release 2017.07.23

All these meta fields must be borrowed from final extractor that actually performs extraction.
This commit fixes extractor id in download archives for url_transparent downloads. Previously, 'transparent' extractor was erroneously
used for extractor archive id, e.g. 'eggheadlesson 4n8ugwwj5t' instead of 'wistia 4n8ugwwj5t'.

.github/ISSUE_TEMPLATE.md

@@ -6,8 +6,8 @@
 
 ---
 
-### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2017.07.02*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
-- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2017.07.02**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2017.07.23*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
+- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2017.07.23**
 
 ### Before submitting an *issue* make sure you have:
 - [ ] At least skimmed through [README](https://github.com/rg3/youtube-dl/blob/master/README.md) and **most notably** [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections
@@ -35,7 +35,7 @@ $ youtube-dl -v <your command line>
 [debug] User config: []
 [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
 [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
-[debug] youtube-dl version 2017.07.02
+[debug] youtube-dl version 2017.07.23
 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2
 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4
 [debug] Proxy map: {}

AUTHORS

@@ -220,3 +220,6 @@ gritstub
 Adam Voss
 Mike Fährmann
 Jan Kundrát
+Giuseppe Fabiano
+Örn Guðjónsson
+Parmjit Virk

ChangeLog

@@ -1,3 +1,80 @@
+version 2017.07.23
+
+Core
+* [YoutubeDL] Improve default format specification (#13704)
+* [YoutubeDL] Do not override id, extractor and extractor_key for
+  url_transparent entities
+* [extractor/common] Fix playlist_from_matches
+
+Extractors
+* [itv] Fix production id extraction (#13671, #13703)
+* [vidio] Make duration non fatal and fix typo
+* [mtv] Skip missing video parts (#13690)
+* [sportbox:embed] Fix extraction
++ [npo] Add support for npo3.nl URLs (#13695)
+* [dramafever] Remove video id from title (#13699)
++ [egghead:lesson] Add support for lessons (#6635)
+* [funnyordie] Extract more metadata (#13677)
+* [youku:show] Fix playlist extraction (#13248)
++ [dispeak] Recognize sevt subdomain (#13276)
+* [adn] Improve error reporting (#13663)
+* [crunchyroll] Relax series and season regex (#13659)
++ [spiegel:article] Add support for nexx iframe embeds (#13029)
++ [nexx:embed] Add support for iframe embeds
+* [nexx] Improve JS embed extraction
++ [pearvideo] Add support for pearvideo.com (#13031)
+
+
+version 2017.07.15
+
+Core
+* [YoutubeDL] Don't expand environment variables in meta fields (#13637)
+
+Extractors
+* [spiegeltv] Delegate extraction to nexx extractor (#13159)
++ [nexx] Add support for nexx.cloud (#10807, #13465)
+* [generic] Fix rutube embeds extraction (#13641)
+* [karrierevideos] Fix title extraction (#13641)
+* [youtube] Don't capture YouTube Red ad for creator meta field (#13621)
+* [slideshare] Fix extraction (#13617)
++ [5tv] Add another video URL pattern (#13354, #13606)
+* [drtv] Make HLS and HDS extraction non fatal
+* [ted] Fix subtitles extraction (#13628, #13629)
+* [vine] Make sure the title won't be empty
++ [twitter] Support HLS streams in vmap URLs
++ [periscope] Support pscp.tv URLs in embedded frames
+* [twitter] Extract mp4 urls via mobile API (#12726)
+* [niconico] Fix authentication error handling (#12486)
+* [giantbomb] Extract m3u8 formats (#13626)
++ [vlive:playlist] Add support for playlists (#13613)
+
+
+version 2017.07.09
+
+Core
++ [extractor/common] Add support for AMP tags in _parse_html5_media_entries
++ [utils] Support attributes with no values in get_elements_by_attribute
+
+Extractors
++ [dailymail] Add support for embeds
++ [joj] Add support for joj.sk (#13268)
+* [abc.net.au:iview] Extract more formats (#13492, #13489)
+* [egghead:course] Fix extraction (#6635, #13370)
++ [cjsw] Add support for cjsw.com (#13525)
++ [eagleplatform] Add support for referrer protected videos (#13557)
++ [eagleplatform] Add support for another embed pattern (#13557)
+* [veoh] Extend URL regular expression (#13601)
+* [npo:live] Fix live stream id extraction (#13568, #13605)
+* [googledrive] Fix height extraction (#13603)
++ [dailymotion] Add support for new layout (#13580)
+- [yam] Remove extractor
+* [xhamster] Extract all formats and fix duration extraction (#13593)
++ [xhamster] Add support for new URL schema (#13593)
+* [espn] Extend URL regular expression (#13244, #13549)
+* [kaltura] Fix typo in subtitles extraction (#13569)
+* [vier] Adapt extraction to redesign (#13575)
+
+
 version 2017.07.02
 
 Core

docs/supportedsites.md

@@ -42,7 +42,7 @@
  - **Allocine**
  - **AlphaPorno**
  - **AMCNetworks**
- - **anderetijden**: npo.nl and ntr.nl
+ - **anderetijden**: npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl
  - **AnimeOnDemand**
  - **anitube.se**
  - **Anvato**
@@ -154,6 +154,7 @@
  - **chirbit**
  - **chirbit:profile**
  - **Cinchcast**
+ - **CJSW**
  - **Clipfish**
  - **cliphunter**
  - **ClipRs**
@@ -237,6 +238,7 @@
  - **EbaumsWorld**
  - **EchoMsk**
  - **egghead:course**: egghead.io course
+ - **egghead:lesson**: egghead.io lesson
  - **eHow**
  - **Einthusan**
  - **eitb.tv**
@@ -369,6 +371,7 @@
  - **Jamendo**
  - **JamendoAlbum**
  - **JeuxVideo**
+ - **Joj**
  - **Jove**
  - **jpopsuki.tv**
  - **JWPlatform**
@@ -519,6 +522,8 @@
  - **NextMedia**: 蘋果日報
  - **NextMediaActionNews**: 蘋果日報 - 動新聞
  - **NextTV**: 壹電視
+ - **Nexx**
+ - **NexxEmbed**
  - **nfb**: National Film Board of Canada
  - **nfl.com**
  - **NhkVod**
@@ -549,7 +554,7 @@
  - **NowTVList**
  - **nowvideo**: NowVideo
  - **Noz**
- - **npo**: npo.nl and ntr.nl
+ - **npo**: npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl
  - **npo.nl:live**
  - **npo.nl:radio**
  - **npo.nl:radio:fragment**
@@ -593,6 +598,7 @@
  - **Patreon**
  - **pbs**: Public Broadcasting Service (PBS) and member stations: PBS: Public Broadcasting Service, APT - Alabama Public Television (WBIQ), GPB/Georgia Public Broadcasting (WGTV), Mississippi Public Broadcasting (WMPN), Nashville Public Television (WNPT), WFSU-TV (WFSU), WSRE (WSRE), WTCI (WTCI), WPBA/Channel 30 (WPBA), Alaska Public Media (KAKM), Arizona PBS (KAET), KNME-TV/Channel 5 (KNME), Vegas PBS (KLVX), AETN/ARKANSAS ETV NETWORK (KETS), KET (WKLE), WKNO/Channel 10 (WKNO), LPB/LOUISIANA PUBLIC BROADCASTING (WLPB), OETA (KETA), Ozarks Public Television (KOZK), WSIU Public Broadcasting (WSIU), KEET TV (KEET), KIXE/Channel 9 (KIXE), KPBS San Diego (KPBS), KQED (KQED), KVIE Public Television (KVIE), PBS SoCal/KOCE (KOCE), ValleyPBS (KVPT), CONNECTICUT PUBLIC TELEVISION (WEDH), KNPB Channel 5 (KNPB), SOPTV (KSYS), Rocky Mountain PBS (KRMA), KENW-TV3 (KENW), KUED Channel 7 (KUED), Wyoming PBS (KCWC), Colorado Public Television / KBDI 12 (KBDI), KBYU-TV (KBYU), Thirteen/WNET New York (WNET), WGBH/Channel 2 (WGBH), WGBY (WGBY), NJTV Public Media NJ (WNJT), WLIW21 (WLIW), mpt/Maryland Public Television (WMPB), WETA Television and Radio (WETA), WHYY (WHYY), PBS 39 (WLVT), WVPT - Your Source for PBS and More! (WVPT), Howard University Television (WHUT), WEDU PBS (WEDU), WGCU Public Media (WGCU), WPBT2 (WPBT), WUCF TV (WUCF), WUFT/Channel 5 (WUFT), WXEL/Channel 42 (WXEL), WLRN/Channel 17 (WLRN), WUSF Public Broadcasting (WUSF), ETV (WRLK), UNC-TV (WUNC), PBS Hawaii - Oceanic Cable Channel 10 (KHET), Idaho Public Television (KAID), KSPS (KSPS), OPB (KOPB), KWSU/Channel 10 & KTNW/Channel 31 (KWSU), WILL-TV (WILL), Network Knowledge - WSEC/Springfield (WSEC), WTTW11 (WTTW), Iowa Public Television/IPTV (KDIN), Nine Network (KETC), PBS39 Fort Wayne (WFWA), WFYI Indianapolis (WFYI), Milwaukee Public Television (WMVS), WNIN (WNIN), WNIT Public Television (WNIT), WPT (WPNE), WVUT/Channel 22 (WVUT), WEIU/Channel 51 (WEIU), WQPT-TV (WQPT), WYCC PBS Chicago (WYCC), WIPB-TV (WIPB), WTIU (WTIU), CET  (WCET), ThinkTVNetwork (WPTD), WBGU-TV (WBGU), WGVU TV (WGVU), NET1 (KUON), Pioneer Public Television (KWCM), SDPB Television (KUSD), TPT (KTCA), KSMQ (KSMQ), KPTS/Channel 8 (KPTS), KTWU/Channel 11 (KTWU), East Tennessee PBS (WSJK), WCTE-TV (WCTE), WLJT, Channel 11 (WLJT), WOSU TV (WOSU), WOUB/WOUC (WOUB), WVPB (WVPB), WKYU-PBS (WKYU), KERA 13 (KERA), MPBN (WCBB), Mountain Lake PBS (WCFE), NHPTV (WENH), Vermont PBS (WETK), witf (WITF), WQED Multimedia (WQED), WMHT Educational Telecommunications (WMHT), Q-TV (WDCQ), WTVS Detroit Public TV (WTVS), CMU Public Television (WCMU), WKAR-TV (WKAR), WNMU-TV Public TV 13 (WNMU), WDSE - WRPT (WDSE), WGTE TV (WGTE), Lakeland Public Television (KAWE), KMOS-TV - Channels 6.1, 6.2 and 6.3 (KMOS), MontanaPBS (KUSM), KRWG/Channel 22 (KRWG), KACV (KACV), KCOS/Channel 13 (KCOS), WCNY/Channel 24 (WCNY), WNED (WNED), WPBS (WPBS), WSKG Public TV (WSKG), WXXI (WXXI), WPSU (WPSU), WVIA Public Media Studios (WVIA), WTVI (WTVI), Western Reserve PBS (WNEO), WVIZ/PBS ideastream (WVIZ), KCTS 9 (KCTS), Basin PBS (KPBT), KUHT / Channel 8 (KUHT), KLRN (KLRN), KLRU (KLRU), WTJX Channel 12 (WTJX), WCVE PBS (WCVE), KBTC Public Television (KBTC)
  - **pcmag**
+ - **PearVideo**
  - **People**
  - **periscope**: Periscope
  - **periscope:user**: Periscope user videos
@@ -769,7 +775,7 @@
  - **tagesschau:player**
  - **Tass**
  - **TastyTrade**
- - **TBS**
+ - **TBS** (Currently broken)
  - **TDSLifeway**
  - **teachertube**: teachertube.com videos
  - **teachertube:user:collection**: teachertube.com user and collection videos
@@ -940,13 +946,14 @@
  - **vk:wallpost**
  - **vlive**
  - **vlive:channel**
+ - **vlive:playlist**
  - **Vodlocker**
  - **VODPl**
  - **VODPlatform**
  - **VoiceRepublic**
  - **VoxMedia**
  - **Vporn**
- - **vpro**: npo.nl and ntr.nl
+ - **vpro**: npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl
  - **Vrak**
  - **VRT**: deredactie.be, sporza.be, cobra.be and cobra.canvas.be
  - **vrv**
@@ -972,7 +979,7 @@
  - **wholecloud**: WholeCloud
  - **Wimp**
  - **Wistia**
- - **wnl**: npo.nl and ntr.nl
+ - **wnl**: npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl
  - **WorldStarHipHop**
  - **wrzuta.pl**
  - **wrzuta.pl:playlist**
@@ -996,7 +1003,6 @@
  - **XVideos**
  - **XXXYMovies**
  - **Yahoo**: Yahoo screen and movies
- - **Yam**: 蕃薯藤yam天空部落
  - **yandexmusic:album**: Яндекс.Музыка - Альбом
  - **yandexmusic:playlist**: Яндекс.Музыка - Плейлист
  - **yandexmusic:track**: Яндекс.Музыка - Трек

test/test_YoutubeDL.py

@@ -41,6 +41,7 @@ def _make_result(formats, **kwargs):
         'id': 'testid',
         'title': 'testttitle',
         'extractor': 'testex',
+        'extractor_key': 'TestEx',
     }
     res.update(**kwargs)
     return res
@@ -448,6 +449,17 @@ class TestFormatSelection(unittest.TestCase):
             pass
         self.assertEqual(ydl.downloaded_info_dicts, [])
 
+    def test_default_format_spec(self):
+        ydl = YDL({'simulate': True})
+        self.assertEqual(ydl._default_format_spec({}), 'bestvideo+bestaudio/best')
+
+        ydl = YDL({'outtmpl': '-'})
+        self.assertEqual(ydl._default_format_spec({}), 'best')
+
+        ydl = YDL({})
+        self.assertEqual(ydl._default_format_spec({}, download=False), 'bestvideo+bestaudio/best')
+        self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best')
+
 
 class TestYoutubeDL(unittest.TestCase):
     def test_subtitles(self):
@@ -527,6 +539,8 @@ class TestYoutubeDL(unittest.TestCase):
             'ext': 'mp4',
             'width': None,
             'height': 1080,
+            'title1': '$PATH',
+            'title2': '%PATH%',
         }
 
         def fname(templ):
@@ -545,10 +559,14 @@ class TestYoutubeDL(unittest.TestCase):
         self.assertEqual(fname('%(height)0 6d.%(ext)s'), ' 01080.mp4')
         self.assertEqual(fname('%(height)0   6d.%(ext)s'), ' 01080.mp4')
         self.assertEqual(fname('%(height)   0   6d.%(ext)s'), ' 01080.mp4')
+        self.assertEqual(fname('%%'), '%')
+        self.assertEqual(fname('%%%%'), '%%')
         self.assertEqual(fname('%%(height)06d.%(ext)s'), '%(height)06d.mp4')
         self.assertEqual(fname('%(width)06d.%(ext)s'), 'NA.mp4')
         self.assertEqual(fname('%(width)06d.%%(ext)s'), 'NA.%(ext)s')
         self.assertEqual(fname('%%(width)06d.%(ext)s'), '%(width)06d.mp4')
+        self.assertEqual(fname('Hello %(title1)s'), 'Hello $PATH')
+        self.assertEqual(fname('Hello %(title2)s'), 'Hello %PATH%')
 
     def test_format_note(self):
         ydl = YoutubeDL()
@@ -755,7 +773,8 @@ class TestYoutubeDL(unittest.TestCase):
                     '_type': 'url_transparent',
                     'url': 'foo2:',
                     'ie_key': 'Foo2',
-                    'title': 'foo1 title'
+                    'title': 'foo1 title',
+                    'id': 'foo1_id',
                 }
 
         class Foo2IE(InfoExtractor):
@@ -781,6 +800,9 @@ class TestYoutubeDL(unittest.TestCase):
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['url'], TEST_URL)
         self.assertEqual(downloaded['title'], 'foo1 title')
+        self.assertEqual(downloaded['id'], 'testid')
+        self.assertEqual(downloaded['extractor'], 'testex')
+        self.assertEqual(downloaded['extractor_key'], 'TestEx')
 
 
 if __name__ == '__main__':

test/test_utils.py

@@ -98,6 +98,7 @@ from youtube_dl.compat import (
     compat_chr,
     compat_etree_fromstring,
     compat_getenv,
+    compat_os_name,
     compat_setenv,
     compat_urlparse,
     compat_parse_qs,
@@ -448,7 +449,9 @@ class TestUtil(unittest.TestCase):
 
     def test_shell_quote(self):
         args = ['ffmpeg', '-i', encodeFilename('ñ€ß\'.mp4')]
-        self.assertEqual(shell_quote(args), """ffmpeg -i 'ñ€ß'"'"'.mp4'""")
+        self.assertEqual(
+            shell_quote(args),
+            """ffmpeg -i 'ñ€ß'"'"'.mp4'""" if compat_os_name != 'nt' else '''ffmpeg -i "ñ€ß'.mp4"''')
 
     def test_str_to_int(self):
         self.assertEqual(str_to_int('123,456'), 123456)
@@ -932,7 +935,7 @@ class TestUtil(unittest.TestCase):
     def test_args_to_str(self):
         self.assertEqual(
             args_to_str(['foo', 'ba/r', '-baz', '2 be', '']),
-            'foo ba/r -baz \'2 be\' \'\''
+            'foo ba/r -baz \'2 be\' \'\'' if compat_os_name != 'nt' else 'foo ba/r -baz "2 be" ""'
         )
 
     def test_parse_filesize(self):
@@ -1228,6 +1231,12 @@ part 3</font></u>
         self.assertEqual(get_element_by_attribute('class', 'foo', html), None)
         self.assertEqual(get_element_by_attribute('class', 'no-such-foo', html), None)
 
+        html = '''
+            <div itemprop="author" itemscope>foo</div>
+        '''
+
+        self.assertEqual(get_element_by_attribute('itemprop', 'author', html), 'foo')
+
     def test_get_elements_by_class(self):
         html = '''
             <span class="foo bar">nice</span><span class="foo bar">also nice</span>

youtube_dl/YoutubeDL.py

@@ -26,6 +26,8 @@ import tokenize
 import traceback
 import random
 
+from string import ascii_letters
+
 from .compat import (
     compat_basestring,
     compat_cookiejar,
@@ -674,7 +676,19 @@ class YoutubeDL(object):
                         FORMAT_RE.format(numeric_field),
                         r'%({0})s'.format(numeric_field), outtmpl)
 
-            filename = expand_path(outtmpl % template_dict)
+            # expand_path translates '%%' into '%' and '$$' into '$'
+            # correspondingly that is not what we want since we need to keep
+            # '%%' intact for template dict substitution step. Working around
+            # with boundary-alike separator hack.
+            sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
+            outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
+
+            # outtmpl should be expand_path'ed before template dict substitution
+            # because meta fields may contain env variables we don't want to
+            # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
+            # title "Hello $PATH", we don't want `$PATH` to be expanded.
+            filename = expand_path(outtmpl).replace(sep, '') % template_dict
+
             # Temporary fix for #4787
             # 'Treat' all problem characters by passing filename through preferredencoding
             # to workaround encoding issues with subprocess on python2 @ Windows
@@ -846,7 +860,7 @@ class YoutubeDL(object):
 
             force_properties = dict(
                 (k, v) for k, v in ie_result.items() if v is not None)
-            for f in ('_type', 'url', 'ie_key'):
+            for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
                 if f in force_properties:
                     del force_properties[f]
             new_result = info.copy()
@@ -1050,6 +1064,25 @@ class YoutubeDL(object):
             return op(actual_value, comparison_value)
         return _filter
 
+    def _default_format_spec(self, info_dict, download=True):
+        req_format_list = []
+
+        def can_have_partial_formats():
+            if self.params.get('simulate', False):
+                return True
+            if not download:
+                return True
+            if self.params.get('outtmpl', DEFAULT_OUTTMPL) == '-':
+                return False
+            if info_dict.get('is_live'):
+                return False
+            merger = FFmpegMergerPP(self)
+            return merger.available and merger.can_merge()
+        if can_have_partial_formats():
+            req_format_list.append('bestvideo+bestaudio')
+        req_format_list.append('best')
+        return '/'.join(req_format_list)
+
     def build_format_selector(self, format_spec):
         def syntax_error(note, start):
             message = (
@@ -1520,14 +1553,10 @@ class YoutubeDL(object):
 
         req_format = self.params.get('format')
         if req_format is None:
-            req_format_list = []
-            if (self.params.get('outtmpl', DEFAULT_OUTTMPL) != '-' and
-                    not info_dict.get('is_live')):
-                merger = FFmpegMergerPP(self)
-                if merger.available and merger.can_merge():
-                    req_format_list.append('bestvideo+bestaudio')
-            req_format_list.append('best')
-            req_format = '/'.join(req_format_list)
+            req_format = self._default_format_spec(info_dict, download=download)
+            if self.params.get('verbose'):
+                self.to_stdout('[debug] Default format spec: %s' % req_format)
+
         format_selector = self.build_format_selector(req_format)
 
         # While in format selection we may need to have an access to the original
@@ -1890,7 +1919,7 @@ class YoutubeDL(object):
                         info_dict.get('protocol') == 'm3u8' and
                         self.params.get('hls_prefer_native')):
                     if fixup_policy == 'warn':
-                        self.report_warning('%s: malformated aac bitstream.' % (
+                        self.report_warning('%s: malformed AAC bitstream detected.' % (
                             info_dict['id']))
                     elif fixup_policy == 'detect_or_warn':
                         fixup_pp = FFmpegFixupM3u8PP(self)
@@ -1899,7 +1928,7 @@ class YoutubeDL(object):
                             info_dict['__postprocessors'].append(fixup_pp)
                         else:
                             self.report_warning(
-                                '%s: malformated aac bitstream. %s'
+                                '%s: malformed AAC bitstream detected. %s'
                                 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
                     else:
                         assert fixup_policy in ('ignore', 'never')

youtube_dl/downloader/ism.py

@@ -98,7 +98,7 @@ def write_piff_header(stream, params):
 
     if is_audio:
         smhd_payload = s88.pack(0)  # balance
-        smhd_payload = u16.pack(0)  # reserved
+        smhd_payload += u16.pack(0)  # reserved
         media_header_box = full_box(b'smhd', 0, 0, smhd_payload)  # Sound Media Header
     else:
         vmhd_payload = u16.pack(0)  # graphics mode
@@ -126,7 +126,6 @@ def write_piff_header(stream, params):
         if fourcc == 'AACL':
             sample_entry_box = box(b'mp4a', sample_entry_payload)
     else:
-        sample_entry_payload = sample_entry_payload
         sample_entry_payload += u16.pack(0)  # pre defined
         sample_entry_payload += u16.pack(0)  # reserved
         sample_entry_payload += u32.pack(0) * 3  # pre defined

youtube_dl/extractor/abc.py

@@ -3,11 +3,13 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
     js_to_json,
     int_or_none,
     parse_iso8601,
+    try_get,
 )
 
 
@@ -124,7 +126,20 @@ class ABCIViewIE(InfoExtractor):
         title = video_params.get('title') or video_params['seriesTitle']
         stream = next(s for s in video_params['playlist'] if s.get('type') == 'program')
 
-        formats = self._extract_akamai_formats(stream['hds-unmetered'], video_id)
+        format_urls = [
+            try_get(stream, lambda x: x['hds-unmetered'], compat_str)]
+
+        # May have higher quality video
+        sd_url = try_get(
+            stream, lambda x: x['streams']['hds']['sd'], compat_str)
+        if sd_url:
+            format_urls.append(sd_url.replace('metered', 'um'))
+
+        formats = []
+        for format_url in format_urls:
+            if format_url:
+                formats.extend(
+                    self._extract_akamai_formats(format_url, video_id))
         self._sort_formats(formats)
 
         subtitles = {}

youtube_dl/extractor/adn.py

@@ -107,11 +107,13 @@ class ADNIE(InfoExtractor):
         metas = options.get('metas') or {}
         title = metas.get('title') or video_info['title']
         links = player_config.get('links') or {}
+        error = None
         if not links:
             links_url = player_config['linksurl']
             links_data = self._download_json(urljoin(
                 self._BASE_URL, links_url), video_id)
             links = links_data.get('links') or {}
+            error = links_data.get('error')
 
         formats = []
         for format_id, qualities in links.items():
@@ -130,7 +132,8 @@ class ADNIE(InfoExtractor):
                     for f in m3u8_formats:
                         f['language'] = 'fr'
                 formats.extend(m3u8_formats)
-        error = options.get('error')
+        if not error:
+            error = options.get('error')
         if not formats and error:
             raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
         self._sort_formats(formats)

youtube_dl/extractor/audioboom.py

@@ -43,7 +43,7 @@ class AudioBoomIE(InfoExtractor):
 
         def from_clip(field):
             if clip:
-                clip.get(field)
+                return clip.get(field)
 
         audio_url = from_clip('clipURLPriorToLoading') or self._og_search_property(
             'audio', webpage, 'audio url')

youtube_dl/extractor/cjsw.py

@@ -0,0 +1,72 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    unescapeHTML,
+)
+
+
+class CJSWIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?cjsw\.com/program/(?P<program>[^/]+)/episode/(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://cjsw.com/program/freshly-squeezed/episode/20170620',
+        'md5': 'cee14d40f1e9433632c56e3d14977120',
+        'info_dict': {
+            'id': '91d9f016-a2e7-46c5-8dcb-7cbcd7437c41',
+            'ext': 'mp3',
+            'title': 'Freshly Squeezed – Episode June 20, 2017',
+            'description': 'md5:c967d63366c3898a80d0c7b0ff337202',
+            'series': 'Freshly Squeezed',
+            'episode_id': '20170620',
+        },
+    }, {
+        # no description
+        'url': 'http://cjsw.com/program/road-pops/episode/20170707/',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        program, episode_id = mobj.group('program', 'id')
+        audio_id = '%s/%s' % (program, episode_id)
+
+        webpage = self._download_webpage(url, episode_id)
+
+        title = unescapeHTML(self._search_regex(
+            (r'<h1[^>]+class=["\']episode-header__title["\'][^>]*>(?P<title>[^<]+)',
+             r'data-audio-title=(["\'])(?P<title>(?:(?!\1).)+)\1'),
+            webpage, 'title', group='title'))
+
+        audio_url = self._search_regex(
+            r'<button[^>]+data-audio-src=(["\'])(?P<url>(?:(?!\1).)+)\1',
+            webpage, 'audio url', group='url')
+
+        audio_id = self._search_regex(
+            r'/([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})\.mp3',
+            audio_url, 'audio id', default=audio_id)
+
+        formats = [{
+            'url': audio_url,
+            'ext': determine_ext(audio_url, 'mp3'),
+            'vcodec': 'none',
+        }]
+
+        description = self._html_search_regex(
+            r'<p>(?P<description>.+?)</p>', webpage, 'description',
+            default=None)
+        series = self._search_regex(
+            r'data-showname=(["\'])(?P<name>(?:(?!\1).)+)\1', webpage,
+            'series', default=program, group='name')
+
+        return {
+            'id': audio_id,
+            'title': title,
+            'description': description,
+            'formats': formats,
+            'series': series,
+            'episode_id': episode_id,
+        }

youtube_dl/extractor/common.py

@@ -730,12 +730,12 @@ class InfoExtractor(object):
             video_info['title'] = video_title
         return video_info
 
-    def playlist_from_matches(self, matches, video_id, video_title, getter=None, ie=None):
-        urlrs = orderedSet(
+    def playlist_from_matches(self, matches, playlist_id=None, playlist_title=None, getter=None, ie=None):
+        urls = orderedSet(
             self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
             for m in matches)
         return self.playlist_result(
-            urlrs, playlist_id=video_id, playlist_title=video_title)
+            urls, playlist_id=playlist_id, playlist_title=playlist_title)
 
     @staticmethod
     def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None):
@@ -2132,15 +2132,18 @@ class InfoExtractor(object):
             return is_plain_url, formats
 
         entries = []
+        # amp-video and amp-audio are very similar to their HTML5 counterparts
+        # so we wll include them right here (see
+        # https://www.ampproject.org/docs/reference/components/amp-video)
         media_tags = [(media_tag, media_type, '')
                       for media_tag, media_type
-                      in re.findall(r'(?s)(<(video|audio)[^>]*/>)', webpage)]
+                      in re.findall(r'(?s)(<(?:amp-)?(video|audio)[^>]*/>)', webpage)]
         media_tags.extend(re.findall(
             # We only allow video|audio followed by a whitespace or '>'.
             # Allowing more characters may end up in significant slow down (see
             # https://github.com/rg3/youtube-dl/issues/11979, example URL:
             # http://www.porntrex.com/maps/videositemap.xml).
-            r'(?s)(<(?P<tag>video|audio)(?:\s+[^>]*)?>)(.*?)</(?P=tag)>', webpage))
+            r'(?s)(<(?P<tag>(?:amp-)?(?:video|audio))(?:\s+[^>]*)?>)(.*?)</(?P=tag)>', webpage))
         for media_tag, media_type, media_content in media_tags:
             media_info = {
                 'formats': [],

youtube_dl/extractor/crunchyroll.py

@@ -510,7 +510,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 
         # webpage provide more accurate data than series_title from XML
         series = self._html_search_regex(
-            r'id=["\']showmedia_about_episode_num[^>]+>\s*<a[^>]+>([^<]+)',
+            r'(?s)<h\d[^>]+\bid=["\']showmedia_about_episode_num[^>]+>(.+?)</h\d',
             webpage, 'series', fatal=False)
         season = xpath_text(metadata, 'series_title')
 
@@ -518,7 +518,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         episode_number = int_or_none(xpath_text(metadata, 'episode_number'))
 
         season_number = int_or_none(self._search_regex(
-            r'(?s)<h4[^>]+id=["\']showmedia_about_episode_num[^>]+>.+?</h4>\s*<h4>\s*Season (\d+)',
+            r'(?s)<h\d[^>]+id=["\']showmedia_about_episode_num[^>]+>.+?</h\d>\s*<h4>\s*Season (\d+)',
             webpage, 'season number', default=None))
 
         return {

youtube_dl/extractor/dailymail.py

@@ -1,6 +1,8 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..compat import compat_str
 from ..utils import (
@@ -12,8 +14,8 @@ from ..utils import (
 
 
 class DailyMailIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/video/[^/]+/video-(?P<id>[0-9]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/(?:video/[^/]+/video-|embed/video/)(?P<id>[0-9]+)'
+    _TESTS = [{
         'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html',
         'md5': 'f6129624562251f628296c3a9ffde124',
         'info_dict': {
@@ -22,7 +24,16 @@ class DailyMailIE(InfoExtractor):
             'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'',
             'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84',
         }
-    }
+    }, {
+        'url': 'http://www.dailymail.co.uk/embed/video/1295863.html',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe\b[^>]+\bsrc=["\'](?P<url>(?:https?:)?//(?:www\.)?dailymail\.co\.uk/embed/video/\d+\.html)',
+            webpage)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

youtube_dl/extractor/dailymotion.py

@@ -147,7 +147,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
         view_count_str = self._search_regex(
             (r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:([\s\d,.]+)"',
              r'video_views_count[^>]+>\s+([\s\d\,.]+)'),
-            webpage, 'view count', fatal=False)
+            webpage, 'view count', default=None)
         if view_count_str:
             view_count_str = re.sub(r'\s', '', view_count_str)
         view_count = str_to_int(view_count_str)
@@ -159,7 +159,9 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
             [r'buildPlayer\(({.+?})\);\n',  # See https://github.com/rg3/youtube-dl/issues/7826
              r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
              r'buildPlayer\(({.+?})\);',
-             r'var\s+config\s*=\s*({.+?});'],
+             r'var\s+config\s*=\s*({.+?});',
+             # New layout regex (see https://github.com/rg3/youtube-dl/issues/13580)
+             r'__PLAYER_CONFIG__\s*=\s*({.+?});'],
             webpage, 'player v5', default=None)
         if player_v5:
             player = self._parse_json(player_v5, video_id)

youtube_dl/extractor/dispeak.py

@@ -13,7 +13,7 @@ from ..utils import (
 
 
 class DigitallySpeakingIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:evt\.dispeak|events\.digitallyspeaking)\.com/(?:[^/]+/)+xml/(?P<id>[^.]+)\.xml'
+    _VALID_URL = r'https?://(?:s?evt\.dispeak|events\.digitallyspeaking)\.com/(?:[^/]+/)+xml/(?P<id>[^.]+)\.xml'
 
     _TESTS = [{
         # From http://gdcvault.com/play/1023460/Tenacious-Design-and-The-Interface
@@ -28,6 +28,10 @@ class DigitallySpeakingIE(InfoExtractor):
         # From http://www.gdcvault.com/play/1014631/Classic-Game-Postmortem-PAC
         'url': 'http://events.digitallyspeaking.com/gdc/sf11/xml/12396_1299111843500GMPX.xml',
         'only_matching': True,
+    }, {
+        # From http://www.gdcvault.com/play/1013700/Advanced-Material
+        'url': 'http://sevt.dispeak.com/ubm/gdc/eur10/xml/11256_1282118587281VNIT.xml',
+        'only_matching': True,
     }]
 
     def _parse_mp4(self, metadata):

youtube_dl/extractor/dramafever.py

@@ -12,6 +12,7 @@ from ..utils import (
     ExtractorError,
     clean_html,
     int_or_none,
+    remove_end,
     sanitized_Request,
     urlencode_postdata
 )
@@ -72,15 +73,15 @@ class DramaFeverIE(DramaFeverBaseIE):
         'url': 'http://www.dramafever.com/drama/4512/1/Cooking_with_Shin/',
         'info_dict': {
             'id': '4512.1',
-            'ext': 'mp4',
-            'title': 'Cooking with Shin 4512.1',
+            'ext': 'flv',
+            'title': 'Cooking with Shin',
             'description': 'md5:a8eec7942e1664a6896fcd5e1287bfd0',
             'episode': 'Episode 1',
             'episode_number': 1,
             'thumbnail': r're:^https?://.*\.jpg',
             'timestamp': 1404336058,
             'upload_date': '20140702',
-            'duration': 343,
+            'duration': 344,
         },
         'params': {
             # m3u8 download
@@ -90,15 +91,15 @@ class DramaFeverIE(DramaFeverBaseIE):
         'url': 'http://www.dramafever.com/drama/4826/4/Mnet_Asian_Music_Awards_2015/?ap=1',
         'info_dict': {
             'id': '4826.4',
-            'ext': 'mp4',
-            'title': 'Mnet Asian Music Awards 2015 4826.4',
+            'ext': 'flv',
+            'title': 'Mnet Asian Music Awards 2015',
             'description': 'md5:3ff2ee8fedaef86e076791c909cf2e91',
             'episode': 'Mnet Asian Music Awards 2015 - Part 3',
             'episode_number': 4,
             'thumbnail': r're:^https?://.*\.jpg',
             'timestamp': 1450213200,
             'upload_date': '20151215',
-            'duration': 5602,
+            'duration': 5359,
         },
         'params': {
             # m3u8 download
@@ -122,6 +123,10 @@ class DramaFeverIE(DramaFeverBaseIE):
                     countries=self._GEO_COUNTRIES)
             raise
 
+        # title is postfixed with video id for some reason, removing
+        if info.get('title'):
+            info['title'] = remove_end(info['title'], video_id).strip()
+
         series_id, episode_number = video_id.split('.')
         episode_info = self._download_json(
             # We only need a single episode info, so restricting page size to one episode

youtube_dl/extractor/drtv.py

@@ -118,7 +118,7 @@ class DRTVIE(InfoExtractor):
                     if target == 'HDS':
                         f4m_formats = self._extract_f4m_formats(
                             uri + '?hdcore=3.3.0&plugin=aasp-3.3.0.99.43',
-                            video_id, preference, f4m_id=format_id)
+                            video_id, preference, f4m_id=format_id, fatal=False)
                         if kind == 'AudioResource':
                             for f in f4m_formats:
                                 f['vcodec'] = 'none'
@@ -126,7 +126,8 @@ class DRTVIE(InfoExtractor):
                     elif target == 'HLS':
                         formats.extend(self._extract_m3u8_formats(
                             uri, video_id, 'mp4', entry_protocol='m3u8_native',
-                            preference=preference, m3u8_id=format_id))
+                            preference=preference, m3u8_id=format_id,
+                            fatal=False))
                     else:
                         bitrate = link.get('Bitrate')
                         if bitrate:

youtube_dl/extractor/eagleplatform.py

@@ -11,6 +11,7 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     int_or_none,
+    unsmuggle_url,
 )
 
 
@@ -50,6 +51,10 @@ class EaglePlatformIE(InfoExtractor):
             'view_count': int,
         },
         'skip': 'Georestricted',
+    }, {
+        # referrer protected video (https://tvrain.ru/lite/teleshow/kak_vse_nachinalos/namin-418921/)
+        'url': 'eagleplatform:tvrainru.media.eagleplatform.com:582306',
+        'only_matching': True,
     }]
 
     @staticmethod
@@ -60,16 +65,40 @@ class EaglePlatformIE(InfoExtractor):
             webpage)
         if mobj is not None:
             return mobj.group('url')
-        # Basic usage embedding (see http://dultonmedia.github.io/eplayer/)
+        PLAYER_JS_RE = r'''
+                        <script[^>]+
+                            src=(?P<qjs>["\'])(?:https?:)?//(?P<host>(?:(?!(?P=qjs)).)+\.media\.eagleplatform\.com)/player/player\.js(?P=qjs)
+                        .+?
+                    '''
+        # "Basic usage" embedding (see http://dultonmedia.github.io/eplayer/)
         mobj = re.search(
             r'''(?xs)
-                    <script[^>]+
-                        src=(?P<q1>["\'])(?:https?:)?//(?P<host>.+?\.media\.eagleplatform\.com)/player/player\.js(?P=q1)
-                    .+?
+                    %s
                     <div[^>]+
-                        class=(?P<q2>["\'])eagleplayer(?P=q2)[^>]+
+                        class=(?P<qclass>["\'])eagleplayer(?P=qclass)[^>]+
                         data-id=["\'](?P<id>\d+)
-            ''', webpage)
+            ''' % PLAYER_JS_RE, webpage)
+        if mobj is not None:
+            return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
+        # Generalization of "Javascript code usage", "Combined usage" and
+        # "Usage without attaching to DOM" embeddings (see
+        # http://dultonmedia.github.io/eplayer/)
+        mobj = re.search(
+            r'''(?xs)
+                    %s
+                    <script>
+                    .+?
+                    new\s+EaglePlayer\(
+                        (?:[^,]+\s*,\s*)?
+                        {
+                            .+?
+                            \bid\s*:\s*["\']?(?P<id>\d+)
+                            .+?
+                        }
+                    \s*\)
+                    .+?
+                    </script>
+            ''' % PLAYER_JS_RE, webpage)
         if mobj is not None:
             return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
 
@@ -79,9 +108,10 @@ class EaglePlatformIE(InfoExtractor):
         if status != 200:
             raise ExtractorError(' '.join(response['errors']), expected=True)
 
-    def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata', *args, **kwargs):
+    def _download_json(self, url_or_request, video_id, *args, **kwargs):
         try:
-            response = super(EaglePlatformIE, self)._download_json(url_or_request, video_id, note)
+            response = super(EaglePlatformIE, self)._download_json(
+                url_or_request, video_id, *args, **kwargs)
         except ExtractorError as ee:
             if isinstance(ee.cause, compat_HTTPError):
                 response = self._parse_json(ee.cause.read().decode('utf-8'), video_id)
@@ -93,11 +123,24 @@ class EaglePlatformIE(InfoExtractor):
         return self._download_json(url_or_request, video_id, note)['data'][0]
 
     def _real_extract(self, url):
+        url, smuggled_data = unsmuggle_url(url, {})
+
         mobj = re.match(self._VALID_URL, url)
         host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id')
 
+        headers = {}
+        query = {
+            'id': video_id,
+        }
+
+        referrer = smuggled_data.get('referrer')
+        if referrer:
+            headers['Referer'] = referrer
+            query['referrer'] = referrer
+
         player_data = self._download_json(
-            'http://%s/api/player_data?id=%s' % (host, video_id), video_id)
+            'http://%s/api/player_data' % host, video_id,
+            headers=headers, query=query)
 
         media = player_data['data']['playlist']['viewports'][0]['medialist'][0]
 

youtube_dl/extractor/egghead.py

@@ -1,15 +1,18 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    try_get,
+    unified_timestamp,
+)
 
 
 class EggheadCourseIE(InfoExtractor):
     IE_DESC = 'egghead.io course'
     IE_NAME = 'egghead:course'
-    _VALID_URL = r'https://egghead\.io/courses/(?P<id>[a-zA-Z_0-9-]+)'
+    _VALID_URL = r'https://egghead\.io/courses/(?P<id>[^/?#&]+)'
     _TEST = {
         'url': 'https://egghead.io/courses/professor-frisby-introduces-composable-functional-javascript',
         'playlist_count': 29,
@@ -22,18 +25,60 @@ class EggheadCourseIE(InfoExtractor):
 
     def _real_extract(self, url):
         playlist_id = self._match_id(url)
-        webpage = self._download_webpage(url, playlist_id)
 
-        title = self._html_search_regex(r'<h1 class="title">([^<]+)</h1>', webpage, 'title')
-        ul = self._search_regex(r'(?s)<ul class="series-lessons-list">(.*?)</ul>', webpage, 'session list')
+        course = self._download_json(
+            'https://egghead.io/api/v1/series/%s' % playlist_id, playlist_id)
 
-        found = re.findall(r'(?s)<a class="[^"]*"\s*href="([^"]+)">\s*<li class="item', ul)
-        entries = [self.url_result(m) for m in found]
+        entries = [
+            self.url_result(
+                'wistia:%s' % lesson['wistia_id'], ie='Wistia',
+                video_id=lesson['wistia_id'], video_title=lesson.get('title'))
+            for lesson in course['lessons'] if lesson.get('wistia_id')]
+
+        return self.playlist_result(
+            entries, playlist_id, course.get('title'),
+            course.get('description'))
+
+
+class EggheadLessonIE(InfoExtractor):
+    IE_DESC = 'egghead.io lesson'
+    IE_NAME = 'egghead:lesson'
+    _VALID_URL = r'https://egghead\.io/lessons/(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'https://egghead.io/lessons/javascript-linear-data-flow-with-container-style-types-box',
+        'info_dict': {
+            'id': 'fv5yotjxcg',
+            'ext': 'mp4',
+            'title': 'Create linear data flow with container style types (Box)',
+            'description': 'md5:9aa2cdb6f9878ed4c39ec09e85a8150e',
+            'thumbnail': r're:^https?:.*\.jpg$',
+            'timestamp': 1481296768,
+            'upload_date': '20161209',
+            'duration': 304,
+            'view_count': 0,
+            'tags': ['javascript', 'free'],
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        lesson_id = self._match_id(url)
+
+        lesson = self._download_json(
+            'https://egghead.io/api/v1/lessons/%s' % lesson_id, lesson_id)
 
         return {
-            '_type': 'playlist',
-            'id': playlist_id,
-            'title': title,
-            'description': self._og_search_description(webpage),
-            'entries': entries,
+            '_type': 'url_transparent',
+            'ie_key': 'Wistia',
+            'url': 'wistia:%s' % lesson['wistia_id'],
+            'id': lesson['wistia_id'],
+            'title': lesson.get('title'),
+            'description': lesson.get('summary'),
+            'thumbnail': lesson.get('thumb_nail'),
+            'timestamp': unified_timestamp(lesson.get('published_at')),
+            'duration': int_or_none(lesson.get('duration')),
+            'view_count': int_or_none(lesson.get('plays_count')),
+            'tags': try_get(lesson, lambda x: x['tag_list'], list),
         }

youtube_dl/extractor/espn.py

@@ -10,7 +10,25 @@ from ..utils import (
 
 
 class ESPNIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:espn\.go|(?:www\.)?espn)\.com/video/clip(?:\?.*?\bid=|/_/id/)(?P<id>\d+)'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:
+                            (?:(?:\w+\.)+)?espn\.go|
+                            (?:www\.)?espn
+                        )\.com/
+                        (?:
+                            (?:
+                                video/clip|
+                                watch/player
+                            )
+                            (?:
+                                \?.*?\bid=|
+                                /_/id/
+                            )
+                        )
+                        (?P<id>\d+)
+                    '''
+
     _TESTS = [{
         'url': 'http://espn.go.com/video/clip?id=10365079',
         'info_dict': {
@@ -25,20 +43,34 @@ class ESPNIE(InfoExtractor):
             'skip_download': True,
         },
     }, {
-        # intl video, from http://www.espnfc.us/video/mls-highlights/150/video/2743663/must-see-moments-best-of-the-mls-season
-        'url': 'http://espn.go.com/video/clip?id=2743663',
+        'url': 'https://broadband.espn.go.com/video/clip?id=18910086',
         'info_dict': {
-            'id': '2743663',
+            'id': '18910086',
             'ext': 'mp4',
-            'title': 'Must-See Moments: Best of the MLS season',
-            'description': 'md5:4c2d7232beaea572632bec41004f0aeb',
-            'timestamp': 1449446454,
-            'upload_date': '20151207',
+            'title': 'Kyrie spins around defender for two',
+            'description': 'md5:2b0f5bae9616d26fba8808350f0d2b9b',
+            'timestamp': 1489539155,
+            'upload_date': '20170315',
         },
         'params': {
             'skip_download': True,
         },
         'expected_warnings': ['Unable to download f4m manifest'],
+    }, {
+        'url': 'http://nonredline.sports.espn.go.com/video/clip?id=19744672',
+        'only_matching': True,
+    }, {
+        'url': 'https://cdn.espn.go.com/video/clip/_/id/19771774',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.espn.com/watch/player?id=19141491',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.espn.com/watch/player?bucketId=257&id=19505875',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.espn.com/watch/player/_/id/19141491',
+        'only_matching': True,
     }, {
         'url': 'http://www.espn.com/video/clip?id=10365079',
         'only_matching': True,

youtube_dl/extractor/extractors.py

@@ -185,6 +185,7 @@ from .chirbit import (
     ChirbitProfileIE,
 )
 from .cinchcast import CinchcastIE
+from .cjsw import CJSWIE
 from .clipfish import ClipfishIE
 from .cliphunter import CliphunterIE
 from .cliprs import ClipRsIE
@@ -297,7 +298,10 @@ from .dw import (
 from .eagleplatform import EaglePlatformIE
 from .ebaumsworld import EbaumsWorldIE
 from .echomsk import EchoMskIE
-from .egghead import EggheadCourseIE
+from .egghead import (
+    EggheadCourseIE,
+    EggheadLessonIE,
+)
 from .ehow import EHowIE
 from .eighttracks import EightTracksIE
 from .einthusan import EinthusanIE
@@ -469,6 +473,7 @@ from .jamendo import (
 )
 from .jeuxvideo import JeuxVideoIE
 from .jove import JoveIE
+from .joj import JojIE
 from .jwplatform import JWPlatformIE
 from .jpopsukitv import JpopsukiIE
 from .kaltura import KalturaIE
@@ -651,6 +656,10 @@ from .nextmedia import (
     AppleDailyIE,
     NextTVIE,
 )
+from .nexx import (
+    NexxIE,
+    NexxEmbedIE,
+)
 from .nfb import NFBIE
 from .nfl import NFLIE
 from .nhk import NhkVodIE
@@ -759,6 +768,7 @@ from .pandoratv import PandoraTVIE
 from .parliamentliveuk import ParliamentLiveUKIE
 from .patreon import PatreonIE
 from .pbs import PBSIE
+from .pearvideo import PearVideoIE
 from .people import PeopleIE
 from .periscope import (
     PeriscopeIE,
@@ -1204,7 +1214,8 @@ from .vk import (
 )
 from .vlive import (
     VLiveIE,
-    VLiveChannelIE
+    VLiveChannelIE,
+    VLivePlaylistIE
 )
 from .vodlocker import VodlockerIE
 from .vodpl import VODPlIE
@@ -1280,7 +1291,6 @@ from .yahoo import (
     YahooIE,
     YahooSearchIE,
 )
-from .yam import YamIE
 from .yandexmusic import (
     YandexMusicTrackIE,
     YandexMusicAlbumIE,

youtube_dl/extractor/fivetv.py

@@ -43,7 +43,7 @@ class FiveTVIE(InfoExtractor):
         'info_dict': {
             'id': 'glavnoe',
             'ext': 'mp4',
-            'title': 'Итоги недели с 8 по 14 июня 2015 года',
+            'title': r're:^Итоги недели с \d+ по \d+ \w+ \d{4} года$',
             'thumbnail': r're:^https?://.*\.jpg$',
         },
     }, {
@@ -70,7 +70,8 @@ class FiveTVIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         video_url = self._search_regex(
-            r'<a[^>]+?href="([^"]+)"[^>]+?class="videoplayer"',
+            [r'<div[^>]+?class="flowplayer[^>]+?data-href="([^"]+)"',
+             r'<a[^>]+?href="([^"]+)"[^>]+?class="videoplayer"'],
             webpage, 'video url')
 
         title = self._og_search_title(webpage, default=None) or self._search_regex(

youtube_dl/extractor/funnyordie.py

@@ -1,10 +1,14 @@
 from __future__ import unicode_literals
 
-import json
 import re
 
 from .common import InfoExtractor
-from ..utils import ExtractorError
+from ..utils import (
+    ExtractorError,
+    float_or_none,
+    int_or_none,
+    unified_timestamp,
+)
 
 
 class FunnyOrDieIE(InfoExtractor):
@@ -18,6 +22,10 @@ class FunnyOrDieIE(InfoExtractor):
             'title': 'Heart-Shaped Box: Literal Video Version',
             'description': 'md5:ea09a01bc9a1c46d9ab696c01747c338',
             'thumbnail': r're:^http:.*\.jpg$',
+            'uploader': 'DASjr',
+            'timestamp': 1317904928,
+            'upload_date': '20111006',
+            'duration': 318.3,
         },
     }, {
         'url': 'http://www.funnyordie.com/embed/e402820827',
@@ -27,6 +35,8 @@ class FunnyOrDieIE(InfoExtractor):
             'title': 'Please Use This Song (Jon Lajoie)',
             'description': 'Please use this to sell something.  www.jonlajoie.com',
             'thumbnail': r're:^http:.*\.jpg$',
+            'timestamp': 1398988800,
+            'upload_date': '20140502',
         },
         'params': {
             'skip_download': True,
@@ -100,15 +110,53 @@ class FunnyOrDieIE(InfoExtractor):
                 'url': 'http://www.funnyordie.com%s' % src,
             }]
 
-        post_json = self._search_regex(
-            r'fb_post\s*=\s*(\{.*?\});', webpage, 'post details')
-        post = json.loads(post_json)
+        timestamp = unified_timestamp(self._html_search_meta(
+            'uploadDate', webpage, 'timestamp', default=None))
+
+        uploader = self._html_search_regex(
+            r'<h\d[^>]+\bclass=["\']channel-preview-name[^>]+>(.+?)</h',
+            webpage, 'uploader', default=None)
+
+        title, description, thumbnail, duration = [None] * 4
+
+        medium = self._parse_json(
+            self._search_regex(
+                r'jsonMedium\s*=\s*({.+?});', webpage, 'JSON medium',
+                default='{}'),
+            video_id, fatal=False)
+        if medium:
+            title = medium.get('title')
+            duration = float_or_none(medium.get('duration'))
+            if not timestamp:
+                timestamp = unified_timestamp(medium.get('publishDate'))
+
+        post = self._parse_json(
+            self._search_regex(
+                r'fb_post\s*=\s*(\{.*?\});', webpage, 'post details',
+                default='{}'),
+            video_id, fatal=False)
+        if post:
+            if not title:
+                title = post.get('name')
+            description = post.get('description')
+            thumbnail = post.get('picture')
+
+        if not title:
+            title = self._og_search_title(webpage)
+        if not description:
+            description = self._og_search_description(webpage)
+        if not duration:
+            duration = int_or_none(self._html_search_meta(
+                ('video:duration', 'duration'), webpage, 'duration', default=False))
 
         return {
             'id': video_id,
-            'title': post['name'],
-            'description': post.get('description'),
-            'thumbnail': post.get('picture'),
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'uploader': uploader,
+            'timestamp': timestamp,
+            'duration': duration,
             'formats': formats,
             'subtitles': subtitles,
         }

youtube_dl/extractor/generic.py

@@ -36,6 +36,10 @@ from .brightcove import (
     BrightcoveLegacyIE,
     BrightcoveNewIE,
 )
+from .nexx import (
+    NexxIE,
+    NexxEmbedIE,
+)
 from .nbc import NBCSportsVPlayerIE
 from .ooyala import OoyalaIE
 from .rutv import RUTVIE
@@ -57,6 +61,7 @@ from .dailymotion import (
     DailymotionIE,
     DailymotionCloudIE,
 )
+from .dailymail import DailyMailIE
 from .onionstudios import OnionStudiosIE
 from .viewlift import ViewLiftEmbedIE
 from .mtv import MTVServicesEmbeddedIE
@@ -91,6 +96,7 @@ from .anvato import AnvatoIE
 from .washingtonpost import WashingtonPostIE
 from .wistia import WistiaIE
 from .mediaset import MediasetIE
+from .joj import JojIE
 
 
 class GenericIE(InfoExtractor):
@@ -759,6 +765,20 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['Dailymotion'],
         },
+        # DailyMail embed
+        {
+            'url': 'http://www.bumm.sk/krimi/2017/07/05/biztonsagi-kamera-buktatta-le-az-agg-ferfit-utlegelo-apolot',
+            'info_dict': {
+                'id': '1495629',
+                'ext': 'mp4',
+                'title': 'Care worker punches elderly dementia patient in head 11 times',
+                'description': 'md5:3a743dee84e57e48ec68bf67113199a5',
+            },
+            'add_ie': ['DailyMail'],
+            'params': {
+                'skip_download': True,
+            },
+        },
         # YouTube embed
         {
             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
@@ -1185,7 +1205,7 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['Kaltura'],
         },
-        # Eagle.Platform embed (generic URL)
+        # EaglePlatform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
@@ -1199,8 +1219,26 @@ class GenericIE(InfoExtractor):
                 'view_count': int,
                 'age_limit': 0,
             },
+            'params': {
+                'skip_download': True,
+            },
         },
-        # ClipYou (Eagle.Platform) embed (custom URL)
+        # referrer protected EaglePlatform embed
+        {
+            'url': 'https://tvrain.ru/lite/teleshow/kak_vse_nachinalos/namin-418921/',
+            'info_dict': {
+                'id': '582306',
+                'ext': 'mp4',
+                'title': 'Стас Намин: «Мы нарушили девственность Кремля»',
+                'thumbnail': r're:^https?://.*\.jpg$',
+                'duration': 3382,
+                'view_count': int,
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
+        # ClipYou (EaglePlatform) embed (custom URL)
         {
             'url': 'http://muz-tv.ru/play/7129/',
             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
@@ -1212,6 +1250,9 @@ class GenericIE(InfoExtractor):
                 'duration': 216,
                 'view_count': int,
             },
+            'params': {
+                'skip_download': True,
+            },
         },
         # Pladform embed
         {
@@ -1512,6 +1553,22 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['BrightcoveLegacy'],
         },
+        # Nexx embed
+        {
+            'url': 'https://www.funk.net/serien/5940e15073f6120001657956/items/593efbb173f6120001657503',
+            'info_dict': {
+                'id': '247746',
+                'ext': 'mp4',
+                'title': "Yesterday's Jam (OV)",
+                'description': 'md5:09bc0984723fed34e2581624a84e05f0',
+                'timestamp': 1492594816,
+                'upload_date': '20170419',
+            },
+            'params': {
+                'format': 'bestvideo',
+                'skip_download': True,
+            },
+        },
         # Facebook <iframe> embed
         {
             'url': 'https://www.hostblogger.de/blog/archives/6181-Auto-jagt-Betonmischer.html',
@@ -1749,6 +1806,26 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': [MediasetIE.ie_key()],
         },
+        {
+            # JOJ.sk embeds
+            'url': 'https://www.noviny.sk/slovensko/238543-slovenskom-sa-prehnala-vlna-silnych-burok',
+            'info_dict': {
+                'id': '238543-slovenskom-sa-prehnala-vlna-silnych-burok',
+                'title': 'Slovenskom sa prehnala vlna silných búrok',
+            },
+            'playlist_mincount': 5,
+            'add_ie': [JojIE.ie_key()],
+        },
+        {
+            # AMP embed (see https://www.ampproject.org/docs/reference/components/amp-video)
+            'url': 'https://tvrain.ru/amp/418921/',
+            'md5': 'cc00413936695987e8de148b67d14f1d',
+            'info_dict': {
+                'id': '418921',
+                'ext': 'mp4',
+                'title': 'Стас Намин: «Мы нарушили девственность Кремля»',
+            },
+        },
         # {
         #     # TODO: find another test
         #     # http://schema.org/VideoObject
@@ -2076,6 +2153,16 @@ class GenericIE(InfoExtractor):
         if bc_urls:
             return self.playlist_from_matches(bc_urls, video_id, video_title, ie='BrightcoveNew')
 
+        # Look for Nexx embeds
+        nexx_urls = NexxIE._extract_urls(webpage)
+        if nexx_urls:
+            return self.playlist_from_matches(nexx_urls, video_id, video_title, ie=NexxIE.ie_key())
+
+        # Look for Nexx iFrame embeds
+        nexx_embed_urls = NexxEmbedIE._extract_urls(webpage)
+        if nexx_embed_urls:
+            return self.playlist_from_matches(nexx_embed_urls, video_id, video_title, ie=NexxEmbedIE.ie_key())
+
         # Look for ThePlatform embeds
         tp_urls = ThePlatformIE._extract_urls(webpage)
         if tp_urls:
@@ -2148,6 +2235,12 @@ class GenericIE(InfoExtractor):
                 return self.playlist_from_matches(
                     playlists, video_id, video_title, lambda p: '//dailymotion.com/playlist/%s' % p)
 
+        # Look for DailyMail embeds
+        dailymail_urls = DailyMailIE._extract_urls(webpage)
+        if dailymail_urls:
+            return self.playlist_from_matches(
+                dailymail_urls, video_id, video_title, ie=DailyMailIE.ie_key())
+
         # Look for embedded Wistia player
         wistia_url = WistiaIE._extract_url(webpage)
         if wistia_url:
@@ -2443,12 +2536,12 @@ class GenericIE(InfoExtractor):
         if kaltura_url:
             return self.url_result(smuggle_url(kaltura_url, {'source_url': url}), KalturaIE.ie_key())
 
-        # Look for Eagle.Platform embeds
+        # Look for EaglePlatform embeds
         eagleplatform_url = EaglePlatformIE._extract_url(webpage)
         if eagleplatform_url:
-            return self.url_result(eagleplatform_url, EaglePlatformIE.ie_key())
+            return self.url_result(smuggle_url(eagleplatform_url, {'referrer': url}), EaglePlatformIE.ie_key())
 
-        # Look for ClipYou (uses Eagle.Platform) embeds
+        # Look for ClipYou (uses EaglePlatform) embeds
         mobj = re.search(
             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
         if mobj is not None:
@@ -2677,7 +2770,7 @@ class GenericIE(InfoExtractor):
         rutube_urls = RutubeIE._extract_urls(webpage)
         if rutube_urls:
             return self.playlist_from_matches(
-                rutube_urls, ie=RutubeIE.ie_key())
+                rutube_urls, video_id, video_title, ie=RutubeIE.ie_key())
 
         # Look for WashingtonPost embeds
         wapo_urls = WashingtonPostIE._extract_urls(webpage)
@@ -2691,6 +2784,12 @@ class GenericIE(InfoExtractor):
             return self.playlist_from_matches(
                 mediaset_urls, video_id, video_title, ie=MediasetIE.ie_key())
 
+        # Look for JOJ.sk embeds
+        joj_urls = JojIE._extract_urls(webpage)
+        if joj_urls:
+            return self.playlist_from_matches(
+                joj_urls, video_id, video_title, ie=JojIE.ie_key())
+
         def merge_dicts(dict1, dict2):
             merged = {}
             for k, v in dict1.items():

youtube_dl/extractor/giantbomb.py

@@ -5,9 +5,10 @@ import json
 
 from .common import InfoExtractor
 from ..utils import (
-    unescapeHTML,
-    qualities,
+    determine_ext,
     int_or_none,
+    qualities,
+    unescapeHTML,
 )
 
 
@@ -15,7 +16,7 @@ class GiantBombIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?giantbomb\.com/videos/(?P<display_id>[^/]+)/(?P<id>\d+-\d+)'
     _TEST = {
         'url': 'http://www.giantbomb.com/videos/quick-look-destiny-the-dark-below/2300-9782/',
-        'md5': '57badeface303ecf6b98b812de1b9018',
+        'md5': 'c8ea694254a59246a42831155dec57ac',
         'info_dict': {
             'id': '2300-9782',
             'display_id': 'quick-look-destiny-the-dark-below',
@@ -51,11 +52,16 @@ class GiantBombIE(InfoExtractor):
         for format_id, video_url in video['videoStreams'].items():
             if format_id == 'f4m_stream':
                 continue
-            if video_url.endswith('.f4m'):
+            ext = determine_ext(video_url)
+            if ext == 'f4m':
                 f4m_formats = self._extract_f4m_formats(video_url + '?hdcore=3.3.1', display_id)
                 if f4m_formats:
                     f4m_formats[0]['quality'] = quality(format_id)
                     formats.extend(f4m_formats)
+            elif ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    video_url, display_id, ext='mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
             else:
                 formats.append({
                     'url': video_url,

youtube_dl/extractor/googledrive.py

@@ -92,7 +92,7 @@ class GoogleDriveIE(InfoExtractor):
             if resolution:
                 f.update({
                     'width': resolution[0],
-                    'height': resolution[0],
+                    'height': resolution[1],
                 })
             formats.append(f)
         self._sort_formats(formats)

youtube_dl/extractor/itv.py

@@ -59,12 +59,18 @@ class ITVIE(InfoExtractor):
         def _add_sub_element(element, name):
             return etree.SubElement(element, _add_ns(name))
 
+        production_id = (
+            params.get('data-video-autoplay-id') or
+            '%s#001' % (
+                params.get('data-video-episode-id') or
+                video_id.replace('a', '/')))
+
         req_env = etree.Element(_add_ns('soapenv:Envelope'))
         _add_sub_element(req_env, 'soapenv:Header')
         body = _add_sub_element(req_env, 'soapenv:Body')
         get_playlist = _add_sub_element(body, ('tem:GetPlaylist'))
         request = _add_sub_element(get_playlist, 'tem:request')
-        _add_sub_element(request, 'itv:ProductionId').text = params['data-video-id']
+        _add_sub_element(request, 'itv:ProductionId').text = production_id
         _add_sub_element(request, 'itv:RequestGuid').text = compat_str(uuid.uuid4()).upper()
         vodcrid = _add_sub_element(request, 'itv:Vodcrid')
         _add_sub_element(vodcrid, 'com:Id')

youtube_dl/extractor/joj.py

@@ -0,0 +1,100 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    int_or_none,
+    js_to_json,
+    try_get,
+)
+
+
+class JojIE(InfoExtractor):
+    _VALID_URL = r'''(?x)
+                    (?:
+                        joj:|
+                        https?://media\.joj\.sk/embed/
+                    )
+                    (?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})
+                '''
+    _TESTS = [{
+        'url': 'https://media.joj.sk/embed/a388ec4c-6019-4a4a-9312-b1bee194e932',
+        'info_dict': {
+            'id': 'a388ec4c-6019-4a4a-9312-b1bee194e932',
+            'ext': 'mp4',
+            'title': 'NOVÉ BÝVANIE',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 3118,
+        }
+    }, {
+        'url': 'joj:a388ec4c-6019-4a4a-9312-b1bee194e932',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe\b[^>]+\bsrc=["\'](?P<url>(?:https?:)?//media\.joj\.sk/embed/[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
+            webpage)
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            'https://media.joj.sk/embed/%s' % video_id, video_id)
+
+        title = self._search_regex(
+            (r'videoTitle\s*:\s*(["\'])(?P<title>(?:(?!\1).)+)\1',
+             r'<title>(?P<title>[^<]+)'), webpage, 'title',
+            default=None, group='title') or self._og_search_title(webpage)
+
+        bitrates = self._parse_json(
+            self._search_regex(
+                r'(?s)bitrates\s*=\s*({.+?});', webpage, 'bitrates',
+                default='{}'),
+            video_id, transform_source=js_to_json, fatal=False)
+
+        formats = []
+        for format_url in try_get(bitrates, lambda x: x['mp4'], list) or []:
+            if isinstance(format_url, compat_str):
+                height = self._search_regex(
+                    r'(\d+)[pP]\.', format_url, 'height', default=None)
+                formats.append({
+                    'url': format_url,
+                    'format_id': '%sp' % height if height else None,
+                    'height': int(height),
+                })
+        if not formats:
+            playlist = self._download_xml(
+                'https://media.joj.sk/services/Video.php?clip=%s' % video_id,
+                video_id)
+            for file_el in playlist.findall('./files/file'):
+                path = file_el.get('path')
+                if not path:
+                    continue
+                format_id = file_el.get('id') or file_el.get('label')
+                formats.append({
+                    'url': 'http://n16.joj.sk/storage/%s' % path.replace(
+                        'dat/', '', 1),
+                    'format_id': format_id,
+                    'height': int_or_none(self._search_regex(
+                        r'(\d+)[pP]', format_id or path, 'height',
+                        default=None)),
+                })
+        self._sort_formats(formats)
+
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        duration = int_or_none(self._search_regex(
+            r'videoDuration\s*:\s*(\d+)', webpage, 'duration', fatal=False))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,
+        }

youtube_dl/extractor/kaltura.py

@@ -324,7 +324,7 @@ class KalturaIE(InfoExtractor):
         if captions:
             for caption in captions.get('objects', []):
                 # Continue if caption is not ready
-                if f.get('status') != 2:
+                if caption.get('status') != 2:
                     continue
                 if not caption.get('id'):
                     continue

youtube_dl/extractor/karrierevideos.py

@@ -48,7 +48,7 @@ class KarriereVideosIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         title = (self._html_search_meta('title', webpage, default=None) or
-                 self._search_regex(r'<h1 class="title">([^<]+)</h1>'))
+                 self._search_regex(r'<h1 class="title">([^<]+)</h1>', webpage, 'video title'))
 
         video_id = self._search_regex(
             r'/config/video/(.+?)\.xml', webpage, 'video id')

youtube_dl/extractor/mtv.py

@@ -83,7 +83,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
                 hls_url = rendition.find('./src').text
                 formats.extend(self._extract_m3u8_formats(
                     hls_url, video_id, ext='mp4', entry_protocol='m3u8_native',
-                    m3u8_id='hls'))
+                    m3u8_id='hls', fatal=False))
             else:
                 # fms
                 try:
@@ -106,7 +106,8 @@ class MTVServicesInfoExtractor(InfoExtractor):
                     }])
                 except (KeyError, TypeError):
                     raise ExtractorError('Invalid rendition field.')
-        self._sort_formats(formats)
+        if formats:
+            self._sort_formats(formats)
         return formats
 
     def _extract_subtitles(self, mdoc, mtvn_id):
@@ -133,8 +134,11 @@ class MTVServicesInfoExtractor(InfoExtractor):
             mediagen_url += 'acceptMethods='
             mediagen_url += 'hls' if use_hls else 'fms'
 
-        mediagen_doc = self._download_xml(mediagen_url, video_id,
-                                          'Downloading video urls')
+        mediagen_doc = self._download_xml(
+            mediagen_url, video_id, 'Downloading video urls', fatal=False)
+
+        if mediagen_doc is False:
+            return None
 
         item = mediagen_doc.find('./video/item')
         if item is not None and item.get('type') == 'text':
@@ -174,6 +178,13 @@ class MTVServicesInfoExtractor(InfoExtractor):
 
         formats = self._extract_video_formats(mediagen_doc, mtvn_id, video_id)
 
+        # Some parts of complete video may be missing (e.g. missing Act 3 in
+        # http://www.southpark.de/alle-episoden/s14e01-sexual-healing)
+        if not formats:
+            return None
+
+        self._sort_formats(formats)
+
         return {
             'title': title,
             'formats': formats,
@@ -205,9 +216,14 @@ class MTVServicesInfoExtractor(InfoExtractor):
         title = xpath_text(idoc, './channel/title')
         description = xpath_text(idoc, './channel/description')
 
+        entries = []
+        for item in idoc.findall('.//item'):
+            info = self._get_video_info(item, use_hls)
+            if info:
+                entries.append(info)
+
         return self.playlist_result(
-            [self._get_video_info(item, use_hls) for item in idoc.findall('.//item')],
-            playlist_title=title, playlist_description=description)
+            entries, playlist_title=title, playlist_description=description)
 
     def _extract_triforce_mgid(self, webpage, data_zone=None, video_id=None):
         triforce_feed = self._parse_json(self._search_regex(

youtube_dl/extractor/nexx.py

@@ -0,0 +1,271 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import hashlib
+import random
+import re
+import time
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    parse_duration,
+    try_get,
+    urlencode_postdata,
+)
+
+
+class NexxIE(InfoExtractor):
+    _VALID_URL = r'https?://api\.nexx(?:\.cloud|cdn\.com)/v3/(?P<domain_id>\d+)/videos/byid/(?P<id>\d+)'
+    _TESTS = [{
+        # movie
+        'url': 'https://api.nexx.cloud/v3/748/videos/byid/128907',
+        'md5': '16746bfc28c42049492385c989b26c4a',
+        'info_dict': {
+            'id': '128907',
+            'ext': 'mp4',
+            'title': 'Stiftung Warentest',
+            'alt_title': 'Wie ein Test abläuft',
+            'description': 'md5:d1ddb1ef63de721132abd38639cc2fd2',
+            'release_year': 2013,
+            'creator': 'SPIEGEL TV',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 2509,
+            'timestamp': 1384264416,
+            'upload_date': '20131112',
+        },
+        'params': {
+            'format': 'bestvideo',
+        },
+    }, {
+        # episode
+        'url': 'https://api.nexx.cloud/v3/741/videos/byid/247858',
+        'info_dict': {
+            'id': '247858',
+            'ext': 'mp4',
+            'title': 'Return of the Golden Child (OV)',
+            'description': 'md5:5d969537509a92b733de21bae249dc63',
+            'release_year': 2017,
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 1397,
+            'timestamp': 1495033267,
+            'upload_date': '20170517',
+            'episode_number': 2,
+            'season_number': 2,
+        },
+        'params': {
+            'format': 'bestvideo',
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://api.nexxcdn.com/v3/748/videos/byid/128907',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        # Reference:
+        # 1. https://nx-s.akamaized.net/files/201510/44.pdf
+
+        entries = []
+
+        # JavaScript Integration
+        mobj = re.search(
+            r'<script\b[^>]+\bsrc=["\']https?://require\.nexx(?:\.cloud|cdn\.com)/(?P<id>\d+)',
+            webpage)
+        if mobj:
+            domain_id = mobj.group('id')
+            for video_id in re.findall(
+                    r'(?is)onPLAYReady.+?_play\.init\s*\(.+?\s*,\s*["\']?(\d+)',
+                    webpage):
+                entries.append(
+                    'https://api.nexx.cloud/v3/%s/videos/byid/%s'
+                    % (domain_id, video_id))
+
+        # TODO: support more embed formats
+
+        return entries
+
+    @staticmethod
+    def _extract_url(webpage):
+        return NexxIE._extract_urls(webpage)[0]
+
+    def _handle_error(self, response):
+        status = int_or_none(try_get(
+            response, lambda x: x['metadata']['status']) or 200)
+        if 200 <= status < 300:
+            return
+        raise ExtractorError(
+            '%s said: %s' % (self.IE_NAME, response['metadata']['errorhint']),
+            expected=True)
+
+    def _call_api(self, domain_id, path, video_id, data=None, headers={}):
+        headers['Content-Type'] = 'application/x-www-form-urlencoded; charset=UTF-8'
+        result = self._download_json(
+            'https://api.nexx.cloud/v3/%s/%s' % (domain_id, path), video_id,
+            'Downloading %s JSON' % path, data=urlencode_postdata(data),
+            headers=headers)
+        self._handle_error(result)
+        return result['result']
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        domain_id, video_id = mobj.group('domain_id', 'id')
+
+        # Reverse engineered from JS code (see getDeviceID function)
+        device_id = '%d:%d:%d%d' % (
+            random.randint(1, 4), int(time.time()),
+            random.randint(1e4, 99999), random.randint(1, 9))
+
+        result = self._call_api(domain_id, 'session/init', video_id, data={
+            'nxp_devh': device_id,
+            'nxp_userh': '',
+            'precid': '0',
+            'playlicense': '0',
+            'screenx': '1920',
+            'screeny': '1080',
+            'playerversion': '6.0.00',
+            'gateway': 'html5',
+            'adGateway': '',
+            'explicitlanguage': 'en-US',
+            'addTextTemplates': '1',
+            'addDomainData': '1',
+            'addAdModel': '1',
+        }, headers={
+            'X-Request-Enable-Auth-Fallback': '1',
+        })
+
+        cid = result['general']['cid']
+
+        # As described in [1] X-Request-Token generation algorithm is
+        # as follows:
+        #   md5( operation + domain_id + domain_secret )
+        # where domain_secret is a static value that will be given by nexx.tv
+        # as per [1]. Here is how this "secret" is generated (reversed
+        # from _play.api.init function, search for clienttoken). So it's
+        # actually not static and not that much of a secret.
+        # 1. https://nexxtvstorage.blob.core.windows.net/files/201610/27.pdf
+        secret = result['device']['clienttoken'][int(device_id[0]):]
+        secret = secret[0:len(secret) - int(device_id[-1])]
+
+        op = 'byid'
+
+        # Reversed from JS code for _play.api.call function (search for
+        # X-Request-Token)
+        request_token = hashlib.md5(
+            ''.join((op, domain_id, secret)).encode('utf-8')).hexdigest()
+
+        video = self._call_api(
+            domain_id, 'videos/%s/%s' % (op, video_id), video_id, data={
+                'additionalfields': 'language,channel,actors,studio,licenseby,slug,subtitle,teaser,description',
+                'addInteractionOptions': '1',
+                'addStatusDetails': '1',
+                'addStreamDetails': '1',
+                'addCaptions': '1',
+                'addScenes': '1',
+                'addHotSpots': '1',
+                'addBumpers': '1',
+                'captionFormat': 'data',
+            }, headers={
+                'X-Request-CID': cid,
+                'X-Request-Token': request_token,
+            })
+
+        general = video['general']
+        title = general['title']
+
+        stream_data = video['streamdata']
+        language = general.get('language_raw') or ''
+
+        # TODO: reverse more cdns and formats
+
+        cdn = stream_data['cdnType']
+        assert cdn == 'azure'
+
+        azure_locator = stream_data['azureLocator']
+
+        AZURE_URL = 'http://nx-p%02d.akamaized.net/'
+
+        for secure in ('s', ''):
+            cdn_shield = stream_data.get('cdnShieldHTTP%s' % secure.upper())
+            if cdn_shield:
+                azure_base = 'http%s://%s' % (secure, cdn_shield)
+                break
+        else:
+            azure_base = AZURE_URL % int(stream_data['azureAccount'].replace('nexxplayplus', ''))
+
+        is_ml = ',' in language
+        azure_m3u8_url = '%s%s/%s_src%s.ism/Manifest(format=m3u8-aapl)' % (
+            azure_base, azure_locator, video_id, ('_manifest' if is_ml else ''))
+
+        protection_token = try_get(
+            video, lambda x: x['protectiondata']['token'], compat_str)
+        if protection_token:
+            azure_m3u8_url += '?hdnts=%s' % protection_token
+
+        formats = self._extract_m3u8_formats(
+            azure_m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
+            m3u8_id='%s-hls' % cdn)
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'alt_title': general.get('subtitle'),
+            'description': general.get('description'),
+            'release_year': int_or_none(general.get('year')),
+            'creator': general.get('studio') or general.get('studio_adref'),
+            'thumbnail': try_get(
+                video, lambda x: x['imagedata']['thumb'], compat_str),
+            'duration': parse_duration(general.get('runtime')),
+            'timestamp': int_or_none(general.get('uploaded')),
+            'episode_number': int_or_none(try_get(
+                video, lambda x: x['episodedata']['episode'])),
+            'season_number': int_or_none(try_get(
+                video, lambda x: x['episodedata']['season'])),
+            'formats': formats,
+        }
+
+
+class NexxEmbedIE(InfoExtractor):
+    _VALID_URL = r'https?://embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'http://embed.nexx.cloud/748/KC1614647Z27Y7T?autoplay=1',
+        'md5': '16746bfc28c42049492385c989b26c4a',
+        'info_dict': {
+            'id': '161464',
+            'ext': 'mp4',
+            'title': 'Nervenkitzel Achterbahn',
+            'alt_title': 'Karussellbauer in Deutschland',
+            'description': 'md5:ffe7b1cc59a01f585e0569949aef73cc',
+            'release_year': 2005,
+            'creator': 'SPIEGEL TV',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 2761,
+            'timestamp': 1394021479,
+            'upload_date': '20140305',
+        },
+        'params': {
+            'format': 'bestvideo',
+            'skip_download': True,
+        },
+    }
+
+    @staticmethod
+    def _extract_urls(webpage):
+        # Reference:
+        # 1. https://nx-s.akamaized.net/files/201510/44.pdf
+
+        # iFrame Embed Integration
+        return [mobj.group('url') for mobj in re.finditer(
+            r'<iframe[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?:(?!\1).)+)\1',
+            webpage)]
+
+    def _real_extract(self, url):
+        embed_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, embed_id)
+
+        return self.url_result(NexxIE._extract_url(webpage), ie=NexxIE.ie_key())

youtube_dl/extractor/niconico.py

@@ -1,23 +1,22 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
 import json
 import datetime
 
 from .common import InfoExtractor
 from ..compat import (
+    compat_parse_qs,
     compat_urlparse,
 )
 from ..utils import (
+    determine_ext,
     ExtractorError,
     int_or_none,
     parse_duration,
     parse_iso8601,
-    sanitized_Request,
-    xpath_text,
-    determine_ext,
     urlencode_postdata,
+    xpath_text,
 )
 
 
@@ -101,19 +100,24 @@ class NiconicoIE(InfoExtractor):
             return True
 
         # Log in
+        login_ok = True
         login_form_strs = {
-            'mail': username,
+            'mail_tel': username,
             'password': password,
         }
-        login_data = urlencode_postdata(login_form_strs)
-        request = sanitized_Request(
-            'https://secure.nicovideo.jp/secure/login', login_data)
-        login_results = self._download_webpage(
-            request, None, note='Logging in', errnote='Unable to log in')
-        if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:
+        urlh = self._request_webpage(
+            'https://account.nicovideo.jp/api/v1/login', None,
+            note='Logging in', errnote='Unable to log in',
+            data=urlencode_postdata(login_form_strs))
+        if urlh is False:
+            login_ok = False
+        else:
+            parts = compat_urlparse.urlparse(urlh.geturl())
+            if compat_parse_qs(parts.query).get('message', [None])[0] == 'cant_login':
+                login_ok = False
+        if not login_ok:
             self._downloader.report_warning('unable to log in: bad username or password')
-            return False
-        return True
+        return login_ok
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

youtube_dl/extractor/npo.py

@@ -28,7 +28,7 @@ class NPOBaseIE(InfoExtractor):
 
 class NPOIE(NPOBaseIE):
     IE_NAME = 'npo'
-    IE_DESC = 'npo.nl and ntr.nl'
+    IE_DESC = 'npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl'
     _VALID_URL = r'''(?x)
                     (?:
                         npo:|
@@ -38,7 +38,7 @@ class NPOIE(NPOBaseIE):
                                 npo\.nl/(?!(?:live|radio)/)(?:[^/]+/){2}|
                                 ntr\.nl/(?:[^/]+/){2,}|
                                 omroepwnl\.nl/video/fragment/[^/]+__|
-                                zapp\.nl/[^/]+/[^/]+/
+                                (?:zapp|npo3)\.nl/(?:[^/]+/){2}
                             )
                         )
                         (?P<id>[^/?#]+)
@@ -146,6 +146,9 @@ class NPOIE(NPOBaseIE):
     }, {
         'url': 'http://www.zapp.nl/beste-vrienden-quiz/extra-video-s/WO_NTR_1067990',
         'only_matching': True,
+    }, {
+        'url': 'https://www.npo3.nl/3onderzoekt/16-09-2015/VPWON_1239870',
+        'only_matching': True,
     }, {
         # live stream
         'url': 'npo:LI_NL1_4188102',
@@ -341,7 +344,7 @@ class NPOLiveIE(NPOBaseIE):
         webpage = self._download_webpage(url, display_id)
 
         live_id = self._search_regex(
-            r'data-prid="([^"]+)"', webpage, 'live id')
+            [r'media-id="([^"]+)"', r'data-prid="([^"]+)"'], webpage, 'live id')
 
         return {
             '_type': 'url_transparent',

youtube_dl/extractor/pearvideo.py

@@ -0,0 +1,63 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    qualities,
+    unified_timestamp,
+)
+
+
+class PearVideoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?pearvideo\.com/video_(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://www.pearvideo.com/video_1076290',
+        'info_dict': {
+            'id': '1076290',
+            'ext': 'mp4',
+            'title': '小浣熊在主人家玻璃上滚石头：没砸',
+            'description': 'md5:01d576b747de71be0ee85eb7cac25f9d',
+            'timestamp': 1494275280,
+            'upload_date': '20170508',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        quality = qualities(
+            ('ldflv', 'ld', 'sdflv', 'sd', 'hdflv', 'hd', 'src'))
+
+        formats = [{
+            'url': mobj.group('url'),
+            'format_id': mobj.group('id'),
+            'quality': quality(mobj.group('id')),
+        } for mobj in re.finditer(
+            r'(?P<id>[a-zA-Z]+)Url\s*=\s*(["\'])(?P<url>(?:https?:)?//.+?)\2',
+            webpage)]
+        self._sort_formats(formats)
+
+        title = self._search_regex(
+            (r'<h1[^>]+\bclass=(["\'])video-tt\1[^>]*>(?P<value>[^<]+)',
+             r'<[^>]+\bdata-title=(["\'])(?P<value>(?:(?!\1).)+)\1'),
+            webpage, 'title', group='value')
+        description = self._search_regex(
+            (r'<div[^>]+\bclass=(["\'])summary\1[^>]*>(?P<value>[^<]+)',
+             r'<[^>]+\bdata-summary=(["\'])(?P<value>(?:(?!\1).)+)\1'),
+            webpage, 'description', default=None,
+            group='value') or self._html_search_meta('Description', webpage)
+        timestamp = unified_timestamp(self._search_regex(
+            r'<div[^>]+\bclass=["\']date["\'][^>]*>([^<]+)',
+            webpage, 'timestamp', fatal=False))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'timestamp': timestamp,
+            'formats': formats,
+        }

youtube_dl/extractor/periscope.py

@@ -49,7 +49,7 @@ class PeriscopeIE(PeriscopeBaseIE):
     @staticmethod
     def _extract_url(webpage):
         mobj = re.search(
-            r'<iframe[^>]+src=([\'"])(?P<url>(?:https?:)?//(?:www\.)?periscope\.tv/(?:(?!\1).)+)\1', webpage)
+            r'<iframe[^>]+src=([\'"])(?P<url>(?:https?:)?//(?:www\.)?(?:periscope|pscp)\.tv/(?:(?!\1).)+)\1', webpage)
         if mobj:
             return mobj.group('url')
 

youtube_dl/extractor/slideshare.py

@@ -31,7 +31,7 @@ class SlideshareIE(InfoExtractor):
         page_title = mobj.group('title')
         webpage = self._download_webpage(url, page_title)
         slideshare_obj = self._search_regex(
-            r'\$\.extend\(slideshare_object,\s*(\{.*?\})\);',
+            r'\$\.extend\(.*?slideshare_object,\s*(\{.*?\})\);',
             webpage, 'slideshare object')
         info = json.loads(slideshare_obj)
         if info['slideshow']['type'] != 'video':

youtube_dl/extractor/spiegel.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from .nexx import NexxEmbedIE
 from .spiegeltv import SpiegeltvIE
 from ..compat import compat_urlparse
 from ..utils import (
@@ -121,6 +122,26 @@ class SpiegelArticleIE(InfoExtractor):
 
         },
         'playlist_count': 6,
+    }, {
+        # Nexx iFrame embed
+        'url': 'http://www.spiegel.de/sptv/spiegeltv/spiegel-tv-ueber-schnellste-katapult-achterbahn-der-welt-taron-a-1137884.html',
+        'info_dict': {
+            'id': '161464',
+            'ext': 'mp4',
+            'title': 'Nervenkitzel Achterbahn',
+            'alt_title': 'Karussellbauer in Deutschland',
+            'description': 'md5:ffe7b1cc59a01f585e0569949aef73cc',
+            'release_year': 2005,
+            'creator': 'SPIEGEL TV',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 2761,
+            'timestamp': 1394021479,
+            'upload_date': '20140305',
+        },
+        'params': {
+            'format': 'bestvideo',
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):
@@ -143,6 +164,9 @@ class SpiegelArticleIE(InfoExtractor):
         entries = [
             self.url_result(compat_urlparse.urljoin(
                 self.http_scheme() + '//spiegel.de/', embed_path))
-            for embed_path in embeds
-        ]
-        return self.playlist_result(entries)
+            for embed_path in embeds]
+        if embeds:
+            return self.playlist_result(entries)
+
+        return self.playlist_from_matches(
+            NexxEmbedIE._extract_urls(webpage), ie=NexxEmbedIE.ie_key())

youtube_dl/extractor/spiegeltv.py

@@ -1,114 +1,17 @@
-# coding: utf-8
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_urlparse
-from ..utils import (
-    determine_ext,
-    float_or_none,
-)
+from .nexx import NexxIE
 
 
 class SpiegeltvIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?spiegel\.tv/(?:#/)?filme/(?P<id>[\-a-z0-9]+)'
-    _TESTS = [{
-        'url': 'http://www.spiegel.tv/filme/flug-mh370/',
-        'info_dict': {
-            'id': 'flug-mh370',
-            'ext': 'm4v',
-            'title': 'Flug MH370',
-            'description': 'Das Rätsel um die Boeing 777 der Malaysia-Airlines',
-            'thumbnail': r're:http://.*\.jpg$',
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        }
-    }, {
-        'url': 'http://www.spiegel.tv/#/filme/alleskino-die-wahrheit-ueber-maenner/',
+    _VALID_URL = r'https?://(?:www\.)?spiegel\.tv/videos/(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://www.spiegel.tv/videos/161681-flug-mh370/',
         'only_matching': True,
-    }]
+    }
 
     def _real_extract(self, url):
-        if '/#/' in url:
-            url = url.replace('/#/', '/')
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-        title = self._html_search_regex(r'<h1.*?>(.*?)</h1>', webpage, 'title')
-
-        apihost = 'http://spiegeltv-ivms2-restapi.s3.amazonaws.com'
-        version_json = self._download_json(
-            '%s/version.json' % apihost, video_id,
-            note='Downloading version information')
-        version_name = version_json['version_name']
-
-        slug_json = self._download_json(
-            '%s/%s/restapi/slugs/%s.json' % (apihost, version_name, video_id),
-            video_id,
-            note='Downloading object information')
-        oid = slug_json['object_id']
-
-        media_json = self._download_json(
-            '%s/%s/restapi/media/%s.json' % (apihost, version_name, oid),
-            video_id, note='Downloading media information')
-        uuid = media_json['uuid']
-        is_wide = media_json['is_wide']
-
-        server_json = self._download_json(
-            'http://spiegeltv-prod-static.s3.amazonaws.com/projectConfigs/projectConfig.json',
-            video_id, note='Downloading server information')
-
-        format = '16x9' if is_wide else '4x3'
-
-        formats = []
-        for streamingserver in server_json['streamingserver']:
-            endpoint = streamingserver.get('endpoint')
-            if not endpoint:
-                continue
-            play_path = 'mp4:%s_spiegeltv_0500_%s.m4v' % (uuid, format)
-            if endpoint.startswith('rtmp'):
-                formats.append({
-                    'url': endpoint,
-                    'format_id': 'rtmp',
-                    'app': compat_urllib_parse_urlparse(endpoint).path[1:],
-                    'play_path': play_path,
-                    'player_path': 'http://prod-static.spiegel.tv/frontend-076.swf',
-                    'ext': 'flv',
-                    'rtmp_live': True,
-                })
-            elif determine_ext(endpoint) == 'm3u8':
-                formats.append({
-                    'url': endpoint.replace('[video]', play_path),
-                    'ext': 'm4v',
-                    'format_id': 'hls',  # Prefer hls since it allows to workaround georestriction
-                    'protocol': 'm3u8',
-                    'preference': 1,
-                    'http_headers': {
-                        'Accept-Encoding': 'deflate',  # gzip causes trouble on the server side
-                    },
-                })
-            else:
-                formats.append({
-                    'url': endpoint,
-                })
-        self._check_formats(formats, video_id)
-
-        thumbnails = []
-        for image in media_json['images']:
-            thumbnails.append({
-                'url': image['url'],
-                'width': image['width'],
-                'height': image['height'],
-            })
-
-        description = media_json['subtitle']
-        duration = float_or_none(media_json.get('duration_in_ms'), scale=1000)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'duration': duration,
-            'thumbnails': thumbnails,
-            'formats': formats,
-        }
+        return self.url_result(
+            'https://api.nexx.cloud/v3/748/videos/byid/%s'
+            % self._match_id(url), ie=NexxIE.ie_key())

youtube_dl/extractor/sportbox.py

@@ -4,7 +4,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import js_to_json
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    js_to_json,
+)
 
 
 class SportBoxEmbedIE(InfoExtractor):
@@ -14,8 +18,10 @@ class SportBoxEmbedIE(InfoExtractor):
         'info_dict': {
             'id': '211355',
             'ext': 'mp4',
-            'title': 'В Новороссийске прошел детский турнир «Поле славы боевой»',
+            'title': '211355',
             'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 292,
+            'view_count': int,
         },
         'params': {
             # m3u8 download
@@ -24,6 +30,9 @@ class SportBoxEmbedIE(InfoExtractor):
     }, {
         'url': 'http://news.sportbox.ru/vdl/player?nid=370908&only_player=1&autostart=false&playeri=2&height=340&width=580',
         'only_matching': True,
+    }, {
+        'url': 'https://news.sportbox.ru/vdl/player/media/193095',
+        'only_matching': True,
     }]
 
     @staticmethod
@@ -37,36 +46,34 @@ class SportBoxEmbedIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
+        wjplayer_data = self._parse_json(
+            self._search_regex(
+                r'(?s)wjplayer\(({.+?})\);', webpage, 'wjplayer settings'),
+            video_id, transform_source=js_to_json)
+
         formats = []
-
-        def cleanup_js(code):
-            # desktop_advert_config contains complex Javascripts and we don't need it
-            return js_to_json(re.sub(r'desktop_advert_config.*', '', code))
-
-        jwplayer_data = self._parse_json(self._search_regex(
-            r'(?s)player\.setup\(({.+?})\);', webpage, 'jwplayer settings'), video_id,
-            transform_source=cleanup_js)
-
-        hls_url = jwplayer_data.get('hls_url')
-        if hls_url:
-            formats.extend(self._extract_m3u8_formats(
-                hls_url, video_id, ext='mp4', m3u8_id='hls'))
-
-        rtsp_url = jwplayer_data.get('rtsp_url')
-        if rtsp_url:
-            formats.append({
-                'url': rtsp_url,
-                'format_id': 'rtsp',
-            })
-
+        for source in wjplayer_data['sources']:
+            src = source.get('src')
+            if not src:
+                continue
+            if determine_ext(src) == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    src, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+            else:
+                formats.append({
+                    'url': src,
+                })
         self._sort_formats(formats)
 
-        title = jwplayer_data['node_title']
-        thumbnail = jwplayer_data.get('image_url')
+        view_count = int_or_none(self._search_regex(
+            r'Просмотров\s*:\s*(\d+)', webpage, 'view count', default=None))
 
         return {
             'id': video_id,
-            'title': title,
-            'thumbnail': thumbnail,
+            'title': video_id,
+            'thumbnail': wjplayer_data.get('poster'),
+            'duration': int_or_none(wjplayer_data.get('duration')),
+            'view_count': view_count,
             'formats': formats,
         }

youtube_dl/extractor/tbs.py

@@ -8,6 +8,9 @@ from ..utils import extract_attributes
 
 
 class TBSIE(TurnerBaseIE):
+    # https://github.com/rg3/youtube-dl/issues/13658
+    _WORKING = False
+
     _VALID_URL = r'https?://(?:www\.)?(?P<site>tbs|tntdrama)\.com/videos/(?:[^/]+/)+(?P<id>[^/?#]+)\.html'
     _TESTS = [{
         'url': 'http://www.tbs.com/videos/people-of-earth/season-1/extras/2007318/theatrical-trailer.html',
@@ -17,7 +20,8 @@ class TBSIE(TurnerBaseIE):
             'ext': 'mp4',
             'title': 'Theatrical Trailer',
             'description': 'Catch the latest comedy from TBS, People of Earth, premiering Halloween night--Monday, October 31, at 9/8c.',
-        }
+        },
+        'skip': 'TBS videos are deleted after a while',
     }, {
         'url': 'http://www.tntdrama.com/videos/good-behavior/season-1/extras/1538823/you-better-run.html',
         'md5': 'ce53c6ead5e9f3280b4ad2031a6fab56',
@@ -26,7 +30,8 @@ class TBSIE(TurnerBaseIE):
             'ext': 'mp4',
             'title': 'You Better Run',
             'description': 'Letty Raines must figure out what she\'s running toward while running away from her past. Good Behavior premieres November 15 at 9/8c.',
-        }
+        },
+        'skip': 'TBS videos are deleted after a while',
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/ted.py

@@ -271,20 +271,22 @@ class TEDIE(InfoExtractor):
         }
 
     def _get_subtitles(self, video_id, talk_info):
-        languages = [lang['languageCode'] for lang in talk_info.get('languages', [])]
-        if languages:
-            sub_lang_list = {}
-            for l in languages:
-                sub_lang_list[l] = [
-                    {
-                        'url': 'http://www.ted.com/talks/subtitles/id/%s/lang/%s/format/%s' % (video_id, l, ext),
-                        'ext': ext,
-                    }
-                    for ext in ['ted', 'srt']
-                ]
-            return sub_lang_list
-        else:
-            return {}
+        sub_lang_list = {}
+        for language in try_get(
+                talk_info,
+                (lambda x: x['downloads']['languages'],
+                 lambda x: x['languages']), list):
+            lang_code = language.get('languageCode') or language.get('ianaCode')
+            if not lang_code:
+                continue
+            sub_lang_list[lang_code] = [
+                {
+                    'url': 'http://www.ted.com/talks/subtitles/id/%s/lang/%s/format/%s' % (video_id, lang_code, ext),
+                    'ext': ext,
+                }
+                for ext in ['ted', 'srt']
+            ]
+        return sub_lang_list
 
     def _watch_info(self, url, name):
         webpage = self._download_webpage(url, name)

youtube_dl/extractor/twitter.py

@@ -7,20 +7,38 @@ from .common import InfoExtractor
 from ..compat import compat_urlparse
 from ..utils import (
     determine_ext,
-    float_or_none,
-    xpath_text,
-    remove_end,
-    int_or_none,
+    dict_get,
     ExtractorError,
+    float_or_none,
+    int_or_none,
+    remove_end,
+    try_get,
+    xpath_text,
 )
 
 from .periscope import PeriscopeIE
 
 
 class TwitterBaseIE(InfoExtractor):
-    def _get_vmap_video_url(self, vmap_url, video_id):
+    def _extract_formats_from_vmap_url(self, vmap_url, video_id):
         vmap_data = self._download_xml(vmap_url, video_id)
-        return xpath_text(vmap_data, './/MediaFile').strip()
+        video_url = xpath_text(vmap_data, './/MediaFile').strip()
+        if determine_ext(video_url) == 'm3u8':
+            return self._extract_m3u8_formats(
+                video_url, video_id, ext='mp4', m3u8_id='hls',
+                entry_protocol='m3u8_native')
+        return [{
+            'url': video_url,
+        }]
+
+    @staticmethod
+    def _search_dimensions_in_video_url(a_format, video_url):
+        m = re.search(r'/(?P<width>\d+)x(?P<height>\d+)/', video_url)
+        if m:
+            a_format.update({
+                'width': int(m.group('width')),
+                'height': int(m.group('height')),
+            })
 
 
 class TwitterCardIE(TwitterBaseIE):
@@ -36,7 +54,8 @@ class TwitterCardIE(TwitterBaseIE):
                 'title': 'Twitter Card',
                 'thumbnail': r're:^https?://.*\.jpg$',
                 'duration': 30.033,
-            }
+            },
+            'skip': 'Video gone',
         },
         {
             'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768',
@@ -48,6 +67,7 @@ class TwitterCardIE(TwitterBaseIE):
                 'thumbnail': r're:^https?://.*\.jpg',
                 'duration': 80.155,
             },
+            'skip': 'Video gone',
         },
         {
             'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977',
@@ -65,7 +85,7 @@ class TwitterCardIE(TwitterBaseIE):
         },
         {
             'url': 'https://twitter.com/i/cards/tfw/v1/665289828897005568',
-            'md5': 'ab2745d0b0ce53319a534fccaa986439',
+            'md5': '6dabeaca9e68cbb71c99c322a4b42a11',
             'info_dict': {
                 'id': 'iBb2x00UVlv',
                 'ext': 'mp4',
@@ -73,16 +93,17 @@ class TwitterCardIE(TwitterBaseIE):
                 'uploader_id': '1189339351084113920',
                 'uploader': 'ArsenalTerje',
                 'title': 'Vine by ArsenalTerje',
+                'timestamp': 1447451307,
             },
             'add_ie': ['Vine'],
         }, {
             'url': 'https://twitter.com/i/videos/tweet/705235433198714880',
-            'md5': '3846d0a07109b5ab622425449b59049d',
+            'md5': '884812a2adc8aaf6fe52b15ccbfa3b88',
             'info_dict': {
                 'id': '705235433198714880',
                 'ext': 'mp4',
                 'title': 'Twitter web player',
-                'thumbnail': r're:^https?://.*\.jpg',
+                'thumbnail': r're:^https?://.*',
             },
         }, {
             'url': 'https://twitter.com/i/videos/752274308186120192',
@@ -90,6 +111,59 @@ class TwitterCardIE(TwitterBaseIE):
         },
     ]
 
+    def _parse_media_info(self, media_info, video_id):
+        formats = []
+        for media_variant in media_info.get('variants', []):
+            media_url = media_variant['url']
+            if media_url.endswith('.m3u8'):
+                formats.extend(self._extract_m3u8_formats(media_url, video_id, ext='mp4', m3u8_id='hls'))
+            elif media_url.endswith('.mpd'):
+                formats.extend(self._extract_mpd_formats(media_url, video_id, mpd_id='dash'))
+            else:
+                vbr = int_or_none(dict_get(media_variant, ('bitRate', 'bitrate')), scale=1000)
+                a_format = {
+                    'url': media_url,
+                    'format_id': 'http-%d' % vbr if vbr else 'http',
+                    'vbr': vbr,
+                }
+                # Reported bitRate may be zero
+                if not a_format['vbr']:
+                    del a_format['vbr']
+
+                self._search_dimensions_in_video_url(a_format, media_url)
+
+                formats.append(a_format)
+        return formats
+
+    def _extract_mobile_formats(self, username, video_id):
+        webpage = self._download_webpage(
+            'https://mobile.twitter.com/%s/status/%s' % (username, video_id),
+            video_id, 'Downloading mobile webpage',
+            headers={
+                # A recent mobile UA is necessary for `gt` cookie
+                'User-Agent': 'Mozilla/5.0 (Android 6.0.1; Mobile; rv:54.0) Gecko/54.0 Firefox/54.0',
+            })
+        main_script_url = self._html_search_regex(
+            r'<script[^>]+src="([^"]+main\.[^"]+)"', webpage, 'main script URL')
+        main_script = self._download_webpage(
+            main_script_url, video_id, 'Downloading main script')
+        bearer_token = self._search_regex(
+            r'BEARER_TOKEN\s*:\s*"([^"]+)"',
+            main_script, 'bearer token')
+        guest_token = self._search_regex(
+            r'document\.cookie\s*=\s*decodeURIComponent\("gt=(\d+)',
+            webpage, 'guest token')
+        api_data = self._download_json(
+            'https://api.twitter.com/2/timeline/conversation/%s.json' % video_id,
+            video_id, 'Downloading mobile API data',
+            headers={
+                'Authorization': 'Bearer ' + bearer_token,
+                'x-guest-token': guest_token,
+            })
+        media_info = try_get(api_data, lambda o: o['globalObjects']['tweets'][video_id]
+                                                  ['extended_entities']['media'][0]['video_info']) or {}
+        return self._parse_media_info(media_info, video_id)
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
@@ -117,14 +191,6 @@ class TwitterCardIE(TwitterBaseIE):
         if periscope_url:
             return self.url_result(periscope_url, PeriscopeIE.ie_key())
 
-        def _search_dimensions_in_video_url(a_format, video_url):
-            m = re.search(r'/(?P<width>\d+)x(?P<height>\d+)/', video_url)
-            if m:
-                a_format.update({
-                    'width': int(m.group('width')),
-                    'height': int(m.group('height')),
-                })
-
         video_url = config.get('video_url') or config.get('playlist', [{}])[0].get('source')
 
         if video_url:
@@ -135,15 +201,14 @@ class TwitterCardIE(TwitterBaseIE):
                     'url': video_url,
                 }
 
-                _search_dimensions_in_video_url(f, video_url)
+                self._search_dimensions_in_video_url(f, video_url)
 
                 formats.append(f)
 
         vmap_url = config.get('vmapUrl') or config.get('vmap_url')
         if vmap_url:
-            formats.append({
-                'url': self._get_vmap_video_url(vmap_url, video_id),
-            })
+            formats.extend(
+                self._extract_formats_from_vmap_url(vmap_url, video_id))
 
         media_info = None
 
@@ -152,29 +217,14 @@ class TwitterCardIE(TwitterBaseIE):
                 media_info = entity['mediaInfo']
 
         if media_info:
-            for media_variant in media_info['variants']:
-                media_url = media_variant['url']
-                if media_url.endswith('.m3u8'):
-                    formats.extend(self._extract_m3u8_formats(media_url, video_id, ext='mp4', m3u8_id='hls'))
-                elif media_url.endswith('.mpd'):
-                    formats.extend(self._extract_mpd_formats(media_url, video_id, mpd_id='dash'))
-                else:
-                    vbr = int_or_none(media_variant.get('bitRate'), scale=1000)
-                    a_format = {
-                        'url': media_url,
-                        'format_id': 'http-%d' % vbr if vbr else 'http',
-                        'vbr': vbr,
-                    }
-                    # Reported bitRate may be zero
-                    if not a_format['vbr']:
-                        del a_format['vbr']
-
-                    _search_dimensions_in_video_url(a_format, media_url)
-
-                    formats.append(a_format)
-
+            formats.extend(self._parse_media_info(media_info, video_id))
             duration = float_or_none(media_info.get('duration', {}).get('nanos'), scale=1e9)
 
+        username = config.get('user', {}).get('screen_name')
+        if username:
+            formats.extend(self._extract_mobile_formats(username, video_id))
+
+        self._remove_duplicate_formats(formats)
         self._sort_formats(formats)
 
         title = self._search_regex(r'<title>([^<]+)</title>', webpage, 'title')
@@ -255,10 +305,10 @@ class TwitterIE(InfoExtractor):
         'info_dict': {
             'id': '700207533655363584',
             'ext': 'mp4',
-            'title': 'JG - BEAT PROD: @suhmeduh #Damndaniel',
-            'description': 'JG on Twitter: "BEAT PROD: @suhmeduh  https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ"',
+            'title': 'Donte - BEAT PROD: @suhmeduh #Damndaniel',
+            'description': 'Donte on Twitter: "BEAT PROD: @suhmeduh  https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ"',
             'thumbnail': r're:^https?://.*\.jpg',
-            'uploader': 'JG',
+            'uploader': 'Donte',
             'uploader_id': 'jaydingeer',
         },
         'params': {
@@ -270,9 +320,11 @@ class TwitterIE(InfoExtractor):
         'info_dict': {
             'id': 'MIOxnrUteUd',
             'ext': 'mp4',
-            'title': 'Dr.Pepperの飲み方 #japanese #バカ #ドクペ #電動ガン',
-            'uploader': 'TAKUMA',
-            'uploader_id': '1004126642786242560',
+            'title': 'FilmDrunk - Vine of the day',
+            'description': 'FilmDrunk on Twitter: "Vine of the day https://t.co/xmTvRdqxWf"',
+            'uploader': 'FilmDrunk',
+            'uploader_id': 'Filmdrunk',
+            'timestamp': 1402826626,
             'upload_date': '20140615',
         },
         'add_ie': ['Vine'],
@@ -294,13 +346,28 @@ class TwitterIE(InfoExtractor):
         'info_dict': {
             'id': '1zqKVVlkqLaKB',
             'ext': 'mp4',
-            'title': 'Sgt Kerry Schmidt - Ontario Provincial Police - Road rage, mischief, assault, rollover and fire in one occurrence',
+            'title': 'Sgt Kerry Schmidt - LIVE on #Periscope: Road rage, mischief, assault, rollover and fire in one occurrence',
+            'description': 'Sgt Kerry Schmidt on Twitter: "LIVE on #Periscope: Road rage, mischief, assault, rollover and fire in one occurrence  https://t.co/EKrVgIXF3s"',
             'upload_date': '20160923',
             'uploader_id': 'OPP_HSD',
-            'uploader': 'Sgt Kerry Schmidt - Ontario Provincial Police',
+            'uploader': 'Sgt Kerry Schmidt',
             'timestamp': 1474613214,
         },
         'add_ie': ['Periscope'],
+    }, {
+        # has mp4 formats via mobile API
+        'url': 'https://twitter.com/news_al3alm/status/852138619213144067',
+        'info_dict': {
+            'id': '852138619213144067',
+            'ext': 'mp4',
+            'title': 'عالم الأخبار - كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة',
+            'description': 'عالم الأخبار on Twitter: "كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة   https://t.co/xg6OhpyKfN"',
+            'uploader': 'عالم الأخبار',
+            'uploader_id': 'news_al3alm',
+        },
+        'params': {
+            'format': 'best[format_id^=http-]',
+        },
     }]
 
     def _real_extract(self, url):
@@ -393,7 +460,7 @@ class TwitterAmplifyIE(TwitterBaseIE):
 
         vmap_url = self._html_search_meta(
             'twitter:amplify:vmap', webpage, 'vmap url')
-        video_url = self._get_vmap_video_url(vmap_url, video_id)
+        formats = self._extract_formats_from_vmap_url(vmap_url, video_id)
 
         thumbnails = []
         thumbnail = self._html_search_meta(
@@ -415,11 +482,10 @@ class TwitterAmplifyIE(TwitterBaseIE):
             })
 
         video_w, video_h = _find_dimension('player')
-        formats = [{
-            'url': video_url,
+        formats[0].update({
             'width': video_w,
             'height': video_h,
-        }]
+        })
 
         return {
             'id': video_id,

youtube_dl/extractor/veoh.py

@@ -12,47 +12,46 @@ from ..utils import (
 
 
 class VeohIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?veoh\.com/(?:watch|iphone/#_Watch)/(?P<id>(?:v|yapi-)[\da-zA-Z]+)'
+    _VALID_URL = r'https?://(?:www\.)?veoh\.com/(?:watch|iphone/#_Watch)/(?P<id>(?:v|e|yapi-)[\da-zA-Z]+)'
 
-    _TESTS = [
-        {
-            'url': 'http://www.veoh.com/watch/v56314296nk7Zdmz3',
-            'md5': '620e68e6a3cff80086df3348426c9ca3',
-            'info_dict': {
-                'id': '56314296',
-                'ext': 'mp4',
-                'title': 'Straight Backs Are Stronger',
-                'uploader': 'LUMOback',
-                'description': 'At LUMOback, we believe straight backs are stronger.  The LUMOback Posture & Movement Sensor:  It gently vibrates when you slouch, inspiring improved posture and mobility.  Use the app to track your data and improve your posture over time. ',
-            },
+    _TESTS = [{
+        'url': 'http://www.veoh.com/watch/v56314296nk7Zdmz3',
+        'md5': '620e68e6a3cff80086df3348426c9ca3',
+        'info_dict': {
+            'id': '56314296',
+            'ext': 'mp4',
+            'title': 'Straight Backs Are Stronger',
+            'uploader': 'LUMOback',
+            'description': 'At LUMOback, we believe straight backs are stronger.  The LUMOback Posture & Movement Sensor:  It gently vibrates when you slouch, inspiring improved posture and mobility.  Use the app to track your data and improve your posture over time. ',
         },
-        {
-            'url': 'http://www.veoh.com/watch/v27701988pbTc4wzN?h1=Chile+workers+cover+up+to+avoid+skin+damage',
-            'md5': '4a6ff84b87d536a6a71e6aa6c0ad07fa',
-            'info_dict': {
-                'id': '27701988',
-                'ext': 'mp4',
-                'title': 'Chile workers cover up to avoid skin damage',
-                'description': 'md5:2bd151625a60a32822873efc246ba20d',
-                'uploader': 'afp-news',
-                'duration': 123,
-            },
-            'skip': 'This video has been deleted.',
+    }, {
+        'url': 'http://www.veoh.com/watch/v27701988pbTc4wzN?h1=Chile+workers+cover+up+to+avoid+skin+damage',
+        'md5': '4a6ff84b87d536a6a71e6aa6c0ad07fa',
+        'info_dict': {
+            'id': '27701988',
+            'ext': 'mp4',
+            'title': 'Chile workers cover up to avoid skin damage',
+            'description': 'md5:2bd151625a60a32822873efc246ba20d',
+            'uploader': 'afp-news',
+            'duration': 123,
         },
-        {
-            'url': 'http://www.veoh.com/watch/v69525809F6Nc4frX',
-            'md5': '4fde7b9e33577bab2f2f8f260e30e979',
-            'note': 'Embedded ooyala video',
-            'info_dict': {
-                'id': '69525809',
-                'ext': 'mp4',
-                'title': 'Doctors Alter Plan For Preteen\'s Weight Loss Surgery',
-                'description': 'md5:f5a11c51f8fb51d2315bca0937526891',
-                'uploader': 'newsy-videos',
-            },
-            'skip': 'This video has been deleted.',
+        'skip': 'This video has been deleted.',
+    }, {
+        'url': 'http://www.veoh.com/watch/v69525809F6Nc4frX',
+        'md5': '4fde7b9e33577bab2f2f8f260e30e979',
+        'note': 'Embedded ooyala video',
+        'info_dict': {
+            'id': '69525809',
+            'ext': 'mp4',
+            'title': 'Doctors Alter Plan For Preteen\'s Weight Loss Surgery',
+            'description': 'md5:f5a11c51f8fb51d2315bca0937526891',
+            'uploader': 'newsy-videos',
         },
-    ]
+        'skip': 'This video has been deleted.',
+    }, {
+        'url': 'http://www.veoh.com/watch/e152215AJxZktGS',
+        'only_matching': True,
+    }]
 
     def _extract_formats(self, source):
         formats = []

youtube_dl/extractor/vh1.py

@@ -121,7 +121,11 @@ class VH1IE(MTVIE):
         idoc = self._download_xml(
             doc_url, video_id,
             'Downloading info', transform_source=fix_xml_ampersands)
-        return self.playlist_result(
-            [self._get_video_info(item) for item in idoc.findall('.//item')],
-            playlist_id=video_id,
-        )
+
+        entries = []
+        for item in idoc.findall('.//item'):
+            info = self._get_video_info(item)
+            if info:
+                entries.append(info)
+
+        return self.playlist_result(entries, playlist_id=video_id)

youtube_dl/extractor/vidio.py

@@ -56,7 +56,8 @@ class VidioIE(InfoExtractor):
         self._sort_formats(formats)
 
         duration = int_or_none(duration or self._search_regex(
-            r'data-video-duration=(["\'])(?P<duartion>\d+)\1', webpage, 'duration'))
+            r'data-video-duration=(["\'])(?P<duration>\d+)\1', webpage,
+            'duration', fatal=False, group='duration'))
         thumbnail = thumbnail or self._og_search_thumbnail(webpage)
 
         like_count = int_or_none(self._search_regex(

youtube_dl/extractor/vier.py

@@ -15,7 +15,21 @@ from ..utils import (
 class VierIE(InfoExtractor):
     IE_NAME = 'vier'
     IE_DESC = 'vier.be and vijf.be'
-    _VALID_URL = r'https?://(?:www\.)?(?P<site>vier|vijf)\.be/(?:[^/]+/videos/(?P<display_id>[^/]+)(?:/(?P<id>\d+))?|video/v3/embed/(?P<embed_id>\d+))'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?(?P<site>vier|vijf)\.be/
+                        (?:
+                            (?:
+                                [^/]+/videos|
+                                video(?:/[^/]+)*
+                            )/
+                            (?P<display_id>[^/]+)(?:/(?P<id>\d+))?|
+                            (?:
+                                video/v3/embed|
+                                embed/video/public
+                            )/(?P<embed_id>\d+)
+                        )
+                    '''
     _NETRC_MACHINE = 'vier'
     _TESTS = [{
         'url': 'http://www.vier.be/planb/videos/het-wordt-warm-de-moestuin/16129',
@@ -83,6 +97,15 @@ class VierIE(InfoExtractor):
     }, {
         'url': 'http://www.vier.be/video/v3/embed/16129',
         'only_matching': True,
+    }, {
+        'url': 'https://www.vijf.be/embed/video/public/4093',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.vier.be/video/blockbusters/in-juli-en-augustus-summer-classics',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.vier.be/video/achter-de-rug/2017/achter-de-rug-seizoen-1-aflevering-6',
+        'only_matching': True,
     }]
 
     def _real_initialize(self):
@@ -133,14 +156,20 @@ class VierIE(InfoExtractor):
         video_id = self._search_regex(
             [r'data-nid="(\d+)"', r'"nid"\s*:\s*"(\d+)"'],
             webpage, 'video id', default=video_id or display_id)
-        application = self._search_regex(
-            [r'data-application="([^"]+)"', r'"application"\s*:\s*"([^"]+)"'],
-            webpage, 'application', default=site + '_vod')
-        filename = self._search_regex(
-            [r'data-filename="([^"]+)"', r'"filename"\s*:\s*"([^"]+)"'],
-            webpage, 'filename')
 
-        playlist_url = 'http://vod.streamcloud.be/%s/_definst_/mp4:%s.mp4/playlist.m3u8' % (application, filename)
+        playlist_url = self._search_regex(
+            r'data-file=(["\'])(?P<url>(?:https?:)?//[^/]+/.+?\.m3u8.*?)\1',
+            webpage, 'm3u8 url', default=None, group='url')
+
+        if not playlist_url:
+            application = self._search_regex(
+                [r'data-application="([^"]+)"', r'"application"\s*:\s*"([^"]+)"'],
+                webpage, 'application', default=site + '_vod')
+            filename = self._search_regex(
+                [r'data-filename="([^"]+)"', r'"filename"\s*:\s*"([^"]+)"'],
+                webpage, 'filename')
+            playlist_url = 'http://vod.streamcloud.be/%s/_definst_/mp4:%s.mp4/playlist.m3u8' % (application, filename)
+
         formats = self._extract_wowza_formats(
             playlist_url, display_id, skip_protocols=['dash'])
         self._sort_formats(formats)

youtube_dl/extractor/vine.py

@@ -92,10 +92,12 @@ class VineIE(InfoExtractor):
 
         username = data.get('username')
 
+        alt_title = 'Vine by %s' % username if username else None
+
         return {
             'id': video_id,
-            'title': data.get('description'),
-            'alt_title': 'Vine by %s' % username if username else None,
+            'title': data.get('description') or alt_title or 'Vine video',
+            'alt_title': alt_title,
             'thumbnail': data.get('thumbnailUrl'),
             'timestamp': unified_timestamp(data.get('created')),
             'uploader': username,

youtube_dl/extractor/vlive.py

@@ -49,6 +49,10 @@ class VLiveIE(InfoExtractor):
         },
     }]
 
+    @classmethod
+    def suitable(cls, url):
+        return False if VLivePlaylistIE.suitable(url) else super(VLiveIE, cls).suitable(url)
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
@@ -261,3 +265,54 @@ class VLiveChannelIE(InfoExtractor):
 
         return self.playlist_result(
             entries, channel_code, channel_name)
+
+
+class VLivePlaylistIE(InfoExtractor):
+    IE_NAME = 'vlive:playlist'
+    _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<video_id>[0-9]+)/playlist/(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://www.vlive.tv/video/22867/playlist/22912',
+        'info_dict': {
+            'id': '22912',
+            'title': 'Valentine Day Message from TWICE'
+        },
+        'playlist_mincount': 9
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id, playlist_id = mobj.group('video_id', 'id')
+
+        VIDEO_URL_TEMPLATE = 'http://www.vlive.tv/video/%s'
+        if self._downloader.params.get('noplaylist'):
+            self.to_screen(
+                'Downloading just video %s because of --no-playlist' % video_id)
+            return self.url_result(
+                VIDEO_URL_TEMPLATE % video_id,
+                ie=VLiveIE.ie_key(), video_id=video_id)
+
+        self.to_screen(
+            'Downloading playlist %s - add --no-playlist to just download video'
+            % playlist_id)
+
+        webpage = self._download_webpage(
+            'http://www.vlive.tv/video/%s/playlist/%s'
+            % (video_id, playlist_id), playlist_id)
+
+        item_ids = self._parse_json(
+            self._search_regex(
+                r'playlistVideoSeqs\s*=\s*(\[[^]]+\])', webpage,
+                'playlist video seqs'),
+            playlist_id)
+
+        entries = [
+            self.url_result(
+                VIDEO_URL_TEMPLATE % item_id, ie=VLiveIE.ie_key(),
+                video_id=compat_str(item_id))
+            for item_id in item_ids]
+
+        playlist_name = self._html_search_regex(
+            r'<div[^>]+class="[^"]*multicam_playlist[^>]*>\s*<h3[^>]+>([^<]+)',
+            webpage, 'playlist title', fatal=False)
+
+        return self.playlist_result(entries, playlist_id, playlist_name)

youtube_dl/extractor/xhamster.py

@@ -3,6 +3,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     clean_html,
     dict_get,
@@ -14,12 +15,21 @@ from ..utils import (
 
 
 class XHamsterIE(InfoExtractor):
-    _VALID_URL = r'(?P<proto>https?)://(?:.+?\.)?xhamster\.com/movies/(?P<id>[0-9]+)/(?P<seo>.*?)\.html(?:\?.*)?'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:.+?\.)?xhamster\.com/
+                        (?:
+                            movies/(?P<id>\d+)/(?P<display_id>[^/]*)\.html|
+                            videos/(?P<display_id_2>[^/]*)-(?P<id_2>\d+)
+                        )
+                    '''
+
     _TESTS = [{
         'url': 'http://xhamster.com/movies/1509445/femaleagent_shy_beauty_takes_the_bait.html',
         'md5': '8281348b8d3c53d39fffb377d24eac4e',
         'info_dict': {
             'id': '1509445',
+            'display_id': 'femaleagent_shy_beauty_takes_the_bait',
             'ext': 'mp4',
             'title': 'FemaleAgent Shy beauty takes the bait',
             'upload_date': '20121014',
@@ -32,6 +42,7 @@ class XHamsterIE(InfoExtractor):
         'url': 'http://xhamster.com/movies/2221348/britney_spears_sexy_booty.html?hd',
         'info_dict': {
             'id': '2221348',
+            'display_id': 'britney_spears_sexy_booty',
             'ext': 'mp4',
             'title': 'Britney Spears  Sexy Booty',
             'upload_date': '20130914',
@@ -66,26 +77,18 @@ class XHamsterIE(InfoExtractor):
         # This video is visible for marcoalfa123456's friends only
         'url': 'https://it.xhamster.com/movies/7263980/la_mia_vicina.html',
         'only_matching': True,
+    }, {
+        # new URL schema
+        'url': 'https://pt.xhamster.com/videos/euro-pedal-pumping-7937821',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        def extract_video_url(webpage, name):
-            return self._search_regex(
-                [r'''file\s*:\s*(?P<q>["'])(?P<mp4>.+?)(?P=q)''',
-                 r'''<a\s+href=(?P<q>["'])(?P<mp4>.+?)(?P=q)\s+class=["']mp4Thumb''',
-                 r'''<video[^>]+file=(?P<q>["'])(?P<mp4>.+?)(?P=q)[^>]*>'''],
-                webpage, name, group='mp4')
-
-        def is_hd(webpage):
-            return '<div class=\'icon iconHD\'' in webpage
-
         mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id') or mobj.group('id_2')
+        display_id = mobj.group('display_id') or mobj.group('display_id_2')
 
-        video_id = mobj.group('id')
-        seo = mobj.group('seo')
-        proto = mobj.group('proto')
-        mrss_url = '%s://xhamster.com/movies/%s/%s.html' % (proto, video_id, seo)
-        webpage = self._download_webpage(mrss_url, video_id)
+        webpage = self._download_webpage(url, video_id)
 
         error = self._html_search_regex(
             r'<div[^>]+id=["\']videoClosed["\'][^>]*>(.+?)</div>',
@@ -99,6 +102,39 @@ class XHamsterIE(InfoExtractor):
              r'<title[^>]*>(.+?)(?:,\s*[^,]*?\s*Porn\s*[^,]*?:\s*xHamster[^<]*| - xHamster\.com)</title>'],
             webpage, 'title')
 
+        formats = []
+        format_urls = set()
+
+        sources = self._parse_json(
+            self._search_regex(
+                r'sources\s*:\s*({.+?})\s*,?\s*\n', webpage, 'sources',
+                default='{}'),
+            video_id, fatal=False)
+        for format_id, format_url in sources.items():
+            if not isinstance(format_url, compat_str):
+                continue
+            if format_url in format_urls:
+                continue
+            format_urls.add(format_url)
+            formats.append({
+                'format_id': format_id,
+                'url': format_url,
+                'height': int_or_none(self._search_regex(
+                    r'^(\d+)[pP]', format_id, 'height', default=None))
+            })
+
+        video_url = self._search_regex(
+            [r'''file\s*:\s*(?P<q>["'])(?P<mp4>.+?)(?P=q)''',
+             r'''<a\s+href=(?P<q>["'])(?P<mp4>.+?)(?P=q)\s+class=["']mp4Thumb''',
+             r'''<video[^>]+file=(?P<q>["'])(?P<mp4>.+?)(?P=q)[^>]*>'''],
+            webpage, 'video url', group='mp4', default=None)
+        if video_url and video_url not in format_urls:
+            formats.append({
+                'url': video_url,
+            })
+
+        self._sort_formats(formats)
+
         # Only a few videos have an description
         mobj = re.search(r'<span>Description: </span>([^<]+)', webpage)
         description = mobj.group(1) if mobj else None
@@ -117,7 +153,8 @@ class XHamsterIE(InfoExtractor):
             webpage, 'thumbnail', fatal=False, group='thumbnail')
 
         duration = parse_duration(self._search_regex(
-            r'Runtime:\s*</span>\s*([\d:]+)', webpage,
+            [r'<[^<]+\bitemprop=["\']duration["\'][^<]+\bcontent=["\'](.+?)["\']',
+             r'Runtime:\s*</span>\s*([\d:]+)'], webpage,
             'duration', fatal=False))
 
         view_count = int_or_none(self._search_regex(
@@ -132,30 +169,6 @@ class XHamsterIE(InfoExtractor):
 
         age_limit = self._rta_search(webpage)
 
-        hd = is_hd(webpage)
-
-        format_id = 'hd' if hd else 'sd'
-
-        video_url = extract_video_url(webpage, format_id)
-        formats = [{
-            'url': video_url,
-            'format_id': 'hd' if hd else 'sd',
-            'preference': 1,
-        }]
-
-        if not hd:
-            mrss_url = self._search_regex(r'<link rel="canonical" href="([^"]+)', webpage, 'mrss_url')
-            webpage = self._download_webpage(mrss_url + '?hd', video_id, note='Downloading HD webpage')
-            if is_hd(webpage):
-                video_url = extract_video_url(webpage, 'hd')
-                formats.append({
-                    'url': video_url,
-                    'format_id': 'hd',
-                    'preference': 2,
-                })
-
-        self._sort_formats(formats)
-
         categories_html = self._search_regex(
             r'(?s)<table.+?(<span>Categories:.+?)</table>', webpage,
             'categories', default=None)
@@ -164,6 +177,7 @@ class XHamsterIE(InfoExtractor):
 
         return {
             'id': video_id,
+            'display_id': display_id,
             'title': title,
             'description': description,
             'upload_date': upload_date,

youtube_dl/extractor/xuite.py

@@ -1,14 +1,13 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import base64
-
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_unquote
 from ..utils import (
     ExtractorError,
+    float_or_none,
+    get_element_by_attribute,
     parse_iso8601,
-    parse_duration,
+    remove_end,
 )
 
 
@@ -24,6 +23,7 @@ class XuiteIE(InfoExtractor):
             'id': '3860914',
             'ext': 'mp3',
             'title': '孤單南半球-歐德陽',
+            'description': '孤單南半球-歐德陽',
             'thumbnail': r're:^https?://.*\.jpg$',
             'duration': 247.246,
             'timestamp': 1314932940,
@@ -44,7 +44,7 @@ class XuiteIE(InfoExtractor):
             'duration': 596.458,
             'timestamp': 1454242500,
             'upload_date': '20160131',
-            'uploader': 'yan12125',
+            'uploader': '屁姥',
             'uploader_id': '12158353',
             'categories': ['個人短片'],
             'description': 'http://download.blender.org/peach/bigbuckbunny_movies/BigBuckBunny_320x180.mp4',
@@ -72,10 +72,10 @@ class XuiteIE(InfoExtractor):
         # from http://forgetfulbc.blogspot.com/2016/06/date.html
         'url': 'http://vlog.xuite.net/embed/cE1xbENoLTI3NDQ3MzM2LmZsdg==?ar=0&as=0',
         'info_dict': {
-            'id': 'cE1xbENoLTI3NDQ3MzM2LmZsdg==',
+            'id': '27447336',
             'ext': 'mp4',
             'title': '男女平權只是口號？專家解釋約會時男生是否該幫女生付錢 (中字)',
-            'description': 'md5:f0abdcb69df300f522a5442ef3146f2a',
+            'description': 'md5:1223810fa123b179083a3aed53574706',
             'timestamp': 1466160960,
             'upload_date': '20160617',
             'uploader': 'B.C. & Lowy',
@@ -86,29 +86,9 @@ class XuiteIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    @staticmethod
-    def base64_decode_utf8(data):
-        return base64.b64decode(data.encode('utf-8')).decode('utf-8')
-
-    @staticmethod
-    def base64_encode_utf8(data):
-        return base64.b64encode(data.encode('utf-8')).decode('utf-8')
-
-    def _extract_flv_config(self, encoded_media_id):
-        flv_config = self._download_xml(
-            'http://vlog.xuite.net/flash/player?media=%s' % encoded_media_id,
-            'flv config')
-        prop_dict = {}
-        for prop in flv_config.findall('./property'):
-            prop_id = self.base64_decode_utf8(prop.attrib['id'])
-            # CDATA may be empty in flv config
-            if not prop.text:
-                continue
-            encoded_content = self.base64_decode_utf8(prop.text)
-            prop_dict[prop_id] = compat_urllib_parse_unquote(encoded_content)
-        return prop_dict
-
     def _real_extract(self, url):
+        # /play/ URLs provide embedded video URL and more metadata
+        url = url.replace('/embed/', '/play/')
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
@@ -121,51 +101,53 @@ class XuiteIE(InfoExtractor):
                 '%s returned error: %s' % (self.IE_NAME, error_msg),
                 expected=True)
 
-        encoded_media_id = self._search_regex(
-            r'attributes\.name\s*=\s*"([^"]+)"', webpage,
-            'encoded media id', default=None)
-        if encoded_media_id is None:
-            video_id = self._html_search_regex(
-                r'data-mediaid="(\d+)"', webpage, 'media id')
-            encoded_media_id = self.base64_encode_utf8(video_id)
-        flv_config = self._extract_flv_config(encoded_media_id)
+        media_info = self._parse_json(self._search_regex(
+            r'var\s+mediaInfo\s*=\s*({.*});', webpage, 'media info'), video_id)
 
-        FORMATS = {
-            'audio': 'mp3',
-            'video': 'mp4',
-        }
+        video_id = media_info['MEDIA_ID']
 
         formats = []
-        for format_tag in ('src', 'hq_src'):
-            video_url = flv_config.get(format_tag)
+        for key in ('html5Url', 'html5HQUrl'):
+            video_url = media_info.get(key)
             if not video_url:
                 continue
             format_id = self._search_regex(
-                r'\bq=(.+?)\b', video_url, 'format id', default=format_tag)
+                r'\bq=(.+?)\b', video_url, 'format id', default=None)
             formats.append({
                 'url': video_url,
-                'ext': FORMATS.get(flv_config['type'], 'mp4'),
+                'ext': 'mp4' if format_id.isnumeric() else format_id,
                 'format_id': format_id,
                 'height': int(format_id) if format_id.isnumeric() else None,
             })
         self._sort_formats(formats)
 
-        timestamp = flv_config.get('publish_datetime')
+        timestamp = media_info.get('PUBLISH_DATETIME')
         if timestamp:
             timestamp = parse_iso8601(timestamp + ' +0800', ' ')
 
-        category = flv_config.get('category')
+        category = media_info.get('catName')
         categories = [category] if category else []
 
+        uploader = media_info.get('NICKNAME')
+        uploader_url = None
+
+        author_div = get_element_by_attribute('itemprop', 'author', webpage)
+        if author_div:
+            uploader = uploader or self._html_search_meta('name', author_div)
+            uploader_url = self._html_search_regex(
+                r'<link[^>]+itemprop="url"[^>]+href="([^"]+)"', author_div,
+                'uploader URL', fatal=False)
+
         return {
             'id': video_id,
-            'title': flv_config['title'],
-            'description': flv_config.get('description'),
-            'thumbnail': flv_config.get('thumb'),
+            'title': media_info['TITLE'],
+            'description': remove_end(media_info.get('metaDesc'), ' (Xuite 影音)'),
+            'thumbnail': media_info.get('ogImageUrl'),
             'timestamp': timestamp,
-            'uploader': flv_config.get('author_name'),
-            'uploader_id': flv_config.get('author_id'),
-            'duration': parse_duration(flv_config.get('duration')),
+            'uploader': uploader,
+            'uploader_id': media_info.get('MEMBER_ID'),
+            'uploader_url': uploader_url,
+            'duration': float_or_none(media_info.get('MEDIA_DURATION'), 1000000),
             'categories': categories,
             'formats': formats,
         }

youtube_dl/extractor/yam.py

@@ -1,123 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..compat import compat_urlparse
-from ..utils import (
-    float_or_none,
-    month_by_abbreviation,
-    ExtractorError,
-    get_element_by_attribute,
-)
-
-
-class YamIE(InfoExtractor):
-    IE_DESC = '蕃薯藤yam天空部落'
-    _VALID_URL = r'https?://mymedia\.yam\.com/m/(?P<id>\d+)'
-
-    _TESTS = [{
-        # An audio hosted on Yam
-        'url': 'http://mymedia.yam.com/m/2283921',
-        'md5': 'c011b8e262a52d5473d9c2e3c9963b9c',
-        'info_dict': {
-            'id': '2283921',
-            'ext': 'mp3',
-            'title': '發現 - 趙薇 京華煙雲主題曲',
-            'description': '發現 - 趙薇 京華煙雲主題曲',
-            'uploader_id': 'princekt',
-            'upload_date': '20080807',
-            'duration': 313.0,
-        }
-    }, {
-        # An external video hosted on YouTube
-        'url': 'http://mymedia.yam.com/m/3599430',
-        'md5': '03127cf10d8f35d120a9e8e52e3b17c6',
-        'info_dict': {
-            'id': 'CNpEoQlrIgA',
-            'ext': 'mp4',
-            'upload_date': '20150306',
-            'uploader': '新莊社大瑜伽社',
-            'description': 'md5:11e2e405311633ace874f2e6226c8b17',
-            'uploader_id': '2323agoy',
-            'title': '20090412陽明山二子坪-1',
-        },
-        'skip': 'Video does not exist',
-    }, {
-        'url': 'http://mymedia.yam.com/m/3598173',
-        'info_dict': {
-            'id': '3598173',
-            'ext': 'mp4',
-        },
-        'skip': 'cause Yam system error',
-    }, {
-        'url': 'http://mymedia.yam.com/m/3599437',
-        'info_dict': {
-            'id': '3599437',
-            'ext': 'mp4',
-        },
-        'skip': 'invalid YouTube URL',
-    }, {
-        'url': 'http://mymedia.yam.com/m/2373534',
-        'md5': '7ff74b91b7a817269d83796f8c5890b1',
-        'info_dict': {
-            'id': '2373534',
-            'ext': 'mp3',
-            'title': '林俊傑&蔡卓妍-小酒窩',
-            'description': 'md5:904003395a0fcce6cfb25028ff468420',
-            'upload_date': '20080928',
-            'uploader_id': 'onliner2',
-        }
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        page = self._download_webpage(url, video_id)
-
-        # Check for errors
-        system_msg = self._html_search_regex(
-            r'系統訊息(?:<br>|\n|\r)*([^<>]+)<br>', page, 'system message',
-            default=None)
-        if system_msg:
-            raise ExtractorError(system_msg, expected=True)
-
-        # Is it hosted externally on YouTube?
-        youtube_url = self._html_search_regex(
-            r'<embed src="(http://www.youtube.com/[^"]+)"',
-            page, 'YouTube url', default=None)
-        if youtube_url:
-            return self.url_result(youtube_url, 'Youtube')
-
-        title = self._html_search_regex(
-            r'<h1[^>]+class="heading"[^>]*>\s*(.+)\s*</h1>', page, 'title')
-
-        api_page = self._download_webpage(
-            'http://mymedia.yam.com/api/a/?pID=' + video_id, video_id,
-            note='Downloading API page')
-        api_result_obj = compat_urlparse.parse_qs(api_page)
-
-        info_table = get_element_by_attribute('class', 'info', page)
-        uploader_id = self._html_search_regex(
-            r'<!-- 發表作者 -->：[\n ]+<a href="/([a-z0-9]+)"',
-            info_table, 'uploader id', fatal=False)
-        mobj = re.search(r'<!-- 發表於 -->(?P<mon>[A-Z][a-z]{2})\s+' +
-                         r'(?P<day>\d{1,2}), (?P<year>\d{4})', page)
-        if mobj:
-            upload_date = '%s%02d%02d' % (
-                mobj.group('year'),
-                month_by_abbreviation(mobj.group('mon')),
-                int(mobj.group('day')))
-        else:
-            upload_date = None
-        duration = float_or_none(api_result_obj['totaltime'][0], scale=1000)
-
-        return {
-            'id': video_id,
-            'url': api_result_obj['mp3file'][0],
-            'title': title,
-            'description': self._html_search_meta('description', page),
-            'duration': duration,
-            'uploader_id': uploader_id,
-            'upload_date': upload_date,
-        }

youtube_dl/extractor/youku.py

@@ -1,7 +1,6 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import itertools
 import random
 import re
 import string
@@ -14,7 +13,6 @@ from ..utils import (
     js_to_json,
     str_or_none,
     strip_jsonp,
-    urljoin,
 )
 
 
@@ -222,17 +220,42 @@ class YoukuShowIE(InfoExtractor):
     _VALID_URL = r'https?://list\.youku\.com/show/id_(?P<id>[0-9a-z]+)\.html'
     IE_NAME = 'youku:show'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://list.youku.com/show/id_zc7c670be07ff11e48b3f.html',
         'info_dict': {
             'id': 'zc7c670be07ff11e48b3f',
-            'title': '花千骨 未删减版',
+            'title': '花千骨 DVD版',
             'description': 'md5:a1ae6f5618571bbeb5c9821f9c81b558',
         },
         'playlist_count': 50,
-    }
+    }, {
+        # Episode number not starting from 1
+        'url': 'http://list.youku.com/show/id_zefbfbd70efbfbd780bef.html',
+        'info_dict': {
+            'id': 'zefbfbd70efbfbd780bef',
+            'title': '超级飞侠3',
+            'description': 'md5:275715156abebe5ccc2a1992e9d56b98',
+        },
+        'playlist_count': 24,
+    }, {
+        # Ongoing playlist. The initial page is the last one
+        'url': 'http://list.youku.com/show/id_za7c275ecd7b411e1a19e.html',
+        'only_matchine': True,
+    }]
 
-    _PAGE_SIZE = 40
+    def _extract_entries(self, playlist_data_url, show_id, note, query):
+        query['callback'] = 'cb'
+        playlist_data = self._download_json(
+            playlist_data_url, show_id, query=query, note=note,
+            transform_source=lambda s: js_to_json(strip_jsonp(s)))['html']
+        drama_list = (get_element_by_class('p-drama-grid', playlist_data) or
+                      get_element_by_class('p-drama-half-row', playlist_data))
+        if drama_list is None:
+            raise ExtractorError('No episodes found')
+        video_urls = re.findall(r'<a[^>]+href="([^"]+)"', drama_list)
+        return playlist_data, [
+            self.url_result(self._proto_relative_url(video_url, 'http:'), YoukuIE.ie_key())
+            for video_url in video_urls]
 
     def _real_extract(self, url):
         show_id = self._match_id(url)
@@ -242,30 +265,29 @@ class YoukuShowIE(InfoExtractor):
         page_config = self._parse_json(self._search_regex(
             r'var\s+PageConfig\s*=\s*({.+});', webpage, 'page config'),
             show_id, transform_source=js_to_json)
-        for idx in itertools.count(0):
-            if idx == 0:
-                playlist_data_url = 'http://list.youku.com/show/module'
-                query = {'id': page_config['showid'], 'tab': 'point'}
-            else:
-                playlist_data_url = 'http://list.youku.com/show/point'
-                query = {
-                    'id': page_config['showid'],
-                    'stage': 'reload_%d' % (self._PAGE_SIZE * idx + 1),
-                }
-            query['callback'] = 'cb'
-            playlist_data = self._download_json(
-                playlist_data_url, show_id, query=query,
+        first_page, initial_entries = self._extract_entries(
+            'http://list.youku.com/show/module', show_id,
+            note='Downloading initial playlist data page',
+            query={
+                'id': page_config['showid'],
+                'tab': 'showInfo',
+            })
+        first_page_reload_id = self._html_search_regex(
+            r'<div[^>]+id="(reload_\d+)', first_page, 'first page reload id')
+        # The first reload_id has the same items as first_page
+        reload_ids = re.findall('<li[^>]+data-id="([^"]+)">', first_page)
+        for idx, reload_id in enumerate(reload_ids):
+            if reload_id == first_page_reload_id:
+                entries.extend(initial_entries)
+                continue
+            _, new_entries = self._extract_entries(
+                'http://list.youku.com/show/episode', show_id,
                 note='Downloading playlist data page %d' % (idx + 1),
-                transform_source=lambda s: js_to_json(strip_jsonp(s)))['html']
-            video_urls = re.findall(
-                r'<div[^>]+class="p-thumb"[^<]+<a[^>]+href="([^"]+)"',
-                playlist_data)
-            new_entries = [
-                self.url_result(urljoin(url, video_url), YoukuIE.ie_key())
-                for video_url in video_urls]
+                query={
+                    'id': page_config['showid'],
+                    'stage': reload_id,
+                })
             entries.extend(new_entries)
-            if len(new_entries) < self._PAGE_SIZE:
-                break
 
         desc = self._html_search_meta('description', webpage, fatal=False)
         playlist_title = desc.split(',')[0] if desc else None

youtube_dl/extractor/youtube.py

@@ -673,6 +673,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             },
         },
         # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
+        # YouTube Red ad is not captured for creator
         {
             'url': '__2ABJjxzNo',
             'info_dict': {
@@ -1649,7 +1650,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             video_webpage, 'license', default=None)
 
         m_music = re.search(
-            r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
+            r'''(?x)
+                <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
+                <ul[^>]*>\s*
+                <li>(?P<title>.+?)
+                by (?P<creator>.+?)
+                (?:
+                    \(.+?\)|
+                    <a[^>]*
+                        (?:
+                            \bhref=["\']/red[^>]*>|             # drop possible
+                            >\s*Listen ad-free with YouTube Red # YouTube Red ad
+                        )
+                    .*?
+                )?</li
+            ''',
             video_webpage)
         if m_music:
             video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))

youtube_dl/postprocessor/ffmpeg.py

@@ -542,7 +542,7 @@ class FFmpegFixupM3u8PP(FFmpegPostProcessor):
             temp_filename = prepend_extension(filename, 'temp')
 
             options = ['-c', 'copy', '-f', 'mp4', '-bsf:a', 'aac_adtstoasc']
-            self._downloader.to_screen('[ffmpeg] Fixing malformated aac bitstream in "%s"' % filename)
+            self._downloader.to_screen('[ffmpeg] Fixing malformed AAC bitstream in "%s"' % filename)
             self.run_ffmpeg(filename, temp_filename, options)
 
             os.remove(encodeFilename(filename))

youtube_dl/utils.py

@@ -365,9 +365,9 @@ def get_elements_by_attribute(attribute, value, html, escape_value=True):
     retlist = []
     for m in re.finditer(r'''(?xs)
         <([a-zA-Z0-9:._-]+)
-         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*?
+         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
          \s+%s=['"]?%s['"]?
-         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*?
+         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
         \s*>
         (?P<content>.*?)
         </\1>

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2017.07.02'
+__version__ = '2017.07.23'