release 2015.01.25

For example for http://www.dr.dk/tv/se/moderne-klassikere/moderne-klassikere-one-republic-apologize#!/, there's a version where everytime someone speaks in English a computer voice translates it.

.travis.yml

@@ -4,6 +4,9 @@ python:
   - "2.7"
   - "3.3"
   - "3.4"
+before_install:
+  - sudo apt-get update -qq
+  - sudo apt-get install -yqq rtmpdump
 script: nosetests test --verbose
 notifications:
   email:

AUTHORS

@@ -102,3 +102,7 @@ Joel Leclerc
 Christopher Krooss
 Ondřej Caletka
 Dinesh S
+Johan K. Jensen
+Yen Chi Hsuan
+Enam Mijbah Noor
+David Luhmer

README.md

@@ -93,6 +93,14 @@ which means you can modify it, redistribute it or use it however you like.
 ## Video Selection:
     --playlist-start NUMBER          playlist video to start at (default is 1)
     --playlist-end NUMBER            playlist video to end at (default is last)
+    --playlist-items ITEM_SPEC       playlist video items to download. Specify
+                                     indices of the videos in the playlist
+                                     seperated by commas like: "--playlist-items
+                                     1,2,5,8" if you want to download videos
+                                     indexed 1, 2, 5, 8 in the playlist. You can
+                                     specify range: "--playlist-items
+                                     1-3,7,10-13", it will download the videos
+                                     at index 1, 2, 3, 7, 10, 11, 12 and 13.
     --match-title REGEX              download only matching titles (regex or
                                      caseless sub-string)
     --reject-title REGEX             skip download for matching titles (regex or
@@ -124,7 +132,8 @@ which means you can modify it, redistribute it or use it however you like.
 ## Download Options:
     -r, --rate-limit LIMIT           maximum download rate in bytes per second
                                      (e.g. 50K or 4.2M)
-    -R, --retries RETRIES            number of retries (default is 10)
+    -R, --retries RETRIES            number of retries (default is 10), or
+                                     "infinite".
     --buffer-size SIZE               size of download buffer (e.g. 1024 or 16K)
                                      (default is 1024)
     --no-resize-buffer               do not automatically adjust the buffer
@@ -132,6 +141,11 @@ which means you can modify it, redistribute it or use it however you like.
                                      automatically resized from an initial value
                                      of SIZE.
     --playlist-reverse               Download playlist videos in reverse order
+    --xattr-set-filesize             (experimental) set file xattribute
+                                     ytdl.filesize with expected filesize
+    --external-downloader COMMAND    (experimental) Use the specified external
+                                     downloader. Currently supports
+                                     aria2c,curl,wget
 
 ## Filesystem Options:
     -a, --batch-file FILE            file containing URLs to download ('-' for
@@ -191,7 +205,6 @@ which means you can modify it, redistribute it or use it however you like.
     --write-info-json                write video metadata to a .info.json file
     --write-annotations              write video annotations to a .annotation
                                      file
-    --write-thumbnail                write thumbnail image to disk
     --load-info FILE                 json file containing the video information
                                      (created with the "--write-json" option)
     --cookies FILE                   file to read cookies from and dump cookie
@@ -206,6 +219,12 @@ which means you can modify it, redistribute it or use it however you like.
     --no-cache-dir                   Disable filesystem caching
     --rm-cache-dir                   Delete all filesystem cache files
 
+## Thumbnail images:
+    --write-thumbnail                write thumbnail image to disk
+    --write-all-thumbnails           write all thumbnail image formats to disk
+    --list-thumbnails                Simulate and list all available thumbnail
+                                     formats
+
 ## Verbosity / Simulation Options:
     -q, --quiet                      activates quiet mode
     --no-warnings                    Ignore warnings
@@ -259,6 +278,8 @@ which means you can modify it, redistribute it or use it however you like.
     --bidi-workaround                Work around terminals that lack
                                      bidirectional text support. Requires bidiv
                                      or fribidi executable in PATH
+    --sleep-interval SECONDS         Number of seconds to sleep before each
+                                     download.
 
 ## Video Format Options:
     -f, --format FORMAT              video format code, specify the order of
@@ -267,10 +288,22 @@ which means you can modify it, redistribute it or use it however you like.
                                      by extension for the extensions aac, m4a,
                                      mp3, mp4, ogg, wav, webm. You can also use
                                      the special names "best", "bestvideo",
-                                     "bestaudio", "worst".  By default, youtube-
-                                     dl will pick the best quality. Use commas
-                                     to download multiple audio formats, such as
-                                     -f
+                                     "bestaudio", "worst".  You can filter the
+                                     video results by putting a condition in
+                                     brackets, as in -f "best[height=720]" (or
+                                     -f "[filesize>10M]").  This works for
+                                     filesize, height, width, tbr, abr, and vbr
+                                     and the comparisons <, <=, >, >=, =, != .
+                                     Formats for which the value is not known
+                                     are excluded unless you put a question mark
+                                     (?) after the operator. You can combine
+                                     format filters, so  -f "[height <=?
+                                     720][tbr>500]" selects up to 720p videos
+                                     (or videos where the height is not known)
+                                     with a bitrate of at least 500 KBit/s. By
+                                     default, youtube-dl will pick the best
+                                     quality. Use commas to download multiple
+                                     audio formats, such as -f
                                      136/137/mp4/bestvideo,140/m4a/bestaudio.
                                      You can merge the video and audio of two
                                      formats into a single file using -f <video-
@@ -304,7 +337,8 @@ which means you can modify it, redistribute it or use it however you like.
 
 ## Authentication Options:
     -u, --username USERNAME          login with this account ID
-    -p, --password PASSWORD          account password
+    -p, --password PASSWORD          account password. If this option is left
+                                     out, youtube-dl will ask interactively.
     -2, --twofactor TWOFACTOR        two-factor auth code
     -n, --netrc                      use .netrc authentication data
     --video-password PASSWORD        video password (vimeo, smotri)
@@ -444,9 +478,15 @@ Apparently YouTube requires you to pass a CAPTCHA test if you download too much.
 
 Once the video is fully downloaded, use any video player, such as [vlc](http://www.videolan.org) or [mplayer](http://www.mplayerhq.hu/).
 
-### The links provided by youtube-dl -g are not working anymore
+### I extracted a video URL with -g, but it does not play on another machine / in my webbrowser.
 
-The URLs youtube-dl outputs require the downloader to have the correct cookies. Use the `--cookies` option to write the required cookies into a file, and advise your downloader to read cookies from that file. Some sites also require a common user agent to be used, use `--dump-user-agent` to see the one in use by youtube-dl.
+It depends a lot on the service. In many cases, requests for the video (to download/play it) must come from the same IP address and with the same cookies.  Use the `--cookies` option to write the required cookies into a file, and advise your downloader to read cookies from that file. Some sites also require a common user agent to be used, use `--dump-user-agent` to see the one in use by youtube-dl.
+
+It may be beneficial to use IPv6; in some cases, the restrictions are only applied to IPv4. Some services (sometimes only for a subset of videos) do not restrict the video URL by IP address, cookie, or user-agent, but these are the exception rather than the rule.
+
+Please bear in mind that some URL protocols are **not** supported by browsers out of the box, including RTMP. If you are using -g, your own downloader must support these as well.
+
+If you want to play the video on a machine that is not running youtube-dl, you can relay the video content from the machine that runs youtube-dl. You can use `-o -` to let youtube-dl stream a video to stdout, or simply allow the player to download the files written by youtube-dl in turn.
 
 ### ERROR: no fmt_url_map or conn information found in video info
 
@@ -481,6 +521,10 @@ To make a different directory work - either for ffmpeg, or for youtube-dl, or fo
 
 From then on, after restarting your shell, you will be able to access both youtube-dl and ffmpeg (and youtube-dl will be able to find ffmpeg) by simply typing `youtube-dl` or `ffmpeg`, no matter what directory you're in.
 
+### How do I put downloads into a specific folder?
+
+Use the `-o` to specify an [output template](#output-template), for example `-o "/home/user/videos/%(title)s-%(id)s.%(ext)s"`. If you want this for all of your downloads, put the option into your [configuration file](#configuration).
+
 ### How can I detect whether a given URL is supported by youtube-dl?
 
 For one, have a look at the [list of supported sites](docs/supportedsites). Note that it can sometimes happen that the site changes its URL scheme (say, from http://example.com/v/1234567 to http://example.com/v/1234567 ) and youtube-dl reports an URL of a service in that list as unsupported. In that case, simply report a bug.
@@ -561,7 +605,7 @@ If you want to add support for a new site, you can follow this quick list (assum
 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 be then be named `TestDownload.test_YourExtractor`, `TestDownload.test_YourExtractor_1`, `TestDownload.test_YourExtractor_2`, etc.
 7. Have a look at [`youtube_dl/common/extractor/common.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py) for possible helper methods and a [detailed description of what your extractor should return](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L38). Add tests and code for as many as you want.
-8. If you can, check the code with [pyflakes](https://pypi.python.org/pypi/pyflakes) (a good idea) and [pep8](https://pypi.python.org/pypi/pep8) (optional, ignore E501).
+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

setup.cfg

@@ -2,5 +2,5 @@
 universal = True
 
 [flake8]
-exclude = youtube_dl/extractor/__init__.py,devscripts/buildserver.py,setup.py,build
+exclude = youtube_dl/extractor/__init__.py,devscripts/buildserver.py,setup.py,build,.git
 ignore = E501

test/helper.py

@@ -140,7 +140,7 @@ def expect_info_dict(self, got_dict, expected_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'))
     missing_keys = set(test_info_dict.keys()) - set(expected_dict.keys())
     if missing_keys:
         def _repr(v):

test/test_YoutubeDL.py

@@ -281,6 +281,61 @@ class TestFormatSelection(unittest.TestCase):
             downloaded = ydl.downloaded_info_dicts[0]
             self.assertEqual(downloaded['format_id'], f1id)
 
+    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')
+
     def test_add_extra_info(self):
         test_dict = {
             'extractor': 'Foo',

test/test_all_urls.py

@@ -14,7 +14,6 @@ from test.helper import gettestcases
 from youtube_dl.extractor import (
     FacebookIE,
     gen_extractors,
-    TwitchIE,
     YoutubeIE,
 )
 
@@ -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')
@@ -115,8 +102,6 @@ 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'])

test/test_utils.py

@@ -28,6 +28,7 @@ from youtube_dl.utils import (
     fix_xml_ampersands,
     InAdvancePagedList,
     intlist_to_bytes,
+    is_html,
     js_to_json,
     limit_length,
     OnDemandPagedList,
@@ -51,6 +52,7 @@ from youtube_dl.utils import (
     urlencode_postdata,
     version_tuple,
     xpath_with_ns,
+    render_table,
 )
 
 
@@ -148,6 +150,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(unified_strdate('8/7/2009'), '20090708')
         self.assertEqual(unified_strdate('Dec 14, 2012'), '20121214')
         self.assertEqual(unified_strdate('2012/10/11 01:56:38 +0000'), '20121011')
+        self.assertEqual(unified_strdate('1968 12 10'), '19681210')
         self.assertEqual(unified_strdate('1968-12-10'), '19681210')
         self.assertEqual(unified_strdate('28/01/2014 21:00:00 +0100'), '20140128')
         self.assertEqual(
@@ -212,6 +215,8 @@ 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)
@@ -414,5 +419,31 @@ ffmpeg version 2.4.4 Copyright (c) 2000-2014 the FFmpeg ...'''), '2.4.4')
         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')
+
+
 if __name__ == '__main__':
     unittest.main()

youtube_dl/YoutubeDL.py

@@ -10,6 +10,7 @@ import io
 import itertools
 import json
 import locale
+import operator
 import os
 import platform
 import re
@@ -49,11 +50,14 @@ from .utils import (
     make_HTTPS_handler,
     MaxDownloadsReached,
     PagedList,
+    parse_filesize,
     PostProcessingError,
     platform_name,
     preferredencoding,
+    render_table,
     SameFileError,
     sanitize_filename,
+    std_headers,
     subtitles_filename,
     takewhile_inclusive,
     UnavailableVideoError,
@@ -71,6 +75,7 @@ from .extractor import get_info_extractor, gen_extractors
 from .downloader import get_suitable_downloader
 from .downloader.rtmp import rtmpdump_version
 from .postprocessor import (
+    FFmpegFixupM4aPP,
     FFmpegFixupStretchedPP,
     FFmpegMergerPP,
     FFmpegPostProcessor,
@@ -132,6 +137,7 @@ class YoutubeDL(object):
     nooverwrites:      Prevent overwriting files.
     playliststart:     Playlist item to start at.
     playlistend:       Playlist item to end at.
+    playlist_items:    Specific indices of playlist to download.
     playlistreverse:   Download playlist items in reverse order.
     matchtitle:        Download only matching titles.
     rejecttitle:       Reject downloads for matching titles.
@@ -141,6 +147,7 @@ class YoutubeDL(object):
     writeinfojson:     Write the video description to a .info.json file
     writeannotations:  Write the video annotations to a .annotations.xml file
     writethumbnail:    Write the thumbnail image to a file
+    write_all_thumbnails:  Write all thumbnail formats to files
     writesubtitles:    Write the video subtitles to a file
     writeautomaticsub: Write the automatic subtitles to a file
     allsubtitles:      Downloads all the subtitles of the video
@@ -191,11 +198,12 @@ class YoutubeDL(object):
                        postprocessor.
     progress_hooks:    A list of functions that get called on download
                        progress, with a dictionary with the entries
-                       * filename: The final filename
-                       * status: One of "downloading" and "finished"
-
-                       The dict may also have some of the following entries:
+                       * status: One of "downloading" and "finished".
+                                 Check this first and ignore unknown values.
 
+                       If status is one of "downloading" or "finished", the
+                       following properties may also be present:
+                       * filename: The final filename (always present)
                        * downloaded_bytes: Bytes on disk
                        * total_bytes: Size of the whole file, None if unknown
                        * tmpfilename: The filename we're currently writing to
@@ -211,16 +219,21 @@ class YoutubeDL(object):
                        - "never": do nothing
                        - "warn": only emit a warning
                        - "detect_or_warn": check whether we can do anything
-                                           about it, warn otherwise
+                                           about it, warn otherwise (default)
     source_address:    (Experimental) Client-side IP address to bind to.
     call_home:         Boolean, true iff we are allowed to contact the
                        youtube-dl servers for debugging.
+    sleep_interval:    Number of seconds to sleep before each download.
+    external_downloader:  Executable of the external downloader to call.
+    listformats:       Print an overview of available video formats and exit.
+    list_thumbnails:   Print a table of all thumbnails and exit.
 
 
     The following parameters are not used by YoutubeDL itself, they are used by
     the FileDownloader:
     nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
-    noresizebuffer, retries, continuedl, noprogress, consoletitle
+    noresizebuffer, retries, continuedl, noprogress, consoletitle,
+    xattr_set_filesize.
 
     The following options are used by the post processors:
     prefer_ffmpeg:     If True, use ffmpeg instead of avconv if both are available,
@@ -693,24 +706,51 @@ class YoutubeDL(object):
             if playlistend == -1:
                 playlistend = None
 
+            playlistitems_str = self.params.get('playlist_items', None)
+            playlistitems = None
+            if playlistitems_str is not None:
+                def iter_playlistitems(format):
+                    for string_segment in format.split(','):
+                        if '-' in string_segment:
+                            start, end = string_segment.split('-')
+                            for item in range(int(start), int(end) + 1):
+                                yield int(item)
+                        else:
+                            yield int(string_segment)
+                playlistitems = iter_playlistitems(playlistitems_str)
+
             ie_entries = ie_result['entries']
             if isinstance(ie_entries, list):
                 n_all_entries = len(ie_entries)
-                entries = ie_entries[playliststart:playlistend]
+                if playlistitems:
+                    entries = [ie_entries[i - 1] for i in playlistitems]
+                else:
+                    entries = ie_entries[playliststart:playlistend]
                 n_entries = len(entries)
                 self.to_screen(
                     "[%s] playlist %s: Collected %d video ids (downloading %d of them)" %
                     (ie_result['extractor'], playlist, n_all_entries, n_entries))
             elif isinstance(ie_entries, PagedList):
-                entries = ie_entries.getslice(
-                    playliststart, playlistend)
+                if playlistitems:
+                    entries = []
+                    for item in playlistitems:
+                        entries.extend(ie_entries.getslice(
+                            item - 1, item
+                        ))
+                else:
+                    entries = ie_entries.getslice(
+                        playliststart, playlistend)
                 n_entries = len(entries)
                 self.to_screen(
                     "[%s] playlist %s: Downloading %d videos" %
                     (ie_result['extractor'], playlist, n_entries))
             else:  # iterable
-                entries = list(itertools.islice(
-                    ie_entries, playliststart, playlistend))
+                if playlistitems:
+                    entry_list = list(ie_entries)
+                    entries = [entry_list[i - 1] for i in playlistitems]
+                else:
+                    entries = list(itertools.islice(
+                        ie_entries, playliststart, playlistend))
                 n_entries = len(entries)
                 self.to_screen(
                     "[%s] playlist %s: Downloading %d videos" %
@@ -768,7 +808,59 @@ class YoutubeDL(object):
         else:
             raise Exception('Invalid result type: %s' % result_type)
 
+    def _apply_format_filter(self, format_spec, available_formats):
+        " Returns a tuple of the remaining format_spec and filtered formats "
+
+        OPERATORS = {
+            '<': operator.lt,
+            '<=': operator.le,
+            '>': operator.gt,
+            '>=': operator.ge,
+            '=': operator.eq,
+            '!=': operator.ne,
+        }
+        operator_rex = re.compile(r'''(?x)\s*\[
+            (?P<key>width|height|tbr|abr|vbr|filesize)
+            \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
+            (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
+            \]$
+            ''' % '|'.join(map(re.escape, OPERATORS.keys())))
+        m = operator_rex.search(format_spec)
+        if not m:
+            raise ValueError('Invalid format specification %r' % format_spec)
+
+        try:
+            comparison_value = int(m.group('value'))
+        except ValueError:
+            comparison_value = parse_filesize(m.group('value'))
+            if comparison_value is None:
+                comparison_value = parse_filesize(m.group('value') + 'B')
+            if comparison_value is None:
+                raise ValueError(
+                    'Invalid value %r in format specification %r' % (
+                        m.group('value'), format_spec))
+        op = OPERATORS[m.group('op')]
+
+        def _filter(f):
+            actual_value = f.get(m.group('key'))
+            if actual_value is None:
+                return m.group('none_inclusive')
+            return op(actual_value, comparison_value)
+        new_formats = [f for f in available_formats if _filter(f)]
+
+        new_format_spec = format_spec[:-len(m.group(0))]
+        if not new_format_spec:
+            new_format_spec = 'best'
+
+        return (new_format_spec, new_formats)
+
     def select_format(self, format_spec, available_formats):
+        while format_spec.endswith(']'):
+            format_spec, available_formats = self._apply_format_filter(
+                format_spec, available_formats)
+        if not available_formats:
+            return None
+
         if format_spec == 'best' or format_spec is None:
             return available_formats[-1]
         elif format_spec == 'worst':
@@ -808,6 +900,42 @@ class YoutubeDL(object):
                 return matches[-1]
         return None
 
+    def _calc_headers(self, info_dict):
+        res = std_headers.copy()
+
+        add_headers = info_dict.get('http_headers')
+        if add_headers:
+            res.update(add_headers)
+
+        cookies = self._calc_cookies(info_dict)
+        if cookies:
+            res['Cookie'] = cookies
+
+        return res
+
+    def _calc_cookies(self, info_dict):
+        class _PseudoRequest(object):
+            def __init__(self, url):
+                self.url = url
+                self.headers = {}
+                self.unverifiable = False
+
+            def add_unredirected_header(self, k, v):
+                self.headers[k] = v
+
+            def get_full_url(self):
+                return self.url
+
+            def is_unverifiable(self):
+                return self.unverifiable
+
+            def has_header(self, h):
+                return h in self.headers
+
+        pr = _PseudoRequest(info_dict['url'])
+        self.cookiejar.add_cookie_header(pr)
+        return pr.headers.get('Cookie')
+
     def process_video_result(self, info_dict, download=True):
         assert info_dict.get('_type', 'video') == 'video'
 
@@ -822,9 +950,14 @@ class YoutubeDL(object):
             info_dict['playlist_index'] = None
 
         thumbnails = info_dict.get('thumbnails')
+        if thumbnails is None:
+            thumbnail = info_dict.get('thumbnail')
+            if thumbnail:
+                thumbnails = [{'url': thumbnail}]
         if thumbnails:
             thumbnails.sort(key=lambda t: (
-                t.get('width'), t.get('height'), t.get('url')))
+                t.get('preference'), t.get('width'), t.get('height'),
+                t.get('id'), t.get('url')))
             for t in thumbnails:
                 if 'width' in t and 'height' in t:
                     t['resolution'] = '%dx%d' % (t['width'], t['height'])
@@ -876,6 +1009,11 @@ class YoutubeDL(object):
             # Automatically determine file extension if missing
             if 'ext' not in format:
                 format['ext'] = determine_ext(format['url']).lower()
+            # Add HTTP headers, so that external programs can use them from the
+            # json output
+            full_format_info = info_dict.copy()
+            full_format_info.update(format)
+            format['http_headers'] = self._calc_headers(full_format_info)
 
         format_limit = self.params.get('format_limit', None)
         if format_limit:
@@ -891,9 +1029,12 @@ class YoutubeDL(object):
             # element in the 'formats' field in info_dict is info_dict itself,
             # wich can't be exported to json
             info_dict['formats'] = formats
-        if self.params.get('listformats', None):
+        if self.params.get('listformats'):
             self.list_formats(info_dict)
             return
+        if self.params.get('list_thumbnails'):
+            self.list_thumbnails(info_dict)
+            return
 
         req_format = self.params.get('format')
         if req_format is None:
@@ -1100,35 +1241,18 @@ class YoutubeDL(object):
                     self.report_error('Cannot write metadata to JSON file ' + infofn)
                     return
 
-        if self.params.get('writethumbnail', False):
-            if info_dict.get('thumbnail') is not None:
-                thumb_format = determine_ext(info_dict['thumbnail'], 'jpg')
-                thumb_filename = os.path.splitext(filename)[0] + '.' + thumb_format
-                if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
-                    self.to_screen('[%s] %s: Thumbnail is already present' %
-                                   (info_dict['extractor'], info_dict['id']))
-                else:
-                    self.to_screen('[%s] %s: Downloading thumbnail ...' %
-                                   (info_dict['extractor'], info_dict['id']))
-                    try:
-                        uf = self.urlopen(info_dict['thumbnail'])
-                        with open(thumb_filename, 'wb') as thumbf:
-                            shutil.copyfileobj(uf, thumbf)
-                        self.to_screen('[%s] %s: Writing thumbnail to: %s' %
-                                       (info_dict['extractor'], info_dict['id'], thumb_filename))
-                    except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-                        self.report_warning('Unable to download thumbnail "%s": %s' %
-                                            (info_dict['thumbnail'], compat_str(err)))
+        self._write_thumbnails(info_dict, filename)
 
         if not self.params.get('skip_download', False):
             try:
                 def dl(name, info):
-                    fd = get_suitable_downloader(info)(self, self.params)
+                    fd = get_suitable_downloader(info, self.params)(self, self.params)
                     for ph in self._progress_hooks:
                         fd.add_progress_hook(ph)
                     if self.params.get('verbose'):
                         self.to_stdout('[debug] Invoking downloader on %r' % info.get('url'))
                     return fd.download(name, info)
+
                 if info_dict.get('requested_formats') is not None:
                     downloaded = []
                     success = True
@@ -1164,11 +1288,12 @@ class YoutubeDL(object):
 
             if success:
                 # Fixup content
+                fixup_policy = self.params.get('fixup')
+                if fixup_policy is None:
+                    fixup_policy = 'detect_or_warn'
+
                 stretched_ratio = info_dict.get('stretched_ratio')
                 if stretched_ratio is not None and stretched_ratio != 1:
-                    fixup_policy = self.params.get('fixup')
-                    if fixup_policy is None:
-                        fixup_policy = 'detect_or_warn'
                     if fixup_policy == 'warn':
                         self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
                             info_dict['id'], stretched_ratio))
@@ -1182,7 +1307,23 @@ class YoutubeDL(object):
                                 '%s: Non-uniform pixel ratio (%s). Install ffmpeg or avconv to fix this automatically.' % (
                                     info_dict['id'], stretched_ratio))
                     else:
-                        assert fixup_policy == 'ignore'
+                        assert fixup_policy in ('ignore', 'never')
+
+                if info_dict.get('requested_formats') is None and info_dict.get('container') == 'm4a_dash':
+                    if fixup_policy == 'warn':
+                        self.report_warning('%s: writing DASH m4a. Only some players support this container.' % (
+                            info_dict['id']))
+                    elif fixup_policy == 'detect_or_warn':
+                        fixup_pp = FFmpegFixupM4aPP(self)
+                        if fixup_pp.available:
+                            info_dict.setdefault('__postprocessors', [])
+                            info_dict['__postprocessors'].append(fixup_pp)
+                        else:
+                            self.report_warning(
+                                '%s: writing DASH m4a. Only some players support this container. Install ffmpeg or avconv to fix this automatically.' % (
+                                    info_dict['id']))
+                    else:
+                        assert fixup_policy in ('ignore', 'never')
 
                 try:
                     self.post_process(filename, info_dict)
@@ -1384,8 +1525,26 @@ class YoutubeDL(object):
         header_line = line({
             'format_id': 'format code', 'ext': 'extension',
             'resolution': 'resolution', 'format_note': 'note'}, idlen=idlen)
-        self.to_screen('[info] Available formats for %s:\n%s\n%s' %
-                       (info_dict['id'], header_line, '\n'.join(formats_s)))
+        self.to_screen(
+            '[info] Available formats for %s:\n%s\n%s' %
+            (info_dict['id'], header_line, '\n'.join(formats_s)))
+
+    def list_thumbnails(self, info_dict):
+        thumbnails = info_dict.get('thumbnails')
+        if not thumbnails:
+            tn_url = info_dict.get('thumbnail')
+            if tn_url:
+                thumbnails = [{'id': '0', 'url': tn_url}]
+            else:
+                self.to_screen(
+                    '[info] No thumbnails present for %s' % info_dict['id'])
+                return
+
+        self.to_screen(
+            '[info] Thumbnails for %s:' % info_dict['id'])
+        self.to_screen(render_table(
+            ['ID', 'width', 'height', 'URL'],
+            [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
 
     def urlopen(self, req):
         """ Start an HTTP download """
@@ -1531,3 +1690,39 @@ class YoutubeDL(object):
         if encoding is None:
             encoding = preferredencoding()
         return encoding
+
+    def _write_thumbnails(self, info_dict, filename):
+        if self.params.get('writethumbnail', False):
+            thumbnails = info_dict.get('thumbnails')
+            if thumbnails:
+                thumbnails = [thumbnails[-1]]
+        elif self.params.get('write_all_thumbnails', False):
+            thumbnails = info_dict.get('thumbnails')
+        else:
+            return
+
+        if not thumbnails:
+            # No thumbnails present, so return immediately
+            return
+
+        for t in thumbnails:
+            thumb_ext = determine_ext(t['url'], 'jpg')
+            suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
+            thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
+            thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext
+
+            if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
+                self.to_screen('[%s] %s: Thumbnail %sis already present' %
+                               (info_dict['extractor'], info_dict['id'], thumb_display_id))
+            else:
+                self.to_screen('[%s] %s: Downloading thumbnail %s...' %
+                               (info_dict['extractor'], info_dict['id'], thumb_display_id))
+                try:
+                    uf = self.urlopen(t['url'])
+                    with open(thumb_filename, 'wb') as thumbf:
+                        shutil.copyfileobj(uf, thumbf)
+                    self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
+                                   (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
+                except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
+                    self.report_warning('Unable to download thumbnail "%s": %s' %
+                                        (t['url'], compat_str(err)))

youtube_dl/__init__.py

@@ -143,10 +143,13 @@ 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:
@@ -238,6 +241,12 @@ def _real_main(argv=None):
             'verboseOutput': opts.verbose,
             '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')
 
     ydl_opts = {
         'usenetrc': opts.usenetrc,
@@ -268,7 +277,7 @@ def _real_main(argv=None):
         'ignoreerrors': opts.ignoreerrors,
         'ratelimit': opts.ratelimit,
         'nooverwrites': opts.nooverwrites,
-        'retries': opts.retries,
+        'retries': opts_retries,
         'buffersize': opts.buffersize,
         'noresizebuffer': opts.noresizebuffer,
         'continuedl': opts.continue_dl,
@@ -286,6 +295,7 @@ 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,
@@ -329,6 +339,11 @@ def _real_main(argv=None):
         '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,
     }
 
     with YoutubeDL(ydl_opts) as ydl:

youtube_dl/downloader/__init__.py

@@ -1,35 +1,41 @@
 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 .rtmp import RtmpFD
-from .f4m import F4mFD
 
 from ..utils import (
-    determine_ext,
+    determine_protocol,
 )
 
+PROTOCOL_MAP = {
+    'rtmp': RtmpFD,
+    'm3u8_native': NativeHlsFD,
+    'm3u8': HlsFD,
+    'mms': MplayerFD,
+    'rtsp': MplayerFD,
+    'f4m': F4mFD,
+}
 
-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
+
+    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
+
+    return PROTOCOL_MAP.get(protocol, HttpFD)
 
-    if url.startswith('rtmp'):
-        return RtmpFD
-    if protocol == 'm3u8_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
 
 __all__ = [
     'get_suitable_downloader',

youtube_dl/downloader/common.py

@@ -25,21 +25,23 @@ 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.
+                        (experimenatal)
 
     Subclasses of this one must re-define the real_download method.
     """
@@ -284,6 +286,7 @@ 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))
@@ -305,6 +308,11 @@ 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):
@@ -319,3 +327,24 @@ class FileDownloader(object):
         # See YoutubeDl.py (search for progress_hooks) for a description of
         # this interface
         self._progress_hooks.append(ph)
+
+    def _debug_cmd(self, args, subprocess_encoding, exe=None):
+        if not self.params.get('verbose', False):
+            return
+
+        if exe is None:
+            exe = os.path.basename(args[0])
+
+        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] %s command line: %s' % (
+            exe, shell_quote(str_args)))

youtube_dl/downloader/external.py

@@ -0,0 +1,117 @@
+from __future__ import unicode_literals
+
+import os.path
+import subprocess
+import sys
+
+from .common import FileDownloader
+from ..utils import (
+    encodeFilename,
+)
+
+
+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 _call_downloader(self, tmpfilename, info_dict):
+        """ Either overwrite this or implement _make_cmd """
+        cmd = self._make_cmd(tmpfilename, info_dict)
+
+        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()
+            cmd = [a.encode(subprocess_encoding, 'ignore') for a in cmd]
+        else:
+            subprocess_encoding = None
+        self._debug_cmd(cmd, subprocess_encoding)
+
+        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, '-o', tmpfilename]
+        for key, val in info_dict['http_headers'].items():
+            cmd += ['--header', '%s: %s' % (key, val)]
+        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 += ['--', info_dict['url']]
+        return cmd
+
+
+class Aria2cFD(ExternalFD):
+    def _make_cmd(self, tmpfilename, info_dict):
+        cmd = [
+            self.exe, '-c',
+            '--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 += ['--', info_dict['url']]
+        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 . """
+    bn = os.path.basename(external_downloader)
+    return _BY_NAME[bn]

youtube_dl/downloader/f4m.py

@@ -177,13 +177,12 @@ 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)))
     return res
 
 

youtube_dl/downloader/http.py

@@ -24,10 +24,6 @@ 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)
@@ -161,6 +157,14 @@ 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:

youtube_dl/downloader/rtmp.py

@@ -104,6 +104,8 @@ class RtmpFD(FileDownloader):
         live = info_dict.get('rtmp_live', False)
         conn = info_dict.get('rtmp_conn', None)
         protocol = info_dict.get('rtmp_protocol', None)
+        no_resume = info_dict.get('no_resume', False)
+        continue_dl = info_dict.get('continuedl', False)
 
         self.report_destination(filename)
         tmpfilename = self.temp_name(filename)
@@ -141,7 +143,12 @@ 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)]
+
+        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 sys.platform == 'win32' and sys.version_info < (3, 0):
             # Windows subprocess module does not actually support Unicode
@@ -152,19 +159,7 @@ class RtmpFD(FileDownloader):
         else:
             subprocess_encoding = None
 
-        if self.params.get('verbose', False):
-            if subprocess_encoding:
-                str_args = [
-                    a.decode(subprocess_encoding) if isinstance(a, bytes) else a
-                    for a in args]
-            else:
-                str_args = args
-            try:
-                import pipes
-                shell_quote = lambda args: ' '.join(map(pipes.quote, str_args))
-            except ImportError:
-                shell_quote = repr
-            self.to_screen('[debug] rtmpdump command line: ' + shell_quote(str_args))
+        self._debug_cmd(args, subprocess_encoding, exe='rtmpdump')
 
         RD_SUCCESS = 0
         RD_FAILED = 1

youtube_dl/extractor/__init__.py

@@ -1,6 +1,7 @@
 from __future__ import unicode_literals
 
 from .abc import ABCIE
+from .abc7news import Abc7NewsIE
 from .academicearth import AcademicEarthCourseIE
 from .addanime import AddAnimeIE
 from .adobetv import AdobeTVIE
@@ -28,7 +29,6 @@ from .arte import (
 from .atresplayer import AtresPlayerIE
 from .atttechchannel import ATTTechChannelIE
 from .audiomack import AudiomackIE, AudiomackAlbumIE
-from .auengine import AUEngineIE
 from .azubu import AzubuIE
 from .bambuser import BambuserIE, BambuserChannelIE
 from .bandcamp import BandcampIE, BandcampAlbumIE
@@ -93,6 +93,7 @@ from .deezer import DeezerPlaylistIE
 from .dfb import DFBIE
 from .dotsub import DotsubIE
 from .dreisat import DreiSatIE
+from .drbonanza import DRBonanzaIE
 from .drtuber import DrTuberIE
 from .drtv import DRTVIE
 from .dvtv import DVTVIE
@@ -174,6 +175,7 @@ from .goshgay import GoshgayIE
 from .grooveshark import GroovesharkIE
 from .groupon import GrouponIE
 from .hark import HarkIE
+from .hearthisat import HearThisAtIE
 from .heise import HeiseIE
 from .hellporno import HellPornoIE
 from .helsinki import HelsinkiIE
@@ -225,6 +227,7 @@ from .livestream import (
     LivestreamOriginalIE,
     LivestreamShortenerIE,
 )
+from .lnkgo import LnkGoIE
 from .lrt import LRTIE
 from .lynda import (
     LyndaIE,
@@ -294,6 +297,7 @@ from .nowness import NownessIE
 from .nowvideo import NowVideoIE
 from .npo import (
     NPOIE,
+    NPOLiveIE,
     TegenlichtVproIE,
 )
 from .nrk import (
@@ -345,6 +349,7 @@ from .rtbf import RTBFIE
 from .rte import RteIE
 from .rtlnl import RtlXlIE
 from .rtlnow import RTLnowIE
+from .rtl2 import RTL2IE
 from .rtp import RTPIE
 from .rts import RTSIE
 from .rtve import RTVEALaCartaIE, RTVELiveIE
@@ -405,6 +410,7 @@ 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 .swrmediathek import SWRMediathekIE
 from .syfy import SyfyIE
@@ -426,6 +432,7 @@ from .telemb import TeleMBIE
 from .teletask import TeleTaskIE
 from .tenplay import TenPlayIE
 from .testurl import TestURLIE
+from .testtube import TestTubeIE
 from .tf1 import TF1IE
 from .theonion import TheOnionIE
 from .theplatform import ThePlatformIE
@@ -454,7 +461,15 @@ from .tvigle import TvigleIE
 from .tvp import TvpIE, TvpSeriesIE
 from .tvplay import TVPlayIE
 from .twentyfourvideo import TwentyFourVideoIE
-from .twitch import TwitchIE
+from .twitch import (
+    TwitchVideoIE,
+    TwitchChapterIE,
+    TwitchVodIE,
+    TwitchProfileIE,
+    TwitchPastBroadcastsIE,
+    TwitchBookmarksIE,
+    TwitchStreamIE,
+)
 from .ubu import UbuIE
 from .udemy import (
     UdemyIE,

youtube_dl/extractor/abc7news.py

@@ -0,0 +1,68 @@
+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')
+        self._sort_formats(formats)
+
+        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/appletrailers.py

@@ -129,7 +129,9 @@ class AppleTrailersIE(InfoExtractor):
                 '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 {

youtube_dl/extractor/atresplayer.py

@@ -3,10 +3,13 @@ from __future__ import unicode_literals
 import time
 import hmac
 
-from .common import InfoExtractor
-from ..utils import (
+from .subtitles import SubtitlesInfoExtractor
+from ..compat import (
     compat_str,
+    compat_urllib_parse,
     compat_urllib_request,
+)
+from ..utils import (
     int_or_none,
     float_or_none,
     xpath_text,
@@ -14,7 +17,7 @@ from ..utils import (
 )
 
 
-class AtresPlayerIE(InfoExtractor):
+class AtresPlayerIE(SubtitlesInfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?atresplayer\.com/television/[^/]+/[^/]+/[^/]+/(?P<id>.+?)_\d+\.html'
     _TESTS = [
         {
@@ -44,6 +47,33 @@ class AtresPlayerIE(InfoExtractor):
     _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'
+
+    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 = compat_urllib_request.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)
 
@@ -65,7 +95,7 @@ class AtresPlayerIE(InfoExtractor):
         for fmt in ['windows', 'android_tablet']:
             request = compat_urllib_request.Request(
                 self._URL_VIDEO_TEMPLATE.format(fmt, episode_id, timestamp_shifted, token))
-            request.add_header('Youtubedl-user-agent', self._USER_AGENT)
+            request.add_header('User-Agent', self._USER_AGENT)
 
             fmt_json = self._download_json(
                 request, video_id, 'Downloading %s video JSON' % fmt)
@@ -75,13 +105,22 @@ class AtresPlayerIE(InfoExtractor):
                 raise ExtractorError(
                     '%s returned error: %s' % (self.IE_NAME, result), expected=True)
 
-            for _, video_url in fmt_json['resultObject'].items():
+            for format_id, video_url in fmt_json['resultObject'].items():
+                if format_id == 'token' or not video_url.startswith('http'):
+                    continue
                 if video_url.endswith('/Manifest'):
-                    formats.extend(self._extract_f4m_formats(video_url[:-9] + '/manifest.f4m', video_id))
+                    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))
                 else:
                     formats.append({
                         'url': video_url,
-                        'format_id': 'android',
+                        'format_id': 'android-%s' % format_id,
                         'preference': 1,
                     })
         self._sort_formats(formats)
@@ -104,6 +143,15 @@ class AtresPlayerIE(InfoExtractor):
         description = xpath_text(art, './description', 'description')
         thumbnail = xpath_text(episode, './media/asset/files/background', 'thumbnail')
 
+        subtitles = {}
+        subtitle = xpath_text(episode, './media/asset/files/subtitle', 'subtitle')
+        if subtitle:
+            subtitles['es'] = subtitle
+
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(video_id, subtitles)
+            return
+
         return {
             'id': video_id,
             'title': title,
@@ -111,4 +159,5 @@ class AtresPlayerIE(InfoExtractor):
             'thumbnail': thumbnail,
             'duration': duration,
             'formats': formats,
+            'subtitles': self.extract_subtitles(video_id, subtitles),
         }

youtube_dl/extractor/audiomack.py

@@ -88,16 +88,21 @@ class AudiomackAlbumIE(InfoExtractor):
         # 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': '9.-heaven-or-hell-chimaca-ft-zuse-prod-by-dj-fu',
-                    'id': '9.-heaven-or-hell-chimaca-ft-zuse-prod-by-dj-fu',
+                    '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': 8,
-                'playlistend': 8,
+                'playliststart': 9,
+                'playlistend': 9,
             }
         }
     ]

youtube_dl/extractor/auengine.py

@@ -1,50 +0,0 @@
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..compat import compat_urllib_parse
-from ..utils import (
-    determine_ext,
-    ExtractorError,
-    remove_end,
-)
-
-
-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>\s*(?P<title>.+?)\s*</title>', webpage, 'title')
-        video_urls = re.findall(r'http://\w+.auengine.com/vod/.*[^\W]', webpage)
-        video_url = compat_urllib_parse.unquote(video_urls[0])
-        thumbnails = re.findall(r'http://\w+.auengine.com/thumb/.*[^\W]', webpage)
-        thumbnail = compat_urllib_parse.unquote(thumbnails[0])
-
-        if not video_url:
-            raise ExtractorError('Could not find video URL')
-
-        ext = '.' + determine_ext(video_url)
-        title = remove_end(title, 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/bandcamp.py

@@ -161,7 +161,8 @@ class BandcampAlbumIE(InfoExtractor):
         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',
             'id': playlist_id,

youtube_dl/extractor/bliptv.py

@@ -199,7 +199,7 @@ class BlipTVIE(SubtitlesInfoExtractor):
         # For some weird reason, blip.tv serves a video instead of subtitles
         # when we request with a common UA
         req = compat_urllib_request.Request(url)
-        req.add_header('Youtubedl-user-agent', 'youtube-dl')
+        req.add_header('User-Agent', 'youtube-dl')
         return self._download_webpage(req, None, note=False)
 
 

youtube_dl/extractor/cliphunter.py

@@ -1,9 +1,7 @@
 from __future__ import unicode_literals
 
-import json
-import re
-
 from .common import InfoExtractor
+from ..utils import determine_ext
 
 
 _translation_table = {
@@ -27,10 +25,10 @@ class CliphunterIE(InfoExtractor):
     '''
     _TEST = {
         'url': 'http://www.cliphunter.com/w/1012420/Fun_Jynx_Maze_solo',
-        'md5': 'a2ba71eebf523859fe527a61018f723e',
+        'md5': 'b7c9bbd4eb3a226ab91093714dcaa480',
         'info_dict': {
             'id': '1012420',
-            'ext': 'mp4',
+            'ext': 'flv',
             'title': 'Fun Jynx Maze solo',
             'thumbnail': 're:^https?://.*\.jpg$',
             'age_limit': 18,
@@ -44,39 +42,31 @@ class CliphunterIE(InfoExtractor):
         video_title = self._search_regex(
             r'mediaTitle = "([^"]+)"', webpage, 'title')
 
-        pl_fiji = self._search_regex(
-            r'pl_fiji = \'([^\']+)\'', webpage, 'video data')
-        pl_c_qual = self._search_regex(
-            r'pl_c_qual = "(.)"', webpage, 'video quality')
-        video_url = _decode(pl_fiji)
-        formats = [{
-            'url': video_url,
-            'format_id': 'default-%s' % pl_c_qual,
-        }]
+        fmts = {}
+        for fmt in ('mp4', 'flv'):
+            fmt_list = self._parse_json(self._search_regex(
+                r'var %sjson\s*=\s*(\[.*?\]);' % fmt, webpage, '%s formats' % fmt), video_id)
+            for f in fmt_list:
+                fmts[f['fname']] = _decode(f['sUrl'])
 
-        qualities_json = self._search_regex(
-            r'var pl_qualities\s*=\s*(.*?);\n', webpage, 'quality info')
-        qualities_data = json.loads(qualities_json)
+        qualities = self._parse_json(self._search_regex(
+            r'var player_btns\s*=\s*(.*?);\n', webpage, 'quality info'), video_id)
 
-        for i, t in enumerate(
-                re.findall(r"pl_fiji_([a-z0-9]+)\s*=\s*'([^']+')", webpage)):
-            quality_id, crypted_url = t
-            video_url = _decode(crypted_url)
+        formats = []
+        for fname, url in fmts.items():
             f = {
-                'format_id': quality_id,
-                'url': video_url,
-                'quality': i,
+                'url': url,
             }
-            if quality_id in qualities_data:
-                qd = qualities_data[quality_id]
-                m = re.match(
-                    r'''(?x)<b>(?P<width>[0-9]+)x(?P<height>[0-9]+)<\\/b>
-                        \s*\(\s*(?P<tbr>[0-9]+)\s*kb\\/s''', qd)
-                if m:
-                    f['width'] = int(m.group('width'))
-                    f['height'] = int(m.group('height'))
-                    f['tbr'] = int(m.group('tbr'))
+            if fname in qualities:
+                qual = qualities[fname]
+                f.update({
+                    'format_id': '%s_%sp' % (determine_ext(url), qual['h']),
+                    'width': qual['w'],
+                    'height': qual['h'],
+                    'tbr': qual['br'],
+                })
             formats.append(f)
+
         self._sort_formats(formats)
 
         thumbnail = self._search_regex(

youtube_dl/extractor/cnn.py

@@ -51,7 +51,7 @@ class CNNIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         path = mobj.group('path')
         page_title = mobj.group('title')
-        info_url = 'http://cnn.com/video/data/3.0/%s/index.xml' % path
+        info_url = 'http://edition.cnn.com/video/data/3.0/%s/index.xml' % path
         info = self._download_xml(info_url, page_title)
 
         formats = []
@@ -143,13 +143,13 @@ class CNNArticleIE(InfoExtractor):
     _VALID_URL = r'https?://(?:(?:edition|www)\.)?cnn\.com/(?!video/)'
     _TEST = {
         'url': 'http://www.cnn.com/2014/12/21/politics/obama-north-koreas-hack-not-war-but-cyber-vandalism/',
-        'md5': '275b326f85d80dff7592a9820f5dc887',
+        'md5': '689034c2a3d9c6dc4aa72d65a81efd01',
         'info_dict': {
-            'id': 'bestoftv/2014/12/21/sotu-crowley-president-obama-north-korea-not-going-to-be-intimidated.cnn',
+            'id': 'bestoftv/2014/12/21/ip-north-korea-obama.cnn',
             'ext': 'mp4',
-            'title': 'Obama: We\'re not going to be intimidated',
-            'description': 'md5:e735586f3dc936075fa654a4d91b21f9',
-            'upload_date': '20141220',
+            'title': 'Obama: Cyberattack not an act of war',
+            'description': 'md5:51ce6750450603795cad0cdfbd7d05c5',
+            'upload_date': '20141221',
         },
         'add_ie': ['CNN'],
     }

youtube_dl/extractor/comedycentral.py

@@ -34,12 +34,12 @@ class ComedyCentralIE(MTVServicesInfoExtractor):
 
 class ComedyCentralShowsIE(MTVServicesInfoExtractor):
     IE_DESC = 'The Daily Show / The Colbert Report'
-    # urls can be abbreviations like :thedailyshow or :colbert
+    # urls can be abbreviations like :thedailyshow
     # urls for episodes like:
     # or urls for clips like: http://www.thedailyshow.com/watch/mon-december-10-2012/any-given-gun-day
     #                     or: http://www.colbertnation.com/the-colbert-report-videos/421667/november-29-2012/moon-shattering-news
     #                     or: http://www.colbertnation.com/the-colbert-report-collections/422008/festival-of-lights/79524
-    _VALID_URL = r'''(?x)^(:(?P<shortname>tds|thedailyshow|cr|colbert|colbertnation|colbertreport)
+    _VALID_URL = r'''(?x)^(:(?P<shortname>tds|thedailyshow)
                       |https?://(:www\.)?
                           (?P<showname>thedailyshow|thecolbertreport)\.(?:cc\.)?com/
                          ((?:full-)?episodes/(?:[0-9a-z]{6}/)?(?P<episode>.*)|

youtube_dl/extractor/common.py

@@ -14,6 +14,7 @@ import xml.etree.ElementTree
 
 from ..compat import (
     compat_cookiejar,
+    compat_HTTPError,
     compat_http_client,
     compat_urllib_error,
     compat_urllib_parse_urlparse,
@@ -26,6 +27,7 @@ from ..utils import (
     compiled_regex_type,
     ExtractorError,
     float_or_none,
+    HEADRequest,
     int_or_none,
     RegexNotFoundError,
     sanitize_filename,
@@ -108,15 +110,17 @@ class InfoExtractor(object):
                                   (quality takes higher priority)
                                  -1 for default (order by other properties),
                                  -2 or smaller for less than default.
-                    * http_referer  HTTP Referer header value to set.
                     * http_method  HTTP method to use for the download.
                     * http_headers  A dictionary of additional HTTP headers
                                  to add to the request.
                     * http_post_data  Additional data to send with a POST
                                  request.
                     * stretched_ratio  If given and not 1, indicates that the
-                                       video's pixels are not square.
-                                       width : height ratio as float.
+                                 video's pixels are not square.
+                                 width : height ratio as float.
+                    * no_resume  The server does not support resuming the
+                                 (HTTP or RTMP) download. Boolean.
+
     url:            Final video URL.
     ext:            Video filename extension.
     format:         The video format, defaults to ext (used for --get-format)
@@ -130,7 +134,9 @@ class InfoExtractor(object):
                     something like "4234987", title "Dancing naked mole rats",
                     and display_id "dancing-naked-mole-rats"
     thumbnails:     A list of dictionaries, with the following entries:
+                        * "id" (optional, string) - Thumbnail format ID
                         * "url"
+                        * "preference" (optional, int) - quality of the image
                         * "width" (optional, int)
                         * "height" (optional, int)
                         * "resolution" (optional, string "{width}x{height"},
@@ -712,6 +718,27 @@ class InfoExtractor(object):
             )
         formats.sort(key=_formats_key)
 
+    def _check_formats(self, formats, video_id):
+        if formats:
+            formats[:] = filter(
+                lambda f: self._is_valid_url(
+                    f['url'], video_id,
+                    item='%s video format' % f.get('format_id') if f.get('format_id') else 'video'),
+                formats)
+
+    def _is_valid_url(self, url, video_id, item='video'):
+        try:
+            self._request_webpage(
+                HEADRequest(url), video_id,
+                'Checking %s URL' % item)
+            return True
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError):
+                self.report_warning(
+                    '%s URL is invalid, skipping' % item, video_id)
+                return False
+            raise
+
     def http_scheme(self):
         """ Either "http:" or "https:", depending on the user's preferences """
         return (

youtube_dl/extractor/drbonanza.py

@@ -0,0 +1,131 @@
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    parse_iso8601,
+)
+
+
+class DRBonanzaIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?dr\.dk/bonanza/(?:[^/]+/)+(?:[^/])+?(?:assetId=(?P<id>\d+))?(?:[#&]|$)'
+
+    _TESTS = [{
+        'url': 'http://www.dr.dk/bonanza/serie/portraetter/Talkshowet.htm?assetId=65517',
+        'md5': 'fe330252ddea607635cf2eb2c99a0af3',
+        'info_dict': {
+            'id': '65517',
+            'ext': 'mp4',
+            'title': 'Talkshowet - Leonard Cohen',
+            'description': 'md5:8f34194fb30cd8c8a30ad8b27b70c0ca',
+            'thumbnail': 're:^https?://.*\.(?:gif|jpg)$',
+            'timestamp': 1295537932,
+            'upload_date': '20110120',
+            'duration': 3664,
+        },
+    }, {
+        'url': 'http://www.dr.dk/bonanza/radio/serie/sport/fodbold.htm?assetId=59410',
+        'md5': '6dfe039417e76795fb783c52da3de11d',
+        'info_dict': {
+            'id': '59410',
+            'ext': 'mp3',
+            'title': 'EM fodbold 1992 Danmark - Tyskland finale Transmission',
+            'description': 'md5:501e5a195749480552e214fbbed16c4e',
+            'thumbnail': 're:^https?://.*\.(?:gif|jpg)$',
+            'timestamp': 1223274900,
+            'upload_date': '20081006',
+            'duration': 7369,
+        },
+    }]
+
+    def _real_extract(self, url):
+        url_id = self._match_id(url)
+        webpage = self._download_webpage(url, url_id)
+
+        if url_id:
+            info = json.loads(self._html_search_regex(r'({.*?%s.*})' % url_id, webpage, 'json'))
+        else:
+            # Just fetch the first video on that page
+            info = json.loads(self._html_search_regex(r'bonanzaFunctions.newPlaylist\(({.*})\)', webpage, 'json'))
+
+        asset_id = str(info['AssetId'])
+        title = info['Title'].rstrip(' \'\"-,.:;!?')
+        duration = int_or_none(info.get('Duration'), scale=1000)
+        # First published online. "FirstPublished" contains the date for original airing.
+        timestamp = parse_iso8601(
+            re.sub(r'\.\d+$', '', info['Created']))
+
+        def parse_filename_info(url):
+            match = re.search(r'/\d+_(?P<width>\d+)x(?P<height>\d+)x(?P<bitrate>\d+)K\.(?P<ext>\w+)$', url)
+            if match:
+                return {
+                    'width': int(match.group('width')),
+                    'height': int(match.group('height')),
+                    'vbr': int(match.group('bitrate')),
+                    'ext': match.group('ext')
+                }
+            match = re.search(r'/\d+_(?P<bitrate>\d+)K\.(?P<ext>\w+)$', url)
+            if match:
+                return {
+                    'vbr': int(match.group('bitrate')),
+                    'ext': match.group(2)
+                }
+            return {}
+
+        video_types = ['VideoHigh', 'VideoMid', 'VideoLow']
+        preferencemap = {
+            'VideoHigh': -1,
+            'VideoMid': -2,
+            'VideoLow': -3,
+            'Audio': -4,
+        }
+
+        formats = []
+        for file in info['Files']:
+            if info['Type'] == "Video":
+                if file['Type'] in video_types:
+                    format = parse_filename_info(file['Location'])
+                    format.update({
+                        'url': file['Location'],
+                        'format_id': file['Type'].replace('Video', ''),
+                        'preference': preferencemap.get(file['Type'], -10),
+                    })
+                    formats.append(format)
+                elif file['Type'] == "Thumb":
+                    thumbnail = file['Location']
+            elif info['Type'] == "Audio":
+                if file['Type'] == "Audio":
+                    format = parse_filename_info(file['Location'])
+                    format.update({
+                        'url': file['Location'],
+                        'format_id': file['Type'],
+                        'vcodec': 'none',
+                    })
+                    formats.append(format)
+                elif file['Type'] == "Thumb":
+                    thumbnail = file['Location']
+
+        description = '%s\n%s\n%s\n' % (
+            info['Description'], info['Actors'], info['Colophon'])
+
+        for f in formats:
+            f['url'] = f['url'].replace('rtmp://vod-bonanza.gss.dr.dk/bonanza/', 'http://vodfiles.dr.dk/')
+            f['url'] = f['url'].replace('mp4:bonanza', 'bonanza')
+        self._sort_formats(formats)
+
+        display_id = re.sub(r'[^\w\d-]', '', re.sub(r' ', '-', title.lower())) + '-' + asset_id
+        display_id = re.sub(r'-+', '-', display_id)
+
+        return {
+            'id': asset_id,
+            'display_id': display_id,
+            'title': title,
+            'formats': formats,
+            'description': description,
+            'thumbnail': thumbnail,
+            'timestamp': timestamp,
+            'duration': duration,
+        }

youtube_dl/extractor/drtv.py

@@ -48,14 +48,20 @@ class DRTVIE(SubtitlesInfoExtractor):
             elif asset['Kind'] == 'VideoResource':
                 duration = asset['DurationInMilliseconds'] / 1000.0
                 restricted_to_denmark = asset['RestrictedToDenmark']
+                spoken_subtitles = asset['Target'] == 'SpokenSubtitles'
                 for link in asset['Links']:
                     target = link['Target']
                     uri = link['Uri']
+                    format_id = target
+                    preference = -1 if target == 'HDS' else -2
+                    if spoken_subtitles:
+                        preference -= 2
+                        format_id += '-spoken-subtitles'
                     formats.append({
                         'url': uri + '?hdcore=3.3.0&plugin=aasp-3.3.0.99.43' if target == 'HDS' else uri,
-                        'format_id': target,
+                        'format_id': format_id,
                         'ext': link['FileFormat'],
-                        'preference': -1 if target == 'HDS' else -2,
+                        'preference': preference,
                     })
                 subtitles_list = asset.get('SubtitlesList')
                 if isinstance(subtitles_list, list):

youtube_dl/extractor/fc2.py

@@ -5,6 +5,7 @@ import hashlib
 
 from .common import InfoExtractor
 from ..compat import (
+    compat_urllib_parse,
     compat_urllib_request,
     compat_urlparse,
 )
@@ -16,7 +17,8 @@ from ..utils import (
 class FC2IE(InfoExtractor):
     _VALID_URL = r'^http://video\.fc2\.com/(?:[^/]+/)?content/(?P<id>[^/]+)'
     IE_NAME = 'fc2'
-    _TEST = {
+    _NETRC_MACHINE = 'fc2'
+    _TESTS = [{
         'url': 'http://video.fc2.com/en/content/20121103kUan1KHs',
         'md5': 'a6ebe8ebe0396518689d963774a54eb7',
         'info_dict': {
@@ -24,12 +26,57 @@ class FC2IE(InfoExtractor):
             'ext': 'flv',
             'title': 'Boxing again with Puff',
         },
-    }
+    }, {
+        'url': 'http://video.fc2.com/en/content/20150125cEva0hDn/',
+        'info_dict': {
+            'id': '20150125cEva0hDn',
+            'ext': 'mp4',
+        },
+        'params': {
+            'username': 'ytdl@yt-dl.org',
+            'password': '(snip)',
+            'skip': 'requires actual password'
+        }
+    }]
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None or password is None:
+            return False
+
+        # Log in
+        login_form_strs = {
+            'email': username,
+            'password': password,
+            'done': 'video',
+            'Submit': ' Login ',
+        }
+
+        # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
+        # chokes on unicode
+        login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
+        login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
+        request = compat_urllib_request.Request(
+            'https://secure.id.fc2.com/index.php?mode=login&switch_language=en', login_data)
+
+        login_results = self._download_webpage(request, None, note='Logging in', errnote='Unable to log in')
+        if 'mode=redirect&login=done' not in login_results:
+            self.report_warning('unable to log in: bad username or password')
+            return False
+
+        # this is also needed
+        login_redir = compat_urllib_request.Request('http://id.fc2.com/?mode=redirect&login=done')
+        self._download_webpage(
+            login_redir, None, note='Login redirect', errnote='Login redirect failed')
+
+        return True
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+        self._login()
         webpage = self._download_webpage(url, video_id)
         self._downloader.cookiejar.clear_session_cookies()  # must clear
+        self._login()
 
         title = self._og_search_title(webpage)
         thumbnail = self._og_search_thumbnail(webpage)
@@ -46,7 +93,12 @@ class FC2IE(InfoExtractor):
         info = compat_urlparse.parse_qs(info_webpage)
 
         if 'err_code' in info:
-            raise ExtractorError('Error code: %s' % info['err_code'][0])
+            # most of the time we can still download wideo even if err_code is 403 or 602
+            self.report_warning(
+                'Error code was: %s... but still trying' % info['err_code'][0])
+
+        if 'filepath' not in info:
+            raise ExtractorError('Cannot download file. Are you logged in?')
 
         video_url = info['filepath'][0] + '?mid=' + info['mid'][0]
         title_info = info.get('title')

youtube_dl/extractor/folketinget.py

@@ -16,6 +16,7 @@ class FolketingetIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?ft\.dk/webtv/video/[^?#]*?\.(?P<id>[0-9]+)\.aspx'
     _TEST = {
         'url': 'http://www.ft.dk/webtv/video/20141/eru/td.1165642.aspx?as=1#player',
+        'md5': '6269e8626fa1a891bf5369b386ae996a',
         'info_dict': {
             'id': '1165642',
             'ext': 'mp4',
@@ -29,9 +30,6 @@ class FolketingetIE(InfoExtractor):
             'upload_date': '20141120',
             'duration': 3960,
         },
-        'params': {
-            'skip_download': 'rtmpdump required',
-        }
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/fourtube.py

@@ -7,10 +7,9 @@ from ..compat import (
     compat_urllib_request,
 )
 from ..utils import (
-    clean_html,
     parse_duration,
+    parse_iso8601,
     str_to_int,
-    unified_strdate,
 )
 
 
@@ -28,68 +27,81 @@ class FourTubeIE(InfoExtractor):
             'uploader': 'WCP Club',
             'uploader_id': 'wcp-club',
             'upload_date': '20131031',
+            'timestamp': 1383263892,
             'duration': 583,
+            'view_count': int,
+            'like_count': int,
+            'categories': list,
         }
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage_url = 'http://www.4tube.com/videos/' + video_id
-        webpage = self._download_webpage(webpage_url, video_id)
+        webpage = self._download_webpage(url, video_id)
 
-        self.report_extraction(video_id)
+        title = self._html_search_meta('name', webpage)
+        timestamp = parse_iso8601(self._html_search_meta(
+            'uploadDate', webpage))
+        thumbnail = self._html_search_meta('thumbnailUrl', webpage)
+        uploader_id = self._html_search_regex(
+            r'<a class="img-avatar" href="[^"]+/channels/([^/"]+)" title="Go to [^"]+ page">',
+            webpage, 'uploader id')
+        uploader = self._html_search_regex(
+            r'<a class="img-avatar" href="[^"]+/channels/[^/"]+" title="Go to ([^"]+) page">',
+            webpage, 'uploader')
 
-        playlist_json = self._html_search_regex(r'var playerConfigPlaylist\s+=\s+([^;]+)', webpage, 'Playlist')
-        media_id = self._search_regex(r'idMedia:\s*(\d+)', playlist_json, 'Media Id')
-        sources = self._search_regex(r'sources:\s*\[([^\]]*)\]', playlist_json, 'Sources').split(',')
-        title = self._search_regex(r'title:\s*"([^"]*)', playlist_json, 'Title')
-        thumbnail_url = self._search_regex(r'image:\s*"([^"]*)', playlist_json, 'Thumbnail', fatal=False)
+        categories_html = self._search_regex(
+            r'(?s)><i class="icon icon-tag"></i>\s*Categories / Tags\s*.*?<ul class="list">(.*?)</ul>',
+            webpage, 'categories', fatal=False)
+        categories = None
+        if categories_html:
+            categories = [
+                c.strip() for c in re.findall(
+                    r'(?s)<li><a.*?>(.*?)</a>', categories_html)]
 
-        uploader_str = self._search_regex(r'<span>Uploaded by</span>(.*?)<span>', webpage, 'uploader', fatal=False)
-        mobj = re.search(r'<a href="/sites/(?P<id>[^"]+)"><strong>(?P<name>[^<]+)</strong></a>', uploader_str)
-        (uploader, uploader_id) = (mobj.group('name'), mobj.group('id')) if mobj else (clean_html(uploader_str), None)
+        view_count = str_to_int(self._search_regex(
+            r'<meta itemprop="interactionCount" content="UserPlays:([0-9,]+)">',
+            webpage, 'view count', fatal=False))
+        like_count = str_to_int(self._search_regex(
+            r'<meta itemprop="interactionCount" content="UserLikes:([0-9,]+)">',
+            webpage, 'like count', fatal=False))
+        duration = parse_duration(self._html_search_meta('duration', webpage))
 
-        upload_date = None
-        view_count = None
-        duration = None
-        description = self._html_search_meta('description', webpage, 'description')
-        if description:
-            upload_date = self._search_regex(r'Published Date: (\d{2} [a-zA-Z]{3} \d{4})', description, 'upload date',
-                                             fatal=False)
-            if upload_date:
-                upload_date = unified_strdate(upload_date)
-            view_count = self._search_regex(r'Views: ([\d,\.]+)', description, 'view count', fatal=False)
-            if view_count:
-                view_count = str_to_int(view_count)
-            duration = parse_duration(self._search_regex(r'Length: (\d+m\d+s)', description, 'duration', fatal=False))
+        params_js = self._search_regex(
+            r'\$\.ajax\(url,\ opts\);\s*\}\s*\}\)\(([0-9,\[\] ]+)\)',
+            webpage, 'initialization parameters'
+        )
+        params = self._parse_json('[%s]' % params_js, video_id)
+        media_id = params[0]
+        sources = ['%s' % p for p in params[2]]
 
-        token_url = "http://tkn.4tube.com/{0}/desktop/{1}".format(media_id, "+".join(sources))
+        token_url = 'http://tkn.4tube.com/{0}/desktop/{1}'.format(
+            media_id, '+'.join(sources))
         headers = {
             b'Content-Type': b'application/x-www-form-urlencoded',
             b'Origin': b'http://www.4tube.com',
         }
         token_req = compat_urllib_request.Request(token_url, b'{}', headers)
         tokens = self._download_json(token_req, video_id)
-
         formats = [{
             'url': tokens[format]['token'],
             'format_id': format + 'p',
             'resolution': format + 'p',
             'quality': int(format),
         } for format in sources]
-
         self._sort_formats(formats)
 
         return {
             'id': video_id,
             'title': title,
             'formats': formats,
-            'thumbnail': thumbnail_url,
+            'categories': categories,
+            'thumbnail': thumbnail,
             'uploader': uploader,
             'uploader_id': uploader_id,
-            'upload_date': upload_date,
+            'timestamp': timestamp,
+            'like_count': like_count,
             'view_count': view_count,
             'duration': duration,
             'age_limit': 18,
-            'webpage_url': webpage_url,
         }

youtube_dl/extractor/gamestar.py

@@ -1,8 +1,6 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
@@ -29,9 +27,7 @@ class GameStarIE(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)
 
         og_title = self._og_search_title(webpage)

youtube_dl/extractor/generic.py

@@ -17,6 +17,7 @@ from ..utils import (
     ExtractorError,
     float_or_none,
     HEADRequest,
+    is_html,
     orderedSet,
     parse_xml,
     smuggle_url,
@@ -361,7 +362,7 @@ class GenericIE(InfoExtractor):
             'info_dict': {
                 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
                 'title': 'Zero Punctuation',
-                'description': 're:'
+                'description': 're:.*groundbreaking video review series.*'
             },
             'playlist_mincount': 11,
         },
@@ -488,6 +489,16 @@ class GenericIE(InfoExtractor):
                 'title': 'Jack Tips: 5 Steps to Permanent Gut Healing',
             }
         },
+        # Cinerama player
+        {
+            'url': 'http://www.abc.net.au/7.30/content/2015/s4164797.htm',
+            'info_dict': {
+                'id': '730m_DandD_1901_512k',
+                'ext': 'mp4',
+                'uploader': 'www.abc.net.au',
+                'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015',
+            }
+        }
     ]
 
     def report_following_redirect(self, new_url):
@@ -647,7 +658,7 @@ class GenericIE(InfoExtractor):
         # Maybe it's a direct link to a video?
         # Be careful not to download the whole thing!
         first_bytes = full_response.read(512)
-        if not re.match(r'^\s*<', first_bytes.decode('utf-8', 'replace')):
+        if not is_html(first_bytes):
             self._downloader.report_warning(
                 'URL could be a direct video link, returning it as such.')
             upload_date = unified_strdate(
@@ -1045,6 +1056,10 @@ class GenericIE(InfoExtractor):
                     \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
                         ["']?url["']?\s*:\s*["']([^"']+)["']
             ''', webpage))
+        if not found:
+            # Cinerama player
+            found = re.findall(
+                r"cinerama\.embedPlayer\(\s*\'[^']+\',\s*'([^']+)'", webpage)
         if not found:
             # Try to find twitter cards info
             found = filter_video(re.findall(

youtube_dl/extractor/hearthisat.py

@@ -0,0 +1,117 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_request,
+    compat_urlparse,
+)
+from ..utils import (
+    HEADRequest,
+    str_to_int,
+    urlencode_postdata,
+    urlhandle_detect_ext,
+)
+
+
+class HearThisAtIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?hearthis\.at/(?P<artist>[^/]+)/(?P<title>[A-Za-z0-9\-]+)/?$'
+    _PLAYLIST_URL = 'https://hearthis.at/playlist.php'
+    _TEST = {
+        'url': 'https://hearthis.at/moofi/dr-kreep',
+        'md5': 'ab6ec33c8fed6556029337c7885eb4e0',
+        'info_dict': {
+            'id': '150939',
+            'ext': 'wav',
+            'title': 'Moofi - Dr. Kreep',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'timestamp': 1421564134,
+            'description': 'Creepy Patch. Mutable Instruments Braids Vowel + Formant Mode.',
+            'upload_date': '20150118',
+            'comment_count': int,
+            'view_count': int,
+            'like_count': int,
+            'duration': 71,
+            'categories': ['Experimental'],
+        }
+    }
+
+    def _real_extract(self, url):
+        m = re.match(self._VALID_URL, url)
+        display_id = '{artist:s} - {title:s}'.format(**m.groupdict())
+
+        webpage = self._download_webpage(url, display_id)
+        track_id = self._search_regex(
+            r'intTrackId\s*=\s*(\d+)', webpage, 'track ID')
+
+        payload = urlencode_postdata({'tracks[]': track_id})
+        req = compat_urllib_request.Request(self._PLAYLIST_URL, payload)
+        req.add_header('Content-type', 'application/x-www-form-urlencoded')
+
+        track = self._download_json(req, track_id, 'Downloading playlist')[0]
+        title = '{artist:s} - {title:s}'.format(**track)
+
+        categories = None
+        if track.get('category'):
+            categories = [track['category']]
+
+        description = self._og_search_description(webpage)
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        meta_span = r'<span[^>]+class="%s".*?</i>([^<]+)</span>'
+        view_count = str_to_int(self._search_regex(
+            meta_span % 'plays_count', webpage, 'view count', fatal=False))
+        like_count = str_to_int(self._search_regex(
+            meta_span % 'likes_count', webpage, 'like count', fatal=False))
+        comment_count = str_to_int(self._search_regex(
+            meta_span % 'comment_count', webpage, 'comment count', fatal=False))
+        duration = str_to_int(self._search_regex(
+            r'data-length="(\d+)', webpage, 'duration', fatal=False))
+        timestamp = str_to_int(self._search_regex(
+            r'<span[^>]+class="calctime"[^>]+data-time="(\d+)', webpage, 'timestamp', fatal=False))
+
+        formats = []
+        mp3_url = self._search_regex(
+            r'(?s)<a class="player-link"\s+(?:[a-zA-Z0-9_:-]+="[^"]+"\s+)*?data-mp3="([^"]+)"',
+            webpage, 'mp3 URL', fatal=False)
+        if mp3_url:
+            formats.append({
+                'format_id': 'mp3',
+                'vcodec': 'none',
+                'acodec': 'mp3',
+                'url': mp3_url,
+            })
+        download_path = self._search_regex(
+            r'<a class="[^"]*download_fct[^"]*"\s+href="([^"]+)"',
+            webpage, 'download URL', default=None)
+        if download_path:
+            download_url = compat_urlparse.urljoin(url, download_path)
+            ext_req = HEADRequest(download_url)
+            ext_handle = self._request_webpage(
+                ext_req, display_id, note='Determining extension')
+            ext = urlhandle_detect_ext(ext_handle)
+            formats.append({
+                'format_id': 'download',
+                'vcodec': 'none',
+                'ext': ext,
+                'url': download_url,
+                'preference': 2,  # Usually better quality
+            })
+        self._sort_formats(formats)
+
+        return {
+            'id': track_id,
+            'display_id': display_id,
+            'title': title,
+            'formats': formats,
+            'thumbnail': thumbnail,
+            'description': description,
+            'duration': duration,
+            'timestamp': timestamp,
+            'view_count': view_count,
+            'comment_count': comment_count,
+            'like_count': like_count,
+            'categories': categories,
+        }

youtube_dl/extractor/krasview.py

@@ -2,18 +2,17 @@
 from __future__ import unicode_literals
 
 import json
-import re
 
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
-    unescapeHTML,
+    js_to_json,
 )
 
 
 class KrasViewIE(InfoExtractor):
     IE_DESC = 'Красвью'
-    _VALID_URL = r'https?://krasview\.ru/video/(?P<id>\d+)'
+    _VALID_URL = r'https?://krasview\.ru/(?:video|embed)/(?P<id>\d+)'
 
     _TEST = {
         'url': 'http://krasview.ru/video/512228',
@@ -29,20 +28,18 @@ class KrasViewIE(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)
 
-        flashvars = json.loads(self._search_regex(
-            r'flashvars\s*:\s*({.+?})\s*}\);', webpage, 'flashvars'))
+        flashvars = json.loads(js_to_json(self._search_regex(
+            r'video_Init\(({.+?})', webpage, 'flashvars')))
 
         video_url = flashvars['url']
-        title = unescapeHTML(flashvars['title'])
-        description = unescapeHTML(flashvars.get('subtitle') or self._og_search_description(webpage, default=None))
-        thumbnail = flashvars['image']
-        duration = int(flashvars['duration'])
-        filesize = int(flashvars['size'])
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage, default=None)
+        thumbnail = flashvars.get('image') or self._og_search_thumbnail(webpage)
+        duration = int_or_none(flashvars.get('duration'))
         width = int_or_none(self._og_search_property('video:width', webpage, 'video width'))
         height = int_or_none(self._og_search_property('video:height', webpage, 'video height'))
 
@@ -53,7 +50,6 @@ class KrasViewIE(InfoExtractor):
             'description': description,
             'thumbnail': thumbnail,
             'duration': duration,
-            'filesize': filesize,
             'width': width,
             'height': height,
         }

youtube_dl/extractor/liveleak.py

@@ -8,20 +8,20 @@ from ..utils import int_or_none
 
 
 class LiveLeakIE(InfoExtractor):
-    _VALID_URL = r'^(?:http://)?(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<video_id>[\w_]+)(?:.*)'
+    _VALID_URL = r'https?://(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<id>[\w_]+)(?:.*)'
     _TESTS = [{
         'url': 'http://www.liveleak.com/view?i=757_1364311680',
-        'md5': '0813c2430bea7a46bf13acf3406992f4',
+        'md5': '50f79e05ba149149c1b4ea961223d5b3',
         'info_dict': {
             'id': '757_1364311680',
-            'ext': 'mp4',
+            'ext': 'flv',
             'description': 'extremely bad day for this guy..!',
             'uploader': 'ljfriel2',
             'title': 'Most unlucky car accident'
         }
     }, {
         'url': 'http://www.liveleak.com/view?i=f93_1390833151',
-        'md5': 'd3f1367d14cc3c15bf24fbfbe04b9abf',
+        'md5': 'b13a29626183c9d33944e6a04f41aafc',
         'info_dict': {
             'id': 'f93_1390833151',
             'ext': 'mp4',
@@ -43,8 +43,7 @@ class LiveLeakIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('video_id')
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
         video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
@@ -81,9 +80,19 @@ class LiveLeakIE(InfoExtractor):
         sources = json.loads(sources_json)
 
         formats = [{
+            'format_id': '%s' % i,
             'format_note': s.get('label'),
             'url': s['file'],
-        } for s in sources]
+        } for i, s in enumerate(sources)]
+        for i, s in enumerate(sources):
+            orig_url = s['file'].replace('.h264_base.mp4', '')
+            if s['file'] != orig_url:
+                formats.append({
+                    'format_id': 'original-%s' % i,
+                    'format_note': s.get('label'),
+                    'url': orig_url,
+                    'preference': 1,
+                })
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/lnkgo.py

@@ -0,0 +1,124 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    js_to_json,
+    unified_strdate,
+)
+
+
+class LnkGoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?lnkgo\.alfa\.lt/visi\-video/(?P<show>[^/]+)/ziurek\-(?P<display_id>[A-Za-z0-9\-]+)'
+    _TESTS = [{
+        'url': 'http://lnkgo.alfa.lt/visi-video/yra-kaip-yra/ziurek-yra-kaip-yra-162',
+        'info_dict': {
+            'id': '46712',
+            'ext': 'mp4',
+            'title': 'Yra kaip yra',
+            'upload_date': '20150107',
+            'description': 'md5:d82a5e36b775b7048617f263a0e3475e',
+            'age_limit': 7,
+            'duration': 3019,
+            'thumbnail': 're:^https?://.*\.jpg$'
+        },
+        'params': {
+            'skip_download': True,  # HLS download
+        },
+    }, {
+        'url': 'http://lnkgo.alfa.lt/visi-video/aktualai-pratesimas/ziurek-nerdas-taiso-kompiuteri-2',
+        'info_dict': {
+            'id': '47289',
+            'ext': 'mp4',
+            'title': 'Nėrdas: Kompiuterio Valymas',
+            'upload_date': '20150113',
+            'description': 'md5:7352d113a242a808676ff17e69db6a69',
+            'age_limit': 18,
+            'duration': 346,
+            'thumbnail': 're:^https?://.*\.jpg$'
+        },
+        'params': {
+            'skip_download': True,  # HLS download
+        },
+    }]
+    _AGE_LIMITS = {
+        'N-7': 7,
+        'N-14': 14,
+        'S': 18,
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('display_id')
+
+        webpage = self._download_webpage(
+            url, display_id, 'Downloading player webpage')
+
+        video_id = self._search_regex(
+            r'data-ep="([^"]+)"', webpage, 'video ID')
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage)
+
+        thumbnail_w = int_or_none(
+            self._og_search_property('image:width', webpage, 'thumbnail width', fatal=False))
+        thumbnail_h = int_or_none(
+            self._og_search_property('image:height', webpage, 'thumbnail height', fatal=False))
+        thumbnail = {
+            'url': self._og_search_thumbnail(webpage),
+        }
+        if thumbnail_w and thumbnail_h:
+            thumbnail.update({
+                'width': thumbnail_w,
+                'height': thumbnail_h,
+            })
+
+        upload_date = unified_strdate(self._search_regex(
+            r'class="meta-item\sair-time">.*?<strong>([^<]+)</strong>', webpage, 'upload date', fatal=False))
+        duration = int_or_none(self._search_regex(
+            r'VideoDuration = "([^"]+)"', webpage, 'duration', fatal=False))
+
+        pg_rating = self._search_regex(
+            r'pgrating="([^"]+)"', webpage, 'PG rating', fatal=False, default='')
+        age_limit = self._AGE_LIMITS.get(pg_rating.upper(), 0)
+
+        sources_js = self._search_regex(
+            r'(?s)sources:\s(\[.*?\]),', webpage, 'sources')
+        sources = self._parse_json(
+            sources_js, video_id, transform_source=js_to_json)
+
+        formats = []
+        for source in sources:
+            if source.get('provider') == 'rtmp':
+                m = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<play_path>.+)$', source['file'])
+                if not m:
+                    continue
+                formats.append({
+                    'format_id': 'rtmp',
+                    'ext': 'flv',
+                    'url': m.group('url'),
+                    'play_path': m.group('play_path'),
+                    'page_url': url,
+                })
+            elif source.get('file').endswith('.m3u8'):
+                formats.append({
+                    'format_id': 'hls',
+                    'ext': source.get('type', 'mp4'),
+                    'url': source['file'],
+                })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'formats': formats,
+            'thumbnails': [thumbnail],
+            'duration': duration,
+            'description': description,
+            'age_limit': age_limit,
+            'upload_date': upload_date,
+        }

youtube_dl/extractor/lynda.py

@@ -85,6 +85,7 @@ class LyndaIE(SubtitlesInfoExtractor):
                 } for format_id, video_url in prioritized_streams['0'].items()
             ])
 
+        self._check_formats(formats, video_id)
         self._sort_formats(formats)
 
         if self._downloader.params.get('listsubtitles', False):

youtube_dl/extractor/mtv.py

@@ -53,7 +53,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
         webpage_url = self._MOBILE_TEMPLATE % mtvn_id
         req = compat_urllib_request.Request(webpage_url)
         # Otherwise we get a webpage that would execute some javascript
-        req.add_header('Youtubedl-user-agent', 'curl/7')
+        req.add_header('User-Agent', 'curl/7')
         webpage = self._download_webpage(req, mtvn_id,
                                          'Downloading mobile page')
         metrics_url = unescapeHTML(self._search_regex(r'<a href="(http://metrics.+?)"', webpage, 'url'))

youtube_dl/extractor/nbc.py

@@ -6,6 +6,7 @@ import json
 from .common import InfoExtractor
 from ..compat import (
     compat_str,
+    compat_HTTPError,
 )
 from ..utils import (
     ExtractorError,
@@ -78,6 +79,16 @@ class NBCNewsIE(InfoExtractor):
             },
             'add_ie': ['ThePlatform'],
         },
+        {
+            'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
+            'md5': 'fdbf39ab73a72df5896b6234ff98518a',
+            'info_dict': {
+                'id': 'Wjf9EDR3A_60',
+                'ext': 'mp4',
+                'title': 'FULL EPISODE: Family Business',
+                'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
+            },
+        },
     ]
 
     def _real_extract(self, url):
@@ -115,10 +126,19 @@ class NBCNewsIE(InfoExtractor):
                 if not base_url:
                     continue
                 playlist_url = base_url + '?form=MPXNBCNewsAPI'
-                all_videos = self._download_json(playlist_url, title)['videos']
 
                 try:
-                    info = next(v for v in all_videos if v['mpxId'] == mpxid)
+                    all_videos = self._download_json(playlist_url, title)
+                except ExtractorError as ee:
+                    if isinstance(ee.cause, compat_HTTPError):
+                        continue
+                    raise
+
+                if not all_videos or 'videos' not in all_videos:
+                    continue
+
+                try:
+                    info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid)
                     break
                 except StopIteration:
                     continue

youtube_dl/extractor/ndtv.py

@@ -27,9 +27,7 @@ class NDTVIE(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)
 
         filename = self._search_regex(

youtube_dl/extractor/npo.py

@@ -1,19 +1,26 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
-    unified_strdate,
+    fix_xml_ampersands,
     parse_duration,
     qualities,
     strip_jsonp,
+    unified_strdate,
     url_basename,
-    fix_xml_ampersands,
 )
 
 
-class NPOIE(InfoExtractor):
+class NPOBaseIE(InfoExtractor):
+    def _get_token(self, video_id):
+        token_page = self._download_webpage(
+            'http://ida.omroep.nl/npoplayer/i.js',
+            video_id, note='Downloading token')
+        return self._search_regex(
+            r'npoplayer\.token = "(.+?)"', token_page, 'token')
+
+
+class NPOIE(NPOBaseIE):
     IE_NAME = 'npo.nl'
     _VALID_URL = r'https?://www\.npo\.nl/[^/]+/[^/]+/(?P<id>[^/?]+)'
 
@@ -67,11 +74,20 @@ class NPOIE(InfoExtractor):
                 'skip_download': True,
             }
         },
+        # non asf in streams
+        {
+            'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771',
+            'md5': 'b3da13de374cbe2d5332a7e910bef97f',
+            'info_dict': {
+                'id': 'WO_NOS_762771',
+                'ext': 'mp4',
+                'title': 'Hoe gaat Europa verder na Parijs?',
+            },
+        },
     ]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
         return self._get_info(video_id)
 
     def _get_info(self, video_id):
@@ -81,12 +97,8 @@ class NPOIE(InfoExtractor):
             # We have to remove the javascript callback
             transform_source=strip_jsonp,
         )
-        token_page = self._download_webpage(
-            'http://ida.omroep.nl/npoplayer/i.js',
-            video_id,
-            note='Downloading token'
-        )
-        token = self._search_regex(r'npoplayer\.token = "(.+?)"', token_page, 'token')
+
+        token = self._get_token(video_id)
 
         formats = []
 
@@ -125,6 +137,12 @@ class NPOIE(InfoExtractor):
                 stream_url = stream.get('url')
                 if not stream_url:
                     continue
+                if '.asf' not in stream_url:
+                    formats.append({
+                        'url': stream_url,
+                        'quality': stream.get('kwaliteit'),
+                    })
+                    continue
                 asx = self._download_xml(
                     stream_url, video_id,
                     'Downloading stream %d ASX playlist' % i,
@@ -154,6 +172,83 @@ class NPOIE(InfoExtractor):
         }
 
 
+class NPOLiveIE(NPOBaseIE):
+    IE_NAME = 'npo.nl:live'
+    _VALID_URL = r'https?://www\.npo\.nl/live/(?P<id>.+)'
+
+    _TEST = {
+        'url': 'http://www.npo.nl/live/npo-1',
+        'info_dict': {
+            'id': 'LI_NEDERLAND1_136692',
+            'display_id': 'npo-1',
+            'ext': 'mp4',
+            'title': 're:^Nederland 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            'description': 'Livestream',
+            'is_live': True,
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        live_id = self._search_regex(
+            r'data-prid="([^"]+)"', webpage, 'live id')
+
+        metadata = self._download_json(
+            'http://e.omroep.nl/metadata/%s' % live_id,
+            display_id, transform_source=strip_jsonp)
+
+        token = self._get_token(display_id)
+
+        formats = []
+
+        streams = metadata.get('streams')
+        if streams:
+            for stream in streams:
+                stream_type = stream.get('type').lower()
+                if stream_type == 'ss':
+                    continue
+                stream_info = self._download_json(
+                    'http://ida.omroep.nl/aapi/?stream=%s&token=%s&type=jsonp'
+                    % (stream.get('url'), token),
+                    display_id, 'Downloading %s JSON' % stream_type)
+                if stream_info.get('error_code', 0) or stream_info.get('errorcode', 0):
+                    continue
+                stream_url = self._download_json(
+                    stream_info['stream'], display_id,
+                    'Downloading %s URL' % stream_type,
+                    transform_source=strip_jsonp)
+                if stream_type == 'hds':
+                    f4m_formats = self._extract_f4m_formats(stream_url, display_id)
+                    # f4m downloader downloads only piece of live stream
+                    for f4m_format in f4m_formats:
+                        f4m_format['preference'] = -1
+                    formats.extend(f4m_formats)
+                elif stream_type == 'hls':
+                    formats.extend(self._extract_m3u8_formats(stream_url, display_id, 'mp4'))
+                else:
+                    formats.append({
+                        'url': stream_url,
+                    })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': live_id,
+            'display_id': display_id,
+            'title': self._live_title(metadata['titel']),
+            'description': metadata['info'],
+            'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
+            'formats': formats,
+            'is_live': True,
+        }
+
+
 class TegenlichtVproIE(NPOIE):
     IE_NAME = 'tegenlicht.vpro.nl'
     _VALID_URL = r'https?://tegenlicht\.vpro\.nl/afleveringen/.*?'

youtube_dl/extractor/pornhub.py

@@ -10,6 +10,7 @@ from ..compat import (
     compat_urllib_request,
 )
 from ..utils import (
+    ExtractorError,
     str_to_int,
 )
 from ..aes import (
@@ -44,6 +45,15 @@ class PornHubIE(InfoExtractor):
         req.add_header('Cookie', 'age_verified=1')
         webpage = self._download_webpage(req, video_id)
 
+        error_msg = self._html_search_regex(
+            r'(?s)<div class="userMessageSection[^"]*".*?>(.*?)</div>',
+            webpage, 'error message', default=None)
+        if error_msg:
+            error_msg = re.sub(r'\s+', ' ', error_msg)
+            raise ExtractorError(
+                'PornHub said: %s' % error_msg,
+                expected=True, video_id=video_id)
+
         video_title = self._html_search_regex(r'<h1 [^>]+>([^<]+)', webpage, 'title')
         video_uploader = self._html_search_regex(
             r'(?s)From:&nbsp;.+?<(?:a href="/users/|a href="/channels/|<span class="username)[^>]+>(.+?)<',

youtube_dl/extractor/rtl2.py

@@ -0,0 +1,72 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+
+class RTL2IE(InfoExtractor):
+    _VALID_URL = r'http?://(?:www\.)?rtl2\.de/[^?#]*?/(?P<id>[^?#/]*?)(?:$|/(?:$|[?#]))'
+    _TESTS = [{
+        'url': 'http://www.rtl2.de/sendung/grip-das-motormagazin/folge/folge-203-0',
+        'md5': 'bfcc179030535b08dc2b36b469b5adc7',
+        'info_dict': {
+            'id': 'folge-203-0',
+            'ext': 'f4v',
+            'title': 'GRIP sucht den Sommerkönig',
+            'description': 'Matthias, Det und Helge treten gegeneinander an.'
+        },
+    }, {
+        'url': 'http://www.rtl2.de/sendung/koeln-50667/video/5512-anna/21040-anna-erwischt-alex/',
+        'md5': 'ffcd517d2805b57ce11a58a2980c2b02',
+        'info_dict': {
+            'id': '21040-anna-erwischt-alex',
+            'ext': 'mp4',
+            'title': 'Anna erwischt Alex!',
+            'description': 'Anna ist Alex\' Tochter bei Köln 50667.'
+        },
+    }]
+
+    def _real_extract(self, url):
+        # Some rtl2 urls have no slash at the end, so append it.
+        if not url.endswith('/'):
+            url += '/'
+
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        vico_id = self._html_search_regex(
+            r'vico_id\s*:\s*([0-9]+)', webpage, 'vico_id')
+        vivi_id = self._html_search_regex(
+            r'vivi_id\s*:\s*([0-9]+)', webpage, 'vivi_id')
+        info_url = 'http://www.rtl2.de/video/php/get_video.php?vico_id=' + vico_id + '&vivi_id=' + vivi_id
+        webpage = self._download_webpage(info_url, '')
+
+        info = self._download_json(info_url, video_id)
+        video_info = info['video']
+        title = video_info['titel']
+        description = video_info.get('beschreibung')
+        thumbnail = video_info.get('image')
+
+        download_url = video_info['streamurl']
+        download_url = download_url.replace('\\', '')
+        stream_url = 'mp4:' + self._html_search_regex(r'ondemand/(.*)', download_url, 'stream URL')
+        rtmp_conn = ["S:connect", "O:1", "NS:pageUrl:" + url, "NB:fpad:0", "NN:videoFunction:1", "O:0"]
+
+        formats = [{
+            'url': download_url,
+            'play_path': stream_url,
+            'player_url': 'http://www.rtl2.de/flashplayer/vipo_player.swf',
+            'page_url': url,
+            'flash_version': 'LNX 11,2,202,429',
+            'rtmp_conn': rtmp_conn,
+            'no_resume': True,
+        }]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'description': description,
+            'formats': formats,
+        }

youtube_dl/extractor/smotri.py

@@ -90,6 +90,21 @@ class SmotriIE(InfoExtractor):
             },
             'skip': 'Video is not approved by moderator',
         },
+        # not approved by moderator, but available
+        {
+            'url': 'http://smotri.com/video/view/?id=v28888533b73',
+            'md5': 'f44bc7adac90af518ef1ecf04893bb34',
+            'info_dict': {
+                'id': 'v28888533b73',
+                'ext': 'mp4',
+                'title': 'Russian Spies Killed By ISIL Child Soldier',
+                'uploader': 'Mopeder',
+                'uploader_id': 'mopeder',
+                'duration': 71,
+                'thumbnail': 'http://frame9.loadup.ru/d7/32/2888853.2.3.jpg',
+                'upload_date': '20150114',
+            },
+        },
         # swf player
         {
             'url': 'http://pics.smotri.com/scrubber_custom8.swf?file=v9188090500',
@@ -146,13 +161,16 @@ class SmotriIE(InfoExtractor):
 
         video = self._download_json(request, video_id, 'Downloading video JSON')
 
-        if video.get('_moderate_no') or not video.get('moderated'):
-            raise ExtractorError('Video %s has not been approved by moderator' % video_id, expected=True)
-
-        if video.get('error'):
-            raise ExtractorError('Video %s does not exist' % video_id, expected=True)
-
         video_url = video.get('_vidURL') or video.get('_vidURL_mp4')
+
+        if not video_url:
+            if video.get('_moderate_no') or not video.get('moderated'):
+                raise ExtractorError(
+                    'Video %s has not been approved by moderator' % video_id, expected=True)
+
+            if video.get('error'):
+                raise ExtractorError('Video %s does not exist' % video_id, expected=True)
+
         title = video['title']
         thumbnail = video['_imgURL']
         upload_date = unified_strdate(video['added'])

youtube_dl/extractor/spiegel.py

@@ -60,23 +60,24 @@ class SpiegelIE(InfoExtractor):
         xml_url = base_url + video_id + '.xml'
         idoc = self._download_xml(xml_url, video_id)
 
-        formats = [
-            {
-                'format_id': n.tag.rpartition('type')[2],
-                'url': base_url + n.find('./filename').text,
-                'width': int(n.find('./width').text),
-                'height': int(n.find('./height').text),
-                'abr': int(n.find('./audiobitrate').text),
-                'vbr': int(n.find('./videobitrate').text),
-                'vcodec': n.find('./codec').text,
-                'acodec': 'MP4A',
-            }
-            for n in list(idoc)
-            # Blacklist type 6, it's extremely LQ and not available on the same server
-            if n.tag.startswith('type') and n.tag != 'type6'
-        ]
+        formats = []
+        for n in list(idoc):
+            if n.tag.startswith('type') and n.tag != 'type6':
+                format_id = n.tag.rpartition('type')[2]
+                video_url = base_url + n.find('./filename').text
+                formats.append({
+                    'format_id': format_id,
+                    'url': video_url,
+                    'width': int(n.find('./width').text),
+                    'height': int(n.find('./height').text),
+                    'abr': int(n.find('./audiobitrate').text),
+                    'vbr': int(n.find('./videobitrate').text),
+                    'vcodec': n.find('./codec').text,
+                    'acodec': 'MP4A',
+                })
         duration = float(idoc[0].findall('./duration')[0].text)
 
+        self._check_formats(formats, video_id)
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/streetvoice.py

@@ -0,0 +1,51 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import unified_strdate
+
+
+class StreetVoiceIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:.+?\.)?streetvoice\.com/[^/]+/songs/(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'http://streetvoice.com/skippylu/songs/94440/',
+        'md5': '15974627fc01a29e492c98593c2fd472',
+        'info_dict': {
+            'id': '94440',
+            'ext': 'mp3',
+            'filesize': 4167053,
+            'title': '輸',
+            'description': 'Crispy脆樂團 - 輸',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 260,
+            'upload_date': '20091018',
+            'uploader': 'Crispy脆樂團',
+            'uploader_id': '627810',
+        }
+    }, {
+        'url': 'http://tw.streetvoice.com/skippylu/songs/94440/',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        song_id = self._match_id(url)
+
+        song = self._download_json(
+            'http://streetvoice.com/music/api/song/%s' % song_id, song_id)
+
+        title = song['name']
+        author = song['musician']['name']
+
+        return {
+            'id': song_id,
+            'url': song['file'],
+            'filesize': song.get('size'),
+            'title': title,
+            'description': '%s - %s' % (author, title),
+            'thumbnail': self._proto_relative_url(song.get('image'), 'http:'),
+            'duration': song.get('length'),
+            'upload_date': unified_strdate(song.get('created_at')),
+            'uploader': author,
+            'uploader_id': compat_str(song['musician']['id']),
+        }

youtube_dl/extractor/testtube.py

@@ -0,0 +1,72 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    qualities,
+)
+
+
+class TestTubeIE(InfoExtractor):
+    _VALID_URL = r'https?://testtube\.com/[^/?#]+/(?P<id>[^/?#]+)'
+    _TESTS = [{
+        'url': 'https://testtube.com/dnews/5-weird-ways-plants-can-eat-animals?utm_source=FB&utm_medium=DNews&utm_campaign=DNewsSocial',
+        'info_dict': {
+            'id': '60163',
+            'display_id': '5-weird-ways-plants-can-eat-animals',
+            'duration': 275,
+            'ext': 'mp4',
+            'title': '5 Weird Ways Plants Can Eat Animals',
+            'description': 'Why have some plants evolved to eat meat?',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'DNews',
+            'uploader_id': 'dnews',
+        },
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+        video_id = self._search_regex(
+            r"player\.loadRevision3Item\('video_id',\s*([0-9]+)\);",
+            webpage, 'video ID')
+
+        all_info = self._download_json(
+            'https://testtube.com/api/getPlaylist.json?api_key=ba9c741bce1b9d8e3defcc22193f3651b8867e62&codecs=h264,vp8,theora&video_id=%s' % video_id,
+            video_id)
+        info = all_info['items'][0]
+
+        formats = []
+        for vcodec, fdatas in info['media'].items():
+            for name, fdata in fdatas.items():
+                formats.append({
+                    'format_id': '%s-%s' % (vcodec, name),
+                    'url': fdata['url'],
+                    'vcodec': vcodec,
+                    'tbr': fdata.get('bitrate'),
+                })
+        self._sort_formats(formats)
+
+        duration = int_or_none(info.get('duration'))
+        images = info.get('images')
+        thumbnails = None
+        preference = qualities(['mini', 'small', 'medium', 'large'])
+        if images:
+            thumbnails = [{
+                'id': thumbnail_id,
+                'url': img_url,
+                'preference': preference(thumbnail_id)
+            } for thumbnail_id, img_url in images.items()]
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': info['title'],
+            'description': info.get('summary'),
+            'thumbnails': thumbnails,
+            'uploader': info.get('show', {}).get('name'),
+            'uploader_id': info.get('show', {}).get('slug'),
+            'duration': duration,
+            'formats': formats,
+        }

youtube_dl/extractor/tinypic.py

@@ -9,17 +9,23 @@ from ..utils import ExtractorError
 class TinyPicIE(InfoExtractor):
     IE_NAME = 'tinypic'
     IE_DESC = 'tinypic.com videos'
-    _VALID_URL = r'http://tinypic\.com/player\.php\?v=(?P<id>[^&]+)&s=\d+'
+    _VALID_URL = r'http://(?:.+?\.)?tinypic\.com/player\.php\?v=(?P<id>[^&]+)&s=\d+'
 
-    _TEST = {
-        'url': 'http://tinypic.com/player.php?v=6xw7tc%3E&s=5#.UtqZmbRFCM8',
-        'md5': '609b74432465364e72727ebc6203f044',
-        'info_dict': {
-            'id': '6xw7tc',
-            'ext': 'flv',
-            'title': 'shadow phenomenon weird',
+    _TESTS = [
+        {
+            'url': 'http://tinypic.com/player.php?v=6xw7tc%3E&s=5#.UtqZmbRFCM8',
+            'md5': '609b74432465364e72727ebc6203f044',
+            'info_dict': {
+                'id': '6xw7tc',
+                'ext': 'flv',
+                'title': 'shadow phenomenon weird',
+            },
+        },
+        {
+            'url': 'http://de.tinypic.com/player.php?v=dy90yh&s=8',
+            'only_matching': True,
         }
-    }
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/tvp.py

@@ -12,61 +12,59 @@ class TvpIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://vod.tvp.pl/filmy-fabularne/filmy-za-darmo/ogniem-i-mieczem/wideo/odc-2/4278035',
+        'md5': 'cdd98303338b8a7f7abab5cd14092bf2',
         'info_dict': {
             'id': '4278035',
             'ext': 'wmv',
             'title': 'Ogniem i mieczem, odc. 2',
-            'description': 'Bohun dowiaduje się o złamaniu przez kniahinię danego mu słowa i wyrusza do Rozłogów. Helenie w ostatniej chwili udaje się uciec dzięki pomocy Zagłoby.',
         },
     }, {
         'url': 'http://vod.tvp.pl/seriale/obyczajowe/czas-honoru/sezon-1-1-13/i-seria-odc-13/194536',
+        'md5': '8aa518c15e5cc32dfe8db400dc921fbb',
         'info_dict': {
             'id': '194536',
             'ext': 'mp4',
             'title': 'Czas honoru, I seria – odc. 13',
-            #  'description': 'WŁADEK\nCzesław prosi Marię o dostarczenie Władkowi zarazki tyfusu. Jeśli zachoruje zostanie przewieziony do szpitala skąd łatwiej będzie go odbić. Czy matka zdecyduje się zarazić syna? Karol odwiedza Wandę przyznaje się, że ją oszukiwał, ale ostrzega też, że grozi jej aresztowanie i nalega, żeby wyjechała z Warszawy. Czy dziewczyna zdecyduje się znów oddalić od ukochanego? Rozpoczyna się akcja odbicia Władka.',
         },
     }, {
         'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176',
+        'md5': 'c3b15ed1af288131115ff17a17c19dda',
         'info_dict': {
             'id': '17916176',
             'ext': 'mp4',
             'title': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata',
         },
-        'params': {
-            # m3u8 download
-            'skip_download': 'true',
-        },
     }, {
         'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272',
+        'md5': 'c3b15ed1af288131115ff17a17c19dda',
         'info_dict': {
             'id': '17834272',
             'ext': 'mp4',
             'title': 'Na sygnale, odc. 39',
-            'description': 'Ekipa Wiktora ratuje młodą matkę, która spadła ze schodów trzymając na rękach noworodka. Okazuje się, że dziewczyna jest surogatką, a biologiczni rodzice dziecka próbują zmusić ją do oddania synka…',
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': 'true',
         },
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+
         webpage = self._download_webpage(
             'http://www.tvp.pl/sess/tvplayer.php?object_id=%s' % video_id, video_id)
 
-        title = self._og_search_title(webpage)
-        series = self._search_regex(
-            r'{name:\s*([\'"])SeriesTitle\1,\s*value:\s*\1(?P<series>.*?)\1},',
+        title = self._search_regex(
+            r'name\s*:\s*([\'"])Title\1\s*,\s*value\s*:\s*\1(?P<title>.+?)\1',
+            webpage, 'title', group='title')
+        series_title = self._search_regex(
+            r'name\s*:\s*([\'"])SeriesTitle\1\s*,\s*value\s*:\s*\1(?P<series>.+?)\1',
             webpage, 'series', group='series', default=None)
-        if series is not None and series not in title:
-            title = '%s, %s' % (series, title)
-        description = self._og_search_description(webpage, default=None)
+        if series_title:
+            title = '%s, %s' % (series_title, title)
+
+        thumbnail = self._search_regex(
+            r"poster\s*:\s*'([^']+)'", webpage, 'thumbnail', default=None)
 
         video_url = self._search_regex(
             r'0:{src:([\'"])(?P<url>.*?)\1', webpage, 'formats', group='url', default=None)
-        if video_url is None:
+        if not video_url:
             video_url = self._download_json(
                 'http://www.tvp.pl/pub/stat/videofileinfo?video_id=%s' % video_id,
                 video_id)['video_url']
@@ -89,8 +87,7 @@ class TvpIE(InfoExtractor):
         return {
             'id': video_id,
             'title': title,
-            'thumbnail': self._og_search_thumbnail(webpage),
-            'description': description,
+            'thumbnail': thumbnail,
             'formats': formats,
         }
 

youtube_dl/extractor/twitch.py

@@ -3,9 +3,11 @@ from __future__ import unicode_literals
 
 import itertools
 import re
+import random
 
 from .common import InfoExtractor
 from ..compat import (
+    compat_str,
     compat_urllib_parse,
     compat_urllib_request,
 )
@@ -15,44 +17,12 @@ from ..utils import (
 )
 
 
-class TwitchIE(InfoExtractor):
-    # TODO: One broadcast may be split into multiple videos. The key
-    # 'broadcast_id' is the same for all parts, and 'broadcast_part'
-    # starts at 1 and increases. Can we treat all parts as one video?
-    _VALID_URL = r"""(?x)^(?:http://)?(?:www\.)?twitch\.tv/
-        (?:
-            (?P<channelid>[^/]+)|
-            (?:(?:[^/]+)/v/(?P<vodid>[^/]+))|
-            (?:(?:[^/]+)/b/(?P<videoid>[^/]+))|
-            (?:(?:[^/]+)/c/(?P<chapterid>[^/]+))
-        )
-        /?(?:\#.*)?$
-        """
-    _PAGE_LIMIT = 100
+class TwitchBaseIE(InfoExtractor):
+    _VALID_URL_BASE = r'https?://(?:www\.)?twitch\.tv'
+
     _API_BASE = 'https://api.twitch.tv'
+    _USHER_BASE = 'http://usher.twitch.tv'
     _LOGIN_URL = 'https://secure.twitch.tv/user/login'
-    _TESTS = [{
-        'url': 'http://www.twitch.tv/riotgames/b/577357806',
-        'info_dict': {
-            'id': 'a577357806',
-            'title': 'Worlds Semifinals - Star Horn Royal Club vs. OMG',
-        },
-        'playlist_mincount': 12,
-    }, {
-        'url': 'http://www.twitch.tv/acracingleague/c/5285812',
-        'info_dict': {
-            'id': 'c5285812',
-            'title': 'ACRL Off Season - Sports Cars @ Nordschleife',
-        },
-        'playlist_mincount': 3,
-    }, {
-        'url': 'http://www.twitch.tv/vanillatv',
-        'info_dict': {
-            'id': 'vanillatv',
-            'title': 'VanillaTV',
-        },
-        'playlist_mincount': 412,
-    }]
 
     def _handle_error(self, response):
         if not isinstance(response, dict):
@@ -64,71 +34,10 @@ class TwitchIE(InfoExtractor):
                 expected=True)
 
     def _download_json(self, url, video_id, note='Downloading JSON metadata'):
-        response = super(TwitchIE, self)._download_json(url, video_id, note)
+        response = super(TwitchBaseIE, self)._download_json(url, video_id, note)
         self._handle_error(response)
         return response
 
-    def _extract_media(self, item, item_id):
-        ITEMS = {
-            'a': 'video',
-            'v': 'vod',
-            'c': 'chapter',
-        }
-        info = self._extract_info(self._download_json(
-            '%s/kraken/videos/%s%s' % (self._API_BASE, item, item_id), item_id,
-            'Downloading %s info JSON' % ITEMS[item]))
-
-        if item == 'v':
-            access_token = self._download_json(
-                '%s/api/vods/%s/access_token' % (self._API_BASE, item_id), item_id,
-                'Downloading %s access token' % ITEMS[item])
-            formats = self._extract_m3u8_formats(
-                'http://usher.twitch.tv/vod/%s?nauth=%s&nauthsig=%s'
-                % (item_id, access_token['token'], access_token['sig']),
-                item_id, 'mp4')
-            info['formats'] = formats
-            return info
-
-        response = self._download_json(
-            '%s/api/videos/%s%s' % (self._API_BASE, item, item_id), item_id,
-            'Downloading %s playlist JSON' % ITEMS[item])
-        entries = []
-        chunks = response['chunks']
-        qualities = list(chunks.keys())
-        for num, fragment in enumerate(zip(*chunks.values()), start=1):
-            formats = []
-            for fmt_num, fragment_fmt in enumerate(fragment):
-                format_id = qualities[fmt_num]
-                fmt = {
-                    'url': fragment_fmt['url'],
-                    'format_id': format_id,
-                    'quality': 1 if format_id == 'live' else 0,
-                }
-                m = re.search(r'^(?P<height>\d+)[Pp]', format_id)
-                if m:
-                    fmt['height'] = int(m.group('height'))
-                formats.append(fmt)
-            self._sort_formats(formats)
-            entry = dict(info)
-            entry['id'] = '%s_%d' % (entry['id'], num)
-            entry['title'] = '%s part %d' % (entry['title'], num)
-            entry['formats'] = formats
-            entries.append(entry)
-        return self.playlist_result(entries, info['id'], info['title'])
-
-    def _extract_info(self, info):
-        return {
-            'id': info['_id'],
-            'title': info['title'],
-            'description': info['description'],
-            'duration': info['length'],
-            'thumbnail': info['preview'],
-            'uploader': info['channel']['display_name'],
-            'uploader_id': info['channel']['name'],
-            'timestamp': parse_iso8601(info['recorded_at']),
-            'view_count': info['views'],
-        }
-
     def _real_initialize(self):
         self._login()
 
@@ -167,81 +76,307 @@ class TwitchIE(InfoExtractor):
             raise ExtractorError(
                 'Unable to login: %s' % m.group('msg').strip(), expected=True)
 
+
+class TwitchItemBaseIE(TwitchBaseIE):
+    def _download_info(self, item, item_id):
+        return self._extract_info(self._download_json(
+            '%s/kraken/videos/%s%s' % (self._API_BASE, item, item_id), item_id,
+            'Downloading %s info JSON' % self._ITEM_TYPE))
+
+    def _extract_media(self, item_id):
+        info = self._download_info(self._ITEM_SHORTCUT, item_id)
+        response = self._download_json(
+            '%s/api/videos/%s%s' % (self._API_BASE, self._ITEM_SHORTCUT, item_id), item_id,
+            'Downloading %s playlist JSON' % self._ITEM_TYPE)
+        entries = []
+        chunks = response['chunks']
+        qualities = list(chunks.keys())
+        for num, fragment in enumerate(zip(*chunks.values()), start=1):
+            formats = []
+            for fmt_num, fragment_fmt in enumerate(fragment):
+                format_id = qualities[fmt_num]
+                fmt = {
+                    'url': fragment_fmt['url'],
+                    'format_id': format_id,
+                    'quality': 1 if format_id == 'live' else 0,
+                }
+                m = re.search(r'^(?P<height>\d+)[Pp]', format_id)
+                if m:
+                    fmt['height'] = int(m.group('height'))
+                formats.append(fmt)
+            self._sort_formats(formats)
+            entry = dict(info)
+            entry['id'] = '%s_%d' % (entry['id'], num)
+            entry['title'] = '%s part %d' % (entry['title'], num)
+            entry['formats'] = formats
+            entries.append(entry)
+        return self.playlist_result(entries, info['id'], info['title'])
+
+    def _extract_info(self, info):
+        return {
+            'id': info['_id'],
+            'title': info['title'],
+            'description': info['description'],
+            'duration': info['length'],
+            'thumbnail': info['preview'],
+            'uploader': info['channel']['display_name'],
+            'uploader_id': info['channel']['name'],
+            'timestamp': parse_iso8601(info['recorded_at']),
+            'view_count': info['views'],
+        }
+
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        if mobj.group('chapterid'):
-            return self._extract_media('c', mobj.group('chapterid'))
+        return self._extract_media(self._match_id(url))
 
-            """
-            webpage = self._download_webpage(url, chapter_id)
-            m = re.search(r'PP\.archive_id = "([0-9]+)";', webpage)
+
+class TwitchVideoIE(TwitchItemBaseIE):
+    IE_NAME = 'twitch:video'
+    _VALID_URL = r'%s/[^/]+/b/(?P<id>[^/]+)' % TwitchBaseIE._VALID_URL_BASE
+    _ITEM_TYPE = 'video'
+    _ITEM_SHORTCUT = 'a'
+
+    _TEST = {
+        'url': 'http://www.twitch.tv/riotgames/b/577357806',
+        'info_dict': {
+            'id': 'a577357806',
+            'title': 'Worlds Semifinals - Star Horn Royal Club vs. OMG',
+        },
+        'playlist_mincount': 12,
+    }
+
+
+class TwitchChapterIE(TwitchItemBaseIE):
+    IE_NAME = 'twitch:chapter'
+    _VALID_URL = r'%s/[^/]+/c/(?P<id>[^/]+)' % TwitchBaseIE._VALID_URL_BASE
+    _ITEM_TYPE = 'chapter'
+    _ITEM_SHORTCUT = 'c'
+
+    _TESTS = [{
+        'url': 'http://www.twitch.tv/acracingleague/c/5285812',
+        'info_dict': {
+            'id': 'c5285812',
+            'title': 'ACRL Off Season - Sports Cars @ Nordschleife',
+        },
+        'playlist_mincount': 3,
+    }, {
+        'url': 'http://www.twitch.tv/tsm_theoddone/c/2349361',
+        'only_matching': True,
+    }]
+
+
+class TwitchVodIE(TwitchItemBaseIE):
+    IE_NAME = 'twitch:vod'
+    _VALID_URL = r'%s/[^/]+/v/(?P<id>[^/]+)' % TwitchBaseIE._VALID_URL_BASE
+    _ITEM_TYPE = 'vod'
+    _ITEM_SHORTCUT = 'v'
+
+    _TEST = {
+        'url': 'http://www.twitch.tv/ksptv/v/3622000',
+        'info_dict': {
+            'id': 'v3622000',
+            'ext': 'mp4',
+            'title': '''KSPTV: Squadcast: "Everyone's on vacation so here's Dahud" Edition!''',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 6951,
+            'timestamp': 1419028564,
+            'upload_date': '20141219',
+            'uploader': 'KSPTV',
+            'uploader_id': 'ksptv',
+            'view_count': int,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        item_id = self._match_id(url)
+        info = self._download_info(self._ITEM_SHORTCUT, item_id)
+        access_token = self._download_json(
+            '%s/api/vods/%s/access_token' % (self._API_BASE, item_id), item_id,
+            'Downloading %s access token' % self._ITEM_TYPE)
+        formats = self._extract_m3u8_formats(
+            '%s/vod/%s?nauth=%s&nauthsig=%s'
+            % (self._USHER_BASE, item_id, access_token['token'], access_token['sig']),
+            item_id, 'mp4')
+        info['formats'] = formats
+        return info
+
+
+class TwitchPlaylistBaseIE(TwitchBaseIE):
+    _PLAYLIST_URL = '%s/kraken/channels/%%s/videos/?offset=%%d&limit=%%d' % TwitchBaseIE._API_BASE
+    _PAGE_LIMIT = 100
+
+    def _extract_playlist(self, channel_id):
+        info = self._download_json(
+            '%s/kraken/channels/%s' % (self._API_BASE, channel_id),
+            channel_id, 'Downloading channel info JSON')
+        channel_name = info.get('display_name') or info.get('name')
+        entries = []
+        offset = 0
+        limit = self._PAGE_LIMIT
+        for counter in itertools.count(1):
+            response = self._download_json(
+                self._PLAYLIST_URL % (channel_id, offset, limit),
+                channel_id, 'Downloading %s videos JSON page %d' % (self._PLAYLIST_TYPE, counter))
+            page_entries = self._extract_playlist_page(response)
+            if not page_entries:
+                break
+            entries.extend(page_entries)
+            offset += limit
+        return self.playlist_result(
+            [self.url_result(entry) for entry in set(entries)],
+            channel_id, channel_name)
+
+    def _extract_playlist_page(self, response):
+        videos = response.get('videos')
+        return [video['url'] for video in videos] if videos else []
+
+    def _real_extract(self, url):
+        return self._extract_playlist(self._match_id(url))
+
+
+class TwitchProfileIE(TwitchPlaylistBaseIE):
+    IE_NAME = 'twitch:profile'
+    _VALID_URL = r'%s/(?P<id>[^/]+)/profile/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
+    _PLAYLIST_TYPE = 'profile'
+
+    _TEST = {
+        'url': 'http://www.twitch.tv/vanillatv/profile',
+        'info_dict': {
+            'id': 'vanillatv',
+            'title': 'VanillaTV',
+        },
+        'playlist_mincount': 412,
+    }
+
+
+class TwitchPastBroadcastsIE(TwitchPlaylistBaseIE):
+    IE_NAME = 'twitch:past_broadcasts'
+    _VALID_URL = r'%s/(?P<id>[^/]+)/profile/past_broadcasts/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
+    _PLAYLIST_URL = TwitchPlaylistBaseIE._PLAYLIST_URL + '&broadcasts=true'
+    _PLAYLIST_TYPE = 'past broadcasts'
+
+    _TEST = {
+        'url': 'http://www.twitch.tv/spamfish/profile/past_broadcasts',
+        'info_dict': {
+            'id': 'spamfish',
+            'title': 'Spamfish',
+        },
+        'playlist_mincount': 54,
+    }
+
+
+class TwitchBookmarksIE(TwitchPlaylistBaseIE):
+    IE_NAME = 'twitch:bookmarks'
+    _VALID_URL = r'%s/(?P<id>[^/]+)/profile/bookmarks/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
+    _PLAYLIST_URL = '%s/api/bookmark/?user=%%s&offset=%%d&limit=%%d' % TwitchBaseIE._API_BASE
+    _PLAYLIST_TYPE = 'bookmarks'
+
+    _TEST = {
+        'url': 'http://www.twitch.tv/ognos/profile/bookmarks',
+        'info_dict': {
+            'id': 'ognos',
+            'title': 'Ognos',
+        },
+        'playlist_mincount': 3,
+    }
+
+    def _extract_playlist_page(self, response):
+        entries = []
+        for bookmark in response.get('bookmarks', []):
+            video = bookmark.get('video')
+            if not video:
+                continue
+            entries.append(video['url'])
+        return entries
+
+
+class TwitchStreamIE(TwitchBaseIE):
+    IE_NAME = 'twitch:stream'
+    _VALID_URL = r'%s/(?P<id>[^/]+)/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
+
+    _TEST = {
+        'url': 'http://www.twitch.tv/shroomztv',
+        'info_dict': {
+            'id': '12772022048',
+            'display_id': 'shroomztv',
+            'ext': 'mp4',
+            'title': 're:^ShroomzTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            'description': 'H1Z1 - lonewolfing with ShroomzTV | A3 Battle Royale later - @ShroomzTV',
+            'is_live': True,
+            'timestamp': 1421928037,
+            'upload_date': '20150122',
+            'uploader': 'ShroomzTV',
+            'uploader_id': 'shroomztv',
+            'view_count': int,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        channel_id = self._match_id(url)
+
+        stream = self._download_json(
+            '%s/kraken/streams/%s' % (self._API_BASE, channel_id), channel_id,
+            'Downloading stream JSON').get('stream')
+
+        # Fallback on profile extraction if stream is offline
+        if not stream:
+            return self.url_result(
+                'http://www.twitch.tv/%s/profile' % channel_id,
+                'TwitchProfile', channel_id)
+
+        access_token = self._download_json(
+            '%s/api/channels/%s/access_token' % (self._API_BASE, channel_id), channel_id,
+            'Downloading channel access token')
+
+        query = {
+            'allow_source': 'true',
+            'p': random.randint(1000000, 10000000),
+            'player': 'twitchweb',
+            'segment_preference': '4',
+            'sig': access_token['sig'],
+            'token': access_token['token'],
+        }
+
+        formats = self._extract_m3u8_formats(
+            '%s/api/channel/hls/%s.m3u8?%s'
+            % (self._USHER_BASE, channel_id, compat_urllib_parse.urlencode(query).encode('utf-8')),
+            channel_id, 'mp4')
+
+        view_count = stream.get('viewers')
+        timestamp = parse_iso8601(stream.get('created_at'))
+
+        channel = stream['channel']
+        title = self._live_title(channel.get('display_name') or channel.get('name'))
+        description = channel.get('status')
+
+        thumbnails = []
+        for thumbnail_key, thumbnail_url in stream['preview'].items():
+            m = re.search(r'(?P<width>\d+)x(?P<height>\d+)\.jpg$', thumbnail_key)
             if not m:
-                raise ExtractorError('Cannot find archive of a chapter')
-            archive_id = m.group(1)
+                continue
+            thumbnails.append({
+                'url': thumbnail_url,
+                'width': int(m.group('width')),
+                'height': int(m.group('height')),
+            })
 
-            api = api_base + '/broadcast/by_chapter/%s.xml' % chapter_id
-            doc = self._download_xml(
-                api, chapter_id,
-                note='Downloading chapter information',
-                errnote='Chapter information download failed')
-            for a in doc.findall('.//archive'):
-                if archive_id == a.find('./id').text:
-                    break
-            else:
-                raise ExtractorError('Could not find chapter in chapter information')
-
-            video_url = a.find('./video_file_url').text
-            video_ext = video_url.rpartition('.')[2] or 'flv'
-
-            chapter_api_url = 'https://api.twitch.tv/kraken/videos/c' + chapter_id
-            chapter_info = self._download_json(
-                chapter_api_url, 'c' + chapter_id,
-                note='Downloading chapter metadata',
-                errnote='Download of chapter metadata failed')
-
-            bracket_start = int(doc.find('.//bracket_start').text)
-            bracket_end = int(doc.find('.//bracket_end').text)
-
-            # TODO determine start (and probably fix up file)
-            #  youtube-dl -v http://www.twitch.tv/firmbelief/c/1757457
-            #video_url += '?start=' + TODO:start_timestamp
-            # bracket_start is 13290, but we want 51670615
-            self._downloader.report_warning('Chapter detected, but we can just download the whole file. '
-                                            'Chapter starts at %s and ends at %s' % (formatSeconds(bracket_start), formatSeconds(bracket_end)))
-
-            info = {
-                'id': 'c' + chapter_id,
-                'url': video_url,
-                'ext': video_ext,
-                'title': chapter_info['title'],
-                'thumbnail': chapter_info['preview'],
-                'description': chapter_info['description'],
-                'uploader': chapter_info['channel']['display_name'],
-                'uploader_id': chapter_info['channel']['name'],
-            }
-            return info
-            """
-        elif mobj.group('videoid'):
-            return self._extract_media('a', mobj.group('videoid'))
-        elif mobj.group('vodid'):
-            return self._extract_media('v', mobj.group('vodid'))
-        elif mobj.group('channelid'):
-            channel_id = mobj.group('channelid')
-            info = self._download_json(
-                '%s/kraken/channels/%s' % (self._API_BASE, channel_id),
-                channel_id, 'Downloading channel info JSON')
-            channel_name = info.get('display_name') or info.get('name')
-            entries = []
-            offset = 0
-            limit = self._PAGE_LIMIT
-            for counter in itertools.count(1):
-                response = self._download_json(
-                    '%s/kraken/channels/%s/videos/?offset=%d&limit=%d'
-                    % (self._API_BASE, channel_id, offset, limit),
-                    channel_id, 'Downloading channel videos JSON page %d' % counter)
-                videos = response['videos']
-                if not videos:
-                    break
-                entries.extend([self.url_result(video['url'], 'Twitch') for video in videos])
-                offset += limit
-            return self.playlist_result(entries, channel_id, channel_name)
+        return {
+            'id': compat_str(stream['_id']),
+            'display_id': channel_id,
+            'title': title,
+            'description': description,
+            'thumbnails': thumbnails,
+            'uploader': channel.get('display_name'),
+            'uploader_id': channel.get('name'),
+            'timestamp': timestamp,
+            'view_count': view_count,
+            'formats': formats,
+            'is_live': True,
+        }

youtube_dl/extractor/ubu.py

@@ -3,50 +3,51 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    int_or_none,
+    qualities,
+)
 
 
 class UbuIE(InfoExtractor):
     _VALID_URL = r'http://(?:www\.)?ubu\.com/film/(?P<id>[\da-z_-]+)\.html'
     _TEST = {
         'url': 'http://ubu.com/film/her_noise.html',
-        'md5': '8edd46ee8aa6b265fb5ed6cf05c36bc9',
+        'md5': '138d5652618bf0f03878978db9bef1ee',
         'info_dict': {
             'id': 'her_noise',
-            'ext': 'mp4',
+            'ext': 'm4v',
             'title': 'Her Noise - The Making Of (2007)',
             'duration': 3600,
         },
     }
 
     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)
 
         title = self._html_search_regex(
             r'<title>.+?Film &amp; Video: ([^<]+)</title>', webpage, 'title')
 
         duration = int_or_none(self._html_search_regex(
-            r'Duration: (\d+) minutes', webpage, 'duration', fatal=False, default=None))
-        if duration:
-            duration *= 60
+            r'Duration: (\d+) minutes', webpage, 'duration', fatal=False),
+            invscale=60)
 
         formats = []
-
         FORMAT_REGEXES = [
-            ['sq', r"'flashvars'\s*,\s*'file=([^']+)'"],
-            ['hq', r'href="(http://ubumexico\.centro\.org\.mx/video/[^"]+)"']
+            ('sq', r"'flashvars'\s*,\s*'file=([^']+)'"),
+            ('hq', r'href="(http://ubumexico\.centro\.org\.mx/video/[^"]+)"'),
         ]
-
+        preference = qualities([fid for fid, _ in FORMAT_REGEXES])
         for format_id, format_regex in FORMAT_REGEXES:
             m = re.search(format_regex, webpage)
             if m:
                 formats.append({
                     'url': m.group(1),
                     'format_id': format_id,
+                    'preference': preference(format_id),
                 })
+        self._sort_formats(formats)
 
         return {
             'id': video_id,

youtube_dl/extractor/videomega.py

@@ -1,11 +1,15 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
+    compat_urllib_request,
 )
 from ..utils import (
+    ExtractorError,
     remove_start,
 )
 
@@ -16,34 +20,40 @@ class VideoMegaIE(InfoExtractor):
         (?:iframe\.php)?\?ref=(?P<id>[A-Za-z0-9]+)
         '''
     _TEST = {
-        'url': 'http://videomega.tv/?ref=GKeGPVedBe',
-        'md5': '240fb5bcf9199961f48eb17839b084d6',
+        'url': 'http://videomega.tv/?ref=QR0HCUHI1661IHUCH0RQ',
+        'md5': 'bf5c2f95c4c917536e80936af7bc51e1',
         'info_dict': {
-            'id': 'GKeGPVedBe',
+            'id': 'QR0HCUHI1661IHUCH0RQ',
             'ext': 'mp4',
-            'title': 'XXL - All Sports United',
+            'title': 'Big Buck Bunny',
             'thumbnail': 're:^https?://.*\.jpg$',
         }
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        url = 'http://videomega.tv/iframe.php?ref={0:}'.format(video_id)
-        webpage = self._download_webpage(url, video_id)
 
-        escaped_data = self._search_regex(
-            r'unescape\("([^"]+)"\)', webpage, 'escaped data')
+        iframe_url = 'http://videomega.tv/iframe.php?ref={0:}'.format(video_id)
+        req = compat_urllib_request.Request(iframe_url)
+        req.add_header('Referer', url)
+        webpage = self._download_webpage(req, video_id)
+
+        try:
+            escaped_data = re.findall(r'unescape\("([^"]+)"\)', webpage)[-1]
+        except IndexError:
+            raise ExtractorError('Unable to extract escaped data')
+
         playlist = compat_urllib_parse.unquote(escaped_data)
 
         thumbnail = self._search_regex(
             r'image:\s*"([^"]+)"', playlist, 'thumbnail', fatal=False)
-        url = self._search_regex(r'file:\s*"([^"]+)"', playlist, 'URL')
+        video_url = self._search_regex(r'file:\s*"([^"]+)"', playlist, 'URL')
         title = remove_start(self._html_search_regex(
             r'<title>(.*?)</title>', webpage, 'title'), 'VideoMega.tv - ')
 
         formats = [{
             'format_id': 'sd',
-            'url': url,
+            'url': video_url,
         }]
         self._sort_formats(formats)
 
@@ -52,4 +62,7 @@ class VideoMegaIE(InfoExtractor):
             'title': title,
             'formats': formats,
             'thumbnail': thumbnail,
+            'http_headers': {
+                'Referer': iframe_url,
+            },
         }

youtube_dl/extractor/videott.py

@@ -13,9 +13,9 @@ from ..utils import (
 class VideoTtIE(InfoExtractor):
     ID_NAME = 'video.tt'
     IE_DESC = 'video.tt - Your True Tube'
-    _VALID_URL = r'http://(?:www\.)?video\.tt/(?:video/|watch_video\.php\?v=)(?P<id>[\da-zA-Z]{9})'
+    _VALID_URL = r'http://(?:www\.)?video\.tt/(?:(?:video|embed)/|watch_video\.php\?v=)(?P<id>[\da-zA-Z]{9})'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.video.tt/watch_video.php?v=amd5YujV8',
         'md5': 'b13aa9e2f267effb5d1094443dff65ba',
         'info_dict': {
@@ -26,7 +26,10 @@ class VideoTtIE(InfoExtractor):
             'upload_date': '20130827',
             'uploader': 'joseph313',
         }
-    }
+    }, {
+        'url': 'http://video.tt/embed/amd5YujV8',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/wdr.py

@@ -169,7 +169,9 @@ class WDRMobileIE(InfoExtractor):
             'title': mobj.group('title'),
             'age_limit': int(mobj.group('age_limit')),
             'url': url,
-            'user_agent': 'mobile',
+            'http_headers': {
+                'User-Agent': 'mobile',
+            },
         }
 
 

youtube_dl/extractor/youtube.py

@@ -264,9 +264,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
 
         # Dash mp4 audio
-        '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50},
-        '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50},
-        '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50},
+        '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
+        '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
+        '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
 
         # Dash webm
         '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
@@ -394,6 +394,23 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'format': '141',
             },
         },
+        # JS player signature function name containing $
+        {
+            'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
+            'info_dict': {
+                'id': 'nfWlot6h_JM',
+                'ext': 'm4a',
+                'title': 'Taylor Swift - Shake It Off',
+                'description': 'md5:2acfda1b285bdd478ccec22f9918199d',
+                'uploader': 'TaylorSwiftVEVO',
+                'uploader_id': 'TaylorSwiftVEVO',
+                'upload_date': '20140818',
+            },
+            'params': {
+                'youtube_include_dash_manifest': True,
+                'format': '141',
+            },
+        },
         # Controversy video
         {
             'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
@@ -588,7 +605,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
 
     def _parse_sig_js(self, jscode):
         funcname = self._search_regex(
-            r'\.sig\|\|([a-zA-Z0-9]+)\(', jscode,
+            r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
             'Initial JS player signature function name')
 
         jsi = JSInterpreter(jscode)
@@ -1665,11 +1682,17 @@ class YoutubeTruncatedURLIE(InfoExtractor):
     IE_NAME = 'youtube:truncated_url'
     IE_DESC = False  # Do not list
     _VALID_URL = r'''(?x)
-        (?:https?://)?[^/]+/watch\?(?:
+        (?:https?://)?
+        (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
+        (?:watch\?(?:
             feature=[a-z_]+|
-            annotation_id=annotation_[^&]+
-        )?$|
-        (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
+            annotation_id=annotation_[^&]+|
+            x-yt-cl=[0-9]+|
+        )?
+        |
+            attribution_link\?a=[^&]+
+        )
+        $
     '''
 
     _TESTS = [{
@@ -1678,6 +1701,12 @@ class YoutubeTruncatedURLIE(InfoExtractor):
     }, {
         'url': 'http://www.youtube.com/watch?',
         'only_matching': True,
+    }, {
+        'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.youtube.com/watch?feature=foo',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -1693,7 +1722,7 @@ class YoutubeTruncatedURLIE(InfoExtractor):
 class YoutubeTruncatedIDIE(InfoExtractor):
     IE_NAME = 'youtube:truncated_id'
     IE_DESC = False  # Do not list
-    _VALID_URL = r'https?://(?:www\.)youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
+    _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
 
     _TESTS = [{
         'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',

youtube_dl/options.py

@@ -5,6 +5,7 @@ import optparse
 import shlex
 import sys
 
+from .downloader.external import list_external_downloaders
 from .compat import (
     compat_expanduser,
     compat_getenv,
@@ -199,6 +200,10 @@ def parseOpts(overrideArguments=None):
         '--playlist-end',
         dest='playlistend', metavar='NUMBER', default=None, type=int,
         help='playlist video to end at (default is last)')
+    selection.add_option(
+        '--playlist-items',
+        dest='playlist_items', metavar='ITEM_SPEC', default=None,
+        help='playlist video items to download. Specify indices of the videos in the playlist seperated by commas like: "--playlist-items 1,2,5,8" if you want to download videos indexed 1, 2, 5, 8 in the playlist. You can specify range: "--playlist-items 1-3,7,10-13", it will download the videos at index 1, 2, 3, 7, 10, 11, 12 and 13.')
     selection.add_option(
         '--match-title',
         dest='matchtitle', metavar='REGEX',
@@ -264,7 +269,7 @@ def parseOpts(overrideArguments=None):
     authentication.add_option(
         '-p', '--password',
         dest='password', metavar='PASSWORD',
-        help='account password')
+        help='account password. If this option is left out, youtube-dl will ask interactively.')
     authentication.add_option(
         '-2', '--twofactor',
         dest='twofactor', metavar='TWOFACTOR',
@@ -289,6 +294,17 @@ def parseOpts(overrideArguments=None):
             'extensions aac, m4a, mp3, mp4, ogg, wav, webm. '
             'You can also use the special names "best",'
             ' "bestvideo", "bestaudio", "worst". '
+            ' You can filter the video results by putting a condition in'
+            ' brackets, as in -f "best[height=720]"'
+            ' (or -f "[filesize>10M]"). '
+            ' This works for filesize, height, width, tbr, abr, and vbr'
+            ' and the comparisons <, <=, >, >=, =, != .'
+            ' Formats for which the value is not known are excluded unless you'
+            ' put a question mark (?) after the operator.'
+            ' You can combine format filters, so  '
+            '-f "[height <=? 720][tbr>500]" '
+            'selects up to 720p videos (or videos where the height is not '
+            'known) with a bitrate of at least 500 KBit/s.'
             ' By default, youtube-dl will pick the best quality.'
             ' Use commas to download multiple audio formats, such as'
             ' -f  136/137/mp4/bestvideo,140/m4a/bestaudio.'
@@ -361,7 +377,7 @@ def parseOpts(overrideArguments=None):
     downloader.add_option(
         '-R', '--retries',
         dest='retries', metavar='RETRIES', default=10,
-        help='number of retries (default is %default)')
+        help='number of retries (default is %default), or "infinite".')
     downloader.add_option(
         '--buffer-size',
         dest='buffersize', metavar='SIZE', default='1024',
@@ -378,6 +394,15 @@ def parseOpts(overrideArguments=None):
         '--playlist-reverse',
         action='store_true',
         help='Download playlist videos in reverse order')
+    downloader.add_option(
+        '--xattr-set-filesize',
+        dest='xattr_set_filesize', action='store_true',
+        help='(experimental) set file xattribute ytdl.filesize with expected filesize')
+    downloader.add_option(
+        '--external-downloader',
+        dest='external_downloader', metavar='COMMAND',
+        help='(experimental) Use the specified external downloader. '
+             'Currently supports %s' % ','.join(list_external_downloaders()))
 
     workarounds = optparse.OptionGroup(parser, 'Workarounds')
     workarounds.add_option(
@@ -410,6 +435,10 @@ def parseOpts(overrideArguments=None):
         '--bidi-workaround',
         dest='bidi_workaround', action='store_true',
         help='Work around terminals that lack bidirectional text support. Requires bidiv or fribidi executable in PATH')
+    workarounds.add_option(
+        '--sleep-interval', metavar='SECONDS',
+        dest='sleep_interval', type=float,
+        help='Number of seconds to sleep before each download.')
 
     verbosity = optparse.OptionGroup(parser, 'Verbosity / Simulation Options')
     verbosity.add_option(
@@ -593,10 +622,6 @@ def parseOpts(overrideArguments=None):
         '--write-annotations',
         action='store_true', dest='writeannotations', default=False,
         help='write video annotations to a .annotation file')
-    filesystem.add_option(
-        '--write-thumbnail',
-        action='store_true', dest='writethumbnail', default=False,
-        help='write thumbnail image to disk')
     filesystem.add_option(
         '--load-info',
         dest='load_info_filename', metavar='FILE',
@@ -616,6 +641,20 @@ def parseOpts(overrideArguments=None):
         action='store_true', dest='rm_cachedir',
         help='Delete all filesystem cache files')
 
+    thumbnail = optparse.OptionGroup(parser, 'Thumbnail images')
+    thumbnail.add_option(
+        '--write-thumbnail',
+        action='store_true', dest='writethumbnail', default=False,
+        help='write thumbnail image to disk')
+    thumbnail.add_option(
+        '--write-all-thumbnails',
+        action='store_true', dest='write_all_thumbnails', default=False,
+        help='write all thumbnail image formats to disk')
+    thumbnail.add_option(
+        '--list-thumbnails',
+        action='store_true', dest='list_thumbnails', default=False,
+        help='Simulate and list all available thumbnail formats')
+
     postproc = optparse.OptionGroup(parser, 'Post-processing Options')
     postproc.add_option(
         '-x', '--extract-audio',
@@ -681,6 +720,7 @@ def parseOpts(overrideArguments=None):
     parser.add_option_group(selection)
     parser.add_option_group(downloader)
     parser.add_option_group(filesystem)
+    parser.add_option_group(thumbnail)
     parser.add_option_group(verbosity)
     parser.add_option_group(workarounds)
     parser.add_option_group(video_format)

youtube_dl/postprocessor/__init__.py

@@ -7,6 +7,7 @@ from .ffmpeg import (
     FFmpegEmbedSubtitlePP,
     FFmpegExtractAudioPP,
     FFmpegFixupStretchedPP,
+    FFmpegFixupM4aPP,
     FFmpegMergerPP,
     FFmpegMetadataPP,
     FFmpegVideoConvertorPP,
@@ -25,6 +26,7 @@ __all__ = [
     'FFmpegAudioFixPP',
     'FFmpegEmbedSubtitlePP',
     'FFmpegExtractAudioPP',
+    'FFmpegFixupM4aPP',
     'FFmpegFixupStretchedPP',
     'FFmpegMergerPP',
     'FFmpegMetadataPP',

youtube_dl/postprocessor/ffmpeg.py

@@ -475,15 +475,21 @@ class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
         filename = information['filepath']
         input_files = [filename] + [subtitles_filename(filename, lang, self._subformat) for lang in sub_langs]
 
-        opts = ['-map', '0:0', '-map', '0:1', '-c:v', 'copy', '-c:a', 'copy']
+        opts = [
+            '-map', '0',
+            '-c', 'copy',
+            # Don't copy the existing subtitles, we may be running the
+            # postprocessor a second time
+            '-map', '-0:s',
+            '-c:s', 'mov_text',
+        ]
         for (i, lang) in enumerate(sub_langs):
-            opts.extend(['-map', '%d:0' % (i + 1), '-c:s:%d' % i, 'mov_text'])
+            opts.extend(['-map', '%d:0' % (i + 1)])
             lang_code = self._conver_lang_code(lang)
             if lang_code is not None:
                 opts.extend(['-metadata:s:s:%d' % i, 'language=%s' % lang_code])
-        opts.extend(['-f', 'mp4'])
 
-        temp_filename = filename + '.temp'
+        temp_filename = prepend_extension(filename, 'temp')
         self._downloader.to_screen('[ffmpeg] Embedding subtitles in \'%s\'' % filename)
         self.run_ffmpeg_multiple_files(input_files, temp_filename, opts)
         os.remove(encodeFilename(filename))
@@ -503,6 +509,10 @@ class FFmpegMetadataPP(FFmpegPostProcessor):
             metadata['artist'] = info['uploader']
         elif info.get('uploader_id') is not None:
             metadata['artist'] = info['uploader_id']
+        if info.get('description') is not None:
+            metadata['description'] = info['description']
+        if info.get('webpage_url') is not None:
+            metadata['comment'] = info['webpage_url']
 
         if not metadata:
             self._downloader.to_screen('[ffmpeg] There isn\'t any metadata to add')
@@ -554,7 +564,7 @@ class FFmpegFixupStretchedPP(FFmpegPostProcessor):
     def run(self, info):
         stretched_ratio = info.get('stretched_ratio')
         if stretched_ratio is None or stretched_ratio == 1:
-            return
+            return True, info
 
         filename = info['filepath']
         temp_filename = prepend_extension(filename, 'temp')
@@ -567,3 +577,21 @@ class FFmpegFixupStretchedPP(FFmpegPostProcessor):
         os.rename(encodeFilename(temp_filename), encodeFilename(filename))
 
         return True, info
+
+
+class FFmpegFixupM4aPP(FFmpegPostProcessor):
+    def run(self, info):
+        if info.get('container') != 'm4a_dash':
+            return True, info
+
+        filename = info['filepath']
+        temp_filename = prepend_extension(filename, 'temp')
+
+        options = ['-c', 'copy', '-f', 'mp4']
+        self._downloader.to_screen('[ffmpeg] Correcting container in "%s"' % filename)
+        self.run_ffmpeg(filename, temp_filename, options)
+
+        os.remove(encodeFilename(filename))
+        os.rename(encodeFilename(temp_filename), encodeFilename(filename))
+
+        return True, info

youtube_dl/utils.py

@@ -399,8 +399,9 @@ def formatSeconds(secs):
 def make_HTTPS_handler(params, **kwargs):
     opts_no_check_certificate = params.get('nocheckcertificate', False)
     if hasattr(ssl, 'create_default_context'):  # Python >= 3.4 or 2.7.9
-        context = ssl.create_default_context(ssl.Purpose.CLIENT_AUTH)
+        context = ssl.create_default_context(ssl.Purpose.SERVER_AUTH)
         if opts_no_check_certificate:
+            context.check_hostname = False
             context.verify_mode = ssl.CERT_NONE
         try:
             return YoutubeDLHTTPSHandler(params, context=context, **kwargs)
@@ -410,25 +411,9 @@ def make_HTTPS_handler(params, **kwargs):
             pass
 
     if sys.version_info < (3, 2):
-        import httplib
-
-        class HTTPSConnectionV3(httplib.HTTPSConnection):
-            def __init__(self, *args, **kwargs):
-                httplib.HTTPSConnection.__init__(self, *args, **kwargs)
-
-            def connect(self):
-                sock = socket.create_connection((self.host, self.port), self.timeout)
-                if getattr(self, '_tunnel_host', False):
-                    self.sock = sock
-                    self._tunnel()
-                try:
-                    self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_TLSv1)
-                except ssl.SSLError:
-                    self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv23)
-
-        return YoutubeDLHTTPSHandler(params, https_conn_class=HTTPSConnectionV3, **kwargs)
+        return YoutubeDLHTTPSHandler(params, **kwargs)
     else:  # Python < 3.4
-        context = ssl.SSLContext(ssl.PROTOCOL_SSLv23)
+        context = ssl.SSLContext(ssl.PROTOCOL_TLSv1)
         context.verify_mode = (ssl.CERT_NONE
                                if opts_no_check_certificate
                                else ssl.CERT_REQUIRED)
@@ -559,7 +544,9 @@ def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs):
                 sock = compat_socket_create_connection(
                     (self.host, self.port), self.timeout, sa)
                 if is_https:
-                    self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file)
+                    self.sock = ssl.wrap_socket(
+                        sock, self.key_file, self.cert_file,
+                        ssl_version=ssl.PROTOCOL_TLSv1)
                 else:
                     self.sock = sock
             hc.connect = functools.partial(_hc_connect, hc)
@@ -611,17 +598,14 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
 
     def http_request(self, req):
         for h, v in std_headers.items():
-            if h not in req.headers:
+            # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275
+            # The dict keys are capitalized because of this bug by urllib
+            if h.capitalize() not in req.headers:
                 req.add_header(h, v)
         if 'Youtubedl-no-compression' in req.headers:
             if 'Accept-encoding' in req.headers:
                 del req.headers['Accept-encoding']
             del req.headers['Youtubedl-no-compression']
-        if 'Youtubedl-user-agent' in req.headers:
-            if 'User-agent' in req.headers:
-                del req.headers['User-agent']
-            req.headers['User-agent'] = req.headers['Youtubedl-user-agent']
-            del req.headers['Youtubedl-user-agent']
 
         if sys.version_info < (2, 7) and '#' in req.get_full_url():
             # Python 2.6 is brain-dead when it comes to fragments
@@ -721,6 +705,7 @@ def unified_strdate(date_str, day_first=True):
         '%b %dst %Y %I:%M%p',
         '%b %dnd %Y %I:%M%p',
         '%b %dth %Y %I:%M%p',
+        '%Y %m %d',
         '%Y-%m-%d',
         '%Y/%m/%d',
         '%Y/%m/%d %H:%M:%S',
@@ -873,6 +858,9 @@ def _windows_write_string(s, out):
     except AttributeError:
         # If the output stream doesn't have a fileno, it's virtual
         return False
+    except io.UnsupportedOperation:
+        # Some strange Windows pseudo files?
+        return False
     if fileno not in WIN_OUTPUT_IDS:
         return False
 
@@ -1269,13 +1257,13 @@ def float_or_none(v, scale=1, invscale=1, default=None):
 
 
 def parse_duration(s):
-    if s is None:
+    if not isinstance(s, basestring if sys.version_info < (3, 0) else compat_str):
         return None
 
     s = s.strip()
 
     m = re.match(
-        r'''(?ix)T?
+        r'''(?ix)(?:P?T)?
         (?:
             (?P<only_mins>[0-9.]+)\s*(?:mins?|minutes?)\s*|
             (?P<only_hours>[0-9.]+)\s*(?:hours?)|
@@ -1610,6 +1598,14 @@ def urlhandle_detect_ext(url_handle):
     except AttributeError:  # Python < 3
         getheader = url_handle.info().getheader
 
+    cd = getheader('Content-Disposition')
+    if cd:
+        m = re.match(r'attachment;\s*filename="(?P<filename>[^"]+)"', cd)
+        if m:
+            e = determine_ext(m.group('filename'), default_ext=None)
+            if e:
+                return e
+
     return getheader('Content-Type').split("/")[1]
 
 
@@ -1621,3 +1617,53 @@ def age_restricted(content_limit, age_limit):
     if content_limit is None:
         return False  # Content available for everyone
     return age_limit < content_limit
+
+
+def is_html(first_bytes):
+    """ Detect whether a file contains HTML by examining its first bytes. """
+
+    BOMS = [
+        (b'\xef\xbb\xbf', 'utf-8'),
+        (b'\x00\x00\xfe\xff', 'utf-32-be'),
+        (b'\xff\xfe\x00\x00', 'utf-32-le'),
+        (b'\xff\xfe', 'utf-16-le'),
+        (b'\xfe\xff', 'utf-16-be'),
+    ]
+    for bom, enc in BOMS:
+        if first_bytes.startswith(bom):
+            s = first_bytes[len(bom):].decode(enc, 'replace')
+            break
+    else:
+        s = first_bytes.decode('utf-8', 'replace')
+
+    return re.match(r'^\s*<', s)
+
+
+def determine_protocol(info_dict):
+    protocol = info_dict.get('protocol')
+    if protocol is not None:
+        return protocol
+
+    url = info_dict['url']
+    if url.startswith('rtmp'):
+        return 'rtmp'
+    elif url.startswith('mms'):
+        return 'mms'
+    elif url.startswith('rtsp'):
+        return 'rtsp'
+
+    ext = determine_ext(url)
+    if ext == 'm3u8':
+        return 'm3u8'
+    elif ext == 'f4m':
+        return 'f4m'
+
+    return compat_urllib_parse_urlparse(url).scheme
+
+
+def render_table(header_row, data):
+    """ Render a list of rows, each as a list of values """
+    table = [header_row] + data
+    max_lens = [max(len(compat_str(v)) for v in col) for col in zip(*table)]
+    format_str = ' '.join('%-' + compat_str(ml + 1) + 's' for ml in max_lens[:-1]) + '%s'
+    return '\n'.join(format_str % tuple(row) for row in table)

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2015.01.11'
+__version__ = '2015.01.25'