .rtve import RTVEALaCartaIE, RTVELiveIE, RTVEInfantilIE
+from .rtvnh import RTVNHIE
 from .ruhd import RUHDIE
+from .ruleporn import RulePornIE
 from .rutube import (
     RutubeIE,
     RutubeChannelIE,
+    RutubeEmbedIE,
     RutubeMovieIE,
     RutubePersonIE,
 )
 from .rutv import RUTVIE
+from .ruutu import RuutuIE
+from .sandia import SandiaIE
+from .safari import (
+    SafariIE,
+    SafariCourseIE,
+)
 from .sapo import SapoIE
 from .savefrom import SaveFromIE
 from .sbs import SBSIE
 from .scivee import SciVeeIE
 from .screencast import ScreencastIE
+from .screencastomatic import ScreencastOMaticIE
+from .screenwavemedia import ScreenwaveMediaIE, TeamFourIE
+from .senateisvp import SenateISVPIE
 from .servingsys import ServingSysIE
 from .sexu import SexuIE
 from .sexykarma import SexyKarmaIE
+from .shahid import ShahidIE
 from .shared import SharedIE
 from .sharesix import ShareSixIE
 from .sina import SinaIE
+from .skynewsarabia import (
+    SkyNewsArabiaIE,
+    SkyNewsArabiaArticleIE,
+)
 from .slideshare import SlideshareIE
 from .slutload import SlutloadIE
 from .smotri import (
@@ -350,34 +634,59 @@ from .smotri import (
     SmotriUserIE,
     SmotriBroadcastIE,
 )
+from .snagfilms import (
+    SnagFilmsIE,
+    SnagFilmsEmbedIE,
+)
 from .snotr import SnotrIE
-from .sockshare import SockshareIE
 from .sohu import SohuIE
 from .soundcloud import (
     SoundcloudIE,
     SoundcloudSetIE,
     SoundcloudUserIE,
-    SoundcloudPlaylistIE
+    SoundcloudPlaylistIE,
+    SoundcloudSearchIE
+)
+from .soundgasm import (
+    SoundgasmIE,
+    SoundgasmProfileIE
 )
-from .soundgasm import SoundgasmIE
 from .southpark import (
     SouthParkIE,
-    SouthparkDeIE,
+    SouthParkDeIE,
+    SouthParkDkIE,
+    SouthParkEsIE,
+    SouthParkNlIE
 )
 from .space import SpaceIE
+from .spankbang import SpankBangIE
 from .spankwire import SpankwireIE
 from .spiegel import SpiegelIE, SpiegelArticleIE
 from .spiegeltv import SpiegeltvIE
 from .spike import SpikeIE
+from .stitcher import StitcherIE
 from .sport5 import Sport5IE
-from .sportbox import SportBoxIE
+from .sportbox import (
+    SportBoxIE,
+    SportBoxEmbedIE,
+)
 from .sportdeutschland import SportDeutschlandIE
+from .srgssr import (
+    SRGSSRIE,
+    SRGSSRPlayIE,
+)
 from .srmediathek import SRMediathekIE
+from .ssa import SSAIE
 from .stanfordoc import StanfordOpenClassroomIE
 from .steam import SteamIE
 from .streamcloud import StreamcloudIE
 from .streamcz import StreamCZIE
+from .streetvoice import StreetVoiceIE
 from .sunporno import SunPornoIE
+from .svt import (
+    SVTIE,
+    SVTPlayIE,
+)
 from .swrmediathek import SWRMediathekIE
 from .syfy import SyfyIE
 from .sztvhu import SztvHuIE
@@ -392,20 +701,36 @@ from .teachingchannel import TeachingChannelIE
 from .teamcoco import TeamcocoIE
 from .techtalks import TechTalksIE
 from .ted import TEDIE
+from .tele13 import Tele13IE
 from .telebruxelles import TeleBruxellesIE
 from .telecinco import TelecincoIE
+from .telegraaf import TelegraafIE
 from .telemb import TeleMBIE
+from .teletask import TeleTaskIE
 from .tenplay import TenPlayIE
 from .testurl import TestURLIE
 from .tf1 import TF1IE
+from .theintercept import TheInterceptIE
 from .theonion import TheOnionIE
-from .theplatform import ThePlatformIE
+from .theplatform import (
+    ThePlatformIE,
+    ThePlatformFeedIE,
+)
 from .thesixtyone import TheSixtyOneIE
+from .thisamericanlife import ThisAmericanLifeIE
 from .thisav import ThisAVIE
 from .tinypic import TinyPicIE
-from .tlc import TlcIE, TlcDeIE
-from .tmz import TMZIE
-from .tnaflix import TNAFlixIE
+from .tlc import TlcDeIE
+from .tmz import (
+    TMZIE,
+    TMZArticleIE,
+)
+from .tnaflix import (
+    TNAFlixIE,
+    EMPFlixIE,
+    MovieFapIE,
+)
+from .toggle import ToggleIE
 from .thvideo import (
     THVideoIE,
     THVideoPlaylistIE
@@ -414,45 +739,99 @@ from .toutv import TouTvIE
 from .toypics import ToypicsUserIE, ToypicsIE
 from .traileraddict import TrailerAddictIE
 from .trilulilu import TriluliluIE
+from .trollvids import TrollvidsIE
 from .trutube import TruTubeIE
 from .tube8 import Tube8IE
-from .tudou import TudouIE
+from .tubitv import TubiTvIE
+from .tudou import (
+    TudouIE,
+    TudouPlaylistIE,
+    TudouAlbumIE,
+)
 from .tumblr import TumblrIE
-from .tunein import TuneInIE
+from .tunein import (
+    TuneInClipIE,
+    TuneInStationIE,
+    TuneInProgramIE,
+    TuneInTopicIE,
+    TuneInShortenerIE,
+)
 from .turbo import TurboIE
 from .tutv import TutvIE
+from .tv2 import (
+    TV2IE,
+    TV2ArticleIE,
+)
+from .tv4 import TV4IE
+from .tvc import (
+    TVCIE,
+    TVCArticleIE,
+)
 from .tvigle import TvigleIE
-from .tvp import TvpIE
+from .tvland import TVLandIE
+from .tvp import TvpIE, TvpSeriesIE
 from .tvplay import TVPlayIE
+from .tweakers import TweakersIE
 from .twentyfourvideo import TwentyFourVideoIE
-from .twitch import TwitchIE
+from .twentymin import TwentyMinutenIE
+from .twentytwotracks import (
+    TwentyTwoTracksIE,
+    TwentyTwoTracksGenreIE
+)
+from .twitch import (
+    TwitchVideoIE,
+    TwitchChapterIE,
+    TwitchVodIE,
+    TwitchProfileIE,
+    TwitchPastBroadcastsIE,
+    TwitchBookmarksIE,
+    TwitchStreamIE,
+)
+from .twitter import TwitterCardIE, TwitterIE
 from .ubu import UbuIE
 from .udemy import (
     UdemyIE,
     UdemyCourseIE
 )
+from .udn import UDNEmbedIE
+from .digiteka import DigitekaIE
 from .unistra import UnistraIE
 from .urort import UrortIE
 from .ustream import UstreamIE, UstreamChannelIE
+from .varzesh3 import Varzesh3IE
 from .vbox7 import Vbox7IE
 from .veehd import VeeHDIE
 from .veoh import VeohIE
+from .vessel import VesselIE
 from .vesti import VestiIE
 from .vevo import VevoIE
-from .vgtv import VGTVIE
+from .vgtv import (
+    BTArticleIE,
+    BTVestlendingenIE,
+    VGTVIE,
+)
 from .vh1 import VH1IE
 from .vice import ViceIE
 from .viddler import ViddlerIE
-from .videobam import VideoBamIE
 from .videodetective import VideoDetectiveIE
-from .videolecturesnet import VideoLecturesNetIE
 from .videofyme import VideofyMeIE
 from .videomega import VideoMegaIE
+from .videomore import (
+    VideomoreIE,
+    VideomoreVideoIE,
+    VideomoreSeasonIE,
+)
 from .videopremium import VideoPremiumIE
 from .videott import VideoTtIE
-from .videoweed import VideoWeedIE
-from .vidme import VidmeIE
+from .vidme import (
+    VidmeIE,
+    VidmeUserIE,
+    VidmeUserLikesIE,
+)
 from .vidzi import VidziIE
+from .vier import VierIE, VierVideosIE
+from .viewster import ViewsterIE
+from .viidea import ViideaIE
 from .vimeo import (
     VimeoIE,
     VimeoAlbumIE,
@@ -468,12 +847,17 @@ from .vine import (
     VineIE,
     VineUserIE,
 )
-from .viki import VikiIE
+from .viki import (
+    VikiIE,
+    VikiChannelIE,
+)
 from .vk import (
     VKIE,
     VKUserVideosIE,
 )
+from .vlive import VLiveIE
 from .vodlocker import VodlockerIE
+from .voicerepublic import VoiceRepublicIE
 from .vporn import VpornIE
 from .vrt import VRTIE
 from .vube import VubeIE
@@ -488,22 +872,43 @@ from .wdr import (
     WDRMobileIE,
     WDRMausIE,
 )
+from .webofstories import (
+    WebOfStoriesIE,
+    WebOfStoriesPlaylistIE,
+)
 from .weibo import WeiboIE
+from .weiqitv import WeiqiTVIE
 from .wimp import WimpIE
 from .wistia import WistiaIE
 from .worldstarhiphop import WorldStarHipHopIE
 from .wrzuta import WrzutaIE
+from .wsj import WSJIE
 from .xbef import XBefIE
 from .xboxclips import XboxClipsIE
-from .xhamster import XHamsterIE
+from .xfileshare import XFileShareIE
+from .xhamster import (
+    XHamsterIE,
+    XHamsterEmbedIE,
+)
 from .xminus import XMinusIE
 from .xnxx import XNXXIE
-from .xvideos import XVideosIE
+from .xstream import XstreamIE
 from .xtube import XTubeUserIE, XTubeIE
+from .xuite import XuiteIE
+from .xvideos import XVideosIE
+from .xxxymovies import XXXYMoviesIE
 from .yahoo import (
     YahooIE,
     YahooSearchIE,
 )
+from .yam import YamIE
+from .yandexmusic import (
+    YandexMusicTrackIE,
+    YandexMusicAlbumIE,
+    YandexMusicPlaylistIE,
+)
+from .yesjapan import YesJapanIE
+from .yinyuetai import YinYueTaiIE
 from .ynet import YnetIE
 from .youjizz import YouJizzIE
 from .youku import YoukuIE
@@ -521,16 +926,19 @@ from .youtube import (
     YoutubeSearchURLIE,
     YoutubeShowIE,
     YoutubeSubscriptionsIE,
-    YoutubeTopListIE,
+    YoutubeTruncatedIDIE,
     YoutubeTruncatedURLIE,
     YoutubeUserIE,
+    YoutubePlaylistsIE,
     YoutubeWatchLaterIE,
 )
-from .zdf import ZDFIE
+from .zapiks import ZapiksIE
+from .zdf import ZDFIE, ZDFChannelIE
 from .zingmp3 import (
     ZingMp3SongIE,
     ZingMp3AlbumIE,
 )
+from .zippcast import ZippCastIE
 
 _ALL_CLASSES = [
     klass
@@ -547,6 +955,17 @@ def gen_extractors():
     return [klass() for klass in _ALL_CLASSES]
 
 
+def list_extractors(age_limit):
+    """
+    Return a list of extractors that are suitable for the given age,
+    sorted by extractor ID.
+    """
+
+    return sorted(
+        filter(lambda ie: ie.is_suitable(age_limit), gen_extractors()),
+        key=lambda ie: ie.IE_NAME.lower())
+
+
 def get_info_extractor(ie_name):
     """Returns the info extractor class with the given ie_name"""
     return globals()[ie_name + 'IE']

youtube_dl/extractor/abc.py

@@ -1,16 +1,20 @@
 from __future__ import unicode_literals
 
 import re
-import json
 
 from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    js_to_json,
+    int_or_none,
+)
 
 
 class ABCIE(InfoExtractor):
     IE_NAME = 'abc.net.au'
-    _VALID_URL = r'http://www\.abc\.net\.au/news/[^/]+/[^/]+/(?P<id>\d+)'
+    _VALID_URL = r'http://www\.abc\.net\.au/news/(?:[^/]+/){1,2}(?P<id>\d+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.abc.net.au/news/2014-11-05/australia-to-staff-ebola-treatment-centre-in-sierra-leone/5868334',
         'md5': 'cb3dd03b18455a661071ee1e28344d9f',
         'info_dict': {
@@ -19,23 +23,67 @@ class ABCIE(InfoExtractor):
             'title': 'Australia to help staff Ebola treatment centre in Sierra Leone',
             'description': 'md5:809ad29c67a05f54eb41f2a105693a67',
         },
-    }
+        'skip': 'this video has expired',
+    }, {
+        'url': 'http://www.abc.net.au/news/2015-08-17/warren-entsch-introduces-same-sex-marriage-bill/6702326',
+        'md5': 'db2a5369238b51f9811ad815b69dc086',
+        'info_dict': {
+            'id': 'NvqvPeNZsHU',
+            'ext': 'mp4',
+            'upload_date': '20150816',
+            'uploader': 'ABC News (Australia)',
+            'description': 'Government backbencher Warren Entsch introduces a cross-party sponsored bill to legalise same-sex marriage, saying the bill is designed to promote "an inclusive Australia, not a divided one.". Read more here: http://ab.co/1Mwc6ef',
+            'uploader_id': 'NewsOnABC',
+            'title': 'Marriage Equality: Warren Entsch introduces same sex marriage bill',
+        },
+        'add_ie': ['Youtube'],
+        'skip': 'Not accessible from Travis CI server',
+    }, {
+        'url': 'http://www.abc.net.au/news/2015-10-23/nab-lifts-interest-rates-following-westpac-and-cba/6880080',
+        'md5': 'b96eee7c9edf4fc5a358a0252881cc1f',
+        'info_dict': {
+            'id': '6880080',
+            'ext': 'mp3',
+            'title': 'NAB lifts interest rates, following Westpac and CBA',
+            'description': 'md5:f13d8edc81e462fce4a0437c7dc04728',
+        },
+    }, {
+        'url': 'http://www.abc.net.au/news/2015-10-19/6866214',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        urls_info_json = self._search_regex(
-            r'inlineVideoData\.push\((.*?)\);', webpage, 'video urls',
-            flags=re.DOTALL)
-        urls_info = json.loads(urls_info_json.replace('\'', '"'))
+        mobj = re.search(
+            r'inline(?P<type>Video|Audio|YouTube)Data\.push\((?P<json_data>[^)]+)\);',
+            webpage)
+        if mobj is None:
+            expired = self._html_search_regex(r'(?s)class="expired-(?:video|audio)".+?<span>(.+?)</span>', webpage, 'expired', None)
+            if expired:
+                raise ExtractorError('%s said: %s' % (self.IE_NAME, expired), expected=True)
+            raise ExtractorError('Unable to extract video urls')
+
+        urls_info = self._parse_json(
+            mobj.group('json_data'), video_id, transform_source=js_to_json)
+
+        if not isinstance(urls_info, list):
+            urls_info = [urls_info]
+
+        if mobj.group('type') == 'YouTube':
+            return self.playlist_result([
+                self.url_result(url_info['url']) for url_info in urls_info])
+
         formats = [{
             'url': url_info['url'],
-            'width': int(url_info['width']),
-            'height': int(url_info['height']),
-            'tbr': int(url_info['bitrate']),
-            'filesize': int(url_info['filesize']),
+            'vcodec': url_info.get('codec') if mobj.group('type') == 'Video' else 'none',
+            'width': int_or_none(url_info.get('width')),
+            'height': int_or_none(url_info.get('height')),
+            'tbr': int_or_none(url_info.get('bitrate')),
+            'filesize': int_or_none(url_info.get('filesize')),
         } for url_info in urls_info]
+
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/abc7news.py

@@ -0,0 +1,67 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import parse_iso8601
+
+
+class Abc7NewsIE(InfoExtractor):
+    _VALID_URL = r'https?://abc7news\.com(?:/[^/]+/(?P<display_id>[^/]+))?/(?P<id>\d+)'
+    _TESTS = [
+        {
+            'url': 'http://abc7news.com/entertainment/east-bay-museum-celebrates-vintage-synthesizers/472581/',
+            'info_dict': {
+                'id': '472581',
+                'display_id': 'east-bay-museum-celebrates-vintage-synthesizers',
+                'ext': 'mp4',
+                'title': 'East Bay museum celebrates history of synthesized music',
+                'description': 'md5:a4f10fb2f2a02565c1749d4adbab4b10',
+                'thumbnail': 're:^https?://.*\.jpg$',
+                'timestamp': 1421123075,
+                'upload_date': '20150113',
+                'uploader': 'Jonathan Bloom',
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://abc7news.com/472581',
+            '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
+
+        webpage = self._download_webpage(url, display_id)
+
+        m3u8 = self._html_search_meta(
+            'contentURL', webpage, 'm3u8 url', fatal=True)
+
+        formats = self._extract_m3u8_formats(m3u8, display_id, 'mp4')
+
+        title = self._og_search_title(webpage).strip()
+        description = self._og_search_description(webpage).strip()
+        thumbnail = self._og_search_thumbnail(webpage)
+        timestamp = parse_iso8601(self._search_regex(
+            r'<div class="meta">\s*<time class="timeago" datetime="([^"]+)">',
+            webpage, 'upload date', fatal=False))
+        uploader = self._search_regex(
+            r'rel="author">([^<]+)</a>',
+            webpage, 'uploader', default=None)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'timestamp': timestamp,
+            'uploader': uploader,
+            'formats': formats,
+        }

youtube_dl/extractor/academicearth.py

@@ -15,7 +15,7 @@ class AcademicEarthCourseIE(InfoExtractor):
             'title': 'Laws of Nature',
             'description': 'Introduce yourself to the laws of nature with these free online college lectures from Yale, Harvard, and MIT.',
         },
-        'playlist_count': 4,
+        'playlist_count': 3,
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/acast.py

@@ -0,0 +1,72 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import int_or_none
+
+
+class ACastIE(InfoExtractor):
+    IE_NAME = 'acast'
+    _VALID_URL = r'https?://(?:www\.)?acast\.com/(?P<channel>[^/]+)/(?P<id>[^/#?]+)'
+    _TEST = {
+        'url': 'https://www.acast.com/condenasttraveler/-where-are-you-taipei-101-taiwan',
+        'md5': 'ada3de5a1e3a2a381327d749854788bb',
+        'info_dict': {
+            'id': '57de3baa-4bb0-487e-9418-2692c1277a34',
+            'ext': 'mp3',
+            'title': '"Where Are You?": Taipei 101, Taiwan',
+            'timestamp': 1196172000000,
+            'description': 'md5:a0b4ef3634e63866b542e5b1199a1a0e',
+            'duration': 211,
+        }
+    }
+
+    def _real_extract(self, url):
+        channel, display_id = re.match(self._VALID_URL, url).groups()
+
+        embed_page = self._download_webpage(
+            re.sub('(?:www\.)?acast\.com', 'embedcdn.acast.com', url), display_id)
+        cast_data = self._parse_json(self._search_regex(
+            r'window\[\'acast/queries\'\]\s*=\s*([^;]+);', embed_page, 'acast data'),
+            display_id)['GetAcast/%s/%s' % (channel, display_id)]
+
+        return {
+            'id': compat_str(cast_data['id']),
+            'display_id': display_id,
+            'url': cast_data['blings'][0]['audio'],
+            'title': cast_data['name'],
+            'description': cast_data.get('description'),
+            'thumbnail': cast_data.get('image'),
+            'timestamp': int_or_none(cast_data.get('publishingDate')),
+            'duration': int_or_none(cast_data.get('duration')),
+        }
+
+
+class ACastChannelIE(InfoExtractor):
+    IE_NAME = 'acast:channel'
+    _VALID_URL = r'https?://(?:www\.)?acast\.com/(?P<id>[^/#?]+)'
+    _TEST = {
+        'url': 'https://www.acast.com/condenasttraveler',
+        'info_dict': {
+            'id': '50544219-29bb-499e-a083-6087f4cb7797',
+            'title': 'Condé Nast Traveler Podcast',
+            'description': 'md5:98646dee22a5b386626ae31866638fbd',
+        },
+        'playlist_mincount': 20,
+    }
+    _API_BASE_URL = 'https://www.acast.com/api/'
+
+    @classmethod
+    def suitable(cls, url):
+        return False if ACastIE.suitable(url) else super(ACastChannelIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        channel_data = self._download_json(self._API_BASE_URL + 'channels/%s' % display_id, display_id)
+        casts = self._download_json(self._API_BASE_URL + 'channels/%s/acasts' % display_id, display_id)
+        entries = [self.url_result('https://www.acast.com/%s/%s' % (display_id, cast['url']), 'ACast') for cast in casts]
+
+        return self.playlist_result(entries, compat_str(channel_data['id']), channel_data['name'], channel_data.get('description'))

youtube_dl/extractor/addanime.py

@@ -11,12 +11,13 @@ from ..compat import (
 )
 from ..utils import (
     ExtractorError,
+    qualities,
 )
 
 
 class AddAnimeIE(InfoExtractor):
-    _VALID_URL = r'^http://(?:\w+\.)?add-anime\.net/watch_video\.php\?(?:.*?)v=(?P<id>[\w_]+)(?:.*)'
-    _TEST = {
+    _VALID_URL = r'http://(?:\w+\.)?add-anime\.net/(?:watch_video\.php\?(?:.*?)v=|video/)(?P<id>[\w_]+)'
+    _TESTS = [{
         'url': 'http://www.add-anime.net/watch_video.php?v=24MR3YO5SAS9',
         'md5': '72954ea10bc979ab5e2eb288b21425a0',
         'info_dict': {
@@ -25,7 +26,10 @@ class AddAnimeIE(InfoExtractor):
             'description': 'One Piece 606',
             'title': 'One Piece 606',
         }
-    }
+    }, {
+        'url': 'http://add-anime.net/video/MDUGWYKNGBD8/One-Piece-687',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -63,8 +67,10 @@ class AddAnimeIE(InfoExtractor):
                 note='Confirming after redirect')
             webpage = self._download_webpage(url, video_id)
 
+        FORMATS = ('normal', 'hq')
+        quality = qualities(FORMATS)
         formats = []
-        for format_id in ('normal', 'hq'):
+        for format_id in FORMATS:
             rex = r"var %s_video_file = '(.*?)';" % re.escape(format_id)
             video_url = self._search_regex(rex, webpage, 'video file URLx',
                                            fatal=False)
@@ -73,6 +79,7 @@ class AddAnimeIE(InfoExtractor):
             formats.append({
                 'format_id': format_id,
                 'url': video_url,
+                'quality': quality(format_id),
             })
         self._sort_formats(formats)
         video_title = self._og_search_title(webpage)

youtube_dl/extractor/adobetv.py

@@ -0,0 +1,194 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    parse_duration,
+    unified_strdate,
+    str_to_int,
+    int_or_none,
+    float_or_none,
+    ISO639Utils,
+    determine_ext,
+)
+
+
+class AdobeTVBaseIE(InfoExtractor):
+    _API_BASE_URL = 'http://tv.adobe.com/api/v4/'
+
+
+class AdobeTVIE(AdobeTVBaseIE):
+    _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?watch/(?P<show_urlname>[^/]+)/(?P<id>[^/]+)'
+
+    _TEST = {
+        'url': 'http://tv.adobe.com/watch/the-complete-picture-with-julieanne-kost/quick-tip-how-to-draw-a-circle-around-an-object-in-photoshop/',
+        'md5': '9bc5727bcdd55251f35ad311ca74fa1e',
+        'info_dict': {
+            'id': '10981',
+            'ext': 'mp4',
+            'title': 'Quick Tip - How to Draw a Circle Around an Object in Photoshop',
+            'description': 'md5:99ec318dc909d7ba2a1f2b038f7d2311',
+            'thumbnail': 're:https?://.*\.jpg$',
+            'upload_date': '20110914',
+            'duration': 60,
+            'view_count': int,
+        },
+    }
+
+    def _real_extract(self, url):
+        language, show_urlname, urlname = re.match(self._VALID_URL, url).groups()
+        if not language:
+            language = 'en'
+
+        video_data = self._download_json(
+            self._API_BASE_URL + 'episode/get/?language=%s&show_urlname=%s&urlname=%s&disclosure=standard' % (language, show_urlname, urlname),
+            urlname)['data'][0]
+
+        formats = [{
+            'url': source['url'],
+            'format_id': source.get('quality_level') or source['url'].split('-')[-1].split('.')[0] or None,
+            'width': int_or_none(source.get('width')),
+            'height': int_or_none(source.get('height')),
+            'tbr': int_or_none(source.get('video_data_rate')),
+        } for source in video_data['videos']]
+        self._sort_formats(formats)
+
+        return {
+            'id': compat_str(video_data['id']),
+            'title': video_data['title'],
+            'description': video_data.get('description'),
+            'thumbnail': video_data.get('thumbnail'),
+            'upload_date': unified_strdate(video_data.get('start_date')),
+            'duration': parse_duration(video_data.get('duration')),
+            'view_count': str_to_int(video_data.get('playcount')),
+            'formats': formats,
+        }
+
+
+class AdobeTVPlaylistBaseIE(AdobeTVBaseIE):
+    def _parse_page_data(self, page_data):
+        return [self.url_result(self._get_element_url(element_data)) for element_data in page_data]
+
+    def _extract_playlist_entries(self, url, display_id):
+        page = self._download_json(url, display_id)
+        entries = self._parse_page_data(page['data'])
+        for page_num in range(2, page['paging']['pages'] + 1):
+            entries.extend(self._parse_page_data(
+                self._download_json(url + '&page=%d' % page_num, display_id)['data']))
+        return entries
+
+
+class AdobeTVShowIE(AdobeTVPlaylistBaseIE):
+    _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?show/(?P<id>[^/]+)'
+
+    _TEST = {
+        'url': 'http://tv.adobe.com/show/the-complete-picture-with-julieanne-kost',
+        'info_dict': {
+            'id': '36',
+            'title': 'The Complete Picture with Julieanne Kost',
+            'description': 'md5:fa50867102dcd1aa0ddf2ab039311b27',
+        },
+        'playlist_mincount': 136,
+    }
+
+    def _get_element_url(self, element_data):
+        return element_data['urls'][0]
+
+    def _real_extract(self, url):
+        language, show_urlname = re.match(self._VALID_URL, url).groups()
+        if not language:
+            language = 'en'
+        query = 'language=%s&show_urlname=%s' % (language, show_urlname)
+
+        show_data = self._download_json(self._API_BASE_URL + 'show/get/?%s' % query, show_urlname)['data'][0]
+
+        return self.playlist_result(
+            self._extract_playlist_entries(self._API_BASE_URL + 'episode/?%s' % query, show_urlname),
+            compat_str(show_data['id']),
+            show_data['show_name'],
+            show_data['show_description'])
+
+
+class AdobeTVChannelIE(AdobeTVPlaylistBaseIE):
+    _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?channel/(?P<id>[^/]+)(?:/(?P<category_urlname>[^/]+))?'
+
+    _TEST = {
+        'url': 'http://tv.adobe.com/channel/development',
+        'info_dict': {
+            'id': 'development',
+        },
+        'playlist_mincount': 96,
+    }
+
+    def _get_element_url(self, element_data):
+        return element_data['url']
+
+    def _real_extract(self, url):
+        language, channel_urlname, category_urlname = re.match(self._VALID_URL, url).groups()
+        if not language:
+            language = 'en'
+        query = 'language=%s&channel_urlname=%s' % (language, channel_urlname)
+        if category_urlname:
+            query += '&category_urlname=%s' % category_urlname
+
+        return self.playlist_result(
+            self._extract_playlist_entries(self._API_BASE_URL + 'show/?%s' % query, channel_urlname),
+            channel_urlname)
+
+
+class AdobeTVVideoIE(InfoExtractor):
+    _VALID_URL = r'https?://video\.tv\.adobe\.com/v/(?P<id>\d+)'
+
+    _TEST = {
+        # From https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners
+        'url': 'https://video.tv.adobe.com/v/2456/',
+        'md5': '43662b577c018ad707a63766462b1e87',
+        'info_dict': {
+            'id': '2456',
+            'ext': 'mp4',
+            'title': 'New experience with Acrobat DC',
+            'description': 'New experience with Acrobat DC',
+            'duration': 248.667,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        video_data = self._download_json(url + '?format=json', video_id)
+
+        formats = [{
+            'format_id': '%s-%s' % (determine_ext(source['src']), source.get('height')),
+            'url': source['src'],
+            'width': int_or_none(source.get('width')),
+            'height': int_or_none(source.get('height')),
+            'tbr': int_or_none(source.get('bitrate')),
+        } for source in video_data['sources']]
+        self._sort_formats(formats)
+
+        # For both metadata and downloaded files the duration varies among
+        # formats. I just pick the max one
+        duration = max(filter(None, [
+            float_or_none(source.get('duration'), scale=1000)
+            for source in video_data['sources']]))
+
+        subtitles = {}
+        for translation in video_data.get('translations', []):
+            lang_id = translation.get('language_w3c') or ISO639Utils.long2short(translation['language_medium'])
+            if lang_id not in subtitles:
+                subtitles[lang_id] = []
+            subtitles[lang_id].append({
+                'url': translation['vttPath'],
+                'ext': 'vtt',
+            })
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': video_data['title'],
+            'description': video_data.get('description'),
+            'thumbnail': video_data['video'].get('poster'),
+            'duration': duration,
+            'subtitles': subtitles,
+        }

youtube_dl/extractor/adultswim.py

@@ -2,11 +2,13 @@
 from __future__ import unicode_literals
 
 import re
-import json
 
 from .common import InfoExtractor
 from ..utils import (
+    determine_ext,
     ExtractorError,
+    float_or_none,
+    xpath_text,
 )
 
 
@@ -36,9 +38,11 @@ class AdultSwimIE(InfoExtractor):
             },
         ],
         'info_dict': {
+            'id': 'rQxZvXQ4ROaSOqq-or2Mow',
             'title': 'Rick and Morty - Pilot',
             'description': "Rick moves in with his daughter's family and establishes himself as a bad influence on his grandson, Morty. "
-        }
+        },
+        'skip': 'This video is only available for registered users',
     }, {
         'url': 'http://www.adultswim.com/videos/playlists/american-parenting/putting-francine-out-of-business/',
         'playlist': [
@@ -53,9 +57,32 @@ class AdultSwimIE(InfoExtractor):
             }
         ],
         'info_dict': {
+            'id': '-t8CamQlQ2aYZ49ItZCFog',
             'title': 'American Dad - Putting Francine Out of Business',
             'description': 'Stan hatches a plan to get Francine out of the real estate business.Watch more American Dad on [adult swim].'
         },
+    }, {
+        'url': 'http://www.adultswim.com/videos/tim-and-eric-awesome-show-great-job/dr-steve-brule-for-your-wine/',
+        'playlist': [
+            {
+                'md5': '3e346a2ab0087d687a05e1e7f3b3e529',
+                'info_dict': {
+                    'id': 'sY3cMUR_TbuE4YmdjzbIcQ-0',
+                    'ext': 'mp4',
+                    'title': 'Tim and Eric Awesome Show Great Job! - Dr. Steve Brule, For Your Wine',
+                    'description': 'Dr. Brule reports live from Wine Country with a special report on wines.  \r\nWatch Tim and Eric Awesome Show Great Job! episode #20, "Embarrassed" on Adult Swim.\r\n\r\n',
+                },
+            }
+        ],
+        'info_dict': {
+            'id': 'sY3cMUR_TbuE4YmdjzbIcQ',
+            'title': 'Tim and Eric Awesome Show Great Job! - Dr. Steve Brule, For Your Wine',
+            'description': 'Dr. Brule reports live from Wine Country with a special report on wines.  \r\nWatch Tim and Eric Awesome Show Great Job! episode #20, "Embarrassed" on Adult Swim.\r\n\r\n',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        }
     }]
 
     @staticmethod
@@ -76,6 +103,7 @@ class AdultSwimIE(InfoExtractor):
             for video in collection.get('videos'):
                 if video.get('slug') == slug:
                     return collection, video
+        return None, None
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -86,30 +114,39 @@ class AdultSwimIE(InfoExtractor):
         webpage = self._download_webpage(url, episode_path)
 
         # Extract the value of `bootstrappedData` from the Javascript in the page.
-        bootstrappedDataJS = self._search_regex(r'var bootstrappedData = ({.*});', webpage, episode_path)
-
-        try:
-            bootstrappedData = json.loads(bootstrappedDataJS)
-        except ValueError as ve:
-            errmsg = '%s: Failed to parse JSON ' % episode_path
-            raise ExtractorError(errmsg, cause=ve)
+        bootstrapped_data = self._parse_json(self._search_regex(
+            r'var bootstrappedData = ({.*});', webpage, 'bootstraped data'), episode_path)
 
         # Downloading videos from a /videos/playlist/ URL needs to be handled differently.
         # NOTE: We are only downloading one video (the current one) not the playlist
         if is_playlist:
-            collections = bootstrappedData['playlists']['collections']
+            collections = bootstrapped_data['playlists']['collections']
             collection = self.find_collection_by_linkURL(collections, show_path)
             video_info = self.find_video_info(collection, episode_path)
 
             show_title = video_info['showTitle']
             segment_ids = [video_info['videoPlaybackID']]
         else:
-            collections = bootstrappedData['show']['collections']
+            collections = bootstrapped_data['show']['collections']
             collection, video_info = self.find_collection_containing_video(collections, episode_path)
+            # Video wasn't found in the collections, let's try `slugged_video`.
+            if video_info is None:
+                if bootstrapped_data.get('slugged_video', {}).get('slug') == episode_path:
+                    video_info = bootstrapped_data['slugged_video']
+                else:
+                    raise ExtractorError('Unable to find video info')
 
-            show = bootstrappedData['show']
+            show = bootstrapped_data['show']
             show_title = show['title']
-            segment_ids = [clip['videoPlaybackID'] for clip in video_info['clips']]
+            stream = video_info.get('stream')
+            clips = [stream] if stream else video_info.get('clips')
+            if not clips:
+                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']
@@ -118,7 +155,7 @@ class AdultSwimIE(InfoExtractor):
 
         entries = []
         for part_num, segment_id in enumerate(segment_ids):
-            segment_url = 'http://www.adultswim.com/videos/api/v0/assets?id=%s&platform=mobile' % segment_id
+            segment_url = 'http://www.adultswim.com/videos/api/v0/assets?id=%s&platform=desktop' % segment_id
 
             segment_title = '%s - %s' % (show_title, episode_title)
             if len(segment_ids) > 1:
@@ -128,22 +165,37 @@ class AdultSwimIE(InfoExtractor):
                 segment_url, segment_title,
                 'Downloading segment information', 'Unable to download segment information')
 
-            segment_duration = idoc.find('.//trt').text.strip()
+            segment_duration = float_or_none(
+                xpath_text(idoc, './/trt', 'segment duration').strip())
 
             formats = []
-            file_els = idoc.findall('.//files/file')
+            file_els = idoc.findall('.//files/file') or idoc.findall('./files/file')
 
+            unique_urls = []
+            unique_file_els = []
             for file_el in file_els:
+                media_url = file_el.text
+                if not media_url or determine_ext(media_url) == 'f4m':
+                    continue
+                if file_el.text not in unique_urls:
+                    unique_urls.append(file_el.text)
+                    unique_file_els.append(file_el)
+
+            for file_el in unique_file_els:
                 bitrate = file_el.attrib.get('bitrate')
                 ftype = file_el.attrib.get('type')
-
-                formats.append({
-                    'format_id': '%s_%s' % (bitrate, ftype),
-                    'url': file_el.text.strip(),
-                    # The bitrate may not be a number (for example: 'iphone')
-                    'tbr': int(bitrate) if bitrate.isdigit() else None,
-                    'quality': 1 if ftype == 'hd' else -1
-                })
+                media_url = file_el.text
+                if determine_ext(media_url) == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        media_url, segment_title, 'mp4', preference=0,
+                        m3u8_id='hls', fatal=False))
+                else:
+                    formats.append({
+                        'format_id': '%s_%s' % (bitrate, ftype),
+                        'url': file_el.text.strip(),
+                        # The bitrate may not be a number (for example: 'iphone')
+                        'tbr': int(bitrate) if bitrate.isdigit() else None,
+                    })
 
             self._sort_formats(formats)
 

youtube_dl/extractor/aenetworks.py

@@ -0,0 +1,66 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import smuggle_url
+
+
+class AENetworksIE(InfoExtractor):
+    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<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',
+            'ext': 'mp4',
+            'title': "Bet You Didn't Know: Valentine's Day",
+            'description': 'md5:7b57ea4829b391995b405fa60bd7b5f7',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        'add_ie': ['ThePlatform'],
+        'expected_warnings': ['JSON-LD'],
+    }, {
+        'url': 'http://www.history.com/shows/mountain-men/season-1/episode-1',
+        'info_dict': {
+            'id': 'eg47EERs_JsZ',
+            'ext': 'mp4',
+            'title': "Winter Is Coming",
+            'description': 'md5:641f424b7a19d8e24f26dea22cf59d74',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        'add_ie': ['ThePlatform'],
+    }, {
+        'url': 'http://www.aetv.com/shows/duck-dynasty/video/inlawful-entry',
+        '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.mylifetime.com/shows/project-runway-junior/video/season-1/episode-6/superstar-clients',
+        'only_matching': True
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        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 = self._search_regex(video_url_re, webpage, 'video url')
+
+        info = self._search_json_ld(webpage, video_id, fatal=False)
+        info.update({
+            '_type': 'url_transparent',
+            'url': smuggle_url(video_url, {'sig': {'key': 'crazyjava', 'secret': 's3cr3t'}}),
+        })
+        return info

youtube_dl/extractor/aftonbladet.py

@@ -1,17 +1,16 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..utils import int_or_none
 
 
 class AftonbladetIE(InfoExtractor):
-    _VALID_URL = r'^http://tv\.aftonbladet\.se/webbtv.+?(?P<video_id>article[0-9]+)\.ab(?:$|[?#])'
+    _VALID_URL = r'http://tv\.aftonbladet\.se/abtv/articles/(?P<id>[0-9]+)'
     _TEST = {
-        'url': 'http://tv.aftonbladet.se/webbtv/nyheter/vetenskap/rymden/article36015.ab',
+        'url': 'http://tv.aftonbladet.se/abtv/articles/36015',
         'info_dict': {
-            'id': 'article36015',
+            'id': '36015',
             'ext': 'mp4',
             'title': 'Vulkanutbrott i rymden - nu släpper NASA bilderna',
             'description': 'Jupiters måne mest aktiv av alla himlakroppar',
@@ -21,15 +20,14 @@ class AftonbladetIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.search(self._VALID_URL, url)
-
-        video_id = mobj.group('video_id')
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
         # find internal video meta data
         meta_url = 'http://aftonbladet-play.drlib.aptoma.no/video/%s.json'
-        internal_meta_id = self._html_search_regex(
-            r'data-aptomaId="([\w\d]+)"', webpage, 'internal_meta_id')
+        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_url = meta_url % internal_meta_id
         internal_meta_json = self._download_json(
             internal_meta_url, video_id, 'Downloading video meta data')
@@ -47,9 +45,9 @@ class AftonbladetIE(InfoExtractor):
             formats.append({
                 'url': 'http://%s:%d/%s/%s' % (p['address'], p['port'], p['path'], p['filename']),
                 'ext': 'mp4',
-                'width': fmt['width'],
-                'height': fmt['height'],
-                'tbr': fmt['bitrate'],
+                'width': int_or_none(fmt.get('width')),
+                'height': int_or_none(fmt.get('height')),
+                'tbr': int_or_none(fmt.get('bitrate')),
                 'protocol': 'http',
             })
         self._sort_formats(formats)
@@ -58,9 +56,9 @@ class AftonbladetIE(InfoExtractor):
             'id': video_id,
             'title': internal_meta_json['title'],
             'formats': formats,
-            'thumbnail': internal_meta_json['imageUrl'],
-            'description': internal_meta_json['shortPreamble'],
-            'timestamp': internal_meta_json['timePublished'],
-            'duration': internal_meta_json['duration'],
-            'view_count': internal_meta_json['views'],
+            'thumbnail': internal_meta_json.get('imageUrl'),
+            'description': internal_meta_json.get('shortPreamble'),
+            'timestamp': int_or_none(internal_meta_json.get('timePublished')),
+            'duration': int_or_none(internal_meta_json.get('duration')),
+            'view_count': int_or_none(internal_meta_json.get('views')),
         }

youtube_dl/extractor/airmozilla.py

@@ -0,0 +1,74 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    parse_duration,
+    parse_iso8601,
+)
+
+
+class AirMozillaIE(InfoExtractor):
+    _VALID_URL = r'https?://air\.mozilla\.org/(?P<id>[0-9a-z-]+)/?'
+    _TEST = {
+        'url': 'https://air.mozilla.org/privacy-lab-a-meetup-for-privacy-minded-people-in-san-francisco/',
+        'md5': '2e3e7486ba5d180e829d453875b9b8bf',
+        'info_dict': {
+            'id': '6x4q2w',
+            'ext': 'mp4',
+            'title': 'Privacy Lab - a meetup for privacy minded people in San Francisco',
+            'thumbnail': 're:https?://vid\.ly/(?P<id>[0-9a-z-]+)/poster',
+            'description': 'Brings together privacy professionals and others interested in privacy at for-profits, non-profits, and NGOs in an effort to contribute to the state of the ecosystem...',
+            'timestamp': 1422487800,
+            'upload_date': '20150128',
+            'location': 'SFO Commons',
+            'duration': 3780,
+            'view_count': int,
+            'categories': ['Main', 'Privacy'],
+        }
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        video_id = self._html_search_regex(r'//vid.ly/(.*?)/embed', webpage, 'id')
+
+        embed_script = self._download_webpage('https://vid.ly/{0}/embed'.format(video_id), video_id)
+        jwconfig = self._search_regex(r'\svar jwconfig = (\{.*?\});\s', embed_script, 'metadata')
+        metadata = self._parse_json(jwconfig, video_id)
+
+        formats = [{
+            'url': source['file'],
+            'ext': source['type'],
+            'format_id': self._search_regex(r'&format=(.*)$', source['file'], 'video format'),
+            'format': source['label'],
+            'height': int(source['label'].rstrip('p')),
+        } for source in metadata['playlist'][0]['sources']]
+        self._sort_formats(formats)
+
+        view_count = int_or_none(self._html_search_regex(
+            r'Views since archived: ([0-9]+)',
+            webpage, 'view count', fatal=False))
+        timestamp = parse_iso8601(self._html_search_regex(
+            r'<time datetime="(.*?)"', webpage, 'timestamp', fatal=False))
+        duration = parse_duration(self._search_regex(
+            r'Duration:\s*(\d+\s*hours?\s*\d+\s*minutes?)',
+            webpage, 'duration', fatal=False))
+
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'formats': formats,
+            'url': self._og_search_url(webpage),
+            'display_id': display_id,
+            'thumbnail': metadata['playlist'][0].get('image'),
+            'description': self._og_search_description(webpage),
+            'timestamp': timestamp,
+            'location': self._html_search_regex(r'Location: (.*)', webpage, 'location', default=None),
+            'duration': duration,
+            'view_count': view_count,
+            'categories': re.findall(r'<a href=".*?" class="channel">(.*?)</a>', webpage),
+        }

youtube_dl/extractor/aljazeera.py

@@ -0,0 +1,36 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+
+class AlJazeeraIE(InfoExtractor):
+    _VALID_URL = r'http://www\.aljazeera\.com/programmes/.*?/(?P<id>[^/]+)\.html'
+
+    _TEST = {
+        'url': 'http://www.aljazeera.com/programmes/the-slum/2014/08/deliverance-201482883754237240.html',
+        'info_dict': {
+            'id': '3792260579001',
+            'ext': 'mp4',
+            'title': 'The Slum - Episode 1: Deliverance',
+            'description': 'As a birth attendant advocating for family planning, Remy is on the frontline of Tondo\'s battle with overcrowding.',
+            'uploader': 'Al Jazeera English',
+        },
+        'add_ie': ['BrightcoveLegacy'],
+        'skip': 'Not accessible from Travis CI server',
+    }
+
+    def _real_extract(self, url):
+        program_name = self._match_id(url)
+        webpage = self._download_webpage(url, program_name)
+        brightcove_id = self._search_regex(
+            r'RenderPagesVideo\(\'(.+?)\'', webpage, 'brightcove id')
+
+        return {
+            '_type': 'url',
+            'url': (
+                'brightcove:'
+                'playerKey=AQ~~%2CAAAAmtVJIFk~%2CTVGOQ5ZTwJbeMWnq5d_H4MOM57xfzApc'
+                '&%40videoPlayer={0}'.format(brightcove_id)
+            ),
+            'ie_key': 'BrightcoveLegacy',
+        }

youtube_dl/extractor/allocine.py

@@ -5,15 +5,16 @@ import re
 import json
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
-    compat_str,
     qualities,
-    determine_ext,
+    unescapeHTML,
+    xpath_element,
 )
 
 
 class AllocineIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?allocine\.fr/(?P<typ>article|video|film)/(fichearticle_gen_carticle=|player_gen_cmedia=|fichefilm_gen_cfilm=)(?P<id>[0-9]+)(?:\.html)?'
+    _VALID_URL = r'https?://(?:www\.)?allocine\.fr/(?P<typ>article|video|film)/(fichearticle_gen_carticle=|player_gen_cmedia=|fichefilm_gen_cfilm=|video-)(?P<id>[0-9]+)(?:\.html)?'
 
     _TESTS = [{
         'url': 'http://www.allocine.fr/article/fichearticle_gen_carticle=18635087.html',
@@ -32,7 +33,7 @@ class AllocineIE(InfoExtractor):
             'id': '19540403',
             'ext': 'mp4',
             'title': 'Planes 2 Bande-annonce VF',
-            'description': 'md5:eeaffe7c2d634525e21159b93acf3b1e',
+            'description': 'Regardez la bande annonce du film Planes 2 (Planes 2 Bande-annonce VF). Planes 2, un film de Roberts Gannaway',
             'thumbnail': 're:http://.*\.jpg',
         },
     }, {
@@ -42,9 +43,12 @@ class AllocineIE(InfoExtractor):
             'id': '19544709',
             'ext': 'mp4',
             'title': 'Dragons 2 - Bande annonce finale VF',
-            'description': 'md5:71742e3a74b0d692c7fce0dd2017a4ac',
+            'description': 'md5:601d15393ac40f249648ef000720e7e3',
             'thumbnail': 're:http://.*\.jpg',
         },
+    }, {
+        'url': 'http://www.allocine.fr/video/video-19550147/',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -57,14 +61,18 @@ class AllocineIE(InfoExtractor):
         if typ == 'film':
             video_id = self._search_regex(r'href="/video/player_gen_cmedia=([0-9]+).+"', webpage, 'video id')
         else:
-            player = self._search_regex(r'data-player=\'([^\']+)\'>', webpage, 'data player')
-
-            player_data = json.loads(player)
-            video_id = compat_str(player_data['refMedia'])
+            player = self._search_regex(r'data-player=\'([^\']+)\'>', webpage, 'data player', default=None)
+            if player:
+                player_data = json.loads(player)
+                video_id = compat_str(player_data['refMedia'])
+            else:
+                model = self._search_regex(r'data-model="([^"]+)">', webpage, 'data model')
+                model_data = self._parse_json(unescapeHTML(model), display_id)
+                video_id = compat_str(model_data['id'])
 
         xml = self._download_xml('http://www.allocine.fr/ws/AcVisiondataV4.ashx?media=%s' % video_id, display_id)
 
-        video = xml.find('.//AcVisionVideo').attrib
+        video = xpath_element(xml, './/AcVisionVideo').attrib
         quality = qualities(['ld', 'md', 'hd'])
 
         formats = []
@@ -75,9 +83,7 @@ class AllocineIE(InfoExtractor):
                     'format_id': format_id,
                     'quality': quality(format_id),
                     'url': v,
-                    'ext': determine_ext(v),
                 })
-
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/alphaporno.py

@@ -0,0 +1,77 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    parse_iso8601,
+    parse_duration,
+    parse_filesize,
+    int_or_none,
+)
+
+
+class AlphaPornoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?alphaporno\.com/videos/(?P<id>[^/]+)'
+    _TEST = {
+        'url': 'http://www.alphaporno.com/videos/sensual-striptease-porn-with-samantha-alexandra/',
+        'md5': 'feb6d3bba8848cd54467a87ad34bd38e',
+        'info_dict': {
+            'id': '258807',
+            'display_id': 'sensual-striptease-porn-with-samantha-alexandra',
+            'ext': 'mp4',
+            'title': 'Sensual striptease porn with Samantha Alexandra',
+            'thumbnail': 're:https?://.*\.jpg$',
+            'timestamp': 1418694611,
+            'upload_date': '20141216',
+            'duration': 387,
+            'filesize_approx': 54120000,
+            'tbr': 1145,
+            'categories': list,
+            'age_limit': 18,
+        }
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        video_id = self._search_regex(
+            r"video_id\s*:\s*'([^']+)'", webpage, 'video id', default=None)
+
+        video_url = self._search_regex(
+            r"video_url\s*:\s*'([^']+)'", webpage, 'video url')
+        ext = self._html_search_meta(
+            'encodingFormat', webpage, 'ext', default='.mp4')[1:]
+
+        title = self._search_regex(
+            [r'<meta content="([^"]+)" itemprop="description">',
+             r'class="title" itemprop="name">([^<]+)<'],
+            webpage, 'title')
+        thumbnail = self._html_search_meta('thumbnail', webpage, 'thumbnail')
+        timestamp = parse_iso8601(self._html_search_meta(
+            'uploadDate', webpage, 'upload date'))
+        duration = parse_duration(self._html_search_meta(
+            'duration', webpage, 'duration'))
+        filesize_approx = parse_filesize(self._html_search_meta(
+            'contentSize', webpage, 'file size'))
+        bitrate = int_or_none(self._html_search_meta(
+            'bitrate', webpage, 'bitrate'))
+        categories = self._html_search_meta(
+            'keywords', webpage, 'categories', default='').split(',')
+
+        age_limit = self._rta_search(webpage)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'url': video_url,
+            'ext': ext,
+            'title': title,
+            'thumbnail': thumbnail,
+            'timestamp': timestamp,
+            'duration': duration,
+            'filesize_approx': filesize_approx,
+            'tbr': bitrate,
+            'categories': categories,
+            'age_limit': age_limit,
+        }

youtube_dl/extractor/amp.py

@@ -0,0 +1,81 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    parse_iso8601,
+)
+
+
+class AMPIE(InfoExtractor):
+    # parse Akamai Adaptive Media Player feed
+    def _extract_feed_info(self, url):
+        item = self._download_json(
+            url, None, 'Downloading Akamai AMP feed',
+            'Unable to download Akamai AMP feed')['channel']['item']
+
+        video_id = item['guid']
+
+        def get_media_node(name, default=None):
+            media_name = 'media-%s' % name
+            media_group = item.get('media-group') or item
+            return media_group.get(media_name) or item.get(media_name) or item.get(name, default)
+
+        thumbnails = []
+        media_thumbnail = get_media_node('thumbnail')
+        if media_thumbnail:
+            if isinstance(media_thumbnail, dict):
+                media_thumbnail = [media_thumbnail]
+            for thumbnail_data in media_thumbnail:
+                thumbnail = thumbnail_data['@attributes']
+                thumbnails.append({
+                    'url': self._proto_relative_url(thumbnail['url'], 'http:'),
+                    'width': int_or_none(thumbnail.get('width')),
+                    'height': int_or_none(thumbnail.get('height')),
+                })
+
+        subtitles = {}
+        media_subtitle = get_media_node('subTitle')
+        if media_subtitle:
+            if isinstance(media_subtitle, dict):
+                media_subtitle = [media_subtitle]
+            for subtitle_data in media_subtitle:
+                subtitle = subtitle_data['@attributes']
+                lang = subtitle.get('lang') or 'en'
+                subtitles[lang] = [{'url': subtitle['href']}]
+
+        formats = []
+        media_content = get_media_node('content')
+        if isinstance(media_content, dict):
+            media_content = [media_content]
+        for media_data in media_content:
+            media = media_data['@attributes']
+            media_type = media['type']
+            if media_type == 'video/f4m':
+                formats.extend(self._extract_f4m_formats(
+                    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':
+                formats.extend(self._extract_m3u8_formats(
+                    media['url'], video_id, 'mp4', m3u8_id='hls', fatal=False))
+            else:
+                formats.append({
+                    'format_id': media_data['media-category']['@attributes']['label'],
+                    'url': media['url'],
+                    'tbr': int_or_none(media.get('bitrate')),
+                    'filesize': int_or_none(media.get('fileSize')),
+                })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': get_media_node('title'),
+            'description': get_media_node('description'),
+            'thumbnails': thumbnails,
+            'timestamp': parse_iso8601(item.get('pubDate'), ' '),
+            'duration': int_or_none(media_content[0].get('@attributes', {}).get('duration')),
+            'subtitles': subtitles,
+            'formats': formats,
+        }

youtube_dl/extractor/anitube.py

@@ -1,11 +1,9 @@
 from __future__ import unicode_literals
 
-import re
-
-from .common import InfoExtractor
+from .nuevo import NuevoBaseIE
 
 
-class AnitubeIE(InfoExtractor):
+class AnitubeIE(NuevoBaseIE):
     IE_NAME = 'anitube.se'
     _VALID_URL = r'https?://(?:www\.)?anitube\.se/video/(?P<id>\d+)'
 
@@ -22,38 +20,11 @@ class AnitubeIE(InfoExtractor):
     }
 
     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)
-        key = self._html_search_regex(
-            r'http://www\.anitube\.se/embed/([A-Za-z0-9_-]*)', webpage, 'key')
+        key = self._search_regex(
+            r'src=["\']https?://[^/]+/embed/([A-Za-z0-9_-]+)', webpage, 'key')
 
-        config_xml = self._download_xml(
-            'http://www.anitube.se/nuevo/econfig.php?key=%s' % key, key)
-
-        video_title = config_xml.find('title').text
-        thumbnail = config_xml.find('image').text
-        duration = float(config_xml.find('duration').text)
-
-        formats = []
-        video_url = config_xml.find('file')
-        if video_url is not None:
-            formats.append({
-                'format_id': 'sd',
-                'url': video_url.text,
-            })
-        video_url = config_xml.find('filehd')
-        if video_url is not None:
-            formats.append({
-                'format_id': 'hd',
-                'url': video_url.text,
-            })
-
-        return {
-            'id': video_id,
-            'title': video_title,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'formats': formats
-        }
+        return self._extract_nuevo(
+            'http://www.anitube.se/nuevo/econfig.php?key=%s' % key, video_id)

youtube_dl/extractor/aol.py

@@ -3,7 +3,6 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from .fivemin import FiveMinIE
 
 
 class AolIE(InfoExtractor):
@@ -42,31 +41,30 @@ class AolIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
-
         playlist_id = mobj.group('playlist_id')
-        if playlist_id and not self._downloader.params.get('noplaylist'):
-            self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
+        if not playlist_id or self._downloader.params.get('noplaylist'):
+            return self.url_result('5min:%s' % video_id)
 
-            webpage = self._download_webpage(url, playlist_id)
-            title = self._html_search_regex(
-                r'<h1 class="video-title[^"]*">(.+?)</h1>', webpage, 'title')
-            playlist_html = self._search_regex(
-                r"(?s)<ul\s+class='video-related[^']*'>(.*?)</ul>", webpage,
-                'playlist HTML')
-            entries = [{
-                '_type': 'url',
-                'url': 'aol-video:%s' % m.group('id'),
-                'ie_key': 'Aol',
-            } for m in re.finditer(
-                r"<a\s+href='.*videoid=(?P<id>[0-9]+)'\s+class='video-thumb'>",
-                playlist_html)]
+        self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
 
-            return {
-                '_type': 'playlist',
-                'id': playlist_id,
-                'display_id': mobj.group('playlist_display_id'),
-                'title': title,
-                'entries': entries,
-            }
+        webpage = self._download_webpage(url, playlist_id)
+        title = self._html_search_regex(
+            r'<h1 class="video-title[^"]*">(.+?)</h1>', webpage, 'title')
+        playlist_html = self._search_regex(
+            r"(?s)<ul\s+class='video-related[^']*'>(.*?)</ul>", webpage,
+            'playlist HTML')
+        entries = [{
+            '_type': 'url',
+            'url': 'aol-video:%s' % m.group('id'),
+            'ie_key': 'Aol',
+        } for m in re.finditer(
+            r"<a\s+href='.*videoid=(?P<id>[0-9]+)'\s+class='video-thumb'>",
+            playlist_html)]
 
-        return FiveMinIE._build_result(video_id)
+        return {
+            '_type': 'playlist',
+            'id': playlist_id,
+            'display_id': mobj.group('playlist_display_id'),
+            'title': title,
+            'entries': entries,
+        }

youtube_dl/extractor/aparat.py

@@ -20,6 +20,7 @@ class AparatIE(InfoExtractor):
             'id': 'wP8On',
             'ext': 'mp4',
             'title': 'تیم گلکسی 11 - زومیت',
+            'age_limit': 0,
         },
         # 'skip': 'Extremely unreliable',
     }
@@ -34,7 +35,8 @@ class AparatIE(InfoExtractor):
                      video_id + '/vt/frame')
         webpage = self._download_webpage(embed_url, video_id)
 
-        video_urls = re.findall(r'fileList\[[0-9]+\]\s*=\s*"([^"]+)"', webpage)
+        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):
             req = HEADRequest(video_url)
             res = self._request_webpage(
@@ -46,7 +48,7 @@ class AparatIE(InfoExtractor):
 
         title = self._search_regex(r'\s+title:\s*"([^"]+)"', webpage, 'title')
         thumbnail = self._search_regex(
-            r'\s+image:\s*"([^"]+)"', webpage, 'thumbnail', fatal=False)
+            r'image:\s*"([^"]+)"', webpage, 'thumbnail', fatal=False)
 
         return {
             'id': video_id,
@@ -54,4 +56,5 @@ class AparatIE(InfoExtractor):
             'url': video_url,
             'ext': 'mp4',
             'thumbnail': thumbnail,
+            'age_limit': self._family_friendly_search(webpage),
         }

youtube_dl/extractor/appleconnect.py

@@ -0,0 +1,50 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    str_to_int,
+    ExtractorError
+)
+
+
+class AppleConnectIE(InfoExtractor):
+    _VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/idsa\.(?P<id>[\w-]+)'
+    _TEST = {
+        'url': 'https://itunes.apple.com/us/post/idsa.4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
+        'md5': '10d0f2799111df4cb1c924520ca78f98',
+        'info_dict': {
+            'id': '4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
+            'ext': 'm4v',
+            'title': 'Energy',
+            'uploader': 'Drake',
+            'thumbnail': 'http://is5.mzstatic.com/image/thumb/Video5/v4/78/61/c5/7861c5fa-ad6d-294b-1464-cf7605b911d6/source/1920x1080sr.jpg',
+            'upload_date': '20150710',
+            'timestamp': 1436545535,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        try:
+            video_json = self._html_search_regex(
+                r'class="auc-video-data">(\{.*?\})', webpage, 'json')
+        except ExtractorError:
+            raise ExtractorError('This post doesn\'t contain a video', expected=True)
+
+        video_data = self._parse_json(video_json, video_id)
+        timestamp = str_to_int(self._html_search_regex(r'data-timestamp="(\d+)"', webpage, 'timestamp'))
+        like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count'))
+
+        return {
+            'id': video_id,
+            'url': video_data['sslSrc'],
+            'title': video_data['title'],
+            'description': video_data['description'],
+            'uploader': video_data['artistName'],
+            'thumbnail': video_data['artworkUrl'],
+            'timestamp': timestamp,
+            'like_count': like_count,
+        }

youtube_dl/extractor/appletrailers.py

@@ -4,63 +4,76 @@ import re
 import json
 
 from .common import InfoExtractor
+from ..compat import compat_urlparse
 from ..utils import (
-    compat_urlparse,
     int_or_none,
 )
 
 
 class AppleTrailersIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/trailers/(?P<company>[^/]+)/(?P<movie>[^/]+)'
-    _TEST = {
-        "url": "http://trailers.apple.com/trailers/wb/manofsteel/",
-        "playlist": [
+    IE_NAME = 'appletrailers'
+    _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/(?:trailers|ca)/(?P<company>[^/]+)/(?P<movie>[^/]+)'
+    _TESTS = [{
+        'url': 'http://trailers.apple.com/trailers/wb/manofsteel/',
+        'info_dict': {
+            'id': 'manofsteel',
+        },
+        'playlist': [
             {
-                "md5": "d97a8e575432dbcb81b7c3acb741f8a8",
-                "info_dict": {
-                    "id": "manofsteel-trailer4",
-                    "ext": "mov",
-                    "duration": 111,
-                    "title": "Trailer 4",
-                    "upload_date": "20130523",
-                    "uploader_id": "wb",
+                'md5': 'd97a8e575432dbcb81b7c3acb741f8a8',
+                'info_dict': {
+                    'id': 'manofsteel-trailer4',
+                    'ext': 'mov',
+                    'duration': 111,
+                    'title': 'Trailer 4',
+                    'upload_date': '20130523',
+                    'uploader_id': 'wb',
                 },
             },
             {
-                "md5": "b8017b7131b721fb4e8d6f49e1df908c",
-                "info_dict": {
-                    "id": "manofsteel-trailer3",
-                    "ext": "mov",
-                    "duration": 182,
-                    "title": "Trailer 3",
-                    "upload_date": "20130417",
-                    "uploader_id": "wb",
+                'md5': 'b8017b7131b721fb4e8d6f49e1df908c',
+                'info_dict': {
+                    'id': 'manofsteel-trailer3',
+                    'ext': 'mov',
+                    'duration': 182,
+                    'title': 'Trailer 3',
+                    'upload_date': '20130417',
+                    'uploader_id': 'wb',
                 },
             },
             {
-                "md5": "d0f1e1150989b9924679b441f3404d48",
-                "info_dict": {
-                    "id": "manofsteel-trailer",
-                    "ext": "mov",
-                    "duration": 148,
-                    "title": "Trailer",
-                    "upload_date": "20121212",
-                    "uploader_id": "wb",
+                'md5': 'd0f1e1150989b9924679b441f3404d48',
+                'info_dict': {
+                    'id': 'manofsteel-trailer',
+                    'ext': 'mov',
+                    'duration': 148,
+                    'title': 'Trailer',
+                    'upload_date': '20121212',
+                    'uploader_id': 'wb',
                 },
             },
             {
-                "md5": "5fe08795b943eb2e757fa95cb6def1cb",
-                "info_dict": {
-                    "id": "manofsteel-teaser",
-                    "ext": "mov",
-                    "duration": 93,
-                    "title": "Teaser",
-                    "upload_date": "20120721",
-                    "uploader_id": "wb",
+                'md5': '5fe08795b943eb2e757fa95cb6def1cb',
+                'info_dict': {
+                    'id': 'manofsteel-teaser',
+                    'ext': 'mov',
+                    'duration': 93,
+                    'title': 'Teaser',
+                    'upload_date': '20120721',
+                    'uploader_id': 'wb',
                 },
             },
         ]
-    }
+    }, {
+        'url': 'http://trailers.apple.com/trailers/magnolia/blackthorn/',
+        'info_dict': {
+            'id': 'blackthorn',
+        },
+        'playlist_mincount': 2,
+    }, {
+        'url': 'http://trailers.apple.com/ca/metropole/autrui/',
+        'only_matching': True,
+    }]
 
     _JSON_RE = r'iTunes.playURL\((.*?)\);'
 
@@ -73,7 +86,7 @@ class AppleTrailersIE(InfoExtractor):
 
         def fix_html(s):
             s = re.sub(r'(?s)<script[^<]*?>.*?</script>', '', s)
-            s = re.sub(r'<img ([^<]*?)>', r'<img \1/>', s)
+            s = re.sub(r'<img ([^<]*?)/?>', r'<img \1/>', s)
             # The ' in the onClick attributes are not escaped, it couldn't be parsed
             # like: http://trailers.apple.com/trailers/wb/gravity/
 
@@ -90,6 +103,9 @@ class AppleTrailersIE(InfoExtractor):
             trailer_info_json = self._search_regex(self._JSON_RE,
                                                    on_click, 'trailer info')
             trailer_info = json.loads(trailer_info_json)
+            first_url = trailer_info.get('url')
+            if not first_url:
+                continue
             title = trailer_info['title']
             video_id = movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', title).lower()
             thumbnail = li.find('.//img').attrib['src']
@@ -101,7 +117,6 @@ class AppleTrailersIE(InfoExtractor):
             if m:
                 duration = 60 * int(m.group('minutes')) + int(m.group('seconds'))
 
-            first_url = trailer_info['url']
             trailer_id = first_url.split('/')[-1].rpartition('_')[0].lower()
             settings_json_url = compat_urlparse.urljoin(url, 'includes/settings/%s.json' % trailer_id)
             settings = self._download_json(settings_json_url, trailer_id, 'Downloading settings json')
@@ -122,14 +137,15 @@ class AppleTrailersIE(InfoExtractor):
             playlist.append({
                 '_type': 'video',
                 'id': video_id,
-                'title': title,
                 'formats': formats,
                 'title': title,
                 'duration': duration,
                 'thumbnail': thumbnail,
                 'upload_date': upload_date,
                 'uploader_id': uploader_id,
-                'user_agent': 'QuickTime compatible (youtube-dl)',
+                'http_headers': {
+                    'User-Agent': 'QuickTime compatible (youtube-dl)',
+                },
             })
 
         return {
@@ -137,3 +153,76 @@ class AppleTrailersIE(InfoExtractor):
             'id': movie,
             'entries': playlist,
         }
+
+
+class AppleTrailersSectionIE(InfoExtractor):
+    IE_NAME = 'appletrailers:section'
+    _SECTIONS = {
+        'justadded': {
+            'feed_path': 'just_added',
+            'title': 'Just Added',
+        },
+        'exclusive': {
+            'feed_path': 'exclusive',
+            'title': 'Exclusive',
+        },
+        'justhd': {
+            'feed_path': 'just_hd',
+            'title': 'Just HD',
+        },
+        'mostpopular': {
+            'feed_path': 'most_pop',
+            'title': 'Most Popular',
+        },
+        'moviestudios': {
+            'feed_path': 'studios',
+            'title': 'Movie Studios',
+        },
+    }
+    _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/#section=(?P<id>%s)' % '|'.join(_SECTIONS)
+    _TESTS = [{
+        'url': 'http://trailers.apple.com/#section=justadded',
+        'info_dict': {
+            'title': 'Just Added',
+            'id': 'justadded',
+        },
+        'playlist_mincount': 80,
+    }, {
+        'url': 'http://trailers.apple.com/#section=exclusive',
+        'info_dict': {
+            'title': 'Exclusive',
+            'id': 'exclusive',
+        },
+        'playlist_mincount': 80,
+    }, {
+        'url': 'http://trailers.apple.com/#section=justhd',
+        'info_dict': {
+            'title': 'Just HD',
+            'id': 'justhd',
+        },
+        'playlist_mincount': 80,
+    }, {
+        'url': 'http://trailers.apple.com/#section=mostpopular',
+        'info_dict': {
+            'title': 'Most Popular',
+            'id': 'mostpopular',
+        },
+        'playlist_mincount': 80,
+    }, {
+        'url': 'http://trailers.apple.com/#section=moviestudios',
+        'info_dict': {
+            'title': 'Movie Studios',
+            'id': 'moviestudios',
+        },
+        'playlist_mincount': 80,
+    }]
+
+    def _real_extract(self, url):
+        section = self._match_id(url)
+        section_data = self._download_json(
+            'http://trailers.apple.com/trailers/home/feeds/%s.json' % self._SECTIONS[section]['feed_path'],
+            section)
+        entries = [
+            self.url_result('http://trailers.apple.com' + e['location'])
+            for e in section_data]
+        return self.playlist_result(entries, section, self._SECTIONS[section]['title'])

youtube_dl/extractor/archiveorg.py

@@ -1,42 +1,48 @@
 from __future__ import unicode_literals
 
-import json
-import re
-
 from .common import InfoExtractor
-from ..utils import (
-    unified_strdate,
-)
+from ..utils import unified_strdate
 
 
 class ArchiveOrgIE(InfoExtractor):
     IE_NAME = 'archive.org'
     IE_DESC = 'archive.org videos'
-    _VALID_URL = r'(?:https?://)?(?:www\.)?archive\.org/details/(?P<id>[^?/]+)(?:[?].*)?$'
-    _TEST = {
-        "url": "http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect",
-        'file': 'XD300-23_68HighlightsAResearchCntAugHumanIntellect.ogv',
+    _VALID_URL = r'https?://(?:www\.)?archive\.org/details/(?P<id>[^?/]+)(?:[?].*)?$'
+    _TESTS = [{
+        'url': 'http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect',
         'md5': '8af1d4cf447933ed3c7f4871162602db',
         'info_dict': {
-            "title": "1968 Demo - FJCC Conference Presentation Reel #1",
-            "description": "Reel 1 of 3: Also known as the \"Mother of All Demos\", Doug Engelbart's presentation at the Fall Joint Computer Conference in San Francisco, December 9, 1968 titled \"A Research Center for Augmenting Human Intellect.\" For this presentation, Doug and his team astonished the audience by not only relating their research, but demonstrating it live. This was the debut of the mouse, interactive computing, hypermedia, computer supported software engineering, video teleconferencing, etc. See also <a href=\"http://dougengelbart.org/firsts/dougs-1968-demo.html\" rel=\"nofollow\">Doug's 1968 Demo page</a> for more background, highlights, links, and the detailed paper published in this conference proceedings. Filmed on 3 reels: Reel 1 | <a href=\"http://www.archive.org/details/XD300-24_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 2</a> | <a href=\"http://www.archive.org/details/XD300-25_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 3</a>",
-            "upload_date": "19681210",
-            "uploader": "SRI International"
+            'id': 'XD300-23_68HighlightsAResearchCntAugHumanIntellect',
+            'ext': 'ogv',
+            'title': '1968 Demo - FJCC Conference Presentation Reel #1',
+            'description': 'md5:1780b464abaca9991d8968c877bb53ed',
+            'upload_date': '19681210',
+            'uploader': 'SRI International'
         }
-    }
+    }, {
+        'url': 'https://archive.org/details/Cops1922',
+        'md5': '18f2a19e6d89af8425671da1cf3d4e04',
+        'info_dict': {
+            'id': 'Cops1922',
+            'ext': 'ogv',
+            'title': 'Buster Keaton\'s "Cops" (1922)',
+            'description': 'md5:70f72ee70882f713d4578725461ffcc3',
+        }
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
-        json_url = url + ('?' if '?' in url else '&') + 'output=json'
-        json_data = self._download_webpage(json_url, video_id)
-        data = json.loads(json_data)
+        json_url = url + ('&' if '?' in url else '?') + 'output=json'
+        data = self._download_json(json_url, video_id)
 
-        title = data['metadata']['title'][0]
-        description = data['metadata']['description'][0]
-        uploader = data['metadata']['creator'][0]
-        upload_date = unified_strdate(data['metadata']['date'][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 = [
             {

youtube_dl/extractor/ard.py

@@ -8,13 +8,14 @@ from .generic import GenericIE
 from ..utils import (
     determine_ext,
     ExtractorError,
+    get_element_by_attribute,
     qualities,
     int_or_none,
     parse_duration,
     unified_strdate,
     xpath_text,
-    parse_xml,
 )
+from ..compat import compat_etree_fromstring
 
 
 class ARDMediathekIE(InfoExtractor):
@@ -22,25 +23,127 @@ class ARDMediathekIE(InfoExtractor):
     _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
 
     _TESTS = [{
-        'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
-        'file': '22429276.mp4',
-        'md5': '469751912f1de0816a9fc9df8336476c',
+        'url': 'http://www.ardmediathek.de/tv/Dokumentation-und-Reportage/Ich-liebe-das-Leben-trotzdem/rbb-Fernsehen/Video?documentId=29582122&bcastId=3822114',
         'info_dict': {
-            'title': 'Vertrauen ist gut, Spionieren ist besser - Geht so deutsch-amerikanische Freundschaft?',
-            'description': 'Das Erste Mediathek [ARD]: Vertrauen ist gut, Spionieren ist besser - Geht so deutsch-amerikanische Freundschaft?, Anne Will, Über die Spionage-Affäre diskutieren Clemens Binninger, Katrin Göring-Eckardt, Georg Mascolo, Andrew B. Denison und Constanze Kurz.. Das Video zur Sendung Anne Will am Mittwoch, 16.07.2014',
-        },
-        'skip': 'Blocked outside of Germany',
-    }, {
-        'url': 'http://www.ardmediathek.de/tv/Tatort/Das-Wunder-von-Wolbeck-Video-tgl-ab-20/Das-Erste/Video?documentId=22490580&bcastId=602916',
-        'info_dict': {
-            'id': '22490580',
+            'id': '29582122',
             'ext': 'mp4',
-            'title': 'Das Wunder von Wolbeck (Video tgl. ab 20 Uhr)',
-            'description': 'Auf einem restaurierten Hof bei Wolbeck wird der Heilpraktiker Raffael Lembeck eines morgens von seiner Frau Stella tot aufgefunden. Das Opfer war offensichtlich in seiner Praxis zu Fall gekommen und ist dann verblutet, erklärt Prof. Boerne am Tatort.',
+            'title': 'Ich liebe das Leben trotzdem',
+            'description': 'md5:45e4c225c72b27993314b31a84a5261c',
+            'duration': 4557,
         },
-        'skip': 'Blocked outside of Germany',
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916',
+        'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e',
+        'info_dict': {
+            'id': '29522730',
+            'ext': 'mp4',
+            'title': 'Tatort: Scheinwelten - Hörfassung (Video tgl. ab 20 Uhr)',
+            'description': 'md5:196392e79876d0ac94c94e8cdb2875f1',
+            'duration': 5252,
+        },
+    }, {
+        # 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',
+        'md5': '219d94d8980b4f538c7fcb0865eb7f2c',
+        'info_dict': {
+            'id': '28488308',
+            'ext': 'mp3',
+            'title': 'Tod eines Fußballers',
+            'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef',
+            'duration': 3240,
+        },
+    }, {
+        'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
+        'only_matching': True,
     }]
 
+    def _extract_media_info(self, media_info_url, webpage, video_id):
+        media_info = self._download_json(
+            media_info_url, video_id, 'Downloading media JSON')
+
+        formats = self._extract_formats(media_info, video_id)
+
+        if not formats:
+            if '"fsk"' in webpage:
+                raise ExtractorError(
+                    'This video is only available after 20:00', expected=True)
+            elif media_info.get('_geoblocked'):
+                raise ExtractorError('This video is not available due to geo restriction', expected=True)
+
+        self._sort_formats(formats)
+
+        duration = int_or_none(media_info.get('_duration'))
+        thumbnail = media_info.get('_previewImage')
+
+        subtitles = {}
+        subtitle_url = media_info.get('_subtitleUrl')
+        if subtitle_url:
+            subtitles['de'] = [{
+                'ext': 'srt',
+                'url': subtitle_url,
+            }]
+
+        return {
+            'id': video_id,
+            'duration': duration,
+            'thumbnail': thumbnail,
+            'formats': formats,
+            'subtitles': subtitles,
+        }
+
+    def _extract_formats(self, media_info, video_id):
+        type_ = media_info.get('_type')
+        media_array = media_info.get('_mediaArray', [])
+        formats = []
+        for num, media in enumerate(media_array):
+            for stream in media.get('_mediaStreamArray', []):
+                stream_urls = stream.get('_stream')
+                if not stream_urls:
+                    continue
+                if not isinstance(stream_urls, list):
+                    stream_urls = [stream_urls]
+                quality = stream.get('_quality')
+                server = stream.get('_server')
+                for stream_url in stream_urls:
+                    ext = determine_ext(stream_url)
+                    if quality != 'auto' and ext in ('f4m', 'm3u8'):
+                        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))
+                    elif ext == 'm3u8':
+                        formats.extend(self._extract_m3u8_formats(
+                            stream_url, video_id, 'mp4', preference=1, m3u8_id='hls', fatal=False))
+                    else:
+                        if server and server.startswith('rtmp'):
+                            f = {
+                                'url': server,
+                                'play_path': stream_url,
+                                'format_id': 'a%s-rtmp-%s' % (num, quality),
+                            }
+                        elif stream_url.startswith('http'):
+                            f = {
+                                'url': stream_url,
+                                'format_id': 'a%s-%s-%s' % (num, ext, quality)
+                            }
+                        else:
+                            continue
+                        m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$', stream_url)
+                        if m:
+                            f.update({
+                                'width': int(m.group('width')),
+                                'height': int(m.group('height')),
+                            })
+                        if type_ == 'audio':
+                            f['vcodec'] = 'none'
+                        formats.append(f)
+        return formats
+
     def _real_extract(self, url):
         # determine video id from url
         m = re.match(self._VALID_URL, url)
@@ -56,8 +159,11 @@ class ARDMediathekIE(InfoExtractor):
         if '>Der gewünschte Beitrag ist nicht mehr verfügbar.<' in webpage:
             raise ExtractorError('Video %s is no longer available' % video_id, expected=True)
 
+        if 'Diese Sendung ist für Jugendliche unter 12 Jahren nicht geeignet. Der Clip ist deshalb nur von 20 bis 6 Uhr verfügbar.' in webpage:
+            raise ExtractorError('This program is only suitable for those aged 12 and older. Video %s is therefore only available between 20 pm and 6 am.' % video_id, expected=True)
+
         if re.search(r'[\?&]rss($|[=&])', url):
-            doc = parse_xml(webpage)
+            doc = compat_etree_fromstring(webpage.encode('utf-8'))
             if doc.tag == 'rss':
                 return GenericIE()._extract_rss(url, video_id, doc)
 
@@ -95,46 +201,22 @@ class ARDMediathekIE(InfoExtractor):
                     'format_id': fid,
                     'url': furl,
                 })
+            self._sort_formats(formats)
+            info = {
+                'formats': formats,
+            }
         else:  # request JSON file
-            media_info = self._download_json(
-                'http://www.ardmediathek.de/play/media/%s' % video_id, video_id)
-            # The second element of the _mediaArray contains the standard http urls
-            streams = media_info['_mediaArray'][1]['_mediaStreamArray']
-            if not streams:
-                if '"fsk"' in webpage:
-                    raise ExtractorError('This video is only available after 20:00')
+            info = self._extract_media_info(
+                'http://www.ardmediathek.de/play/media/%s' % video_id, webpage, video_id)
 
-            formats = []
-            for s in streams:
-                if type(s['_stream']) == list:
-                    for index, url in enumerate(s['_stream'][::-1]):
-                        quality = s['_quality'] + index
-                        formats.append({
-                            'quality': quality,
-                            'url': url,
-                            'format_id': '%s-%s' % (determine_ext(url), quality)
-                        })
-                    continue
-
-                format = {
-                    'quality': s['_quality'],
-                    'url': s['_stream'],
-                }
-
-                format['format_id'] = '%s-%s' % (
-                    determine_ext(format['url']), format['quality'])
-
-                formats.append(format)
-
-        self._sort_formats(formats)
-
-        return {
+        info.update({
             'id': video_id,
             'title': title,
             'description': description,
-            'formats': formats,
             'thumbnail': thumbnail,
-        }
+        })
+
+        return info
 
 
 class ARDIE(InfoExtractor):
@@ -192,3 +274,41 @@ 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/arte.py

@@ -4,12 +4,16 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_parse_qs,
+    compat_urllib_parse_urlparse,
+)
 from ..utils import (
     find_xpath_attr,
     unified_strdate,
-    get_element_by_id,
     get_element_by_attribute,
     int_or_none,
+    NO_DEFAULT,
     qualities,
 )
 
@@ -37,7 +41,7 @@ class ArteTvIE(InfoExtractor):
             config_xml_url, video_id, note='Downloading configuration')
 
         formats = [{
-            'forma_id': q.attrib['quality'],
+            'format_id': q.attrib['quality'],
             # The playpath starts at 'mp4:', if we don't manually
             # split the url, rtmpdump will incorrectly parse them
             'url': q.text.split('mp4:', 1)[0],
@@ -65,9 +69,13 @@ class ArteTVPlus7IE(InfoExtractor):
     def _extract_url_info(cls, url):
         mobj = re.match(cls._VALID_URL, url)
         lang = mobj.group('lang')
-        # This is not a real id, it can be for example AJT for the news
-        # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
-        video_id = mobj.group('id')
+        query = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
+        if 'vid' in query:
+            video_id = query['vid'][0]
+        else:
+            # This is not a real id, it can be for example AJT for the news
+            # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
+            video_id = mobj.group('id')
         return video_id, lang
 
     def _real_extract(self, url):
@@ -76,9 +84,30 @@ class ArteTVPlus7IE(InfoExtractor):
         return self._extract_from_webpage(webpage, video_id, lang)
 
     def _extract_from_webpage(self, webpage, video_id, lang):
+        patterns_templates = (r'arte_vp_url=["\'](.*?%s.*?)["\']', r'data-url=["\']([^"]+%s[^"]+)["\']')
+        ids = (video_id, '')
+        # some pages contain multiple videos (like
+        # http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D),
+        # so we first try to look for json URLs that contain the video id from
+        # the 'vid' parameter.
+        patterns = [t % re.escape(_id) for _id in ids for t in patterns_templates]
         json_url = self._html_search_regex(
-            [r'arte_vp_url=["\'](.*?)["\']', r'data-url=["\']([^"]+)["\']'],
-            webpage, 'json vp url')
+            patterns, webpage, 'json vp url', default=None)
+        if not json_url:
+            def find_iframe_url(webpage, default=NO_DEFAULT):
+                return self._html_search_regex(
+                    r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1',
+                    webpage, 'iframe url', group='url', default=default)
+
+            iframe_url = find_iframe_url(webpage, None)
+            if not iframe_url:
+                embed_url = self._html_search_regex(
+                    r'arte_vp_url_oembed=\'([^\']+?)\'', webpage, 'embed url')
+                player = self._download_json(
+                    embed_url, video_id, 'Downloading player page')
+                iframe_url = find_iframe_url(player['html'])
+            json_url = compat_parse_qs(
+                compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0]
         return self._extract_from_json_url(json_url, video_id, lang)
 
     def _extract_from_json_url(self, json_url, video_id, lang):
@@ -133,7 +162,7 @@ class ArteTVPlus7IE(InfoExtractor):
                 'width': int_or_none(f.get('width')),
                 'height': int_or_none(f.get('height')),
                 'tbr': int_or_none(f.get('bitrate')),
-                'quality': qfunc(f['quality']),
+                'quality': qfunc(f.get('quality')),
                 'source_preference': source_pref,
             }
 
@@ -146,6 +175,7 @@ class ArteTVPlus7IE(InfoExtractor):
 
             formats.append(format)
 
+        self._check_formats(formats, video_id)
         self._sort_formats(formats)
 
         info_dict['formats'] = formats
@@ -179,23 +209,19 @@ class ArteTVCreativeIE(ArteTVPlus7IE):
 
 class ArteTVFutureIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:future'
-    _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(thema|sujet)/.*?#article-anchor-(?P<id>\d+)'
+    _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(?P<id>.+)'
 
-    _TEST = {
-        'url': 'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081',
+    _TESTS = [{
+        'url': 'http://future.arte.tv/fr/info-sciences/les-ecrevisses-aussi-sont-anxieuses',
         'info_dict': {
-            'id': '5201',
+            'id': '050940-028-A',
             'ext': 'mp4',
-            'title': 'Les champignons au secours de la planète',
-            'upload_date': '20131101',
+            'title': 'Les écrevisses aussi peuvent être anxieuses',
         },
-    }
-
-    def _real_extract(self, url):
-        anchor_id, lang = self._extract_url_info(url)
-        webpage = self._download_webpage(url, anchor_id)
-        row = get_element_by_id(anchor_id, webpage)
-        return self._extract_from_webpage(row, anchor_id, lang)
+    }, {
+        'url': 'http://future.arte.tv/fr/la-science-est-elle-responsable',
+        'only_matching': True,
+    }]
 
 
 class ArteTVDDCIE(ArteTVPlus7IE):
@@ -233,6 +259,23 @@ class ArteTVConcertIE(ArteTVPlus7IE):
     }
 
 
+class ArteTVCinemaIE(ArteTVPlus7IE):
+    IE_NAME = 'arte.tv:cinema'
+    _VALID_URL = r'https?://cinema\.arte\.tv/(?P<lang>de|fr)/(?P<id>.+)'
+
+    _TEST = {
+        'url': 'http://cinema.arte.tv/de/node/38291',
+        'md5': '6b275511a5107c60bacbeeda368c3aa1',
+        'info_dict': {
+            'id': '055876-000_PWA12025-D',
+            'ext': 'mp4',
+            'title': 'Tod auf dem Nil',
+            'upload_date': '20160122',
+            'description': 'md5:7f749bbb77d800ef2be11d54529b96bc',
+        },
+    }
+
+
 class ArteTVEmbedIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:embed'
     _VALID_URL = r'''(?x)

youtube_dl/extractor/atresplayer.py

@@ -0,0 +1,199 @@
+from __future__ import unicode_literals
+
+import time
+import hmac
+import hashlib
+import re
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_str,
+    compat_urllib_parse,
+)
+from ..utils import (
+    int_or_none,
+    float_or_none,
+    sanitized_Request,
+    xpath_text,
+    ExtractorError,
+)
+
+
+class AtresPlayerIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?atresplayer\.com/television/[^/]+/[^/]+/[^/]+/(?P<id>.+?)_\d+\.html'
+    _NETRC_MACHINE = 'atresplayer'
+    _TESTS = [
+        {
+            'url': 'http://www.atresplayer.com/television/programas/el-club-de-la-comedia/temporada-4/capitulo-10-especial-solidario-nochebuena_2014122100174.html',
+            'md5': 'efd56753cda1bb64df52a3074f62e38a',
+            'info_dict': {
+                'id': 'capitulo-10-especial-solidario-nochebuena',
+                'ext': 'mp4',
+                'title': 'Especial Solidario de Nochebuena',
+                'description': 'md5:e2d52ff12214fa937107d21064075bf1',
+                'duration': 5527.6,
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+            'skip': 'This video is only available for registered users'
+        },
+        {
+            'url': 'http://www.atresplayer.com/television/especial/videoencuentros/temporada-1/capitulo-112-david-bustamante_2014121600375.html',
+            'md5': '0d0e918533bbd4b263f2de4d197d4aac',
+            'info_dict': {
+                'id': 'capitulo-112-david-bustamante',
+                'ext': 'flv',
+                'title': 'David Bustamante',
+                'description': 'md5:f33f1c0a05be57f6708d4dd83a3b81c6',
+                'duration': 1439.0,
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+        },
+        {
+            'url': 'http://www.atresplayer.com/television/series/el-secreto-de-puente-viejo/el-chico-de-los-tres-lunares/capitulo-977-29-12-14_2014122400174.html',
+            'only_matching': True,
+        },
+    ]
+
+    _USER_AGENT = 'Dalvik/1.6.0 (Linux; U; Android 4.3; GT-I9300 Build/JSS15J'
+    _MAGIC = 'QWtMLXs414Yo+c#_+Q#K@NN)'
+    _TIMESTAMP_SHIFT = 30000
+
+    _TIME_API_URL = 'http://servicios.atresplayer.com/api/admin/time.json'
+    _URL_VIDEO_TEMPLATE = 'https://servicios.atresplayer.com/api/urlVideo/{1}/{0}/{1}|{2}|{3}.json'
+    _PLAYER_URL_TEMPLATE = 'https://servicios.atresplayer.com/episode/getplayer.json?episodePk=%s'
+    _EPISODE_URL_TEMPLATE = 'http://www.atresplayer.com/episodexml/%s'
+
+    _LOGIN_URL = 'https://servicios.atresplayer.com/j_spring_security_check'
+
+    _ERRORS = {
+        'UNPUBLISHED': 'We\'re sorry, but this video is not yet available.',
+        'DELETED': 'This video has expired and is no longer available for online streaming.',
+        'GEOUNPUBLISHED': 'We\'re sorry, but this video is not available in your region due to right restrictions.',
+        # 'PREMIUM': 'PREMIUM',
+    }
+
+    def _real_initialize(self):
+        self._login()
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None:
+            return
+
+        login_form = {
+            'j_username': username,
+            'j_password': password,
+        }
+
+        request = sanitized_Request(
+            self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
+        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        response = self._download_webpage(
+            request, None, 'Logging in as %s' % username)
+
+        error = self._html_search_regex(
+            r'(?s)<ul class="list_error">(.+?)</ul>', response, 'error', default=None)
+        if error:
+            raise ExtractorError(
+                'Unable to login: %s' % error, expected=True)
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        episode_id = self._search_regex(
+            r'episode="([^"]+)"', webpage, 'episode id')
+
+        request = sanitized_Request(
+            self._PLAYER_URL_TEMPLATE % episode_id,
+            headers={'User-Agent': self._USER_AGENT})
+        player = self._download_json(request, episode_id, 'Downloading player JSON')
+
+        episode_type = player.get('typeOfEpisode')
+        error_message = self._ERRORS.get(episode_type)
+        if error_message:
+            raise ExtractorError(
+                '%s returned error: %s' % (self.IE_NAME, error_message), expected=True)
+
+        formats = []
+        video_url = player.get('urlVideo')
+        if video_url:
+            format_info = {
+                'url': video_url,
+                'format_id': 'http',
+            }
+            mobj = re.search(r'(?P<bitrate>\d+)K_(?P<width>\d+)x(?P<height>\d+)', video_url)
+            if mobj:
+                format_info.update({
+                    'width': int_or_none(mobj.group('width')),
+                    'height': int_or_none(mobj.group('height')),
+                    'tbr': int_or_none(mobj.group('bitrate')),
+                })
+            formats.append(format_info)
+
+        timestamp = int_or_none(self._download_webpage(
+            self._TIME_API_URL,
+            video_id, 'Downloading timestamp', fatal=False), 1000, time.time())
+        timestamp_shifted = compat_str(timestamp + self._TIMESTAMP_SHIFT)
+        token = hmac.new(
+            self._MAGIC.encode('ascii'),
+            (episode_id + timestamp_shifted).encode('utf-8'), hashlib.md5
+        ).hexdigest()
+
+        request = sanitized_Request(
+            self._URL_VIDEO_TEMPLATE.format('windows', episode_id, timestamp_shifted, token),
+            headers={'User-Agent': self._USER_AGENT})
+
+        fmt_json = self._download_json(
+            request, video_id, 'Downloading windows video JSON')
+
+        result = fmt_json.get('resultDes')
+        if result.lower() != 'ok':
+            raise ExtractorError(
+                '%s returned error: %s' % (self.IE_NAME, result), expected=True)
+
+        for format_id, video_url in fmt_json['resultObject'].items():
+            if format_id == 'token' or not video_url.startswith('http'):
+                continue
+            if 'geodeswowsmpra3player' in video_url:
+                f4m_path = video_url.split('smil:', 1)[-1].split('free_', 1)[0]
+                f4m_url = 'http://drg.antena3.com/{0}hds/es/sd.f4m'.format(f4m_path)
+                # this videos are protected by DRM, the f4m downloader doesn't support them
+                continue
+            else:
+                f4m_url = video_url[:-9] + '/manifest.f4m'
+            formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False))
+        self._sort_formats(formats)
+
+        path_data = player.get('pathData')
+
+        episode = self._download_xml(
+            self._EPISODE_URL_TEMPLATE % path_data, video_id,
+            'Downloading episode XML')
+
+        duration = float_or_none(xpath_text(
+            episode, './media/asset/info/technical/contentDuration', 'duration'))
+
+        art = episode.find('./media/asset/info/art')
+        title = xpath_text(art, './name', 'title')
+        description = xpath_text(art, './description', 'description')
+        thumbnail = xpath_text(episode, './media/asset/files/background', 'thumbnail')
+
+        subtitles = {}
+        subtitle_url = xpath_text(episode, './media/asset/files/subtitle', 'subtitle')
+        if subtitle_url:
+            subtitles['es'] = [{
+                'ext': 'srt',
+                'url': subtitle_url,
+            }]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,
+            'subtitles': subtitles,
+        }

youtube_dl/extractor/atttechchannel.py

@@ -0,0 +1,55 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import unified_strdate
+
+
+class ATTTechChannelIE(InfoExtractor):
+    _VALID_URL = r'https?://techchannel\.att\.com/play-video\.cfm/([^/]+/)*(?P<id>.+)'
+    _TEST = {
+        'url': 'http://techchannel.att.com/play-video.cfm/2014/1/27/ATT-Archives-The-UNIX-System-Making-Computers-Easier-to-Use',
+        'info_dict': {
+            'id': '11316',
+            'display_id': 'ATT-Archives-The-UNIX-System-Making-Computers-Easier-to-Use',
+            'ext': 'flv',
+            'title': 'AT&T Archives : The UNIX System: Making Computers Easier to Use',
+            'description': 'A 1982 film about UNIX is the foundation for software in use around Bell Labs and AT&T.',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'upload_date': '20140127',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        video_url = self._search_regex(
+            r"url\s*:\s*'(rtmp://[^']+)'",
+            webpage, 'video URL')
+
+        video_id = self._search_regex(
+            r'mediaid\s*=\s*(\d+)',
+            webpage, 'video id', fatal=False)
+
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage)
+        thumbnail = self._og_search_thumbnail(webpage)
+        upload_date = unified_strdate(self._search_regex(
+            r'[Rr]elease\s+date:\s*(\d{1,2}/\d{1,2}/\d{4})',
+            webpage, 'upload date', fatal=False), False)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'url': video_url,
+            'ext': 'flv',
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'upload_date': upload_date,
+        }

youtube_dl/extractor/audimedia.py

@@ -0,0 +1,80 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    parse_iso8601,
+    sanitized_Request,
+)
+
+
+class AudiMediaIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?audimedia\.tv/(?:en|de)/vid/(?P<id>[^/?#]+)'
+    _TEST = {
+        'url': 'https://audimedia.tv/en/vid/60-seconds-of-audi-sport-104-2015-wec-bahrain-rookie-test',
+        'md5': '79a8b71c46d49042609795ab59779b66',
+        'info_dict': {
+            'id': '1565',
+            'ext': 'mp4',
+            'title': '60 Seconds of Audi Sport 104/2015 - WEC Bahrain, Rookie Test',
+            'description': 'md5:60e5d30a78ced725f7b8d34370762941',
+            'upload_date': '20151124',
+            'timestamp': 1448354940,
+            'duration': 74022,
+            'view_count': int,
+        }
+    }
+    # extracted from https://audimedia.tv/assets/embed/embedded-player.js (dataSourceAuthToken)
+    _AUTH_TOKEN = 'e25b42847dba18c6c8816d5d8ce94c326e06823ebf0859ed164b3ba169be97f2'
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        raw_payload = self._search_regex(r'<script[^>]+class="amtv-embed"[^>]+id="([^"]+)"', webpage, 'raw payload')
+        _, stage_mode, video_id, lang = raw_payload.split('-')
+
+        # TODO: handle s and e stage_mode (live streams and ended live streams)
+        if stage_mode not in ('s', 'e'):
+            request = sanitized_Request(
+                'https://audimedia.tv/api/video/v1/videos/%s?embed[]=video_versions&embed[]=thumbnail_image&where[content_language_iso]=%s' % (video_id, lang),
+                headers={'X-Auth-Token': self._AUTH_TOKEN})
+            json_data = self._download_json(request, video_id)['results']
+            formats = []
+
+            stream_url_hls = json_data.get('stream_url_hls')
+            if stream_url_hls:
+                formats.extend(self._extract_m3u8_formats(
+                    stream_url_hls, video_id, 'mp4',
+                    entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
+
+            stream_url_hds = json_data.get('stream_url_hds')
+            if stream_url_hds:
+                formats.extend(self._extract_f4m_formats(
+                    stream_url_hds + '?hdcore=3.4.0',
+                    video_id, f4m_id='hds', fatal=False))
+
+            for video_version in json_data.get('video_versions'):
+                video_version_url = video_version.get('download_url') or video_version.get('stream_url')
+                if not video_version_url:
+                    continue
+                formats.append({
+                    'url': video_version_url,
+                    'width': int_or_none(video_version.get('width')),
+                    'height': int_or_none(video_version.get('height')),
+                    'abr': int_or_none(video_version.get('audio_bitrate')),
+                    'vbr': int_or_none(video_version.get('video_bitrate')),
+                })
+            self._sort_formats(formats)
+
+            return {
+                'id': video_id,
+                'title': json_data['title'],
+                'description': json_data.get('subtitle'),
+                'thumbnail': json_data.get('thumbnail_image', {}).get('file'),
+                'timestamp': parse_iso8601(json_data.get('publication_date')),
+                'duration': int_or_none(json_data.get('duration')),
+                'view_count': int_or_none(json_data.get('view_count')),
+                'formats': formats,
+            }

youtube_dl/extractor/audiomack.py

@@ -1,11 +1,15 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import itertools
+import time
+
 from .common import InfoExtractor
 from .soundcloud import SoundcloudIE
-from ..utils import ExtractorError
-
-import time
+from ..utils import (
+    ExtractorError,
+    url_basename,
+)
 
 
 class AudiomackIE(InfoExtractor):
@@ -17,12 +21,13 @@ class AudiomackIE(InfoExtractor):
             'url': 'http://www.audiomack.com/song/roosh-williams/extraordinary',
             'info_dict':
             {
-                'id': 'roosh-williams/extraordinary',
+                'id': '310086',
                 'ext': 'mp3',
-                'title': 'Roosh Williams - Extraordinary'
+                'uploader': 'Roosh Williams',
+                'title': 'Extraordinary'
             }
         },
-        # hosted on soundcloud via audiomack
+        # audiomack wrapper around soundcloud song
         {
             'add_ie': ['Soundcloud'],
             'url': 'http://www.audiomack.com/song/xclusiveszone/take-kare',
@@ -38,32 +43,102 @@ class AudiomackIE(InfoExtractor):
     ]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        # URLs end with [uploader name]/[uploader title]
+        # this title is whatever the user types in, and is rarely
+        # the proper song title.  Real metadata is in the api response
+        album_url_tag = self._match_id(url)
 
+        # Request the extended version of the api for extra fields like artist and title
         api_response = self._download_json(
-            "http://www.audiomack.com/api/music/url/song/%s?_=%d" % (
-                video_id, time.time()),
-            video_id)
+            'http://www.audiomack.com/api/music/url/song/%s?extended=1&_=%d' % (
+                album_url_tag, time.time()),
+            album_url_tag)
 
-        if "url" not in api_response:
-            raise ExtractorError("Unable to deduce api url of song")
-        realurl = api_response["url"]
+        # API is inconsistent with errors
+        if 'url' not in api_response or not api_response['url'] or 'error' in api_response:
+            raise ExtractorError('Invalid url %s' % url)
 
         # Audiomack wraps a lot of soundcloud tracks in their branded wrapper
-        # - if so, pass the work off to the soundcloud extractor
-        if SoundcloudIE.suitable(realurl):
-            return {'_type': 'url', 'url': realurl, 'ie_key': 'Soundcloud'}
-
-        webpage = self._download_webpage(url, video_id)
-        artist = self._html_search_regex(
-            r'<span class="artist">(.*?)</span>', webpage, "artist")
-        songtitle = self._html_search_regex(
-            r'<h1 class="profile-title song-title"><span class="artist">.*?</span>(.*?)</h1>',
-            webpage, "title")
-        title = artist + " - " + songtitle
+        # if so, pass the work off to the soundcloud extractor
+        if SoundcloudIE.suitable(api_response['url']):
+            return {'_type': 'url', 'url': api_response['url'], 'ie_key': 'Soundcloud'}
 
         return {
-            'id': video_id,
-            'title': title,
-            'url': realurl,
+            'id': api_response.get('id', album_url_tag),
+            'uploader': api_response.get('artist'),
+            'title': api_response.get('title'),
+            'url': api_response['url'],
         }
+
+
+class AudiomackAlbumIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?audiomack\.com/album/(?P<id>[\w/-]+)'
+    IE_NAME = 'audiomack:album'
+    _TESTS = [
+        # Standard album playlist
+        {
+            'url': 'http://www.audiomack.com/album/flytunezcom/tha-tour-part-2-mixtape',
+            'playlist_count': 15,
+            'info_dict':
+            {
+                'id': '812251',
+                'title': 'Tha Tour: Part 2 (Official Mixtape)'
+            }
+        },
+        # Album playlist ripped from fakeshoredrive with no metadata
+        {
+            'url': 'http://www.audiomack.com/album/fakeshoredrive/ppp-pistol-p-project',
+            'info_dict': {
+                'title': 'PPP (Pistol P Project)',
+                'id': '837572',
+            },
+            'playlist': [{
+                'info_dict': {
+                    'title': 'PPP (Pistol P Project) - 9. Heaven or Hell (CHIMACA) ft Zuse (prod by DJ FU)',
+                    'id': '837577',
+                    'ext': 'mp3',
+                    'uploader': 'Lil Herb a.k.a. G Herbo',
+                }
+            }],
+            'params': {
+                'playliststart': 9,
+                'playlistend': 9,
+            }
+        }
+    ]
+
+    def _real_extract(self, url):
+        # URLs end with [uploader name]/[uploader title]
+        # this title is whatever the user types in, and is rarely
+        # the proper song title.  Real metadata is in the api response
+        album_url_tag = self._match_id(url)
+        result = {'_type': 'playlist', 'entries': []}
+        # There is no one endpoint for album metadata - instead it is included/repeated in each song's metadata
+        # Therefore we don't know how many songs the album has and must infi-loop until failure
+        for track_no in itertools.count():
+            # Get song's metadata
+            api_response = self._download_json(
+                'http://www.audiomack.com/api/music/url/album/%s/%d?extended=1&_=%d'
+                % (album_url_tag, track_no, time.time()), album_url_tag,
+                note='Querying song information (%d)' % (track_no + 1))
+
+            # Total failure, only occurs when url is totally wrong
+            # Won't happen in middle of valid playlist (next case)
+            if 'url' not in api_response or 'error' in api_response:
+                raise ExtractorError('Invalid url for track %d of album url %s' % (track_no, url))
+            # URL is good but song id doesn't exist - usually means end of playlist
+            elif not api_response['url']:
+                break
+            else:
+                # Pull out the album metadata and add to result (if it exists)
+                for resultkey, apikey in [('id', 'album_id'), ('title', 'album_title')]:
+                    if apikey in api_response and resultkey not in result:
+                        result[resultkey] = api_response[apikey]
+                song_id = url_basename(api_response['url']).rpartition('.')[0]
+                result['entries'].append({
+                    'id': api_response.get('id', song_id),
+                    'uploader': api_response.get('artist'),
+                    'title': api_response.get('title', song_id),
+                    'url': api_response['url'],
+                })
+        return result

youtube_dl/extractor/auengine.py

@@ -1,54 +0,0 @@
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..utils import (
-    compat_urllib_parse,
-    determine_ext,
-    ExtractorError,
-)
-
-
-class AUEngineIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?auengine\.com/embed\.php\?.*?file=(?P<id>[^&]+).*?'
-
-    _TEST = {
-        'url': 'http://auengine.com/embed.php?file=lfvlytY6&w=650&h=370',
-        'md5': '48972bdbcf1a3a2f5533e62425b41d4f',
-        'info_dict': {
-            'id': 'lfvlytY6',
-            'ext': 'mp4',
-            'title': '[Commie]The Legend of the Legendary Heroes - 03 - Replication Eye (Alpha Stigma)[F9410F5A]'
-        }
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, video_id)
-        title = self._html_search_regex(r'<title>(?P<title>.+?)</title>', webpage, 'title')
-        title = title.strip()
-        links = re.findall(r'\s(?:file|url):\s*["\']([^\'"]+)["\']', webpage)
-        links = map(compat_urllib_parse.unquote, links)
-
-        thumbnail = None
-        video_url = None
-        for link in links:
-            if link.endswith('.png'):
-                thumbnail = link
-            elif '/videos/' in link:
-                video_url = link
-        if not video_url:
-            raise ExtractorError('Could not find video URL')
-        ext = '.' + determine_ext(video_url)
-        if ext == title[-len(ext):]:
-            title = title[:-len(ext)]
-
-        return {
-            'id': video_id,
-            'url': video_url,
-            'title': title,
-            'thumbnail': thumbnail,
-            'http_referer': 'http://www.auengine.com/flowplayer/flowplayer.commercial-3.2.14.swf',
-        }

youtube_dl/extractor/azubu.py

@@ -3,7 +3,11 @@ from __future__ import unicode_literals
 import json
 
 from .common import InfoExtractor
-from ..utils import float_or_none
+from ..utils import (
+    ExtractorError,
+    float_or_none,
+    sanitized_Request,
+)
 
 
 class AzubuIE(InfoExtractor):
@@ -91,3 +95,37 @@ class AzubuIE(InfoExtractor):
             'view_count': view_count,
             'formats': formats,
         }
+
+
+class AzubuLiveIE(InfoExtractor):
+    _VALID_URL = r'http://www.azubu.tv/(?P<id>[^/]+)$'
+
+    _TEST = {
+        'url': 'http://www.azubu.tv/MarsTVMDLen',
+        'only_matching': True,
+    }
+
+    def _real_extract(self, url):
+        user = self._match_id(url)
+
+        info = self._download_json(
+            'http://api.azubu.tv/public/modules/last-video/{0}/info'.format(user),
+            user)['data']
+        if info['type'] != 'STREAM':
+            raise ExtractorError('{0} is not streaming live'.format(user), expected=True)
+
+        req = sanitized_Request(
+            'https://edge-elb.api.brightcove.com/playback/v1/accounts/3361910549001/videos/ref:' + info['reference_id'])
+        req.add_header('Accept', 'application/json;pk=BCpkADawqM1gvI0oGWg8dxQHlgT8HkdE2LnAlWAZkOlznO39bSZX726u4JqnDsK3MDXcO01JxXK2tZtJbgQChxgaFzEVdHRjaDoxaOu8hHOO8NYhwdxw9BzvgkvLUlpbDNUuDoc4E4wxDToV')
+        bc_info = self._download_json(req, user)
+        m3u8_url = next(source['src'] for source in bc_info['sources'] if source['container'] == 'M2TS')
+        formats = self._extract_m3u8_formats(m3u8_url, user, ext='mp4')
+
+        return {
+            'id': info['id'],
+            'title': self._live_title(info['title']),
+            'uploader_id': user,
+            'formats': formats,
+            'is_live': True,
+            'thumbnail': bc_info['poster'],
+        }

youtube_dl/extractor/baidu.py

@@ -0,0 +1,56 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import unescapeHTML
+
+
+class BaiduVideoIE(InfoExtractor):
+    IE_DESC = '百度视频'
+    _VALID_URL = r'http://v\.baidu\.com/(?P<type>[a-z]+)/(?P<id>\d+)\.htm'
+    _TESTS = [{
+        'url': 'http://v.baidu.com/comic/1069.htm?frp=bdbrand&q=%E4%B8%AD%E5%8D%8E%E5%B0%8F%E5%BD%93%E5%AE%B6',
+        'info_dict': {
+            'id': '1069',
+            'title': '中华小当家 TV版国语',
+            'description': 'md5:51be07afe461cf99fa61231421b5397c',
+        },
+        'playlist_count': 52,
+    }, {
+        'url': 'http://v.baidu.com/show/11595.htm?frp=bdbrand',
+        'info_dict': {
+            'id': '11595',
+            'title': 're:^奔跑吧兄弟',
+            'description': 'md5:1bf88bad6d850930f542d51547c089b8',
+        },
+        'playlist_mincount': 12,
+    }]
+
+    def _call_api(self, path, category, playlist_id, note):
+        return self._download_json('http://app.video.baidu.com/%s/?worktype=adnative%s&id=%s' % (
+            path, category, playlist_id), playlist_id, note)
+
+    def _real_extract(self, url):
+        category, playlist_id = re.match(self._VALID_URL, url).groups()
+        if category == 'show':
+            category = 'tvshow'
+        if category == 'tv':
+            category = 'tvplay'
+
+        playlist_detail = self._call_api(
+            'xqinfo', category, playlist_id, 'Download playlist JSON metadata')
+
+        playlist_title = playlist_detail['title']
+        playlist_description = unescapeHTML(playlist_detail.get('intro'))
+
+        episodes_detail = self._call_api(
+            'xqsingle', category, playlist_id, 'Download episodes JSON metadata')
+
+        entries = [self.url_result(
+            episode['url'], video_title=episode['title']
+        ) for episode in episodes_detail['videos']]
+
+        return self.playlist_result(
+            entries, playlist_id, playlist_title, playlist_description)

youtube_dl/extractor/bambuser.py

@@ -1,12 +1,18 @@
 from __future__ import unicode_literals
 
 import re
-import json
 import itertools
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_parse,
+    compat_str,
+)
 from ..utils import (
-    compat_urllib_request,
+    ExtractorError,
+    int_or_none,
+    float_or_none,
+    sanitized_Request,
 )
 
 
@@ -14,6 +20,8 @@ class BambuserIE(InfoExtractor):
     IE_NAME = 'bambuser'
     _VALID_URL = r'https?://bambuser\.com/v/(?P<id>\d+)'
     _API_KEY = '005f64509e19a868399060af746a00aa'
+    _LOGIN_URL = 'https://bambuser.com/user'
+    _NETRC_MACHINE = 'bambuser'
 
     _TEST = {
         'url': 'http://bambuser.com/v/4050584',
@@ -26,6 +34,9 @@ class BambuserIE(InfoExtractor):
             'duration': 3741,
             'uploader': 'pixelversity',
             'uploader_id': '344706',
+            'timestamp': 1382976692,
+            'upload_date': '20131028',
+            'view_count': int,
         },
         'params': {
             # It doesn't respect the 'Range' header, it would download the whole video
@@ -34,23 +45,60 @@ class BambuserIE(InfoExtractor):
         },
     }
 
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None:
+            return
+
+        login_form = {
+            'form_id': 'user_login',
+            'op': 'Log in',
+            'name': username,
+            'pass': password,
+        }
+
+        request = sanitized_Request(
+            self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
+        request.add_header('Referer', self._LOGIN_URL)
+        response = self._download_webpage(
+            request, None, 'Logging in as %s' % username)
+
+        login_error = self._html_search_regex(
+            r'(?s)<div class="messages error">(.+?)</div>',
+            response, 'login error', default=None)
+        if login_error:
+            raise ExtractorError(
+                'Unable to login: %s' % login_error, expected=True)
+
+    def _real_initialize(self):
+        self._login()
+
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        info_url = ('http://player-c.api.bambuser.com/getVideo.json?'
-                    '&api_key=%s&vid=%s' % (self._API_KEY, video_id))
-        info_json = self._download_webpage(info_url, video_id)
-        info = json.loads(info_json)['result']
+        video_id = self._match_id(url)
+
+        info = self._download_json(
+            'http://player-c.api.bambuser.com/getVideo.json?api_key=%s&vid=%s'
+            % (self._API_KEY, video_id), video_id)
+
+        error = info.get('error')
+        if error:
+            raise ExtractorError(
+                '%s returned error: %s' % (self.IE_NAME, error), expected=True)
+
+        result = info['result']
 
         return {
             'id': video_id,
-            'title': info['title'],
-            'url': info['url'],
-            'thumbnail': info.get('preview'),
-            'duration': int(info['length']),
-            'view_count': int(info['views_total']),
-            'uploader': info['username'],
-            'uploader_id': info['uid'],
+            'title': result['title'],
+            'url': result['url'],
+            'thumbnail': result.get('preview'),
+            'duration': int_or_none(result.get('length')),
+            'uploader': result.get('username'),
+            'uploader_id': compat_str(result.get('owner', {}).get('uid')),
+            'timestamp': int_or_none(result.get('created')),
+            'fps': float_or_none(result.get('framerate')),
+            'view_count': int_or_none(result.get('views_total')),
+            'comment_count': int_or_none(result.get('comment_count')),
         }
 
 
@@ -78,7 +126,7 @@ class BambuserChannelIE(InfoExtractor):
                 '&sort=created&access_mode=0%2C1%2C2&limit={count}'
                 '&method=broadcast&format=json&vid_older_than={last}'
             ).format(user=user, count=self._STEP, last=last_id)
-            req = compat_urllib_request.Request(req_url)
+            req = sanitized_Request(req_url)
             # Without setting this header, we wouldn't get any result
             req.add_header('Referer', 'http://bambuser.com/channel/%s' % user)
             data = self._download_json(

youtube_dl/extractor/bandcamp.py

@@ -4,10 +4,14 @@ import json
 import re
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
     compat_str,
     compat_urlparse,
+)
+from ..utils import (
     ExtractorError,
+    float_or_none,
+    int_or_none,
 )
 
 
@@ -50,11 +54,11 @@ class BandcampIE(InfoExtractor):
                     ext, abr_str = format_id.split('-', 1)
                     formats.append({
                         'format_id': format_id,
-                        'url': format_url,
+                        'url': self._proto_relative_url(format_url, 'http:'),
                         'ext': ext,
                         'vcodec': 'none',
                         'acodec': ext,
-                        'abr': int(abr_str),
+                        'abr': int_or_none(abr_str),
                     })
 
                 self._sort_formats(formats)
@@ -63,33 +67,36 @@ class BandcampIE(InfoExtractor):
                     'id': compat_str(data['id']),
                     'title': data['title'],
                     'formats': formats,
-                    'duration': float(data['duration']),
+                    'duration': float_or_none(data.get('duration')),
                 }
             else:
                 raise ExtractorError('No free songs found')
 
         download_link = m_download.group(1)
         video_id = self._search_regex(
-            r'var TralbumData = {.*?id: (?P<id>\d+),?$',
-            webpage, 'video id', flags=re.MULTILINE | re.DOTALL)
+            r'(?ms)var TralbumData = .*?[{,]\s*id: (?P<id>\d+),?$',
+            webpage, 'video id')
 
         download_webpage = self._download_webpage(download_link, video_id, 'Downloading free downloads page')
         # We get the dictionary of the track from some javascript code
-        info = re.search(r'items: (.*?),$', download_webpage, re.MULTILINE).group(1)
-        info = json.loads(info)[0]
+        all_info = self._parse_json(self._search_regex(
+            r'(?sm)items: (.*?),$', download_webpage, 'items'), video_id)
+        info = all_info[0]
         # We pick mp3-320 for now, until format selection can be easily implemented.
         mp3_info = info['downloads']['mp3-320']
         # If we try to use this url it says the link has expired
         initial_url = mp3_info['url']
-        re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
-        m_url = re.match(re_url, initial_url)
+        m_url = re.match(
+            r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$',
+            initial_url)
         # We build the url we will use to get the final track url
         # This url is build in Bandcamp in the script download_bunde_*.js
         request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), video_id, m_url.group('ts'))
         final_url_webpage = self._download_webpage(request_url, video_id, 'Requesting download url')
         # If we could correctly generate the .rand field the url would be
         # in the "download_url" key
-        final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
+        final_url = self._proto_relative_url(self._search_regex(
+            r'"retry_url":"(.+?)"', final_url_webpage, 'final video URL'), 'http:')
 
         return {
             'id': video_id,
@@ -104,7 +111,7 @@ class BandcampIE(InfoExtractor):
 
 class BandcampAlbumIE(InfoExtractor):
     IE_NAME = 'Bandcamp:album'
-    _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<title>[^?#]+))'
+    _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<album_id>[^?#]+)|/?(?:$|[?#]))'
 
     _TESTS = [{
         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
@@ -128,36 +135,49 @@ class BandcampAlbumIE(InfoExtractor):
         ],
         'info_dict': {
             'title': 'Jazz Format Mixtape vol.1',
+            'id': 'jazz-format-mixtape-vol-1',
+            'uploader_id': 'blazo',
         },
         'params': {
             'playlistend': 2
         },
-        'skip': 'Bandcamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test'
+        'skip': 'Bandcamp imposes download limits.'
     }, {
         'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
         'info_dict': {
             'title': 'Hierophany of the Open Grave',
+            'uploader_id': 'nightbringer',
+            'id': 'hierophany-of-the-open-grave',
         },
         'playlist_mincount': 9,
+    }, {
+        'url': 'http://dotscale.bandcamp.com',
+        'info_dict': {
+            'title': 'Loom',
+            'id': 'dotscale',
+            'uploader_id': 'dotscale',
+        },
+        'playlist_mincount': 7,
     }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        playlist_id = mobj.group('subdomain')
-        title = mobj.group('title')
-        display_id = title or playlist_id
-        webpage = self._download_webpage(url, display_id)
+        uploader_id = mobj.group('subdomain')
+        album_id = mobj.group('album_id')
+        playlist_id = album_id or uploader_id
+        webpage = self._download_webpage(url, playlist_id)
         tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
         if not tracks_paths:
             raise ExtractorError('The page doesn\'t contain any tracks')
         entries = [
             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
             for t_path in tracks_paths]
-        title = self._search_regex(r'album_title : "(.*?)"', webpage, 'title')
+        title = self._search_regex(
+            r'album_title\s*:\s*"(.*?)"', webpage, 'title', fatal=False)
         return {
             '_type': 'playlist',
+            'uploader_id': uploader_id,
             'id': playlist_id,
-            'display_id': display_id,
             'title': title,
             'entries': entries,
         }

youtube_dl/extractor/bbc.py

@@ -0,0 +1,961 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    float_or_none,
+    int_or_none,
+    parse_duration,
+    parse_iso8601,
+    remove_end,
+    unescapeHTML,
+)
+from ..compat import (
+    compat_etree_fromstring,
+    compat_HTTPError,
+)
+
+
+class BBCCoUkIE(InfoExtractor):
+    IE_NAME = 'bbc.co.uk'
+    IE_DESC = 'BBC iPlayer'
+    _ID_REGEX = r'[pb][\da-z]{7}'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?bbc\.co\.uk/
+                        (?:
+                            programmes/(?!articles/)|
+                            iplayer(?:/[^/]+)?/(?:episode/|playlist/)|
+                            music/clips[/#]|
+                            radio/player/
+                        )
+                        (?P<id>%s)
+                    ''' % _ID_REGEX
+
+    _MEDIASELECTOR_URLS = [
+        # Provides HQ HLS streams with even better quality that pc mediaset but fails
+        # with geolocation in some cases when it's even not geo restricted at all (e.g.
+        # http://www.bbc.co.uk/programmes/b06bp7lf). Also may fail with selectionunavailable.
+        'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
+        'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s',
+    ]
+
+    _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection'
+    _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist'
+
+    _NAMESPACES = (
+        _MEDIASELECTION_NS,
+        _EMP_PLAYLIST_NS,
+    )
+
+    _TESTS = [
+        {
+            'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
+            'info_dict': {
+                'id': 'b039d07m',
+                'ext': 'flv',
+                'title': 'Leonard Cohen, Kaleidoscope - BBC Radio 4',
+                'description': 'The Canadian poet and songwriter reflects on his musical career.',
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            }
+        },
+        {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
+            'info_dict': {
+                'id': 'b00yng1d',
+                'ext': 'flv',
+                'title': 'The Man in Black: Series 3: The Printed Name',
+                'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
+                'duration': 1800,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Episode is no longer available on BBC iPlayer Radio',
+        },
+        {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
+            'info_dict': {
+                'id': 'b00yng1d',
+                'ext': 'flv',
+                'title': 'The Voice UK: Series 3: Blind Auditions 5',
+                'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.",
+                'duration': 5100,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
+        },
+        {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
+            'info_dict': {
+                'id': 'b03k3pb7',
+                'ext': 'flv',
+                'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
+                'description': '2. Invasion',
+                'duration': 3600,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
+        }, {
+            'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
+            'info_dict': {
+                'id': 'b04v209v',
+                'ext': 'flv',
+                'title': 'Pete Tong, The Essential New Tune Special',
+                'description': "Pete has a very special mix - all of 2014's Essential New Tunes!",
+                'duration': 10800,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Episode is no longer available on BBC iPlayer Radio',
+        }, {
+            'url': 'http://www.bbc.co.uk/music/clips/p022h44b',
+            'note': 'Audio',
+            'info_dict': {
+                'id': 'p022h44j',
+                'ext': 'flv',
+                'title': 'BBC Proms Music Guides, Rachmaninov: Symphonic Dances',
+                'description': "In this Proms Music Guide, Andrew McGregor looks at Rachmaninov's Symphonic Dances.",
+                'duration': 227,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            }
+        }, {
+            'url': 'http://www.bbc.co.uk/music/clips/p025c0zz',
+            'note': 'Video',
+            'info_dict': {
+                'id': 'p025c103',
+                'ext': 'flv',
+                'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
+                'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
+                'duration': 226,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            }
+        }, {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls',
+            'info_dict': {
+                'id': 'p02n76xf',
+                'ext': 'flv',
+                'title': 'Natural World, 2015-2016: 2. Super Powered Owls',
+                'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d',
+                'duration': 3540,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'geolocation',
+        }, {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
+            'info_dict': {
+                'id': 'b05zmgw1',
+                'ext': 'flv',
+                'description': 'Kirsty Wark and Morgan Quaintance visit the Royal Academy as it prepares for its annual artistic extravaganza, meeting people who have come together to make the show unique.',
+                'title': 'Royal Academy Summer Exhibition',
+                'duration': 3540,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'geolocation',
+        }, {
+            # iptv-all mediaset fails with geolocation however there is no geo restriction
+            # for this programme at all
+            'url': 'http://www.bbc.co.uk/programmes/b06rkn85',
+            'info_dict': {
+                'id': 'b06rkms3',
+                'ext': 'flv',
+                'title': "Best of the Mini-Mixes 2015: Part 3, Annie Mac's Friday Night - BBC Radio 1",
+                'description': "Annie has part three in the Best of the Mini-Mixes 2015, plus the year's Most Played!",
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        }, {
+            # compact player (https://github.com/rg3/youtube-dl/issues/8147)
+            'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player',
+            'info_dict': {
+                'id': 'p028bfkj',
+                'ext': 'flv',
+                'title': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
+                'description': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+        }, {
+            'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
+            'only_matching': True,
+        }, {
+            'url': 'http://www.bbc.co.uk/music/clips#p02frcc3',
+            'only_matching': True,
+        }, {
+            'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
+            'only_matching': True,
+        }, {
+            'url': 'http://www.bbc.co.uk/radio/player/p03cchwf',
+            'only_matching': True,
+        }
+    ]
+
+    class MediaSelectionError(Exception):
+        def __init__(self, id):
+            self.id = id
+
+    def _extract_asx_playlist(self, connection, programme_id):
+        asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
+        return [ref.get('href') for ref in asx.findall('./Entry/ref')]
+
+    def _extract_connection(self, connection, programme_id):
+        formats = []
+        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)
+
+    def _findall_ns(self, element, xpath):
+        elements = []
+        for ns in self._NAMESPACES:
+            elements.extend(element.findall(xpath % ns))
+        return elements
+
+    def _extract_medias(self, media_selection):
+        error = media_selection.find('./{%s}error' % self._MEDIASELECTION_NS)
+        if error is None:
+            media_selection.find('./{%s}error' % self._EMP_PLAYLIST_NS)
+        if error is not None:
+            raise BBCCoUkIE.MediaSelectionError(error.get('id'))
+        return self._findall_ns(media_selection, './{%s}media')
+
+    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,
+                })
+            formats.extend(conn_formats)
+        return formats
+
+    def _get_subtitles(self, media, programme_id):
+        subtitles = {}
+        for connection in self._extract_connections(media):
+            captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
+            lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
+            subtitles[lang] = [
+                {
+                    'url': connection.get('href'),
+                    'ext': 'ttml',
+                },
+            ]
+        return subtitles
+
+    def _raise_extractor_error(self, media_selection_error):
+        raise ExtractorError(
+            '%s returned error: %s' % (self.IE_NAME, media_selection_error.id),
+            expected=True)
+
+    def _download_media_selector(self, programme_id):
+        last_exception = None
+        for mediaselector_url in self._MEDIASELECTOR_URLS:
+            try:
+                return self._download_media_selector_url(
+                    mediaselector_url % programme_id, programme_id)
+            except BBCCoUkIE.MediaSelectionError as e:
+                if e.id in ('notukerror', 'geolocation', 'selectionunavailable'):
+                    last_exception = e
+                    continue
+                self._raise_extractor_error(e)
+        self._raise_extractor_error(last_exception)
+
+    def _download_media_selector_url(self, url, programme_id=None):
+        try:
+            media_selection = self._download_xml(
+                url, programme_id, 'Downloading media selection XML')
+        except ExtractorError as ee:
+            if isinstance(ee.cause, compat_HTTPError) and ee.cause.code in (403, 404):
+                media_selection = compat_etree_fromstring(ee.cause.read().decode('utf-8'))
+            else:
+                raise
+        return self._process_media_selector(media_selection, programme_id)
+
+    def _process_media_selector(self, media_selection, programme_id):
+        formats = []
+        subtitles = None
+
+        for media in self._extract_medias(media_selection):
+            kind = media.get('kind')
+            if kind == 'audio':
+                formats.extend(self._extract_audio(media, programme_id))
+            elif kind == 'video':
+                formats.extend(self._extract_video(media, programme_id))
+            elif kind == 'captions':
+                subtitles = self.extract_subtitles(media, programme_id)
+        return formats, subtitles
+
+    def _download_playlist(self, playlist_id):
+        try:
+            playlist = self._download_json(
+                'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id,
+                playlist_id, 'Downloading playlist JSON')
+
+            version = playlist.get('defaultAvailableVersion')
+            if version:
+                smp_config = version['smpConfig']
+                title = smp_config['title']
+                description = smp_config['summary']
+                for item in smp_config['items']:
+                    kind = item['kind']
+                    if kind != 'programme' and kind != 'radioProgramme':
+                        continue
+                    programme_id = item.get('vpid')
+                    duration = int_or_none(item.get('duration'))
+                    formats, subtitles = self._download_media_selector(programme_id)
+                return programme_id, title, description, duration, formats, subtitles
+        except ExtractorError as ee:
+            if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404):
+                raise
+
+        # fallback to legacy playlist
+        return self._process_legacy_playlist(playlist_id)
+
+    def _process_legacy_playlist_url(self, url, display_id):
+        playlist = self._download_legacy_playlist_url(url, display_id)
+        return self._extract_from_legacy_playlist(playlist, display_id)
+
+    def _process_legacy_playlist(self, playlist_id):
+        return self._process_legacy_playlist_url(
+            'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id)
+
+    def _download_legacy_playlist_url(self, url, playlist_id=None):
+        return self._download_xml(
+            url, playlist_id, 'Downloading legacy playlist XML')
+
+    def _extract_from_legacy_playlist(self, playlist, playlist_id):
+        no_items = playlist.find('./{%s}noItems' % self._EMP_PLAYLIST_NS)
+        if no_items is not None:
+            reason = no_items.get('reason')
+            if reason == 'preAvailability':
+                msg = 'Episode %s is not yet available' % playlist_id
+            elif reason == 'postAvailability':
+                msg = 'Episode %s is no longer available' % playlist_id
+            elif reason == 'noMedia':
+                msg = 'Episode %s is not currently available' % playlist_id
+            else:
+                msg = 'Episode %s is not available: %s' % (playlist_id, reason)
+            raise ExtractorError(msg, expected=True)
+
+        for item in self._extract_items(playlist):
+            kind = item.get('kind')
+            if kind != 'programme' and kind != 'radioProgramme':
+                continue
+            title = playlist.find('./{%s}title' % self._EMP_PLAYLIST_NS).text
+            description_el = playlist.find('./{%s}summary' % self._EMP_PLAYLIST_NS)
+            description = description_el.text if description_el is not None else None
+
+            def get_programme_id(item):
+                def get_from_attributes(item):
+                    for p in('identifier', 'group'):
+                        value = item.get(p)
+                        if value and re.match(r'^[pb][\da-z]{7}$', value):
+                            return value
+                get_from_attributes(item)
+                mediator = item.find('./{%s}mediator' % self._EMP_PLAYLIST_NS)
+                if mediator is not None:
+                    return get_from_attributes(mediator)
+
+            programme_id = get_programme_id(item)
+            duration = int_or_none(item.get('duration'))
+
+            if programme_id:
+                formats, subtitles = self._download_media_selector(programme_id)
+            else:
+                formats, subtitles = self._process_media_selector(item, playlist_id)
+                programme_id = playlist_id
+
+        return programme_id, title, description, duration, formats, subtitles
+
+    def _real_extract(self, url):
+        group_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, group_id, 'Downloading video page')
+
+        programme_id = None
+        duration = None
+
+        tviplayer = self._search_regex(
+            r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
+            webpage, 'player', default=None)
+
+        if tviplayer:
+            player = self._parse_json(tviplayer, group_id).get('player', {})
+            duration = int_or_none(player.get('duration'))
+            programme_id = player.get('vpid')
+
+        if not programme_id:
+            programme_id = self._search_regex(
+                r'"vpid"\s*:\s*"(%s)"' % self._ID_REGEX, webpage, 'vpid', fatal=False, default=None)
+
+        if programme_id:
+            formats, subtitles = self._download_media_selector(programme_id)
+            title = self._og_search_title(webpage, default=None) or self._html_search_regex(
+                (r'<h2[^>]+id="parent-title"[^>]*>(.+?)</h2>',
+                 r'<div[^>]+class="info"[^>]*>\s*<h1>(.+?)</h1>'), webpage, 'title')
+            description = self._search_regex(
+                (r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
+                 r'<div[^>]+class="info_+synopsis"[^>]*>([^<]+)</div>'),
+                webpage, 'description', default=None)
+            if not description:
+                description = self._html_search_meta('description', webpage)
+        else:
+            programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
+
+        self._sort_formats(formats)
+
+        return {
+            'id': programme_id,
+            'title': title,
+            'description': description,
+            'thumbnail': self._og_search_thumbnail(webpage, default=None),
+            'duration': duration,
+            'formats': formats,
+            'subtitles': subtitles,
+        }
+
+
+class BBCIE(BBCCoUkIE):
+    IE_NAME = 'bbc'
+    IE_DESC = 'BBC'
+    _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)'
+
+    _MEDIASELECTOR_URLS = [
+        # Provides HQ HLS streams but fails with geolocation in some cases when it's
+        # even not geo restricted at all
+        'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/iptv-all/vpid/%s',
+        # Provides more formats, namely direct mp4 links, but fails on some videos with
+        # notukerror for non UK (?) users (e.g.
+        # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
+        'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s',
+        # Provides fewer formats, but works everywhere for everybody (hopefully)
+        'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s',
+    ]
+
+    _TESTS = [{
+        # article with multiple videos embedded with data-playable containing vpids
+        'url': 'http://www.bbc.com/news/world-europe-32668511',
+        'info_dict': {
+            'id': 'world-europe-32668511',
+            'title': 'Russia stages massive WW2 parade despite Western boycott',
+            'description': 'md5:00ff61976f6081841f759a08bf78cc9c',
+        },
+        'playlist_count': 2,
+    }, {
+        # article with multiple videos embedded with data-playable (more videos)
+        'url': 'http://www.bbc.com/news/business-28299555',
+        'info_dict': {
+            'id': 'business-28299555',
+            'title': 'Farnborough Airshow: Video highlights',
+            'description': 'BBC reports and video highlights at the Farnborough Airshow.',
+        },
+        'playlist_count': 9,
+        'skip': 'Save time',
+    }, {
+        # article with multiple videos embedded with `new SMP()`
+        # broken
+        'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
+        'info_dict': {
+            'id': '3662a707-0af9-3149-963f-47bea720b460',
+            'title': 'BBC Blogs - Adam Curtis - BUGGER',
+        },
+        'playlist_count': 18,
+    }, {
+        # single video embedded with data-playable containing vpid
+        'url': 'http://www.bbc.com/news/world-europe-32041533',
+        'info_dict': {
+            'id': 'p02mprgb',
+            'ext': 'mp4',
+            'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
+            'description': 'md5:2868290467291b37feda7863f7a83f54',
+            'duration': 47,
+            'timestamp': 1427219242,
+            'upload_date': '20150324',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        }
+    }, {
+        # article with single video embedded with data-playable containing XML playlist
+        # with direct video links as progressiveDownloadUrl (for now these are extracted)
+        # and playlist with f4m and m3u8 as streamingUrl
+        'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu',
+        'info_dict': {
+            'id': '150615_telabyad_kentin_cogu',
+            'ext': 'mp4',
+            'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde",
+            'timestamp': 1434397334,
+            'upload_date': '20150615',
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }, {
+        # single video embedded with data-playable containing XML playlists (regional section)
+        'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw',
+        'info_dict': {
+            'id': '150619_video_honduras_militares_hospitales_corrupcion_aw',
+            'ext': 'mp4',
+            'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción',
+            'timestamp': 1434713142,
+            'upload_date': '20150619',
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }, {
+        # single video from video playlist embedded with vxp-playlist-data JSON
+        'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376',
+        'info_dict': {
+            'id': 'p02w6qjc',
+            'ext': 'mp4',
+            'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
+            'duration': 56,
+            'description': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }, {
+        # single video story with digitalData
+        'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret',
+        'info_dict': {
+            'id': 'p02q6gc4',
+            'ext': 'flv',
+            'title': 'Sri Lanka’s spicy secret',
+            'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.',
+            'timestamp': 1437674293,
+            'upload_date': '20150723',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        }
+    }, {
+        # single video story without digitalData
+        'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star',
+        'info_dict': {
+            'id': 'p018zqqg',
+            'ext': 'mp4',
+            'title': 'Hyundai Santa Fe Sport: Rock star',
+            'description': 'md5:b042a26142c4154a6e472933cf20793d',
+            'timestamp': 1415867444,
+            'upload_date': '20141113',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        }
+    }, {
+        # single video with playlist.sxml URL in playlist param
+        'url': 'http://www.bbc.com/sport/0/football/33653409',
+        'info_dict': {
+            'id': 'p02xycnp',
+            'ext': 'mp4',
+            'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?',
+            'description': 'BBC Sport\'s David Ornstein has the latest transfer gossip, including rumours of a Manchester United return for Cristiano Ronaldo.',
+            'duration': 140,
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        }
+    }, {
+        # article with multiple videos embedded with playlist.sxml in playlist param
+        'url': 'http://www.bbc.com/sport/0/football/34475836',
+        'info_dict': {
+            'id': '34475836',
+            'title': 'What Liverpool can expect from Klopp',
+        },
+        'playlist_count': 3,
+    }, {
+        # single video with playlist URL from weather section
+        'url': 'http://www.bbc.com/weather/features/33601775',
+        'only_matching': True,
+    }, {
+        # custom redirection to www.bbc.com
+        'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if BBCCoUkIE.suitable(url) or BBCCoUkArticleIE.suitable(url) 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.
+        # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
+        # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml
+        source_files = media_meta.get('sourceFiles')
+        if source_files:
+            return [{
+                'url': f['url'],
+                'format_id': format_id,
+                'ext': f.get('encoding'),
+                'tbr': float_or_none(f.get('bitrate'), 1000),
+                'filesize': int_or_none(f.get('filesize')),
+            } for format_id, f in source_files.items() if f.get('url')], []
+
+        programme_id = media_meta.get('externalId')
+        if programme_id:
+            return self._download_media_selector(programme_id)
+
+        # Process playlist.sxml as legacy playlist
+        href = media_meta.get('href')
+        if href:
+            playlist = self._download_legacy_playlist_url(href)
+            _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id)
+            return formats, subtitles
+
+        return [], []
+
+    def _extract_from_playlist_sxml(self, url, playlist_id, timestamp):
+        programme_id, title, description, duration, formats, subtitles = \
+            self._process_legacy_playlist_url(url, playlist_id)
+        self._sort_formats(formats)
+        return {
+            'id': programme_id,
+            'title': title,
+            'description': description,
+            'duration': duration,
+            'timestamp': timestamp,
+            'formats': formats,
+            'subtitles': subtitles,
+        }
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        json_ld_info = self._search_json_ld(webpage, playlist_id, default=None)
+        timestamp = json_ld_info.get('timestamp')
+        playlist_title = json_ld_info.get('title')
+        playlist_description = json_ld_info.get('description')
+
+        if not timestamp:
+            timestamp = parse_iso8601(self._search_regex(
+                [r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
+                 r'itemprop="datePublished"[^>]+datetime="([^"]+)"',
+                 r'"datePublished":\s*"([^"]+)'],
+                webpage, 'date', default=None))
+
+        entries = []
+
+        # article with multiple videos embedded with playlist.sxml (e.g.
+        # http://www.bbc.com/sport/0/football/34475836)
+        playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage)
+        playlists.extend(re.findall(r'data-media-id="([^"]+/playlist\.sxml)"', webpage))
+        if playlists:
+            entries = [
+                self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
+                for playlist_url in playlists]
+
+        # news article with multiple videos embedded with data-playable
+        data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage)
+        if data_playables:
+            for _, data_playable_json in data_playables:
+                data_playable = self._parse_json(
+                    unescapeHTML(data_playable_json), playlist_id, fatal=False)
+                if not data_playable:
+                    continue
+                settings = data_playable.get('settings', {})
+                if settings:
+                    # data-playable with video vpid in settings.playlistObject.items (e.g.
+                    # http://www.bbc.com/news/world-us-canada-34473351)
+                    playlist_object = settings.get('playlistObject', {})
+                    if playlist_object:
+                        items = playlist_object.get('items')
+                        if items and isinstance(items, list):
+                            title = playlist_object['title']
+                            description = playlist_object.get('summary')
+                            duration = int_or_none(items[0].get('duration'))
+                            programme_id = items[0].get('vpid')
+                            formats, subtitles = self._download_media_selector(programme_id)
+                            self._sort_formats(formats)
+                            entries.append({
+                                'id': programme_id,
+                                'title': title,
+                                'description': description,
+                                'timestamp': timestamp,
+                                'duration': duration,
+                                'formats': formats,
+                                'subtitles': subtitles,
+                            })
+                    else:
+                        # data-playable without vpid but with a playlist.sxml URLs
+                        # in otherSettings.playlist (e.g.
+                        # 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))
+
+        if entries:
+            playlist_title = playlist_title or remove_end(self._og_search_title(webpage), ' - BBC News')
+            playlist_description = playlist_description or self._og_search_description(webpage, default=None)
+            return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
+
+        # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
+        programme_id = self._search_regex(
+            [r'data-video-player-vpid="(%s)"' % self._ID_REGEX,
+             r'<param[^>]+name="externalIdentifier"[^>]+value="(%s)"' % self._ID_REGEX,
+             r'videoId\s*:\s*["\'](%s)["\']' % self._ID_REGEX],
+            webpage, 'vpid', default=None)
+
+        if programme_id:
+            formats, subtitles = self._download_media_selector(programme_id)
+            self._sort_formats(formats)
+            # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
+            digital_data = self._parse_json(
+                self._search_regex(
+                    r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'),
+                programme_id, fatal=False)
+            page_info = digital_data.get('page', {}).get('pageInfo', {})
+            title = page_info.get('pageName') or self._og_search_title(webpage)
+            description = page_info.get('description') or self._og_search_description(webpage)
+            timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp
+            return {
+                'id': programme_id,
+                'title': title,
+                'description': description,
+                'timestamp': timestamp,
+                'formats': formats,
+                'subtitles': subtitles,
+            }
+
+        playlist_title = self._html_search_regex(
+            r'<title>(.*?)(?:\s*-\s*BBC [^ ]+)?</title>', webpage, 'playlist title')
+        playlist_description = self._og_search_description(webpage, default=None)
+
+        def extract_all(pattern):
+            return list(filter(None, map(
+                lambda s: self._parse_json(s, playlist_id, fatal=False),
+                re.findall(pattern, webpage))))
+
+        # Multiple video article (e.g.
+        # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460)
+        EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+%s(?:\b[^"]+)?' % self._ID_REGEX
+        entries = []
+        for match in extract_all(r'new\s+SMP\(({.+?})\)'):
+            embed_url = match.get('playerSettings', {}).get('externalEmbedUrl')
+            if embed_url and re.match(EMBED_URL, embed_url):
+                entries.append(embed_url)
+        entries.extend(re.findall(
+            r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage))
+        if entries:
+            return self.playlist_result(
+                [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)
+        medias = extract_all(r"data-media-meta='({[^']+})'")
+
+        if not medias:
+            # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international)
+            media_asset = self._search_regex(
+                r'mediaAssetPage\.init\(\s*({.+?}), "/',
+                webpage, 'media asset', default=None)
+            if media_asset:
+                media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False)
+                medias = []
+                for video in media_asset_page.get('videos', {}).values():
+                    medias.extend(video.values())
+
+        if not medias:
+            # Multiple video playlist with single `now playing` entry (e.g.
+            # http://www.bbc.com/news/video_and_audio/must_see/33767813)
+            vxp_playlist = self._parse_json(
+                self._search_regex(
+                    r'<script[^>]+class="vxp-playlist-data"[^>]+type="application/json"[^>]*>([^<]+)</script>',
+                    webpage, 'playlist data'),
+                playlist_id)
+            playlist_medias = []
+            for item in vxp_playlist:
+                media = item.get('media')
+                if not media:
+                    continue
+                playlist_medias.append(media)
+                # Download single video if found media with asset id matching the video id from URL
+                if item.get('advert', {}).get('assetId') == playlist_id:
+                    medias = [media]
+                    break
+            # Fallback to the whole playlist
+            if not medias:
+                medias = playlist_medias
+
+        entries = []
+        for num, media_meta in enumerate(medias, start=1):
+            formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id)
+            if not formats:
+                continue
+            self._sort_formats(formats)
+
+            video_id = media_meta.get('externalId')
+            if not video_id:
+                video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num)
+
+            title = media_meta.get('caption')
+            if not title:
+                title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num)
+
+            duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration'))
+
+            images = []
+            for image in media_meta.get('images', {}).values():
+                images.extend(image.values())
+            if 'image' in media_meta:
+                images.append(media_meta['image'])
+
+            thumbnails = [{
+                'url': image.get('href'),
+                'width': int_or_none(image.get('width')),
+                'height': int_or_none(image.get('height')),
+            } for image in images]
+
+            entries.append({
+                'id': video_id,
+                'title': title,
+                'thumbnails': thumbnails,
+                'duration': duration,
+                'timestamp': timestamp,
+                'formats': formats,
+                'subtitles': subtitles,
+            })
+
+        return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
+
+
+class BBCCoUkArticleIE(InfoExtractor):
+    _VALID_URL = 'http://www.bbc.co.uk/programmes/articles/(?P<id>[a-zA-Z0-9]+)'
+    IE_NAME = 'bbc.co.uk:article'
+    IE_DESC = 'BBC articles'
+
+    _TEST = {
+        'url': 'http://www.bbc.co.uk/programmes/articles/3jNQLTMrPlYGTBn0WV6M2MS/not-your-typical-role-model-ada-lovelace-the-19th-century-programmer',
+        'info_dict': {
+            'id': '3jNQLTMrPlYGTBn0WV6M2MS',
+            'title': 'Calculating Ada: The Countess of Computing - Not your typical role model: Ada Lovelace the 19th century programmer - BBC Four',
+            'description': 'Hannah Fry reveals some of her surprising discoveries about Ada Lovelace during filming.',
+        },
+        'playlist_count': 4,
+        'add_ie': ['BBCCoUk'],
+    }
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage).strip()
+
+        entries = [self.url_result(programme_url) for programme_url in re.findall(
+            r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)]
+
+        return self.playlist_result(entries, playlist_id, title, description)

youtube_dl/extractor/bbccouk.py

@@ -1,262 +0,0 @@
-from __future__ import unicode_literals
-
-import xml.etree.ElementTree
-
-from .subtitles import SubtitlesInfoExtractor
-from ..utils import ExtractorError
-from ..compat import compat_HTTPError
-
-
-class BBCCoUkIE(SubtitlesInfoExtractor):
-    IE_NAME = 'bbc.co.uk'
-    IE_DESC = 'BBC iPlayer'
-    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:programmes|iplayer/episode)/(?P<id>[\da-z]{8})'
-
-    _TESTS = [
-        {
-            'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
-            'info_dict': {
-                'id': 'b039d07m',
-                'ext': 'flv',
-                'title': 'Kaleidoscope: Leonard Cohen',
-                'description': 'md5:db4755d7a665ae72343779f7dacb402c',
-                'duration': 1740,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            }
-        },
-        {
-            'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
-            'info_dict': {
-                'id': 'b00yng1d',
-                'ext': 'flv',
-                'title': 'The Man in Black: Series 3: The Printed Name',
-                'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
-                'duration': 1800,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            },
-            'skip': 'Episode is no longer available on BBC iPlayer Radio',
-        },
-        {
-            'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
-            'info_dict': {
-                'id': 'b00yng1d',
-                'ext': 'flv',
-                'title': 'The Voice UK: Series 3: Blind Auditions 5',
-                'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.",
-                'duration': 5100,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            },
-            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
-        },
-        {
-            'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
-            'info_dict': {
-                'id': 'b03k3pb7',
-                'ext': 'flv',
-                'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction",
-                'description': '2. Invasion',
-                'duration': 3600,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            },
-            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
-        },
-    ]
-
-    def _extract_asx_playlist(self, connection, programme_id):
-        asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
-        return [ref.get('href') for ref in asx.findall('./Entry/ref')]
-
-    def _extract_connection(self, connection, programme_id):
-        formats = []
-        protocol = connection.get('protocol')
-        supplier = connection.get('supplier')
-        if protocol == 'http':
-            href = connection.get('href')
-            # ASX playlist
-            if supplier == 'asx':
-                for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
-                    formats.append({
-                        'url': ref,
-                        'format_id': 'ref%s_%s' % (i, supplier),
-                    })
-            # Direct link
-            else:
-                formats.append({
-                    'url': href,
-                    'format_id': supplier,
-                })
-        elif protocol == 'rtmp':
-            application = connection.get('application', 'ondemand')
-            auth_string = connection.get('authString')
-            identifier = connection.get('identifier')
-            server = connection.get('server')
-            formats.append({
-                'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
-                'play_path': identifier,
-                'app': '%s?%s' % (application, auth_string),
-                'page_url': 'http://www.bbc.co.uk',
-                'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
-                'rtmp_live': False,
-                'ext': 'flv',
-                'format_id': supplier,
-            })
-        return formats
-
-    def _extract_items(self, playlist):
-        return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
-
-    def _extract_medias(self, media_selection):
-        error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error')
-        if error is not None:
-            raise ExtractorError(
-                '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True)
-        return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
-
-    def _extract_connections(self, media):
-        return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
-
-    def _extract_video(self, media, programme_id):
-        formats = []
-        vbr = int(media.get('bitrate'))
-        vcodec = media.get('encoding')
-        service = media.get('service')
-        width = int(media.get('width'))
-        height = int(media.get('height'))
-        file_size = int(media.get('media_file_size'))
-        for connection in self._extract_connections(media):
-            conn_formats = self._extract_connection(connection, programme_id)
-            for format in conn_formats:
-                format.update({
-                    'format_id': '%s_%s' % (service, format['format_id']),
-                    'width': width,
-                    'height': height,
-                    'vbr': vbr,
-                    'vcodec': vcodec,
-                    'filesize': file_size,
-                })
-            formats.extend(conn_formats)
-        return formats
-
-    def _extract_audio(self, media, programme_id):
-        formats = []
-        abr = int(media.get('bitrate'))
-        acodec = media.get('encoding')
-        service = media.get('service')
-        for connection in self._extract_connections(media):
-            conn_formats = self._extract_connection(connection, programme_id)
-            for format in conn_formats:
-                format.update({
-                    'format_id': '%s_%s' % (service, format['format_id']),
-                    'abr': abr,
-                    'acodec': acodec,
-                })
-            formats.extend(conn_formats)
-        return formats
-
-    def _extract_captions(self, media, programme_id):
-        subtitles = {}
-        for connection in self._extract_connections(media):
-            captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
-            lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
-            ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
-            srt = ''
-            for pos, p in enumerate(ps):
-                srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'),
-                                                          p.text.strip() if p.text is not None else '')
-            subtitles[lang] = srt
-        return subtitles
-
-    def _download_media_selector(self, programme_id):
-        try:
-            media_selection = self._download_xml(
-                'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s' % programme_id,
-                programme_id, 'Downloading media selection XML')
-        except ExtractorError as ee:
-            if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
-                media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().encode('utf-8'))
-            else:
-                raise
-
-        formats = []
-        subtitles = None
-
-        for media in self._extract_medias(media_selection):
-            kind = media.get('kind')
-            if kind == 'audio':
-                formats.extend(self._extract_audio(media, programme_id))
-            elif kind == 'video':
-                formats.extend(self._extract_video(media, programme_id))
-            elif kind == 'captions':
-                subtitles = self._extract_captions(media, programme_id)
-
-        return formats, subtitles
-
-    def _real_extract(self, url):
-        group_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, group_id, 'Downloading video page')
-
-        programme_id = self._search_regex(
-            r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False)
-        if programme_id:
-            player = self._download_json(
-                'http://www.bbc.co.uk/iplayer/episode/%s.json' % group_id,
-                group_id)['jsConf']['player']
-            title = player['title']
-            description = player['subtitle']
-            duration = player['duration']
-            formats, subtitles = self._download_media_selector(programme_id)
-        else:
-            playlist = self._download_xml(
-                'http://www.bbc.co.uk/iplayer/playlist/%s' % group_id,
-                group_id, 'Downloading playlist XML')
-
-            no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
-            if no_items is not None:
-                reason = no_items.get('reason')
-                if reason == 'preAvailability':
-                    msg = 'Episode %s is not yet available' % group_id
-                elif reason == 'postAvailability':
-                    msg = 'Episode %s is no longer available' % group_id
-                elif reason == 'noMedia':
-                    msg = 'Episode %s is not currently available' % group_id
-                else:
-                    msg = 'Episode %s is not available: %s' % (group_id, reason)
-                raise ExtractorError(msg, expected=True)
-
-            for item in self._extract_items(playlist):
-                kind = item.get('kind')
-                if kind != 'programme' and kind != 'radioProgramme':
-                    continue
-                title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
-                description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
-                programme_id = item.get('identifier')
-                duration = int(item.get('duration'))
-                formats, subtitles = self._download_media_selector(programme_id)
-
-        if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(programme_id, subtitles)
-            return
-
-        self._sort_formats(formats)
-
-        return {
-            'id': programme_id,
-            'title': title,
-            'description': description,
-            'duration': duration,
-            'formats': formats,
-            'subtitles': subtitles,
-        }

youtube_dl/extractor/beatportpro.py

@@ -0,0 +1,103 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import int_or_none
+
+
+class BeatportProIE(InfoExtractor):
+    _VALID_URL = r'https?://pro\.beatport\.com/track/(?P<display_id>[^/]+)/(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'https://pro.beatport.com/track/synesthesia-original-mix/5379371',
+        'md5': 'b3c34d8639a2f6a7f734382358478887',
+        'info_dict': {
+            'id': '5379371',
+            'display_id': 'synesthesia-original-mix',
+            'ext': 'mp4',
+            'title': 'Froxic - Synesthesia (Original Mix)',
+        },
+    }, {
+        'url': 'https://pro.beatport.com/track/love-and-war-original-mix/3756896',
+        'md5': 'e44c3025dfa38c6577fbaeb43da43514',
+        'info_dict': {
+            'id': '3756896',
+            'display_id': 'love-and-war-original-mix',
+            'ext': 'mp3',
+            'title': 'Wolfgang Gartner - Love & War (Original Mix)',
+        },
+    }, {
+        'url': 'https://pro.beatport.com/track/birds-original-mix/4991738',
+        'md5': 'a1fd8e8046de3950fd039304c186c05f',
+        'info_dict': {
+            'id': '4991738',
+            'display_id': 'birds-original-mix',
+            'ext': 'mp4',
+            'title': "Tos, Middle Milk, Mumblin' Johnsson - Birds (Original Mix)",
+        }
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        track_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        playables = self._parse_json(
+            self._search_regex(
+                r'window\.Playables\s*=\s*({.+?});', webpage,
+                'playables info', flags=re.DOTALL),
+            track_id)
+
+        track = next(t for t in playables['tracks'] if t['id'] == int(track_id))
+
+        title = ', '.join((a['name'] for a in track['artists'])) + ' - ' + track['name']
+        if track['mix']:
+            title += ' (' + track['mix'] + ')'
+
+        formats = []
+        for ext, info in track['preview'].items():
+            if not info['url']:
+                continue
+            fmt = {
+                'url': info['url'],
+                'ext': ext,
+                'format_id': ext,
+                'vcodec': 'none',
+            }
+            if ext == 'mp3':
+                fmt['preference'] = 0
+                fmt['acodec'] = 'mp3'
+                fmt['abr'] = 96
+                fmt['asr'] = 44100
+            elif ext == 'mp4':
+                fmt['preference'] = 1
+                fmt['acodec'] = 'aac'
+                fmt['abr'] = 96
+                fmt['asr'] = 44100
+            formats.append(fmt)
+        self._sort_formats(formats)
+
+        images = []
+        for name, info in track['images'].items():
+            image_url = info.get('url')
+            if name == 'dynamic' or not image_url:
+                continue
+            image = {
+                'id': name,
+                'url': image_url,
+                'height': int_or_none(info.get('height')),
+                'width': int_or_none(info.get('width')),
+            }
+            images.append(image)
+
+        return {
+            'id': compat_str(track.get('id')) or track_id,
+            'display_id': track.get('slug') or display_id,
+            'title': title,
+            'formats': formats,
+            'thumbnails': images,
+        }

youtube_dl/extractor/beeg.py

@@ -1,65 +1,105 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..compat import (
+    compat_chr,
+    compat_ord,
+    compat_urllib_parse_unquote,
+)
+from ..utils import (
+    int_or_none,
+    parse_iso8601,
+)
 
 
 class BeegIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?beeg\.com/(?P<id>\d+)'
     _TEST = {
         'url': 'http://beeg.com/5416503',
-        'md5': '634526ae978711f6b748fe0dd6c11f57',
+        'md5': '46c384def73b33dbc581262e5ee67cef',
         'info_dict': {
             'id': '5416503',
             'ext': 'mp4',
             'title': 'Sultry Striptease',
-            'description': 'md5:6db3c6177972822aaba18652ff59c773',
-            'categories': list,  # NSFW
-            'thumbnail': 're:https?://.*\.jpg$',
+            'description': 'md5:d22219c09da287c14bed3d6c37ce4bc2',
+            'timestamp': 1391813355,
+            'upload_date': '20140207',
+            'duration': 383,
+            'tags': list,
             'age_limit': 18,
         }
     }
 
     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)
+        video = self._download_json(
+            'https://api.beeg.com/api/v5/video/%s' % video_id, video_id)
 
-        quality_arr = self._search_regex(
-            r'(?s)var\s+qualityArr\s*=\s*{\s*(.+?)\s*}', webpage, 'quality formats')
+        def split(o, e):
+            def cut(s, x):
+                n.append(s[:x])
+                return s[x:]
+            n = []
+            r = len(o) % e
+            if r > 0:
+                o = cut(o, r)
+            while len(o) > e:
+                o = cut(o, e)
+            n.append(o)
+            return n
 
-        formats = [{
-            'url': fmt[1],
-            'format_id': fmt[0],
-            'height': int(fmt[0][:-1]),
-        } for fmt in re.findall(r"'([^']+)'\s*:\s*'([^']+)'", quality_arr)]
+        def decrypt_key(key):
+            # Reverse engineered from http://static.beeg.com/cpl/1105.js
+            a = '5ShMcIQlssOd7zChAIOlmeTZDaUxULbJRnywYaiB'
+            e = compat_urllib_parse_unquote(key)
+            o = ''.join([
+                compat_chr(compat_ord(e[n]) - compat_ord(a[n % len(a)]) % 21)
+                for n in range(len(e))])
+            return ''.join(split(o, 3)[::-1])
 
+        def decrypt_url(encrypted_url):
+            encrypted_url = self._proto_relative_url(
+                encrypted_url.replace('{DATA_MARKERS}', ''), 'https:')
+            key = self._search_regex(
+                r'/key=(.*?)%2Cend=', encrypted_url, 'key', default=None)
+            if not key:
+                return encrypted_url
+            return encrypted_url.replace(key, decrypt_key(key))
+
+        formats = []
+        for format_id, video_url in video.items():
+            if not video_url:
+                continue
+            height = self._search_regex(
+                r'^(\d+)[pP]$', format_id, 'height', default=None)
+            if not height:
+                continue
+            formats.append({
+                'url': decrypt_url(video_url),
+                'format_id': format_id,
+                'height': int(height),
+            })
         self._sort_formats(formats)
 
-        title = self._html_search_regex(
-            r'<title>([^<]+)\s*-\s*beeg\.?</title>', webpage, 'title')
+        title = video['title']
+        video_id = video.get('id') or video_id
+        display_id = video.get('code')
+        description = video.get('desc')
 
-        description = self._html_search_regex(
-            r'<meta name="description" content="([^"]*)"',
-            webpage, 'description', fatal=False)
-        thumbnail = self._html_search_regex(
-            r'\'previewer.url\'\s*:\s*"([^"]*)"',
-            webpage, 'thumbnail', fatal=False)
+        timestamp = parse_iso8601(video.get('date'), ' ')
+        duration = int_or_none(video.get('duration'))
 
-        categories_str = self._html_search_regex(
-            r'<meta name="keywords" content="([^"]+)"', webpage, 'categories', fatal=False)
-        categories = (
-            None if categories_str is None
-            else categories_str.split(','))
+        tags = [tag.strip() for tag in video['tags'].split(',')] if video.get('tags') else None
 
         return {
             'id': video_id,
+            'display_id': display_id,
             'title': title,
             'description': description,
-            'thumbnail': thumbnail,
-            'categories': categories,
+            'timestamp': timestamp,
+            'duration': duration,
+            'tags': tags,
             'formats': formats,
             'age_limit': 18,
         }

youtube_dl/extractor/behindkink.py

@@ -10,15 +10,15 @@ from ..utils import url_basename
 class BehindKinkIE(InfoExtractor):
     _VALID_URL = r'http://(?:www\.)?behindkink\.com/(?P<year>[0-9]{4})/(?P<month>[0-9]{2})/(?P<day>[0-9]{2})/(?P<id>[^/#?_]+)'
     _TEST = {
-        'url': 'http://www.behindkink.com/2014/08/14/ab1576-performers-voice-finally-heard-the-bill-is-killed/',
-        'md5': '41ad01222b8442089a55528fec43ec01',
+        'url': 'http://www.behindkink.com/2014/12/05/what-are-you-passionate-about-marley-blaze/',
+        'md5': '507b57d8fdcd75a41a9a7bdb7989c762',
         'info_dict': {
-            'id': '36370',
+            'id': '37127',
             'ext': 'mp4',
-            'title': 'AB1576 - PERFORMERS VOICE FINALLY HEARD - THE BILL IS KILLED!',
-            'description': 'The adult industry voice was finally heard as Assembly Bill 1576 remained\xa0 in suspense today at the Senate Appropriations Hearing. AB1576 was, among other industry damaging issues, a condom mandate...',
-            'upload_date': '20140814',
-            'thumbnail': 'http://www.behindkink.com/wp-content/uploads/2014/08/36370_AB1576_Win.jpg',
+            'title': 'What are you passionate about – Marley Blaze',
+            'description': 'md5:aee8e9611b4ff70186f752975d9b94b4',
+            'upload_date': '20141205',
+            'thumbnail': 'http://www.behindkink.com/wp-content/uploads/2014/12/blaze-1.jpg',
             'age_limit': 18,
         }
     }
@@ -26,26 +26,19 @@ class BehindKinkIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         display_id = mobj.group('id')
-        year = mobj.group('year')
-        month = mobj.group('month')
-        day = mobj.group('day')
-        upload_date = year + month + day
 
         webpage = self._download_webpage(url, display_id)
 
         video_url = self._search_regex(
-            r"'file':\s*'([^']+)'",
-            webpage, 'URL base')
-
-        video_id = url_basename(video_url)
-        video_id = video_id.split('_')[0]
+            r'<source src="([^"]+)"', webpage, 'video URL')
+        video_id = url_basename(video_url).split('_')[0]
+        upload_date = mobj.group('year') + mobj.group('month') + mobj.group('day')
 
         return {
             'id': video_id,
-            'url': video_url,
-            'ext': 'mp4',
-            'title': self._og_search_title(webpage),
             'display_id': display_id,
+            'url': video_url,
+            'title': self._og_search_title(webpage),
             'thumbnail': self._og_search_thumbnail(webpage),
             'description': self._og_search_description(webpage),
             'upload_date': upload_date,

youtube_dl/extractor/bet.py

@@ -1,8 +1,8 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_urllib_parse_unquote
 from ..utils import (
-    compat_urllib_parse,
     xpath_text,
     xpath_with_ns,
     int_or_none,
@@ -16,11 +16,11 @@ class BetIE(InfoExtractor):
         {
             'url': 'http://www.bet.com/news/politics/2014/12/08/in-bet-exclusive-obama-talks-race-and-racism.html',
             'info_dict': {
-                'id': '417cd61c-c793-4e8e-b006-e445ecc45add',
+                'id': 'news/national/2014/a-conversation-with-president-obama',
                 'display_id': 'in-bet-exclusive-obama-talks-race-and-racism',
                 'ext': 'flv',
-                'title': 'BET News Presents: A Conversation With President Obama',
-                'description': 'md5:5a88d8ae912c1b33e090290af7ec33c6',
+                'title': 'A Conversation With President Obama',
+                'description': 'md5:699d0652a350cf3e491cd15cc745b5da',
                 'duration': 1534,
                 'timestamp': 1418075340,
                 'upload_date': '20141208',
@@ -35,7 +35,7 @@ class BetIE(InfoExtractor):
         {
             'url': 'http://www.bet.com/video/news/national/2014/justice-for-ferguson-a-community-reacts.html',
             'info_dict': {
-                'id': '4160e53b-ad41-43b1-980f-8d85f63121f4',
+                'id': 'news/national/2014/justice-for-ferguson-a-community-reacts',
                 'display_id': 'justice-for-ferguson-a-community-reacts',
                 'ext': 'flv',
                 'title': 'Justice for Ferguson: A Community Reacts',
@@ -55,13 +55,15 @@ class BetIE(InfoExtractor):
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
-
         webpage = self._download_webpage(url, display_id)
 
-        media_url = compat_urllib_parse.unquote(self._search_regex(
+        media_url = compat_urllib_parse_unquote(self._search_regex(
             [r'mediaURL\s*:\s*"([^"]+)"', r"var\s+mrssMediaUrl\s*=\s*'([^']+)'"],
             webpage, 'media URL'))
 
+        video_id = self._search_regex(
+            r'/video/(.*)/_jcr_content/', media_url, 'video id')
+
         mrss = self._download_xml(media_url, display_id)
 
         item = mrss.find('./channel/item')
@@ -76,8 +78,6 @@ class BetIE(InfoExtractor):
         description = xpath_text(
             item, './description', 'description', fatal=False)
 
-        video_id = xpath_text(item, './guid', 'video id', fatal=False)
-
         timestamp = parse_iso8601(xpath_text(
             item, xpath_with_ns('./dc:date', NS_MAP),
             'upload date', fatal=False))

youtube_dl/extractor/bigflix.py

@@ -0,0 +1,85 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import base64
+import re
+
+from .common import InfoExtractor
+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',
+            'formats': 'mincount:2',
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }, {
+        # multiple formats
+        'url': 'http://www.bigflix.com/Malayalam-movies/Drama-movies/Indian-Rupee/15967',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._html_search_regex(
+            r'<div[^>]+class=["\']pagetitle["\'][^>]*>(.+?)</div>',
+            webpage, 'title')
+
+        def decode_url(quoted_b64_url):
+            return base64.b64decode(compat_urllib_parse_unquote(
+                quoted_b64_url).encode('ascii')).decode('utf-8')
+
+        formats = []
+        for height, encoded_url in re.findall(
+                r'ContentURL_(\d{3,4})[pP][^=]+=([^&]+)', webpage):
+            video_url = decode_url(encoded_url)
+            f = {
+                'url': video_url,
+                'format_id': '%sp' % height,
+                'height': int(height),
+            }
+            if video_url.startswith('rtmp'):
+                f['ext'] = 'flv'
+            formats.append(f)
+
+        file_url = self._search_regex(
+            r'file=([^&]+)', webpage, 'video url', default=None)
+        if file_url:
+            video_url = decode_url(file_url)
+            if all(f['url'] != video_url for f in formats):
+                formats.append({
+                    'url': decode_url(file_url),
+                })
+
+        self._sort_formats(formats)
+
+        description = self._html_search_meta('description', webpage)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'formats': formats
+        }