diff options
57 files changed, 1327 insertions, 616 deletions
diff --git a/.github/ISSUE_TEMPLATE.md b/.github/ISSUE_TEMPLATE.md index a2fe59f80..c03092442 100644 --- a/.github/ISSUE_TEMPLATE.md +++ b/.github/ISSUE_TEMPLATE.md @@ -6,8 +6,8 @@ --- -### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.08.28*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected. -- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.08.28** +### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.09.04.1*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected. +- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.09.04.1** ### Before submitting an *issue* make sure you have: - [ ] At least skimmed through [README](https://github.com/rg3/youtube-dl/blob/master/README.md) and **most notably** [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections @@ -35,7 +35,7 @@ $ youtube-dl -v <your command line> [debug] User config: [] [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj'] [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251 -[debug] youtube-dl version 2016.08.28 +[debug] youtube-dl version 2016.09.04.1 [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: {} @@ -182,3 +182,4 @@ Rob van Bekkum Petr Zvoníček Pratyush Singh Aleksander Nitecki +Sebastian Blunt @@ -1,6 +1,85 @@ version <unreleased> Extractors ++ [tvnoe] New extractor (#10524) + + +version 2016.09.04.1 + +Core +* In DASH downloader if the first segment fails, abort the whole download + process to prevent throttling (#10497) ++ Add support for --skip-unavailable-fragments and --fragment retries in + hlsnative downloader (#10165, #10448). ++ Add support for --skip-unavailable-fragments in DASH downloader ++ Introduce --skip-unavailable-fragments option for fragment based downloaders + that allows to skip fragments unavailable due to a HTTP error +* Fix extraction of video/audio entries with src attribute in + _parse_html5_media_entries (#10540) + +Extractors +* [theplatform] Relax URL regular expression (#10546) +* [youtube:playlist] Extend URL regular expression +* [rottentomatoes] Delegate extraction to internetvideoarchive extractor +* [internetvideoarchive] Extract all formats +* [pornvoisines] Fix extraction (#10469) +* [rottentomatoes] Fix extraction (#10467) +* [espn] Extend URL regular expression (#10549) +* [vimple] Extend URL regular expression (#10547) +* [youtube:watchlater] Fix extraction (#10544) +* [youjizz] Fix extraction (#10437) ++ [foxnews] Add support for FoxNews Insider (#10445) ++ [fc2] Recognize Flash player URLs (#10512) + + +version 2016.09.03 + +Core +* Restore usage of NAME attribute from EXT-X-MEDIA tag for formats codes in + _extract_m3u8_formats (#10522) +* Handle semicolon in mimetype2ext + +Extractors ++ [youtube] Add support for rental videos' previews (#10532) +* [youtube:playlist] Fallback to video extraction for video/playlist URLs when + no playlist is actually served (#10537) ++ [drtv] Add support for dr.dk/nyheder (#10536) ++ [facebook:plugins:video] Add extractor (#10530) ++ [go] Add extractor for *.go.com sites +* [adobepass] Check for authz_token expiration (#10527) +* [nytimes] improve extraction +* [thestar] Fix extraction (#10465) +* [glide] Fix extraction (#10478) +- [exfm] Remove extractor (#10482) +* [youporn] Fix categories and tags extraction (#10521) ++ [curiositystream] Add extractor for app.curiositystream.com +- [thvideo] Remove extractor (#10464) +* [movingimage] Fix for the new site name (#10466) ++ [cbs] Add support for once formats (#10515) +* [limelight] Skip ism snd duplicate manifests ++ [porncom] Extract categories and tags (#10510) ++ [facebook] Extract timestamp (#10508) ++ [yahoo] Extract more formats + + +version 2016.08.31 + +Extractors +* [soundcloud] Fix URL regular expression to avoid clashes with sets (#10505) +* [bandcamp:album] Fix title extraction (#10455) +* [pyvideo] Fix extraction (#10468) ++ [ctv] Add support for tsn.ca, bnn.ca and thecomedynetwork.ca (#10016) +* [9c9media] Extract more metadata +* [9c9media] Fix multiple stacks extraction (#10016) +* [adultswim] Improve video info extraction (#10492) +* [vodplatform] Improve embed regular expression +- [played] Remove extractor (#10470) ++ [tbs] Add extractor for tbs.com and tntdrama.com (#10222) ++ [cartoonnetwork] Add extractor for cartoonnetwork.com (#10110) +* [adultswim] Rework in terms of turner extractor +* [cnn] Rework in terms of turner extractor +* [nba] Rework in terms of turner extractor ++ [turner] Add base extractor for Turner Broadcasting System based sites * [bilibili] Fix extraction (#10375) * [openload] Fix extraction (#10408) @@ -89,6 +89,8 @@ which means you can modify it, redistribute it or use it however you like. --mark-watched Mark videos watched (YouTube only) --no-mark-watched Do not mark videos watched (YouTube only) --no-color Do not emit color codes in output + --abort-on-unavailable-fragment Abort downloading when some fragment is not + available ## Network Options: --proxy URL Use the specified HTTP/HTTPS/SOCKS proxy. @@ -173,7 +175,10 @@ which means you can modify it, redistribute it or use it however you like. -R, --retries RETRIES Number of retries (default is 10), or "infinite". --fragment-retries RETRIES Number of retries for a fragment (default - is 10), or "infinite" (DASH only) + is 10), or "infinite" (DASH and hlsnative + only) + --skip-unavailable-fragments Skip unavailable fragments (DASH and + hlsnative only) --buffer-size SIZE Size of download buffer (e.g. 1024 or 16K) (default is 1024) --no-resize-buffer Do not automatically adjust the buffer diff --git a/docs/supportedsites.md b/docs/supportedsites.md index bf08697be..9e21016f7 100644 --- a/docs/supportedsites.md +++ b/docs/supportedsites.md @@ -13,6 +13,8 @@ - **5min** - **8tracks** - **91porn** + - **9c9media** + - **9c9media:stack** - **9gag** - **9now.com.au** - **abc.net.au** @@ -89,7 +91,7 @@ - **Bet** - **Bigflix** - **Bild**: Bild.de - - **BiliBili** (Currently broken) + - **BiliBili** - **BioBioChileTV** - **BIQLE** - **BleacherReport** @@ -115,6 +117,7 @@ - **Canvas** - **CarambaTV** - **CarambaTVPage** + - **CartoonNetwork** - **cbc.ca** - **cbc.ca:player** - **cbc.ca:watch** @@ -168,6 +171,8 @@ - **CTVNews** - **culturebox.francetvinfo.fr** - **CultureUnplugged** + - **curiositystream** + - **curiositystream:collection** - **CWTV** - **DailyMail** - **dailymotion** @@ -220,13 +225,14 @@ - **EsriVideo** - **Europa** - **EveryonesMixtape** - - **exfm**: ex.fm - **ExpoTV** - **ExtremeTube** - **EyedoTV** - **facebook** + - **FacebookPluginsVideo** - **faz.net** - **fc2** + - **fc2:embed** - **Fczenit** - **features.aol.com** - **fernsehkritik.tv** @@ -240,6 +246,7 @@ - **FOX** - **Foxgay** - **FoxNews**: Fox News and Fox Business Video + - **foxnews:insider** - **FoxSports** - **france2.fr:generation-quoi** - **FranceCulture** @@ -268,6 +275,7 @@ - **Glide**: Glide mobile video messages (glide.me) - **Globo** - **GloboArticle** + - **Go** - **GodTube** - **GodTV** - **Golem** @@ -403,6 +411,7 @@ - **MovieClips** - **MovieFap** - **Moviezine** + - **MovingImage** - **MPORA** - **MSN** - **mtg**: MTG services @@ -459,7 +468,6 @@ - **nick.de** - **niconico**: ニコニコ動画 - **NiconicoPlaylist** - - **NineCNineMedia** - **Nintendo** - **njoy**: N-JOY - **njoy:embed** @@ -517,7 +525,6 @@ - **Pinkbike** - **Pladform** - **play.fm** - - **played.to** - **PlaysTV** - **Playtvak**: Playtvak.cz, iDNES.cz and Lidovky.cz - **Playvid** @@ -658,7 +665,6 @@ - **sr:mediathek**: Saarländischer Rundfunk - **SRGSSR** - **SRGSSRPlay**: srf.ch, rts.ch, rsi.ch, rtr.ch and swissinfo.ch play sites - - **SSA** - **stanfordoc**: Stanford Open ClassRoom - **Steam** - **Stitcher** @@ -675,6 +681,7 @@ - **Tagesschau** - **tagesschau:player** - **Tass** + - **TBS** - **TDSLifeway** - **teachertube**: teachertube.com videos - **teachertube:user:collection**: teachertube.com user and collection videos @@ -700,8 +707,6 @@ - **TheStar** - **ThisAmericanLife** - **ThisAV** - - **THVideo** - - **THVideoPlaylist** - **tinypic**: tinypic.com videos - **tlc.de** - **TMZ** diff --git a/test/test_utils.py b/test/test_utils.py index d16ea7f77..405c5d351 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -39,6 +39,7 @@ from youtube_dl.utils import ( is_html, js_to_json, limit_length, + mimetype2ext, ohdave_rsa_encrypt, OnDemandPagedList, orderedSet, @@ -625,6 +626,14 @@ class TestUtil(unittest.TestCase): limit_length('foo bar baz asd', 12).startswith('foo bar')) self.assertTrue('...' in limit_length('foo bar baz asd', 12)) + def test_mimetype2ext(self): + self.assertEqual(mimetype2ext(None), None) + self.assertEqual(mimetype2ext('video/x-flv'), 'flv') + self.assertEqual(mimetype2ext('application/x-mpegURL'), 'm3u8') + self.assertEqual(mimetype2ext('text/vtt'), 'vtt') + self.assertEqual(mimetype2ext('text/vtt;charset=utf-8'), 'vtt') + self.assertEqual(mimetype2ext('text/html; charset=utf-8'), 'html') + def test_parse_codecs(self): self.assertEqual(parse_codecs(''), {}) self.assertEqual(parse_codecs('avc1.77.30, mp4a.40.2'), { diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index a9730292c..42128272a 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -318,6 +318,7 @@ def _real_main(argv=None): 'nooverwrites': opts.nooverwrites, 'retries': opts.retries, 'fragment_retries': opts.fragment_retries, + 'skip_unavailable_fragments': opts.skip_unavailable_fragments, 'buffersize': opts.buffersize, 'noresizebuffer': opts.noresizebuffer, 'continuedl': opts.continue_dl, diff --git a/youtube_dl/downloader/dash.py b/youtube_dl/downloader/dash.py index 8bbab9dbc..41fc9cfc2 100644 --- a/youtube_dl/downloader/dash.py +++ b/youtube_dl/downloader/dash.py @@ -38,8 +38,10 @@ class DashSegmentsFD(FragmentFD): segments_filenames = [] fragment_retries = self.params.get('fragment_retries', 0) + skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True) - def append_url_to_file(target_url, tmp_filename, segment_name): + def process_segment(segment, tmp_filename, fatal): + target_url, segment_name = segment target_filename = '%s-%s' % (tmp_filename, segment_name) count = 0 while count <= fragment_retries: @@ -52,26 +54,35 @@ class DashSegmentsFD(FragmentFD): down.close() segments_filenames.append(target_sanitized) break - except (compat_urllib_error.HTTPError, ) as err: + except compat_urllib_error.HTTPError as err: # YouTube may often return 404 HTTP error for a fragment causing the # whole download to fail. However if the same fragment is immediately # retried with the same request data this usually succeeds (1-2 attemps # is usually enough) thus allowing to download the whole file successfully. - # So, we will retry all fragments that fail with 404 HTTP error for now. - if err.code != 404: - raise - # Retry fragment + # To be future-proof we will retry all fragments that fail with any + # HTTP error. count += 1 if count <= fragment_retries: - self.report_retry_fragment(segment_name, count, fragment_retries) + self.report_retry_fragment(err, segment_name, count, fragment_retries) if count > fragment_retries: + if not fatal: + self.report_skip_fragment(segment_name) + return True self.report_error('giving up after %s fragment retries' % fragment_retries) return False - - if initialization_url: - append_url_to_file(initialization_url, ctx['tmpfilename'], 'Init') - for i, segment_url in enumerate(segment_urls): - append_url_to_file(segment_url, ctx['tmpfilename'], 'Seg%d' % i) + return True + + segments_to_download = [(initialization_url, 'Init')] if initialization_url else [] + segments_to_download.extend([ + (segment_url, 'Seg%d' % i) + for i, segment_url in enumerate(segment_urls)]) + + for i, segment in enumerate(segments_to_download): + # In DASH, the first segment contains necessary headers to + # generate a valid MP4 file, so always abort for the first segment + fatal = i == 0 or not skip_unavailable_fragments + if not process_segment(segment, ctx['tmpfilename'], fatal): + return False self._finish_frag_download(ctx) diff --git a/youtube_dl/downloader/external.py b/youtube_dl/downloader/external.py index 17f12e970..0aeae3b8f 100644 --- a/youtube_dl/downloader/external.py +++ b/youtube_dl/downloader/external.py @@ -223,7 +223,8 @@ class FFmpegFD(ExternalFD): if proxy.startswith('socks'): self.report_warning( - '%s does not support SOCKS proxies. Downloading may fail.' % self.get_basename()) + '%s does not support SOCKS proxies. Downloading is likely to fail. ' + 'Consider adding --hls-prefer-native to your command.' % self.get_basename()) # Since December 2015 ffmpeg supports -http_proxy option (see # http://git.videolan.org/?p=ffmpeg.git;a=commit;h=b4eb1f29ebddd60c41a2eb39f5af701e38e0d3fd) diff --git a/youtube_dl/downloader/fragment.py b/youtube_dl/downloader/fragment.py index ba903ae10..84aacf7db 100644 --- a/youtube_dl/downloader/fragment.py +++ b/youtube_dl/downloader/fragment.py @@ -6,6 +6,7 @@ import time from .common import FileDownloader from .http import HttpFD from ..utils import ( + error_to_compat_str, encodeFilename, sanitize_open, ) @@ -22,13 +23,19 @@ class FragmentFD(FileDownloader): Available options: - fragment_retries: Number of times to retry a fragment for HTTP error (DASH only) + fragment_retries: Number of times to retry a fragment for HTTP error (DASH + and hlsnative only) + skip_unavailable_fragments: + Skip unavailable fragments (DASH and hlsnative only) """ - def report_retry_fragment(self, fragment_name, count, retries): + def report_retry_fragment(self, err, fragment_name, count, retries): self.to_screen( - '[download] Got server HTTP error. Retrying fragment %s (attempt %d of %s)...' - % (fragment_name, count, self.format_retries(retries))) + '[download] Got server HTTP error: %s. Retrying fragment %s (attempt %d of %s)...' + % (error_to_compat_str(err), fragment_name, count, self.format_retries(retries))) + + def report_skip_fragment(self, fragment_name): + self.to_screen('[download] Skipping fragment %s...' % fragment_name) def _prepare_and_start_frag_download(self, ctx): self._prepare_frag_download(ctx) diff --git a/youtube_dl/downloader/hls.py b/youtube_dl/downloader/hls.py index baaff44d5..5d70abf62 100644 --- a/youtube_dl/downloader/hls.py +++ b/youtube_dl/downloader/hls.py @@ -13,6 +13,7 @@ from .fragment import FragmentFD from .external import FFmpegFD from ..compat import ( + compat_urllib_error, compat_urlparse, compat_struct_pack, ) @@ -83,6 +84,10 @@ class HlsFD(FragmentFD): self._prepare_and_start_frag_download(ctx) + fragment_retries = self.params.get('fragment_retries', 0) + skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True) + test = self.params.get('test', False) + extra_query = None extra_param_to_segment_url = info_dict.get('extra_param_to_segment_url') if extra_param_to_segment_url: @@ -99,15 +104,37 @@ class HlsFD(FragmentFD): line if re.match(r'^https?://', line) else compat_urlparse.urljoin(man_url, line)) - frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i) + frag_name = 'Frag%d' % i + frag_filename = '%s-%s' % (ctx['tmpfilename'], frag_name) if extra_query: frag_url = update_url_query(frag_url, extra_query) - success = ctx['dl'].download(frag_filename, {'url': frag_url}) - if not success: + count = 0 + while count <= fragment_retries: + try: + success = ctx['dl'].download(frag_filename, {'url': frag_url}) + if not success: + return False + down, frag_sanitized = sanitize_open(frag_filename, 'rb') + frag_content = down.read() + down.close() + break + except compat_urllib_error.HTTPError as err: + # Unavailable (possibly temporary) fragments may be served. + # First we try to retry then either skip or abort. + # See https://github.com/rg3/youtube-dl/issues/10165, + # https://github.com/rg3/youtube-dl/issues/10448). + count += 1 + if count <= fragment_retries: + self.report_retry_fragment(err, frag_name, count, fragment_retries) + if count > fragment_retries: + if skip_unavailable_fragments: + i += 1 + media_sequence += 1 + self.report_skip_fragment(frag_name) + continue + self.report_error( + 'giving up after %s fragment retries' % fragment_retries) return False - down, frag_sanitized = sanitize_open(frag_filename, 'rb') - frag_content = down.read() - down.close() if decrypt_info['METHOD'] == 'AES-128': iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', media_sequence) frag_content = AES.new( @@ -115,7 +142,7 @@ class HlsFD(FragmentFD): ctx['dest_stream'].write(frag_content) frags_filenames.append(frag_sanitized) # We only download the first fragment during the test - if self.params.get('test', False): + if test: break i += 1 media_sequence += 1 diff --git a/youtube_dl/extractor/adobepass.py b/youtube_dl/extractor/adobepass.py index 9e3a3e362..68ec37e00 100644 --- a/youtube_dl/extractor/adobepass.py +++ b/youtube_dl/extractor/adobepass.py @@ -37,6 +37,10 @@ class AdobePassIE(InfoExtractor): return self._search_regex( '<%s>(.+?)</%s>' % (tag, tag), xml_str, tag) + def is_expired(token, date_ele): + token_expires = unified_timestamp(re.sub(r'[_ ]GMT', '', xml_text(token, date_ele))) + return token_expires and token_expires <= int(time.time()) + mvpd_headers = { 'ap_42': 'anonymous', 'ap_11': 'Linux i686', @@ -47,11 +51,8 @@ class AdobePassIE(InfoExtractor): guid = xml_text(resource, 'guid') requestor_info = self._downloader.cache.load('mvpd', requestor_id) or {} authn_token = requestor_info.get('authn_token') - if authn_token: - token_expires = unified_timestamp(re.sub(r'[_ ]GMT', '', xml_text(authn_token, 'simpleTokenExpires'))) - if token_expires and token_expires <= int(time.time()): - authn_token = None - requestor_info = {} + if authn_token and is_expired(authn_token, 'simpleTokenExpires'): + authn_token = None if not authn_token: # TODO add support for other TV Providers mso_id = 'DTV' @@ -98,6 +99,8 @@ class AdobePassIE(InfoExtractor): self._downloader.cache.store('mvpd', requestor_id, requestor_info) authz_token = requestor_info.get(guid) + if authz_token and is_expired(authz_token, 'simpleTokenTTL'): + authz_token = None if not authz_token: authorize = self._download_webpage( self._SERVICE_PROVIDER_TEMPLATE % 'authorize', video_id, diff --git a/youtube_dl/extractor/adultswim.py b/youtube_dl/extractor/adultswim.py index ef3cc2a61..5d0bf5a68 100644 --- a/youtube_dl/extractor/adultswim.py +++ b/youtube_dl/extractor/adultswim.py @@ -4,7 +4,10 @@ from __future__ import unicode_literals import re from .turner import TurnerBaseIE -from ..utils import ExtractorError +from ..utils import ( + ExtractorError, + int_or_none, +) class AdultSwimIE(TurnerBaseIE): @@ -144,7 +147,10 @@ class AdultSwimIE(TurnerBaseIE): if bootstrapped_data.get('slugged_video', {}).get('slug') == episode_path: video_info = bootstrapped_data['slugged_video'] if not video_info: - video_info = bootstrapped_data.get('heroMetadata', {}).get('trailer').get('video') + video_info = bootstrapped_data.get( + 'heroMetadata', {}).get('trailer', {}).get('video') + if not video_info: + video_info = bootstrapped_data.get('onlineOriginals', [None])[0] if not video_info: raise ExtractorError('Unable to find video info') @@ -167,8 +173,9 @@ class AdultSwimIE(TurnerBaseIE): episode_id = video_info['id'] episode_title = video_info['title'] - episode_description = video_info['description'] - episode_duration = video_info.get('duration') + episode_description = video_info.get('description') + episode_duration = int_or_none(video_info.get('duration')) + view_count = int_or_none(video_info.get('views')) entries = [] for part_num, segment_id in enumerate(segment_ids): @@ -197,5 +204,6 @@ class AdultSwimIE(TurnerBaseIE): 'entries': entries, 'title': '%s - %s' % (show_title, episode_title), 'description': episode_description, - 'duration': episode_duration + 'duration': episode_duration, + 'view_count': view_count, } diff --git a/youtube_dl/extractor/bandcamp.py b/youtube_dl/extractor/bandcamp.py index 991ab0676..249c3d956 100644 --- a/youtube_dl/extractor/bandcamp.py +++ b/youtube_dl/extractor/bandcamp.py @@ -162,6 +162,15 @@ class BandcampAlbumIE(InfoExtractor): 'uploader_id': 'dotscale', }, 'playlist_mincount': 7, + }, { + # with escaped quote in title + 'url': 'https://jstrecords.bandcamp.com/album/entropy-ep', + 'info_dict': { + 'title': '"Entropy" EP', + 'uploader_id': 'jstrecords', + 'id': 'entropy-ep', + }, + 'playlist_mincount': 3, }] def _real_extract(self, url): @@ -176,8 +185,11 @@ class BandcampAlbumIE(InfoExtractor): entries = [ self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key()) for t_path in tracks_paths] - title = self._search_regex( - r'album_title\s*:\s*"(.*?)"', webpage, 'title', fatal=False) + title = self._html_search_regex( + r'album_title\s*:\s*"((?:\\.|[^"\\])+?)"', + webpage, 'title', fatal=False) + if title: + title = title.replace(r'\"', '"') return { '_type': 'playlist', 'uploader_id': uploader_id, diff --git a/youtube_dl/extractor/cbs.py b/youtube_dl/extractor/cbs.py index c72ed2dbb..3f4dea40c 100644 --- a/youtube_dl/extractor/cbs.py +++ b/youtube_dl/extractor/cbs.py @@ -51,7 +51,7 @@ class CBSIE(CBSBaseIE): path = 'dJ5BDC/media/guid/2198311517/' + guid smil_url = 'http://link.theplatform.com/s/%s?mbr=true' % path formats, subtitles = self._extract_theplatform_smil(smil_url + '&manifest=m3u', guid) - for r in ('HLS&formats=M3U', 'RTMP', 'WIFI', '3G'): + for r in ('OnceURL&formats=M3U', 'HLS&formats=M3U', 'RTMP', 'WIFI', '3G'): try: tp_formats, _ = self._extract_theplatform_smil(smil_url + '&assetTypes=' + r, guid, 'Downloading %s SMIL data' % r.split('&')[0]) formats.extend(tp_formats) diff --git a/youtube_dl/extractor/cnn.py b/youtube_dl/extractor/cnn.py index 1bf87f6ea..5fc311f53 100644 --- a/youtube_dl/extractor/cnn.py +++ b/youtube_dl/extractor/cnn.py @@ -15,7 +15,7 @@ class CNNIE(TurnerBaseIE): 'url': 'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn', 'md5': '3e6121ea48df7e2259fe73a0628605c4', 'info_dict': { - 'id': 'nadal-1-on-1', + 'id': 'sports/2013/06/09/nadal-1-on-1.cnn', 'ext': 'mp4', 'title': 'Nadal wins 8th French Open title', 'description': 'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.', @@ -27,7 +27,7 @@ class CNNIE(TurnerBaseIE): 'url': 'http://edition.cnn.com/video/?/video/us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fcnn_topstories+%28RSS%3A+Top+Stories%29', 'md5': 'b5cc60c60a3477d185af8f19a2a26f4e', 'info_dict': { - 'id': 'sot-student-gives-epic-speech', + 'id': 'us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology', 'ext': 'mp4', 'title': "Student's epic speech stuns new freshmen", 'description': "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\"", @@ -38,7 +38,7 @@ class CNNIE(TurnerBaseIE): 'url': 'http://www.cnn.com/video/data/2.0/video/living/2014/12/22/growing-america-nashville-salemtown-board-episode-1.hln.html', 'md5': 'f14d02ebd264df951feb2400e2c25a1b', 'info_dict': { - 'id': 'growing-america-nashville-salemtown-board-episode-1', + 'id': 'living/2014/12/22/growing-america-nashville-salemtown-board-episode-1.hln', 'ext': 'mp4', 'title': 'Nashville Ep. 1: Hand crafted skateboards', 'description': 'md5:e7223a503315c9f150acac52e76de086', @@ -49,7 +49,7 @@ class CNNIE(TurnerBaseIE): 'url': 'http://money.cnn.com/video/news/2016/08/19/netflix-stunning-stats.cnnmoney/index.html', 'md5': '52a515dc1b0f001cd82e4ceda32be9d1', 'info_dict': { - 'id': 'netflix-stunning-stats', + 'id': '/video/news/2016/08/19/netflix-stunning-stats.cnnmoney', 'ext': 'mp4', 'title': '5 stunning stats about Netflix', 'description': 'Did you know that Netflix has more than 80 million members? Here are five facts about the online video distributor that you probably didn\'t know.', @@ -83,6 +83,10 @@ class CNNIE(TurnerBaseIE): }, } + def _extract_timestamp(self, video_data): + # TODO: fix timestamp extraction + return None + def _real_extract(self, url): sub_domain, path, page_title = re.match(self._VALID_URL, url).groups() if sub_domain not in ('money', 'edition'): @@ -108,6 +112,7 @@ class CNNBlogsIE(InfoExtractor): 'description': 'Glenn Greenwald responds to comments made this week on Capitol Hill that journalists could be criminal accessories.', 'upload_date': '20140209', }, + 'expected_warnings': ['Failed to download m3u8 information'], 'add_ie': ['CNN'], } @@ -130,9 +135,10 @@ class CNNArticleIE(InfoExtractor): 'id': 'bestoftv/2014/12/21/ip-north-korea-obama.cnn', 'ext': 'mp4', 'title': 'Obama: Cyberattack not an act of war', - 'description': 'md5:51ce6750450603795cad0cdfbd7d05c5', + 'description': 'md5:0a802a40d2376f60e6b04c8d5bcebc4b', 'upload_date': '20141221', }, + 'expected_warnings': ['Failed to download m3u8 information'], 'add_ie': ['CNN'], } diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index da0af29ec..6edd5a769 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -1163,13 +1163,6 @@ class InfoExtractor(object): m3u8_id=None, note=None, errnote=None, fatal=True, live=False): - formats = [self._m3u8_meta_format(m3u8_url, ext, preference, m3u8_id)] - - format_url = lambda u: ( - u - if re.match(r'^https?://', u) - else compat_urlparse.urljoin(m3u8_url, u)) - res = self._download_webpage_handle( m3u8_url, video_id, note=note or 'Downloading m3u8 information', @@ -1180,6 +1173,13 @@ class InfoExtractor(object): m3u8_doc, urlh = res m3u8_url = urlh.geturl() + formats = [self._m3u8_meta_format(m3u8_url, ext, preference, m3u8_id)] + + format_url = lambda u: ( + u + if re.match(r'^https?://', u) + else compat_urlparse.urljoin(m3u8_url, u)) + # We should try extracting formats only from master playlists [1], i.e. # playlists that describe available qualities. On the other hand media # playlists [2] should be returned as is since they contain just the media @@ -1201,7 +1201,8 @@ class InfoExtractor(object): 'protocol': entry_protocol, 'preference': preference, }] - last_info = None + last_info = {} + last_media = {} for line in m3u8_doc.splitlines(): if line.startswith('#EXT-X-STREAM-INF:'): last_info = parse_m3u8_attributes(line) @@ -1224,23 +1225,24 @@ class InfoExtractor(object): 'protocol': entry_protocol, 'preference': preference, }) + else: + # When there is no URI in EXT-X-MEDIA let this tag's + # data be used by regular URI lines below + last_media = media elif line.startswith('#') or not line.strip(): continue else: - if last_info is None: - formats.append({'url': format_url(line)}) - continue tbr = int_or_none(last_info.get('AVERAGE-BANDWIDTH') or last_info.get('BANDWIDTH'), scale=1000) format_id = [] if m3u8_id: format_id.append(m3u8_id) + # Despite specification does not mention NAME attribute for + # EXT-X-STREAM-INF it still sometimes may be present + stream_name = last_info.get('NAME') or last_media.get('NAME') # Bandwidth of live streams may differ over time thus making # format_id unpredictable. So it's better to keep provided # format_id intact. if not live: - # Despite specification does not mention NAME attribute for - # EXT-X-STREAM-INF it still sometimes may be present - stream_name = last_info.get('NAME') format_id.append(stream_name if stream_name else '%d' % (tbr if tbr else len(formats))) f = { 'format_id': '-'.join(format_id), @@ -1269,6 +1271,7 @@ class InfoExtractor(object): f.update(parse_codecs(last_info.get('CODECS'))) formats.append(f) last_info = {} + last_media = {} return formats @staticmethod @@ -1746,7 +1749,7 @@ class InfoExtractor(object): media_attributes = extract_attributes(media_tag) src = media_attributes.get('src') if src: - _, formats = _media_formats(src) + _, formats = _media_formats(src, media_type) media_info['formats'].extend(formats) media_info['thumbnail'] = media_attributes.get('poster') if media_content: diff --git a/youtube_dl/extractor/ctv.py b/youtube_dl/extractor/ctv.py index 5807fbac9..a1fe86316 100644 --- a/youtube_dl/extractor/ctv.py +++ b/youtube_dl/extractor/ctv.py @@ -1,11 +1,13 @@ # coding: utf-8 from __future__ import unicode_literals +import re + from .common import InfoExtractor class CTVIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?ctv\.ca/video/player\?vid=(?P<id>[0-9.]+)' + _VALID_URL = r'https?://(?:www\.)?(?P<domain>ctv|tsn|bnn|thecomedynetwork)\.ca/.*?(?:\bvid=|-vid|~|%7E)(?P<id>[0-9.]+)' _TESTS = [{ 'url': 'http://www.ctv.ca/video/player?vid=706966', 'md5': 'ff2ebbeae0aa2dcc32a830c3fd69b7b0', @@ -18,13 +20,27 @@ class CTVIE(InfoExtractor): 'timestamp': 1442624700, }, 'expected_warnings': ['HTTP Error 404'], + }, { + 'url': 'http://www.thecomedynetwork.ca/video/player?vid=923582', + 'only_matching': True, + }, { + 'url': 'http://www.tsn.ca/video/expectations-high-for-milos-raonic-at-us-open~939549', + 'only_matching': True, + }, { + 'url': 'http://www.bnn.ca/video/berman-s-call-part-two-viewer-questions~939654', + 'only_matching': True, + }, { + 'url': 'http://www.ctv.ca/YourMorning/Video/S1E6-Monday-August-29-2016-vid938009', + 'only_matching': True, }] def _real_extract(self, url): - video_id = self._match_id(url) + domain, video_id = re.match(self._VALID_URL, url).groups() + if domain == 'thecomedynetwork': + domain = 'comedy' return { '_type': 'url_transparent', 'id': video_id, - 'url': '9c9media:ctv_web:%s' % video_id, + 'url': '9c9media:%s_web:%s' % (domain, video_id), 'ie_key': 'NineCNineMedia', } diff --git a/youtube_dl/extractor/curiositystream.py b/youtube_dl/extractor/curiositystream.py new file mode 100644 index 000000000..e3c99468c --- /dev/null +++ b/youtube_dl/extractor/curiositystream.py @@ -0,0 +1,120 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + int_or_none, + urlencode_postdata, + compat_str, + ExtractorError, +) + + +class CuriosityStreamBaseIE(InfoExtractor): + _NETRC_MACHINE = 'curiositystream' + _auth_token = None + _API_BASE_URL = 'https://api.curiositystream.com/v1/' + + def _handle_errors(self, result): + error = result.get('error', {}).get('message') + if error: + if isinstance(error, dict): + error = ', '.join(error.values()) + raise ExtractorError( + '%s said: %s' % (self.IE_NAME, error), expected=True) + + def _call_api(self, path, video_id): + headers = {} + if self._auth_token: + headers['X-Auth-Token'] = self._auth_token + result = self._download_json( + self._API_BASE_URL + path, video_id, headers=headers) + self._handle_errors(result) + return result['data'] + + def _real_initialize(self): + (email, password) = self._get_login_info() + if email is None: + return + result = self._download_json( + self._API_BASE_URL + 'login', None, data=urlencode_postdata({ + 'email': email, + 'password': password, + })) + self._handle_errors(result) + self._auth_token = result['message']['auth_token'] + + def _extract_media_info(self, media): + video_id = compat_str(media['id']) + limelight_media_id = media['limelight_media_id'] + title = media['title'] + + subtitles = {} + for closed_caption in media.get('closed_captions', []): + sub_url = closed_caption.get('file') + if not sub_url: + continue + lang = closed_caption.get('code') or closed_caption.get('language') or 'en' + subtitles.setdefault(lang, []).append({ + 'url': sub_url, + }) + + return { + '_type': 'url_transparent', + 'id': video_id, + 'url': 'limelight:media:' + limelight_media_id, + 'title': title, + 'description': media.get('description'), + 'thumbnail': media.get('image_large') or media.get('image_medium') or media.get('image_small'), + 'duration': int_or_none(media.get('duration')), + 'tags': media.get('tags'), + 'subtitles': subtitles, + 'ie_key': 'LimelightMedia', + } + + +class CuriosityStreamIE(CuriosityStreamBaseIE): + IE_NAME = 'curiositystream' + _VALID_URL = r'https?://app\.curiositystream\.com/video/(?P<id>\d+)' + _TEST = { + 'url': 'https://app.curiositystream.com/video/2', + 'md5': 'a0074c190e6cddaf86900b28d3e9ee7a', + 'info_dict': { + 'id': '2', + 'ext': 'mp4', + 'title': 'How Did You Develop The Internet?', + 'description': 'Vint Cerf, Google\'s Chief Internet Evangelist, describes how he and Bob Kahn created the internet.', + 'timestamp': 1448388615, + 'upload_date': '20151124', + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + media = self._call_api('media/' + video_id, video_id) + return self._extract_media_info(media) + + +class CuriosityStreamCollectionIE(CuriosityStreamBaseIE): + IE_NAME = 'curiositystream:collection' + _VALID_URL = r'https?://app\.curiositystream\.com/collection/(?P<id>\d+)' + _TEST = { + 'url': 'https://app.curiositystream.com/collection/2', + 'info_dict': { + 'id': '2', + 'title': 'Curious Minds: The Internet', + 'description': 'How is the internet shaping our lives in the 21st Century?', + }, + 'playlist_mincount': 17, + } + + def _real_extract(self, url): + collection_id = self._match_id(url) + collection = self._call_api( + 'collections/' + collection_id, collection_id) + entries = [] + for media in collection.get('media', []): + entries.append(self._extract_media_info(media)) + return self.playlist_result( + entries, collection_id, + collection.get('title'), collection.get('description')) diff --git a/youtube_dl/extractor/drtv.py b/youtube_dl/extractor/drtv.py index 2d74ff855..88d096b30 100644 --- a/youtube_dl/extractor/drtv.py +++ b/youtube_dl/extractor/drtv.py @@ -4,26 +4,45 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( ExtractorError, + int_or_none, + float_or_none, + mimetype2ext, parse_iso8601, + remove_end, ) class DRTVIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?dr\.dk/tv/se/(?:[^/]+/)*(?P<id>[\da-z-]+)(?:[/#?]|$)' + _VALID_URL = r'https?://(?:www\.)?dr\.dk/(?:tv/se|nyheder)/(?:[^/]+/)*(?P<id>[\da-z-]+)(?:[/#?]|$)' - _TEST = { - 'url': 'https://www.dr.dk/tv/se/boern/ultra/panisk-paske/panisk-paske-5', - 'md5': 'dc515a9ab50577fa14cc4e4b0265168f', + _TESTS = [{ + 'url': 'https://www.dr.dk/tv/se/boern/ultra/klassen-ultra/klassen-darlig-taber-10', + 'md5': '25e659cccc9a2ed956110a299fdf5983', 'info_dict': { - 'id': 'panisk-paske-5', + 'id': 'klassen-darlig-taber-10', 'ext': 'mp4', - 'title': 'Panisk Påske (5)', - 'description': 'md5:ca14173c5ab24cd26b0fcc074dff391c', - 'timestamp': 1426984612, - 'upload_date': '20150322', - 'duration': 1455, + 'title': 'Klassen - Dårlig taber (10)', + 'description': 'md5:815fe1b7fa656ed80580f31e8b3c79aa', + 'timestamp': 1471991907, + 'upload_date': '20160823', + 'duration': 606.84, }, - } + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'https://www.dr.dk/nyheder/indland/live-christianias-rydning-af-pusher-street-er-i-gang', + 'md5': '2c37175c718155930f939ef59952474a', + 'info_dict': { + 'id': 'christiania-pusher-street-ryddes-drdkrjpo', + 'ext': 'mp4', + 'title': 'LIVE Christianias rydning af Pusher Street er i gang', + 'description': '- Det er det fedeste, der er sket i 20 år, fortæller christianit til DR Nyheder.', + 'timestamp': 1472800279, + 'upload_date': '20160902', + 'duration': 131.4, + }, + }] def _real_extract(self, url): video_id = self._match_id(url) @@ -35,7 +54,8 @@ class DRTVIE(InfoExtractor): 'Video %s is not available' % video_id, expected=True) video_id = self._search_regex( - r'data-(?:material-identifier|episode-slug)="([^"]+)"', + (r'data-(?:material-identifier|episode-slug)="([^"]+)"', + r'data-resource="[^>"]+mu/programcard/expanded/([^"]+)"'), webpage, 'video id') programcard = self._download_json( @@ -43,9 +63,12 @@ class DRTVIE(InfoExtractor): video_id, 'Downloading video JSON') data = programcard['Data'][0] - title = data['Title'] - description = data['Description'] - timestamp = parse_iso8601(data['CreatedTime']) + title = remove_end(self._og_search_title( + webpage, default=None), ' | TV | DR') or data['Title'] + description = self._og_search_description( + webpage, default=None) or data.get('Description') + + timestamp = parse_iso8601(data.get('CreatedTime')) thumbnail = None duration = None @@ -56,16 +79,18 @@ class DRTVIE(InfoExtractor): subtitles = {} for asset in data['Assets']: - if asset['Kind'] == 'Image': - thumbnail = asset['Uri'] - elif asset['Kind'] == 'VideoResource': - duration = asset['DurationInMilliseconds'] / 1000.0 - restricted_to_denmark = asset['RestrictedToDenmark'] - spoken_subtitles = asset['Target'] == 'SpokenSubtitles' - for link in asset['Links']: - uri = link['Uri'] - target = link['Target'] - format_id = target + if asset.get('Kind') == 'Image': + thumbnail = asset.get('Uri') + elif asset.get('Kind') == 'VideoResource': + duration = float_or_none(asset.get('DurationInMilliseconds'), 1000) + restricted_to_denmark = asset.get('RestrictedToDenmark') + spoken_subtitles = asset.get('Target') == 'SpokenSubtitles' + for link in asset.get('Links', []): + uri = link.get('Uri') + if not uri: + continue + target = link.get('Target') + format_id = target or '' preference = None if spoken_subtitles: preference = -1 @@ -76,8 +101,8 @@ class DRTVIE(InfoExtractor): video_id, preference, f4m_id=format_id)) elif target == 'HLS': formats.extend(self._extract_m3u8_formats( - uri, video_id, 'mp4', preference=preference, - m3u8_id=format_id)) + uri, video_id, 'mp4', entry_protocol='m3u8_native', + preference=preference, m3u8_id=format_id)) else: bitrate = link.get('Bitrate') if bitrate: @@ -85,7 +110,7 @@ class DRTVIE(InfoExtractor): formats.append({ 'url': uri, 'format_id': format_id, - 'tbr': bitrate, + 'tbr': int_or_none(bitrate), 'ext': link.get('FileFormat'), }) subtitles_list = asset.get('SubtitlesList') @@ -94,12 +119,18 @@ class DRTVIE(InfoExtractor): 'Danish': 'da', } for subs in subtitles_list: - lang = subs['Language'] - subtitles[LANGS.get(lang, lang)] = [{'url': subs['Uri'], 'ext': 'vtt'}] + if not subs.get('Uri'): + continue + lang = subs.get('Language') or 'da' + subtitles.setdefault(LANGS.get(lang, lang), []).append({ + 'url': subs['Uri'], + 'ext': mimetype2ext(subs.get('MimeType')) or 'vtt' + }) if not formats and restricted_to_denmark: - raise ExtractorError( - 'Unfortunately, DR is not allowed to show this program outside Denmark.', expected=True) + self.raise_geo_restricted( + 'Unfortunately, DR is not allowed to show this program outside Denmark.', + expected=True) self._sort_formats(formats) diff --git a/youtube_dl/extractor/espn.py b/youtube_dl/extractor/espn.py index 66c08bec4..6d10f8e68 100644 --- a/youtube_dl/extractor/espn.py +++ b/youtube_dl/extractor/espn.py @@ -5,7 +5,7 @@ from ..utils import remove_end class ESPNIE(InfoExtractor): - _VALID_URL = r'https?://espn\.go\.com/(?:[^/]+/)*(?P<id>[^/]+)' + _VALID_URL = r'https?://(?:espn\.go|(?:www\.)?espn)\.com/(?:[^/]+/)*(?P<id>[^/]+)' _TESTS = [{ 'url': 'http://espn.go.com/video/clip?id=10365079', 'md5': '60e5d097a523e767d06479335d1bdc58', @@ -47,6 +47,9 @@ class ESPNIE(InfoExtractor): }, { 'url': 'http://espn.go.com/nba/playoffs/2015/story/_/id/12887571/john-wall-washington-wizards-no-swelling-left-hand-wrist-game-5-return', 'only_matching': True, + }, { + 'url': 'http://www.espn.com/video/clip?id=10365079', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/exfm.py b/youtube_dl/extractor/exfm.py deleted file mode 100644 index 09ed4f2b5..000000000 --- a/youtube_dl/extractor/exfm.py +++ /dev/null @@ -1,58 +0,0 @@ -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor - - -class ExfmIE(InfoExtractor): - IE_NAME = 'exfm' - IE_DESC = 'ex.fm' - _VALID_URL = r'https?://(?:www\.)?ex\.fm/song/(?P<id>[^/]+)' - _SOUNDCLOUD_URL = r'http://(?:www\.)?api\.soundcloud\.com/tracks/([^/]+)/stream' - _TESTS = [ - { - 'url': 'http://ex.fm/song/eh359', - 'md5': 'e45513df5631e6d760970b14cc0c11e7', - 'info_dict': { - 'id': '44216187', - 'ext': 'mp3', - 'title': 'Test House "Love Is Not Enough" (Extended Mix) DeadJournalist Exclusive', - 'uploader': 'deadjournalist', - 'upload_date': '20120424', - 'description': 'Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive', - }, - 'note': 'Soundcloud song', - 'skip': 'The site is down too often', - }, - { - 'url': 'http://ex.fm/song/wddt8', - 'md5': '966bd70741ac5b8570d8e45bfaed3643', - 'info_dict': { - 'id': 'wddt8', - 'ext': 'mp3', - 'title': 'Safe and Sound', - 'uploader': 'Capital Cities', - }, - 'skip': 'The site is down too often', - }, - ] - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - song_id = mobj.group('id') - info_url = 'http://ex.fm/api/v3/song/%s' % song_id - info = self._download_json(info_url, song_id)['song'] - song_url = info['url'] - if re.match(self._SOUNDCLOUD_URL, song_url) is not None: - self.to_screen('Soundcloud song detected') - return self.url_result(song_url.replace('/stream', ''), 'Soundcloud') - return { - 'id': song_id, - 'url': song_url, - 'ext': 'mp3', - 'title': info['title'], - 'thumbnail': info['image']['large'], - 'uploader': info['artist'], - 'view_count': info['loved_count'], - } diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 20e85703f..e47adc26c 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -194,6 +194,10 @@ from .ctsnews import CtsNewsIE from .ctv import CTVIE from .ctvnews import CTVNewsIE from .cultureunplugged import CultureUnpluggedIE +from .curiositystream import ( + CuriosityStreamIE, + CuriosityStreamCollectionIE, +) from .cwtv import CWTVIE from .dailymail import DailyMailIE from .dailymotion import ( @@ -257,13 +261,18 @@ from .espn import ESPNIE from .esri import EsriVideoIE from .europa import EuropaIE from .everyonesmixtape import EveryonesMixtapeIE -from .exfm import ExfmIE from .expotv import ExpoTVIE from .extremetube import ExtremeTubeIE from .eyedotv import EyedoTVIE -from .facebook import FacebookIE +from .facebook import ( + FacebookIE, + FacebookPluginsVideoIE, +) from .faz import FazIE -from .fc2 import FC2IE +from .fc2 import ( + FC2IE, + FC2EmbedIE, +) from .fczenit import FczenitIE from .firstpost import FirstpostIE from .firsttv import FirstTVIE @@ -278,7 +287,10 @@ from .formula1 import Formula1IE from .fourtube import FourTubeIE from .fox import FOXIE from .foxgay import FoxgayIE -from .foxnews import FoxNewsIE +from .foxnews import ( + FoxNewsIE, + FoxNewsInsiderIE, +) from .foxsports import FoxSportsIE from .franceculture import FranceCultureIE from .franceinter import FranceInterIE @@ -315,6 +327,7 @@ from .globo import ( GloboIE, GloboArticleIE, ) +from .go import GoIE from .godtube import GodTubeIE from .godtv import GodTVIE from .golem import GolemIE @@ -486,6 +499,7 @@ from .motherless import MotherlessIE from .motorsport import MotorsportIE from .movieclips import MovieClipsIE from .moviezine import MoviezineIE +from .movingimage import MovingImageIE from .msn import MSNIE from .mtv import ( MTVIE, @@ -554,7 +568,10 @@ from .nick import ( NickDeIE, ) from .niconico import NiconicoIE, NiconicoPlaylistIE -from .ninecninemedia import NineCNineMediaIE +from .ninecninemedia import ( + NineCNineMediaStackIE, + NineCNineMediaIE, +) from .ninegag import NineGagIE from .ninenow import NineNowIE from .nintendo import NintendoIE @@ -803,7 +820,6 @@ from .srgssr import ( SRGSSRPlayIE, ) from .srmediathek import SRMediathekIE -from .ssa import SSAIE from .stanfordoc import StanfordOpenClassroomIE from .steam import SteamIE from .streamable import StreamableIE @@ -866,10 +882,6 @@ from .tnaflix import ( MovieFapIE, ) from .toggle import ToggleIE -from .thvideo import ( - THVideoIE, - THVideoPlaylistIE -) from .toutv import TouTvIE from .toypics import ToypicsUserIE, ToypicsIE from .traileraddict import TrailerAddictIE @@ -904,6 +916,7 @@ from .tvc import ( ) from .tvigle import TvigleIE from .tvland import TVLandIE +from .tvnoe import TVNoeIE from .tvp import ( TVPEmbedIE, TVPIE, diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py index 0fb781a73..3a220e995 100644 --- a/youtube_dl/extractor/facebook.py +++ b/youtube_dl/extractor/facebook.py @@ -15,6 +15,7 @@ from ..compat import ( from ..utils import ( error_to_compat_str, ExtractorError, + int_or_none, limit_length, sanitized_Request, urlencode_postdata, @@ -62,6 +63,8 @@ class FacebookIE(InfoExtractor): 'ext': 'mp4', 'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam', 'uploader': 'Tennis on Facebook', + 'upload_date': '20140908', + 'timestamp': 1410199200, } }, { 'note': 'Video without discernible title', @@ -71,6 +74,8 @@ class FacebookIE(InfoExtractor): 'ext': 'mp4', 'title': 'Facebook video #274175099429670', 'uploader': 'Asif Nawab Butt', + 'upload_date': '20140506', + 'timestamp': 1399398998, }, 'expected_warnings': [ 'title' @@ -78,12 +83,14 @@ class FacebookIE(InfoExtractor): }, { 'note': 'Video with DASH manifest', 'url': 'https://www.facebook.com/video.php?v=957955867617029', - 'md5': '54706e4db4f5ad58fbad82dde1f1213f', + 'md5': 'b2c28d528273b323abe5c6ab59f0f030', 'info_dict': { 'id': '957955867617029', 'ext': 'mp4', 'title': 'When you post epic content on instagram.com/433 8 million followers, this is ...', 'uploader': 'Demy de Zeeuw', + 'upload_date': '20160110', + 'timestamp': 1452431627, }, }, { 'url': 'https://www.facebook.com/maxlayn/posts/10153807558977570', @@ -306,12 +313,16 @@ class FacebookIE(InfoExtractor): if not video_title: video_title = 'Facebook video #%s' % video_id uploader = clean_html(get_element_by_id('fbPhotoPageAuthorName', webpage)) + timestamp = int_or_none(self._search_regex( + r'<abbr[^>]+data-utime=["\'](\d+)', webpage, + 'timestamp', default=None)) info_dict = { 'id': video_id, 'title': video_title, 'formats': formats, 'uploader': uploader, + 'timestamp': timestamp, } return webpage, info_dict @@ -340,3 +351,32 @@ class FacebookIE(InfoExtractor): self._VIDEO_PAGE_TEMPLATE % video_id, video_id, fatal_if_no_video=True) return info_dict + + +class FacebookPluginsVideoIE(InfoExtractor): + _VALID_URL = r'https?://(?:[\w-]+\.)?facebook\.com/plugins/video\.php\?.*?\bhref=(?P<id>https.+)' + + _TESTS = [{ + 'url': 'https://www.facebook.com/plugins/video.php?href=https%3A%2F%2Fwww.facebook.com%2Fgov.sg%2Fvideos%2F10154383743583686%2F&show_text=0&width=560', + 'md5': '5954e92cdfe51fe5782ae9bda7058a07', + 'info_dict': { + 'id': '10154383743583686', + 'ext': 'mp4', + 'title': 'What to do during the haze?', + 'uploader': 'Gov.sg', + 'upload_date': '20160826', + 'timestamp': 1472184808, + }, + 'add_ie': [FacebookIE.ie_key()], + }, { + 'url': 'https://www.facebook.com/plugins/video.php?href=https%3A%2F%2Fwww.facebook.com%2Fvideo.php%3Fv%3D10204634152394104', + 'only_matching': True, + }, { + 'url': 'https://www.facebook.com/plugins/video.php?href=https://www.facebook.com/gov.sg/videos/10154383743583686/&show_text=0&width=560', + 'only_matching': True, + }] + + def _real_extract(self, url): + return self.url_result( + compat_urllib_parse_unquote(self._match_id(url)), + FacebookIE.ie_key()) diff --git a/youtube_dl/extractor/fc2.py b/youtube_dl/extractor/fc2.py index c7d69ff1f..c032d4d02 100644 --- a/youtube_dl/extractor/fc2.py +++ b/youtube_dl/extractor/fc2.py @@ -1,10 +1,12 @@ -#! -*- coding: utf-8 -*- +# coding: utf-8 from __future__ import unicode_literals import hashlib +import re from .common import InfoExtractor from ..compat import ( + compat_parse_qs, compat_urllib_request, compat_urlparse, ) @@ -16,7 +18,7 @@ from ..utils import ( class FC2IE(InfoExtractor): - _VALID_URL = r'^https?://video\.fc2\.com/(?:[^/]+/)*content/(?P<id>[^/]+)' + _VALID_URL = r'^(?:https?://video\.fc2\.com/(?:[^/]+/)*content/|fc2:)(?P<id>[^/]+)' IE_NAME = 'fc2' _NETRC_MACHINE = 'fc2' _TESTS = [{ @@ -75,12 +77,17 @@ class FC2IE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) self._login() - webpage = self._download_webpage(url, video_id) - self._downloader.cookiejar.clear_session_cookies() # must clear - self._login() - - title = self._og_search_title(webpage) - thumbnail = self._og_search_thumbnail(webpage) + webpage = None + if not url.startswith('fc2:'): + webpage = self._download_webpage(url, video_id) + self._downloader.cookiejar.clear_session_cookies() # must clear + self._login() + + title = 'FC2 video %s' % video_id + thumbnail = None + if webpage is not None: + title = self._og_search_title(webpage) + thumbnail = self._og_search_thumbnail(webpage) refer = url.replace('/content/', '/a/content/') if '/a/content/' not in url else url mimi = hashlib.md5((video_id + '_gGddgPfeaf_gzyr').encode('utf-8')).hexdigest() @@ -113,3 +120,41 @@ class FC2IE(InfoExtractor): 'ext': 'flv', 'thumbnail': thumbnail, } + + +class FC2EmbedIE(InfoExtractor): + _VALID_URL = r'https?://video\.fc2\.com/flv2\.swf\?(?P<query>.+)' + IE_NAME = 'fc2:embed' + + _TEST = { + 'url': 'http://video.fc2.com/flv2.swf?t=201404182936758512407645&i=20130316kwishtfitaknmcgd76kjd864hso93htfjcnaogz629mcgfs6rbfk0hsycma7shkf85937cbchfygd74&i=201403223kCqB3Ez&d=2625&sj=11&lang=ja&rel=1&from=11&cmt=1&tk=TlRBM09EQTNNekU9&tl=プリズン･ブレイク%20S1-01%20マイケル%20【吹替】', + 'md5': 'b8aae5334cb691bdb1193a88a6ab5d5a', + 'info_dict': { + 'id': '201403223kCqB3Ez', + 'ext': 'flv', + 'title': 'プリズン･ブレイク S1-01 マイケル 【吹替】', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + query = compat_parse_qs(mobj.group('query')) + + video_id = query['i'][-1] + title = query.get('tl', ['FC2 video %s' % video_id])[0] + + sj = query.get('sj', [None])[0] + thumbnail = None + if sj: + # See thumbnailImagePath() in ServerConst.as of flv2.swf + thumbnail = 'http://video%s-thumbnail.fc2.com/up/pic/%s.jpg' % ( + sj, '/'.join((video_id[:6], video_id[6:8], video_id[-2], video_id[-1], video_id))) + + return { + '_type': 'url_transparent', + 'ie_key': FC2IE.ie_key(), + 'url': 'fc2:%s' % video_id, + 'title': title, + 'thumbnail': thumbnail, + } diff --git a/youtube_dl/extractor/foxnews.py b/youtube_dl/extractor/foxnews.py index b04da2415..5c7acd795 100644 --- a/youtube_dl/extractor/foxnews.py +++ b/youtube_dl/extractor/foxnews.py @@ -3,11 +3,12 @@ from __future__ import unicode_literals import re from .amp import AMPIE +from .common import InfoExtractor class FoxNewsIE(AMPIE): IE_DESC = 'Fox News and Fox Business Video' - _VALID_URL = r'https?://(?P<host>video\.fox(?:news|business)\.com)/v/(?:video-embed\.html\?video_id=)?(?P<id>\d+)' + _VALID_URL = r'https?://(?P<host>video\.(?:insider\.)?fox(?:news|business)\.com)/v/(?:video-embed\.html\?video_id=)?(?P<id>\d+)' _TESTS = [ { 'url': 'http://video.foxnews.com/v/3937480/frozen-in-time/#sp=show-clips', @@ -49,6 +50,11 @@ class FoxNewsIE(AMPIE): 'url': 'http://video.foxbusiness.com/v/4442309889001', 'only_matching': True, }, + { + # From http://insider.foxnews.com/2016/08/25/univ-wisconsin-student-group-pushing-silence-certain-words + 'url': 'http://video.insider.foxnews.com/v/video-embed.html?video_id=5099377331001&autoplay=true&share_url=http://insider.foxnews.com/2016/08/25/univ-wisconsin-student-group-pushing-silence-certain-words&share_title=Student%20Group:%20Saying%20%27Politically%20Correct,%27%20%27Trash%27%20and%20%27Lame%27%20Is%20Offensive&share=true', + 'only_matching': True, + }, ] def _real_extract(self, url): @@ -58,3 +64,43 @@ class FoxNewsIE(AMPIE): 'http://%s/v/feed/video/%s.js?template=fox' % (host, video_id)) info['id'] = video_id return info + + +class FoxNewsInsiderIE(InfoExtractor): + _VALID_URL = r'https?://insider\.foxnews\.com/([^/]+/)+(?P<id>[a-z-]+)' + IE_NAME = 'foxnews:insider' + + _TEST = { + 'url': 'http://insider.foxnews.com/2016/08/25/univ-wisconsin-student-group-pushing-silence-certain-words', + 'md5': 'a10c755e582d28120c62749b4feb4c0c', + 'info_dict': { + 'id': '5099377331001', + 'display_id': 'univ-wisconsin-student-group-pushing-silence-certain-words', + 'ext': 'mp4', + 'title': 'Student Group: Saying \'Politically Correct,\' \'Trash\' and \'Lame\' Is Offensive', + 'description': 'Is campus censorship getting out of control?', + 'timestamp': 1472168725, + 'upload_date': '20160825', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + 'add_ie': [FoxNewsIE.ie_key()], + } + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + embed_url = self._html_search_meta('embedUrl', webpage, 'embed URL') + + title = self._og_search_title(webpage) + description = self._og_search_description(webpage) + + return { + '_type': 'url_transparent', + 'ie_key': FoxNewsIE.ie_key(), + 'url': embed_url, + 'display_id': display_id, + 'title': title, + 'description': description, + } diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index c6e655c84..24b217715 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -2243,11 +2243,11 @@ class GenericIE(InfoExtractor): # Look for VODPlatform embeds mobj = re.search( - r'<iframe[^>]+src=[\'"]((?:https?:)?//(?:www\.)?vod-platform\.net/embed/[^/?#]+)', + r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vod-platform\.net/[eE]mbed/.+?)\1', webpage) if mobj is not None: return self.url_result( - self._proto_relative_url(unescapeHTML(mobj.group(1))), 'VODPlatform') + self._proto_relative_url(unescapeHTML(mobj.group('url'))), 'VODPlatform') # Look for Instagram embeds instagram_embed_url = InstagramIE._extract_embed_url(webpage) diff --git a/youtube_dl/extractor/glide.py b/youtube_dl/extractor/glide.py index 62ff84835..f0d951396 100644 --- a/youtube_dl/extractor/glide.py +++ b/youtube_dl/extractor/glide.py @@ -2,7 +2,6 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import unified_strdate class GlideIE(InfoExtractor): @@ -14,10 +13,8 @@ class GlideIE(InfoExtractor): 'info_dict': { 'id': 'UZF8zlmuQbe4mr+7dCiQ0w==', 'ext': 'mp4', - 'title': 'Damon Timm\'s Glide message', + 'title': "Damon's Glide message", 'thumbnail': 're:^https?://.*?\.cloudfront\.net/.*\.jpg$', - 'uploader': 'Damon Timm', - 'upload_date': '20140919', } } @@ -27,7 +24,8 @@ class GlideIE(InfoExtractor): webpage = self._download_webpage(url, video_id) title = self._html_search_regex( - r'<title>(.+?)</title>', webpage, 'title') + r'<title>(.+?)</title>', webpage, + 'title', default=None) or self._og_search_title(webpage) video_url = self._proto_relative_url(self._search_regex( r'<source[^>]+src=(["\'])(?P<url>.+?)\1', webpage, 'video URL', default=None, @@ -36,18 +34,10 @@ class GlideIE(InfoExtractor): r'<img[^>]+id=["\']video-thumbnail["\'][^>]+src=(["\'])(?P<url>.+?)\1', webpage, 'thumbnail url', default=None, group='url')) or self._og_search_thumbnail(webpage) - uploader = self._search_regex( - r'<div[^>]+class=["\']info-name["\'][^>]*>([^<]+)', - webpage, 'uploader', fatal=False) - upload_date = unified_strdate(self._search_regex( - r'<div[^>]+class="info-date"[^>]*>([^<]+)', - webpage, 'upload date', fatal=False)) return { 'id': video_id, 'title': title, 'url': video_url, 'thumbnail': thumbnail, - 'uploader': uploader, - 'upload_date': upload_date, } diff --git a/youtube_dl/extractor/go.py b/youtube_dl/extractor/go.py new file mode 100644 index 000000000..6a437c54d --- /dev/null +++ b/youtube_dl/extractor/go.py @@ -0,0 +1,101 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + int_or_none, + determine_ext, + parse_age_limit, +) + + +class GoIE(InfoExtractor): + _BRANDS = { + 'abc': '001', + 'freeform': '002', + 'watchdisneychannel': '004', + 'watchdisneyjunior': '008', + 'watchdisneyxd': '009', + } + _VALID_URL = r'https?://(?:(?P<sub_domain>%s)\.)?go\.com/.*?vdka(?P<id>\w+)' % '|'.join(_BRANDS.keys()) + _TESTS = [{ + 'url': 'http://abc.go.com/shows/castle/video/most-recent/vdka0_g86w5onx', + 'info_dict': { + 'id': '0_g86w5onx', + 'ext': 'mp4', + 'title': 'Sneak Peek: Language Arts', + 'description': 'md5:7dcdab3b2d17e5217c953256af964e9c', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, { + 'url': 'http://abc.go.com/shows/after-paradise/video/most-recent/vdka3335601', + 'only_matching': True, + }] + + def _real_extract(self, url): + sub_domain, video_id = re.match(self._VALID_URL, url).groups() + video_data = self._download_json( + 'http://api.contents.watchabc.go.com/vp2/ws/contents/3000/videos/%s/001/-1/-1/-1/%s/-1/-1.json' % (self._BRANDS[sub_domain], video_id), + video_id)['video'][0] + title = video_data['title'] + + formats = [] + for asset in video_data.get('assets', {}).get('asset', []): + asset_url = asset.get('value') + if not asset_url: + continue + format_id = asset.get('format') + ext = determine_ext(asset_url) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + asset_url, video_id, 'mp4', m3u8_id=format_id or 'hls', fatal=False)) + else: + formats.append({ + 'format_id': format_id, + 'url': asset_url, + 'ext': ext, + }) + self._sort_formats(formats) + + subtitles = {} + for cc in video_data.get('closedcaption', {}).get('src', []): + cc_url = cc.get('value') + if not cc_url: + continue + ext = determine_ext(cc_url) + if ext == 'xml': + ext = 'ttml' + subtitles.setdefault(cc.get('lang'), []).append({ + 'url': cc_url, + 'ext': ext, + }) + + thumbnails = [] + for thumbnail in video_data.get('thumbnails', {}).get('thumbnail', []): + thumbnail_url = thumbnail.get('value') + if not thumbnail_url: + continue + thumbnails.append({ + 'url': thumbnail_url, + 'width': int_or_none(thumbnail.get('width')), + 'height': int_or_none(thumbnail.get('height')), + }) + + return { + 'id': video_id, + 'title': title, + 'description': video_data.get('longdescription') or video_data.get('description'), + 'duration': int_or_none(video_data.get('duration', {}).get('value'), 1000), + 'age_limit': parse_age_limit(video_data.get('tvrating', {}).get('rating')), + 'episode_number': int_or_none(video_data.get('episodenumber')), + 'series': video_data.get('show', {}).get('title'), + 'season_number': int_or_none(video_data.get('season', {}).get('num')), + 'thumbnails': thumbnails, + 'formats': formats, + 'subtitles': subtitles, + } diff --git a/youtube_dl/extractor/internetvideoarchive.py b/youtube_dl/extractor/internetvideoarchive.py index 45add007f..76cc5ec3e 100644 --- a/youtube_dl/extractor/internetvideoarchive.py +++ b/youtube_dl/extractor/internetvideoarchive.py @@ -48,13 +48,23 @@ class InternetVideoArchiveIE(InfoExtractor): # There are multiple videos in the playlist whlie only the first one # matches the video played in browsers video_info = configuration['playlist'][0] + title = video_info['title'] formats = [] for source in video_info['sources']: file_url = source['file'] if determine_ext(file_url) == 'm3u8': - formats.extend(self._extract_m3u8_formats( - file_url, video_id, ext='mp4', m3u8_id='hls')) + m3u8_formats = self._extract_m3u8_formats( + file_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + file_url = m3u8_formats[0]['url'] + formats.extend(self._extract_f4m_formats( + file_url.replace('.m3u8', '.f4m'), + video_id, f4m_id='hds', fatal=False)) + formats.extend(self._extract_mpd_formats( + file_url.replace('.m3u8', '.mpd'), + video_id, mpd_id='dash', fatal=False)) else: a_format = { 'url': file_url, @@ -70,7 +80,6 @@ class InternetVideoArchiveIE(InfoExtractor): self._sort_formats(formats) - title = video_info['title'] description = video_info.get('description') thumbnail = video_info.get('image') else: diff --git a/youtube_dl/extractor/kusi.py b/youtube_dl/extractor/kusi.py index 12cc56e44..2e66e8cf9 100644 --- a/youtube_dl/extractor/kusi.py +++ b/youtube_dl/extractor/kusi.py @@ -18,31 +18,20 @@ from ..utils import ( class KUSIIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?kusi\.com/(?P<path>story/.+|video\?clipId=(?P<clipId>\d+))' _TESTS = [{ - 'url': 'http://www.kusi.com/story/31183873/turko-files-case-closed-put-on-hold', - 'md5': 'f926e7684294cf8cb7bdf8858e1b3988', + 'url': 'http://www.kusi.com/story/32849881/turko-files-refused-to-help-it-aint-right', + 'md5': '4e76ce8e53660ce9697d06c0ba6fc47d', 'info_dict': { - 'id': '12203019', + 'id': '12689020', 'ext': 'mp4', - 'title': 'Turko Files: Case Closed! & Put On Hold!', - 'duration': 231.0, - 'upload_date': '20160210', - 'timestamp': 1455087571, + 'title': "Turko Files: Refused to Help, It Ain't Right!", + 'duration': 223.586, + 'upload_date': '20160826', + 'timestamp': 1472233118, 'thumbnail': 're:^https?://.*\.jpg$' }, }, { 'url': 'http://kusi.com/video?clipId=12203019', - 'info_dict': { - 'id': '12203019', - 'ext': 'mp4', - 'title': 'Turko Files: Case Closed! & Put On Hold!', - 'duration': 231.0, - 'upload_date': '20160210', - 'timestamp': 1455087571, - 'thumbnail': 're:^https?://.*\.jpg$' - }, - 'params': { - 'skip_download': True, # Same as previous one - }, + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/limelight.py b/youtube_dl/extractor/limelight.py index a425bafe3..6752ffee2 100644 --- a/youtube_dl/extractor/limelight.py +++ b/youtube_dl/extractor/limelight.py @@ -34,11 +34,12 @@ class LimelightBaseIE(InfoExtractor): def _extract_info(self, streams, mobile_urls, properties): video_id = properties['media_id'] formats = [] - + urls = [] for stream in streams: stream_url = stream.get('url') - if not stream_url or stream.get('drmProtected'): + if not stream_url or stream.get('drmProtected') or stream_url in urls: continue + urls.append(stream_url) ext = determine_ext(stream_url) if ext == 'f4m': formats.extend(self._extract_f4m_formats( @@ -58,9 +59,11 @@ class LimelightBaseIE(InfoExtractor): format_id = 'rtmp' if stream.get('videoBitRate'): format_id += '-%d' % int_or_none(stream['videoBitRate']) + http_url = 'http://%s/%s' % (rtmp.group('host').replace('csl.', 'cpl.'), rtmp.group('playpath')[4:]) + urls.append(http_url) http_fmt = fmt.copy() http_fmt.update({ - 'url': 'http://%s/%s' % (rtmp.group('host').replace('csl.', 'cpl.'), rtmp.group('playpath')[4:]), + 'url': http_url, 'format_id': format_id.replace('rtmp', 'http'), }) formats.append(http_fmt) @@ -76,8 +79,9 @@ class LimelightBaseIE(InfoExtractor): for mobile_url in mobile_urls: media_url = mobile_url.get('mobileUrl') format_id = mobile_url.get('targetMediaPlatform') - if not media_url or format_id == 'Widevine': + if not media_url or format_id in ('Widevine', 'SmoothStreaming') or media_url in urls: continue + urls.append(media_url) ext = determine_ext(media_url) if ext == 'm3u8': formats.extend(self._extract_m3u8_formats( diff --git a/youtube_dl/extractor/ssa.py b/youtube_dl/extractor/movingimage.py index 54d1843f2..bb789c32e 100644 --- a/youtube_dl/extractor/ssa.py +++ b/youtube_dl/extractor/movingimage.py @@ -7,22 +7,19 @@ from ..utils import ( ) -class SSAIE(InfoExtractor): - _VALID_URL = r'https?://ssa\.nls\.uk/film/(?P<id>\d+)' +class MovingImageIE(InfoExtractor): + _VALID_URL = r'https?://movingimage\.nls\.uk/film/(?P<id>\d+)' _TEST = { - 'url': 'http://ssa.nls.uk/film/3561', + 'url': 'http://movingimage.nls.uk/film/3561', + 'md5': '4caa05c2b38453e6f862197571a7be2f', 'info_dict': { 'id': '3561', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'SHETLAND WOOL', 'description': 'md5:c5afca6871ad59b4271e7704fe50ab04', 'duration': 900, 'thumbnail': 're:^https?://.*\.jpg$', }, - 'params': { - # rtmp download - 'skip_download': True, - }, } def _real_extract(self, url): @@ -30,10 +27,9 @@ class SSAIE(InfoExtractor): webpage = self._download_webpage(url, video_id) - streamer = self._search_regex( - r"'streamer'\s*,\S*'(rtmp[^']+)'", webpage, 'streamer') - play_path = self._search_regex( - r"'file'\s*,\s*'([^']+)'", webpage, 'file').rpartition('.')[0] + formats = self._extract_m3u8_formats( + self._html_search_regex(r'file\s*:\s*"([^"]+)"', webpage, 'm3u8 manifest URL'), + video_id, ext='mp4', entry_protocol='m3u8_native') def search_field(field_name, fatal=False): return self._search_regex( @@ -44,13 +40,11 @@ class SSAIE(InfoExtractor): description = unescapeHTML(search_field('Description')) duration = parse_duration(search_field('Running time')) thumbnail = self._search_regex( - r"'image'\s*,\s*'([^']+)'", webpage, 'thumbnails', fatal=False) + r"image\s*:\s*'([^']+)'", webpage, 'thumbnail', fatal=False) return { 'id': video_id, - 'url': streamer, - 'play_path': play_path, - 'ext': 'flv', + 'formats': formats, 'title': title, 'description': description, 'duration': duration, diff --git a/youtube_dl/extractor/myvidster.py b/youtube_dl/extractor/myvidster.py index 731c24542..2117d302d 100644 --- a/youtube_dl/extractor/myvidster.py +++ b/youtube_dl/extractor/myvidster.py @@ -13,7 +13,7 @@ class MyVidsterIE(InfoExtractor): 'id': '3685814', 'title': 'md5:7d8427d6d02c4fbcef50fe269980c749', 'upload_date': '20141027', - 'uploader_id': 'utkualp', + 'uploader': 'utkualp', 'ext': 'mp4', 'age_limit': 18, }, diff --git a/youtube_dl/extractor/nba.py b/youtube_dl/extractor/nba.py index aabd5b670..53561961c 100644 --- a/youtube_dl/extractor/nba.py +++ b/youtube_dl/extractor/nba.py @@ -39,18 +39,19 @@ class NBAIE(TurnerBaseIE): 'url': 'http://watch.nba.com/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba', 'md5': 'b2b39b81cf28615ae0c3360a3f9668c4', 'info_dict': { - 'id': '0041400301-cle-atl-recap', + 'id': 'channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba', 'ext': 'mp4', 'title': 'Hawks vs. Cavaliers Game 1', 'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d', 'duration': 228, 'timestamp': 1432134543, 'upload_date': '20150520', - } + }, + 'expected_warnings': ['Unable to download f4m manifest'], }, { 'url': 'http://www.nba.com/clippers/news/doc-rivers-were-not-trading-blake', 'info_dict': { - 'id': '1455672027478-Doc_Feb16_720', + 'id': 'teams/clippers/2016/02/17/1455672027478-Doc_Feb16_720.mov-297324', 'ext': 'mp4', 'title': 'Practice: Doc Rivers - 2/16/16', 'description': 'Head Coach Doc Rivers addresses the media following practice.', @@ -61,6 +62,7 @@ class NBAIE(TurnerBaseIE): # m3u8 download 'skip_download': True, }, + 'expected_warnings': ['Unable to download f4m manifest'], }, { 'url': 'http://www.nba.com/timberwolves/wiggins-shootaround#', 'info_dict': { @@ -75,7 +77,7 @@ class NBAIE(TurnerBaseIE): }, { 'url': 'http://www.nba.com/timberwolves/wiggins-shootaround#', 'info_dict': { - 'id': 'Wigginsmp4-3462601', + 'id': 'teams/timberwolves/2014/12/12/Wigginsmp4-3462601', 'ext': 'mp4', 'title': 'Shootaround Access - Dec. 12 | Andrew Wiggins', 'description': 'Wolves rookie Andrew Wiggins addresses the media after Friday\'s shootaround.', @@ -87,6 +89,7 @@ class NBAIE(TurnerBaseIE): # m3u8 download 'skip_download': True, }, + 'expected_warnings': ['Unable to download f4m manifest'], }] _PAGE_SIZE = 30 diff --git a/youtube_dl/extractor/ninecninemedia.py b/youtube_dl/extractor/ninecninemedia.py index d889245ad..ec4d675e2 100644 --- a/youtube_dl/extractor/ninecninemedia.py +++ b/youtube_dl/extractor/ninecninemedia.py @@ -4,40 +4,36 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_str from ..utils import ( parse_iso8601, - parse_duration, - ExtractorError + float_or_none, + ExtractorError, + int_or_none, ) -class NineCNineMediaIE(InfoExtractor): - _VALID_URL = r'9c9media:(?P<destination_code>[^:]+):(?P<id>\d+)' +class NineCNineMediaBaseIE(InfoExtractor): + _API_BASE_TEMPLATE = 'http://capi.9c9media.com/destinations/%s/platforms/desktop/contents/%s/' + + +class NineCNineMediaStackIE(NineCNineMediaBaseIE): + IE_NAME = '9c9media:stack' + _VALID_URL = r'9c9media:stack:(?P<destination_code>[^:]+):(?P<content_id>\d+):(?P<content_package>\d+):(?P<id>\d+)' def _real_extract(self, url): - destination_code, video_id = re.match(self._VALID_URL, url).groups() - api_base_url = 'http://capi.9c9media.com/destinations/%s/platforms/desktop/contents/%s/' % (destination_code, video_id) - content = self._download_json(api_base_url, video_id, query={ - '$include': '[contentpackages]', - }) - title = content['Name'] - if len(content['ContentPackages']) > 1: - raise ExtractorError('multiple content packages') - content_package = content['ContentPackages'][0] - stacks_base_url = api_base_url + 'contentpackages/%s/stacks/' % content_package['Id'] - stacks = self._download_json(stacks_base_url, video_id)['Items'] - if len(stacks) > 1: - raise ExtractorError('multiple stacks') - stack = stacks[0] - stack_base_url = '%s%s/manifest.' % (stacks_base_url, stack['Id']) + destination_code, content_id, package_id, stack_id = re.match(self._VALID_URL, url).groups() + stack_base_url_template = self._API_BASE_TEMPLATE + 'contentpackages/%s/stacks/%s/manifest.' + stack_base_url = stack_base_url_template % (destination_code, content_id, package_id, stack_id) + formats = [] formats.extend(self._extract_m3u8_formats( - stack_base_url + 'm3u8', video_id, 'mp4', + stack_base_url + 'm3u8', stack_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) formats.extend(self._extract_f4m_formats( - stack_base_url + 'f4m', video_id, + stack_base_url + 'f4m', stack_id, f4m_id='hds', fatal=False)) - mp4_url = self._download_webpage(stack_base_url + 'pd', video_id, fatal=False) + mp4_url = self._download_webpage(stack_base_url + 'pd', stack_id, fatal=False) if mp4_url: formats.append({ 'url': mp4_url, @@ -46,10 +42,86 @@ class NineCNineMediaIE(InfoExtractor): self._sort_formats(formats) return { - 'id': video_id, - 'title': title, - 'description': content.get('Desc') or content.get('ShortDesc'), - 'timestamp': parse_iso8601(content.get('BroadcastDateTime')), - 'duration': parse_duration(content.get('BroadcastTime')), + 'id': stack_id, 'formats': formats, } + + +class NineCNineMediaIE(NineCNineMediaBaseIE): + IE_NAME = '9c9media' + _VALID_URL = r'9c9media:(?P<destination_code>[^:]+):(?P<id>\d+)' + + def _real_extract(self, url): + destination_code, content_id = re.match(self._VALID_URL, url).groups() + api_base_url = self._API_BASE_TEMPLATE % (destination_code, content_id) + content = self._download_json(api_base_url, content_id, query={ + '$include': '[Media,Season,ContentPackages]', + }) + title = content['Name'] + if len(content['ContentPackages']) > 1: + raise ExtractorError('multiple content packages') + content_package = content['ContentPackages'][0] + package_id = content_package['Id'] + content_package_url = api_base_url + 'contentpackages/%s/' % package_id + content_package = self._download_json(content_package_url, content_id) + + if content_package.get('Constraints', {}).get('Security', {}).get('Type') == 'adobe-drm': + raise ExtractorError('This video is DRM protected.', expected=True) + + stacks = self._download_json(content_package_url + 'stacks/', package_id)['Items'] + multistacks = len(stacks) > 1 + + thumbnails = [] + for image in content.get('Images', []): + image_url = image.get('Url') + if not image_url: + continue + thumbnails.append({ + 'url': image_url, + 'width': int_or_none(image.get('Width')), + 'height': int_or_none(image.get('Height')), + }) + + tags, categories = [], [] + for source_name, container in (('Tags', tags), ('Genres', categories)): + for e in content.get(source_name, []): + e_name = e.get('Name') + if not e_name: + continue + container.append(e_name) + + description = content.get('Desc') or content.get('ShortDesc') + season = content.get('Season', {}) + base_info = { + 'description': description, + 'timestamp': parse_iso8601(content.get('BroadcastDateTime')), + 'episode_number': int_or_none(content.get('Episode')), + 'season': season.get('Name'), + 'season_number': season.get('Number'), + 'season_id': season.get('Id'), + 'series': content.get('Media', {}).get('Name'), + 'tags': tags, + 'categories': categories, + } + + entries = [] + for stack in stacks: + stack_id = compat_str(stack['Id']) + entry = { + '_type': 'url_transparent', + 'url': '9c9media:stack:%s:%s:%s:%s' % (destination_code, content_id, package_id, stack_id), + 'id': stack_id, + 'title': '%s_part%s' % (title, stack['Name']) if multistacks else title, + 'duration': float_or_none(stack.get('Duration')), + 'ie_key': 'NineCNineMediaStack', + } + entry.update(base_info) + entries.append(entry) + + return { + '_type': 'multi_video', + 'id': content_id, + 'title': title, + 'description': description, + 'entries': entries, + } diff --git a/youtube_dl/extractor/nytimes.py b/youtube_dl/extractor/nytimes.py index 681683e86..142c34256 100644 --- a/youtube_dl/extractor/nytimes.py +++ b/youtube_dl/extractor/nytimes.py @@ -1,26 +1,37 @@ from __future__ import unicode_literals +import hmac +import hashlib +import base64 + from .common import InfoExtractor from ..utils import ( float_or_none, int_or_none, parse_iso8601, + mimetype2ext, + determine_ext, ) class NYTimesBaseIE(InfoExtractor): + _SECRET = b'pX(2MbU2);4N{7J8)>YwKRJ+/pQ3JkiU2Q^V>mFYv6g6gYvt6v' + def _extract_video_from_id(self, video_id): - video_data = self._download_json( - 'http://www.nytimes.com/svc/video/api/v2/video/%s' % video_id, - video_id, 'Downloading video JSON') + # Authorization generation algorithm is reverse engineered from `signer` in + # http://graphics8.nytimes.com/video/vhs/vhs-2.x.min.js + path = '/svc/video/api/v3/video/' + video_id + hm = hmac.new(self._SECRET, (path + ':vhs').encode(), hashlib.sha512).hexdigest() + video_data = self._download_json('http://www.nytimes.com' + path, video_id, 'Downloading video JSON', headers={ + 'Authorization': 'NYTV ' + base64.b64encode(hm.encode()).decode(), + 'X-NYTV': 'vhs', + }, fatal=False) + if not video_data: + video_data = self._download_json( + 'http://www.nytimes.com/svc/video/api/v2/video/' + video_id, + video_id, 'Downloading video JSON') title = video_data['headline'] - description = video_data.get('summary') - duration = float_or_none(video_data.get('duration'), 1000) - - uploader = video_data.get('byline') - publication_date = video_data.get('publication_date') - timestamp = parse_iso8601(publication_date[:-8]) if publication_date else None def get_file_size(file_size): if isinstance(file_size, int): @@ -28,35 +39,59 @@ class NYTimesBaseIE(InfoExtractor): elif isinstance(file_size, dict): return int(file_size.get('value', 0)) else: - return 0 - - formats = [ - { - 'url': video['url'], - 'format_id': video.get('type'), - 'vcodec': video.get('video_codec'), - 'width': int_or_none(video.get('width')), - 'height': int_or_none(video.get('height')), - 'filesize': get_file_size(video.get('fileSize')), - } for video in video_data['renditions'] if video.get('url') - ] + return None + + urls = [] + formats = [] + for video in video_data.get('renditions', []): + video_url = video.get('url') + format_id = video.get('type') + if not video_url or format_id == 'thumbs' or video_url in urls: + continue + urls.append(video_url) + ext = mimetype2ext(video.get('mimetype')) or determine_ext(video_url) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + video_url, video_id, 'mp4', 'm3u8_native', + m3u8_id=format_id or 'hls', fatal=False)) + elif ext == 'mpd': + continue + # formats.extend(self._extract_mpd_formats( + # video_url, video_id, format_id or 'dash', fatal=False)) + else: + formats.append({ + 'url': video_url, + 'format_id': format_id, + 'vcodec': video.get('videoencoding') or video.get('video_codec'), + 'width': int_or_none(video.get('width')), + 'height': int_or_none(video.get('height')), + 'filesize': get_file_size(video.get('file_size') or video.get('fileSize')), + 'tbr': int_or_none(video.get('bitrate'), 1000), + 'ext': ext, + }) self._sort_formats(formats) - thumbnails = [ - { - 'url': 'http://www.nytimes.com/%s' % image['url'], + thumbnails = [] + for image in video_data.get('images', []): + image_url = image.get('url') + if not image_url: + continue + thumbnails.append({ + 'url': 'http://www.nytimes.com/' + image_url, 'width': int_or_none(image.get('width')), 'height': int_or_none(image.get('height')), - } for image in video_data.get('images', []) if image.get('url') - ] + }) + + publication_date = video_data.get('publication_date') + timestamp = parse_iso8601(publication_date[:-8]) if publication_date else None return { 'id': video_id, 'title': title, - 'description': description, + 'description': video_data.get('summary'), 'timestamp': timestamp, - 'uploader': uploader, - 'duration': duration, + 'uploader': video_data.get('byline'), + 'duration': float_or_none(video_data.get('duration'), 1000), 'formats': formats, 'thumbnails': thumbnails, } @@ -67,7 +102,7 @@ class NYTimesIE(NYTimesBaseIE): _TESTS = [{ 'url': 'http://www.nytimes.com/video/opinion/100000002847155/verbatim-what-is-a-photocopier.html?playlistId=100000001150263', - 'md5': '18a525a510f942ada2720db5f31644c0', + 'md5': 'd665342765db043f7e225cff19df0f2d', 'info_dict': { 'id': '100000002847155', 'ext': 'mov', diff --git a/youtube_dl/extractor/porncom.py b/youtube_dl/extractor/porncom.py index 4baf79688..d85e0294d 100644 --- a/youtube_dl/extractor/porncom.py +++ b/youtube_dl/extractor/porncom.py @@ -26,6 +26,8 @@ class PornComIE(InfoExtractor): 'duration': 551, 'view_count': int, 'age_limit': 18, + 'categories': list, + 'tags': list, }, }, { 'url': 'http://se.porn.com/videos/marsha-may-rides-seth-on-top-of-his-thick-cock-2658067', @@ -75,7 +77,14 @@ class PornComIE(InfoExtractor): self._sort_formats(formats) view_count = str_to_int(self._search_regex( - r'class=["\']views["\'][^>]*><p>([\d,.]+)', webpage, 'view count')) + r'class=["\']views["\'][^>]*><p>([\d,.]+)', webpage, + 'view count', fatal=False)) + + def extract_list(kind): + s = self._search_regex( + r'(?s)<p[^>]*>%s:(.+?)</p>' % kind.capitalize(), + webpage, kind, fatal=False) + return re.findall(r'<a[^>]+>([^<]+)</a>', s or '') return { 'id': video_id, @@ -86,4 +95,6 @@ class PornComIE(InfoExtractor): 'view_count': view_count, 'formats': formats, 'age_limit': 18, + 'categories': extract_list('categories'), + 'tags': extract_list('tags'), } diff --git a/youtube_dl/extractor/pornovoisines.py b/youtube_dl/extractor/pornovoisines.py index 6b51e5c54..58f557e39 100644 --- a/youtube_dl/extractor/pornovoisines.py +++ b/youtube_dl/extractor/pornovoisines.py @@ -2,7 +2,6 @@ from __future__ import unicode_literals import re -import random from .common import InfoExtractor from ..utils import ( @@ -13,61 +12,69 @@ from ..utils import ( class PornoVoisinesIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?pornovoisines\.com/showvideo/(?P<id>\d+)/(?P<display_id>[^/]+)' - - _VIDEO_URL_TEMPLATE = 'http://stream%d.pornovoisines.com' \ - '/static/media/video/transcoded/%s-640x360-1000-trscded.mp4' - - _SERVER_NUMBERS = (1, 2) + _VALID_URL = r'https?://(?:www\.)?pornovoisines\.com/videos/show/(?P<id>\d+)/(?P<display_id>[^/.]+)' _TEST = { - 'url': 'http://www.pornovoisines.com/showvideo/1285/recherche-appartement/', - 'md5': '5ac670803bc12e9e7f9f662ce64cf1d1', + 'url': 'http://www.pornovoisines.com/videos/show/919/recherche-appartement.html', + 'md5': '6f8aca6a058592ab49fe701c8ba8317b', 'info_dict': { - 'id': '1285', + 'id': '919', 'display_id': 'recherche-appartement', 'ext': 'mp4', 'title': 'Recherche appartement', - 'description': 'md5:819ea0b785e2a04667a1a01cdc89594e', + 'description': 'md5:fe10cb92ae2dd3ed94bb4080d11ff493', 'thumbnail': 're:^https?://.*\.jpg$', 'upload_date': '20140925', 'duration': 120, 'view_count': int, 'average_rating': float, - 'categories': ['Débutantes', 'Scénario', 'Sodomie'], + 'categories': ['Débutante', 'Débutantes', 'Scénario', 'Sodomie'], 'age_limit': 18, + 'subtitles': { + 'fr': [{ + 'ext': 'vtt', + }] + }, } } - @classmethod - def build_video_url(cls, num): - return cls._VIDEO_URL_TEMPLATE % (random.choice(cls._SERVER_NUMBERS), num) - def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') display_id = mobj.group('display_id') - webpage = self._download_webpage(url, video_id) + settings_url = self._download_json( + 'http://www.pornovoisines.com/api/video/%s/getsettingsurl/' % video_id, + video_id, note='Getting settings URL')['video_settings_url'] + settings = self._download_json(settings_url, video_id)['data'] + + formats = [] + for kind, data in settings['variants'].items(): + if kind == 'HLS': + formats.extend(self._extract_m3u8_formats( + data, video_id, ext='mp4', entry_protocol='m3u8_native', m3u8_id='hls')) + elif kind == 'MP4': + for item in data: + formats.append({ + 'url': item['url'], + 'height': item.get('height'), + 'bitrate': item.get('bitrate'), + }) + self._sort_formats(formats) - video_url = self.build_video_url(video_id) + webpage = self._download_webpage(url, video_id) - title = self._html_search_regex( - r'<h1>(.+?)</h1>', webpage, 'title', flags=re.DOTALL) - description = self._html_search_regex( - r'<article id="descriptif">(.+?)</article>', - webpage, 'description', fatal=False, flags=re.DOTALL) + title = self._og_search_title(webpage) + description = self._og_search_description(webpage) - thumbnail = self._search_regex( - r'<div id="mediaspace%s">\s*<img src="/?([^"]+)"' % video_id, - webpage, 'thumbnail', fatal=False) - if thumbnail: - thumbnail = 'http://www.pornovoisines.com/%s' % thumbnail + # The webpage has a bug - there's no space between "thumb" and src= + thumbnail = self._html_search_regex( + r'<img[^>]+class=([\'"])thumb\1[^>]*src=([\'"])(?P<url>[^"]+)\2', + webpage, 'thumbnail', fatal=False, group='url') upload_date = unified_strdate(self._search_regex( - r'Publié le ([\d-]+)', webpage, 'upload date', fatal=False)) - duration = int_or_none(self._search_regex( - 'Durée (\d+)', webpage, 'duration', fatal=False)) + r'Le\s*<b>([\d/]+)', webpage, 'upload date', fatal=False)) + duration = settings.get('main', {}).get('duration') view_count = int_or_none(self._search_regex( r'(\d+) vues', webpage, 'view count', fatal=False)) average_rating = self._search_regex( @@ -75,15 +82,19 @@ class PornoVoisinesIE(InfoExtractor): if average_rating: average_rating = float_or_none(average_rating.replace(',', '.')) - categories = self._html_search_meta( - 'keywords', webpage, 'categories', fatal=False) + categories = self._html_search_regex( + r'(?s)Catégories\s*:\s*<b>(.+?)</b>', webpage, 'categories', fatal=False) if categories: categories = [category.strip() for category in categories.split(',')] + subtitles = {'fr': [{ + 'url': subtitle, + } for subtitle in settings.get('main', {}).get('vtt_tracks', {}).values()]} + return { 'id': video_id, 'display_id': display_id, - 'url': video_url, + 'formats': formats, 'title': title, 'description': description, 'thumbnail': thumbnail, @@ -93,4 +104,5 @@ class PornoVoisinesIE(InfoExtractor): 'average_rating': average_rating, 'categories': categories, 'age_limit': 18, + 'subtitles': subtitles, } diff --git a/youtube_dl/extractor/pyvideo.py b/youtube_dl/extractor/pyvideo.py index cc0416cb8..b8ac93a62 100644 --- a/youtube_dl/extractor/pyvideo.py +++ b/youtube_dl/extractor/pyvideo.py @@ -1,59 +1,72 @@ from __future__ import unicode_literals import re -import os from .common import InfoExtractor +from ..compat import compat_str +from ..utils import int_or_none class PyvideoIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?pyvideo\.org/video/(?P<id>\d+)/(.*)' - - _TESTS = [ - { - 'url': 'http://pyvideo.org/video/1737/become-a-logging-expert-in-30-minutes', - 'md5': '520915673e53a5c5d487c36e0c4d85b5', - 'info_dict': { - 'id': '24_4WWkSmNo', - 'ext': 'webm', - 'title': 'Become a logging expert in 30 minutes', - 'description': 'md5:9665350d466c67fb5b1598de379021f7', - 'upload_date': '20130320', - 'uploader': 'Next Day Video', - 'uploader_id': 'NextDayVideo', - }, - 'add_ie': ['Youtube'], + _VALID_URL = r'https?://(?:www\.)?pyvideo\.org/(?P<category>[^/]+)/(?P<id>[^/?#&.]+)' + + _TESTS = [{ + 'url': 'http://pyvideo.org/pycon-us-2013/become-a-logging-expert-in-30-minutes.html', + 'info_dict': { + 'id': 'become-a-logging-expert-in-30-minutes', }, - { - 'url': 'http://pyvideo.org/video/2542/gloriajw-spotifywitherikbernhardsson182m4v', - 'md5': '5fe1c7e0a8aa5570330784c847ff6d12', - 'info_dict': { - 'id': '2542', - 'ext': 'm4v', - 'title': 'Gloriajw-SpotifyWithErikBernhardsson182', - }, + 'playlist_count': 2, + }, { + 'url': 'http://pyvideo.org/pygotham-2012/gloriajw-spotifywitherikbernhardsson182m4v.html', + 'md5': '5fe1c7e0a8aa5570330784c847ff6d12', + 'info_dict': { + 'id': '2542', + 'ext': 'm4v', + 'title': 'Gloriajw-SpotifyWithErikBernhardsson182.m4v', }, - ] + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) + category = mobj.group('category') video_id = mobj.group('id') - webpage = self._download_webpage(url, video_id) + entries = [] - m_youtube = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', webpage) - if m_youtube is not None: - return self.url_result(m_youtube.group(1), 'Youtube') + data = self._download_json( + 'https://raw.githubusercontent.com/pyvideo/data/master/%s/videos/%s.json' + % (category, video_id), video_id, fatal=False) - title = self._html_search_regex( - r'<div class="section">\s*<h3(?:\s+class="[^"]*"[^>]*)?>([^>]+?)</h3>', - webpage, 'title', flags=re.DOTALL) - video_url = self._search_regex( - [r'<source src="(.*?)"', r'<dt>Download</dt>.*?<a href="(.+?)"'], - webpage, 'video url', flags=re.DOTALL) + if data: + for video in data['videos']: + video_url = video.get('url') + if video_url: + if video.get('type') == 'youtube': + entries.append(self.url_result(video_url, 'Youtube')) + else: + entries.append({ + 'id': compat_str(data.get('id') or video_id), + 'url': video_url, + 'title': data['title'], + 'description': data.get('description') or data.get('summary'), + 'thumbnail': data.get('thumbnail_url'), + 'duration': int_or_none(data.get('duration')), + }) + else: + webpage = self._download_webpage(url, video_id) + title = self._og_search_title(webpage) + media_urls = self._search_regex( + r'(?s)Media URL:(.+?)</li>', webpage, 'media urls') + for m in re.finditer( + r'<a[^>]+href=(["\'])(?P<url>http.+?)\1', media_urls): + media_url = m.group('url') + if re.match(r'https?://www\.youtube\.com/watch\?v=.*', media_url): + entries.append(self.url_result(media_url, 'Youtube')) + else: + entries.append({ + 'id': video_id, + 'url': media_url, + 'title': title, + }) - return { - 'id': video_id, - 'title': os.path.splitext(title)[0], - 'url': video_url, - } + return self.playlist_result(entries, video_id) diff --git a/youtube_dl/extractor/rottentomatoes.py b/youtube_dl/extractor/rottentomatoes.py index f9cd48790..23abf7a27 100644 --- a/youtube_dl/extractor/rottentomatoes.py +++ b/youtube_dl/extractor/rottentomatoes.py @@ -1,7 +1,6 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import compat_urlparse from .internetvideoarchive import InternetVideoArchiveIE @@ -11,21 +10,23 @@ class RottenTomatoesIE(InfoExtractor): _TEST = { 'url': 'http://www.rottentomatoes.com/m/toy_story_3/trailers/11028566/', 'info_dict': { - 'id': '613340', + 'id': '11028566', 'ext': 'mp4', 'title': 'Toy Story 3', + 'description': 'From the creators of the beloved TOY STORY films, comes a story that will reunite the gang in a whole new way.', + 'thumbnail': 're:^https?://.*\.jpg$', }, } def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - og_video = self._og_search_video_url(webpage) - query = compat_urlparse.urlparse(og_video).query + iva_id = self._search_regex(r'publishedid=(\d+)', webpage, 'internet video archive id') return { '_type': 'url_transparent', - 'url': InternetVideoArchiveIE._build_xml_url(query), + 'url': 'http://video.internetvideoarchive.net/player/6/configuration.ashx?domain=www.videodetective.com&customerid=69249&playerid=641&publishedid=' + iva_id, 'ie_key': InternetVideoArchiveIE.ie_key(), + 'id': video_id, 'title': self._og_search_title(webpage), } diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index aeae931a2..9635c2b49 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -32,7 +32,7 @@ class SoundcloudIE(InfoExtractor): _VALID_URL = r'''(?x)^(?:https?://)? (?:(?:(?:www\.|m\.)?soundcloud\.com/ (?P<uploader>[\w\d-]+)/ - (?!(?:tracks|sets(?:/[^/?#]+)?|reposts|likes|spotlight)/?(?:$|[?#])) + (?!(?:tracks|sets(?:/.+?)?|reposts|likes|spotlight)/?(?:$|[?#])) (?P<title>[\w\d-]+)/? (?P<token>[^?]+?)?(?:[?].*)?$) |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+) @@ -265,6 +265,9 @@ class SoundcloudSetIE(SoundcloudIE): 'title': 'The Royal Concept EP', }, 'playlist_mincount': 6, + }, { + 'url': 'https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep/token', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/southpark.py b/youtube_dl/extractor/southpark.py index a147f7db1..e2a9e45ac 100644 --- a/youtube_dl/extractor/southpark.py +++ b/youtube_dl/extractor/southpark.py @@ -35,6 +35,7 @@ class SouthParkEsIE(SouthParkIE): 'description': 'Cartman Consigue Una Sonda Anal', }, 'playlist_count': 4, + 'skip': 'Geo-restricted', }] diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index 23067e8c6..6febf805b 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -96,7 +96,7 @@ class ThePlatformBaseIE(OnceIE): class ThePlatformIE(ThePlatformBaseIE, AdobePassIE): _VALID_URL = r'''(?x) (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/ - (?:(?:(?:[^/]+/)+select/)?(?P<media>media/(?:guid/\d+/)?)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))? + (?:(?:(?:[^/]+/)+select/)?(?P<media>media/(?:guid/\d+/)?)?|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))? |theplatform:)(?P<id>[^/\?&]+)''' _TESTS = [{ @@ -116,6 +116,7 @@ class ThePlatformIE(ThePlatformBaseIE, AdobePassIE): # rtmp download 'skip_download': True, }, + 'skip': '404 Not Found', }, { # from http://www.cnet.com/videos/tesla-model-s-a-second-step-towards-a-cleaner-motoring-future/ 'url': 'http://link.theplatform.com/s/kYEXFC/22d_qsQ6MIRT', diff --git a/youtube_dl/extractor/thestar.py b/youtube_dl/extractor/thestar.py index ba1380abc..c3f118894 100644 --- a/youtube_dl/extractor/thestar.py +++ b/youtube_dl/extractor/thestar.py @@ -2,8 +2,6 @@ from __future__ import unicode_literals from .common import InfoExtractor -from .brightcove import BrightcoveLegacyIE -from ..compat import compat_parse_qs class TheStarIE(InfoExtractor): @@ -30,6 +28,9 @@ class TheStarIE(InfoExtractor): def _real_extract(self, url): display_id = self._match_id(url) webpage = self._download_webpage(url, display_id) - brightcove_legacy_url = BrightcoveLegacyIE._extract_brightcove_url(webpage) - brightcove_id = compat_parse_qs(brightcove_legacy_url)['@videoPlayer'][0] - return self.url_result(self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id, 'BrightcoveNew', brightcove_id) + brightcove_id = self._search_regex( + r'mainartBrightcoveVideoId["\']?\s*:\s*["\']?(\d+)', + webpage, 'brightcove id') + return self.url_result( + self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id, + 'BrightcoveNew', brightcove_id) diff --git a/youtube_dl/extractor/thvideo.py b/youtube_dl/extractor/thvideo.py deleted file mode 100644 index 406f4a826..000000000 --- a/youtube_dl/extractor/thvideo.py +++ /dev/null @@ -1,84 +0,0 @@ -# coding: utf-8 -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor -from ..utils import ( - unified_strdate -) - - -class THVideoIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?thvideo\.tv/(?:v/th|mobile\.php\?cid=)(?P<id>[0-9]+)' - _TEST = { - 'url': 'http://thvideo.tv/v/th1987/', - 'md5': 'fa107b1f73817e325e9433505a70db50', - 'info_dict': { - 'id': '1987', - 'ext': 'mp4', - 'title': '【动画】秘封活动记录 ～ The Sealed Esoteric History.分镜稿预览', - 'display_id': 'th1987', - 'thumbnail': 'http://thvideo.tv/uploadfile/2014/0722/20140722013459856.jpg', - 'description': '社团京都幻想剧团的第一个东方二次同人动画作品「秘封活动记录 ～ The Sealed Esoteric History.」 本视频是该动画第一期的分镜草稿...', - 'upload_date': '20140722' - } - } - - def _real_extract(self, url): - video_id = self._match_id(url) - - # extract download link from mobile player page - webpage_player = self._download_webpage( - 'http://thvideo.tv/mobile.php?cid=%s-0' % (video_id), - video_id, note='Downloading video source page') - video_url = self._html_search_regex( - r'<source src="(.*?)" type', webpage_player, 'video url') - - # extract video info from main page - webpage = self._download_webpage( - 'http://thvideo.tv/v/th%s' % (video_id), video_id) - title = self._og_search_title(webpage) - display_id = 'th%s' % video_id - thumbnail = self._og_search_thumbnail(webpage) - description = self._og_search_description(webpage) - upload_date = unified_strdate(self._html_search_regex( - r'span itemprop="datePublished" content="(.*?)">', webpage, - 'upload date', fatal=False)) - - return { - 'id': video_id, - 'ext': 'mp4', - 'url': video_url, - 'title': title, - 'display_id': display_id, - 'thumbnail': thumbnail, - 'description': description, - 'upload_date': upload_date - } - - -class THVideoPlaylistIE(InfoExtractor): - _VALID_URL = r'http?://(?:www\.)?thvideo\.tv/mylist(?P<id>[0-9]+)' - _TEST = { - 'url': 'http://thvideo.tv/mylist2', - 'info_dict': { - 'id': '2', - 'title': '幻想万華鏡', - }, - 'playlist_mincount': 23, - } - - def _real_extract(self, url): - playlist_id = self._match_id(url) - - webpage = self._download_webpage(url, playlist_id) - list_title = self._html_search_regex( - r'<h1 class="show_title">(.*?)<b id', webpage, 'playlist title', - fatal=False) - - entries = [ - self.url_result('http://thvideo.tv/v/th' + id, 'THVideo') - for id in re.findall(r'<dd><a href="http://thvideo.tv/v/th(\d+)/" target=', webpage)] - - return self.playlist_result(entries, playlist_id, list_title) diff --git a/youtube_dl/extractor/turner.py b/youtube_dl/extractor/turner.py index 108caa9d8..b59dafda6 100644 --- a/youtube_dl/extractor/turner.py +++ b/youtube_dl/extractor/turner.py @@ -4,6 +4,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_str from ..utils import ( xpath_text, int_or_none, @@ -16,9 +17,12 @@ from ..utils import ( class TurnerBaseIE(InfoExtractor): + def _extract_timestamp(self, video_data): + return int_or_none(xpath_attr(video_data, 'dateCreated', 'uts')) + def _extract_cvp_info(self, data_src, video_id, path_data={}): video_data = self._download_xml(data_src, video_id) - video_id = video_data.attrib['id'].split('/')[-1].split('.')[0] + video_id = video_data.attrib['id'] title = xpath_text(video_data, 'headline', fatal=True) # rtmp_src = xpath_text(video_data, 'akamai/src') # if rtmp_src: @@ -30,11 +34,11 @@ class TurnerBaseIE(InfoExtractor): tokens = {} urls = [] formats = [] - rex = re.compile(r'''(?x) - (?P<width>[0-9]+)x(?P<height>[0-9]+) - (?:_(?P<bitrate>[0-9]+))? - ''') - for video_file in video_data.findall('files/file'): + rex = re.compile( + r'(?P<width>[0-9]+)x(?P<height>[0-9]+)(?:_(?P<bitrate>[0-9]+))?') + # Possible formats locations: files/file, files/groupFiles/files + # and maybe others + for video_file in video_data.findall('.//file'): video_url = video_file.text.strip() if not video_url: continue @@ -84,12 +88,14 @@ class TurnerBaseIE(InfoExtractor): if video_url in urls: continue urls.append(video_url) - format_id = video_file.attrib['bitrate'] + format_id = video_file.get('bitrate') if ext == 'smil': - formats.extend(self._extract_smil_formats(video_url, video_id, fatal=False)) + formats.extend(self._extract_smil_formats( + video_url, video_id, fatal=False)) elif ext == 'm3u8': m3u8_formats = self._extract_m3u8_formats( - video_url, video_id, 'mp4', m3u8_id=format_id, fatal=False) + video_url, video_id, 'mp4', m3u8_id=format_id or 'hls', + fatal=False) if m3u8_formats: # Sometimes final URLs inside m3u8 are unsigned, let's fix this # ourselves @@ -103,7 +109,7 @@ class TurnerBaseIE(InfoExtractor): elif ext == 'f4m': formats.extend(self._extract_f4m_formats( update_url_query(video_url, {'hdcore': '3.7.0'}), - video_id, f4m_id=format_id, fatal=False)) + video_id, f4m_id=format_id or 'hds', fatal=False)) else: f = { 'format_id': format_id, @@ -117,29 +123,31 @@ class TurnerBaseIE(InfoExtractor): 'height': int(mobj.group('height')), 'tbr': int_or_none(mobj.group('bitrate')), }) - elif format_id.isdigit(): - f['tbr'] = int(format_id) - else: - mobj = re.match(r'ios_(audio|[0-9]+)$', format_id) - if mobj: - if mobj.group(1) == 'audio': - f.update({ - 'vcodec': 'none', - 'ext': 'm4a', - }) - else: - f['tbr'] = int(mobj.group(1)) + elif isinstance(format_id, compat_str): + if format_id.isdigit(): + f['tbr'] = int(format_id) + else: + mobj = re.match(r'ios_(audio|[0-9]+)$', format_id) + if mobj: + if mobj.group(1) == 'audio': + f.update({ + 'vcodec': 'none', + 'ext': 'm4a', + }) + else: + f['tbr'] = int(mobj.group(1)) formats.append(f) self._sort_formats(formats) subtitles = {} for source in video_data.findall('closedCaptions/source'): for track in source.findall('track'): - source_url = source.get('url') - if not source_url: + track_url = track.get('url') + if not isinstance(track_url, compat_str) or track_url.endswith('/big'): continue - subtitles.set_default(source.get('lang') or source.get('label') or 'en', []).append({ - 'url': source_url, + lang = track.get('lang') or track.get('label') or 'en' + subtitles.setdefault(lang, []).append({ + 'url': track_url, 'ext': { 'scc': 'scc', 'webvtt': 'vtt', @@ -154,10 +162,6 @@ class TurnerBaseIE(InfoExtractor): 'height': int_or_none(image.get('height')), } for image in video_data.findall('images/image')] - timestamp = None - if 'cnn.com' not in data_src: - timestamp = int_or_none(xpath_attr(video_data, 'dateCreated', 'uts')) - return { 'id': video_id, 'title': title, @@ -166,7 +170,7 @@ class TurnerBaseIE(InfoExtractor): 'thumbnails': thumbnails, 'description': xpath_text(video_data, 'description'), 'duration': parse_duration(xpath_text(video_data, 'length') or xpath_text(video_data, 'trt')), - 'timestamp': timestamp, + 'timestamp': self._extract_timestamp(video_data), 'upload_date': xpath_attr(video_data, 'metas', 'version'), 'series': xpath_text(video_data, 'showTitle'), 'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')), diff --git a/youtube_dl/extractor/tvnoe.py b/youtube_dl/extractor/tvnoe.py new file mode 100644 index 000000000..1cd3e6a58 --- /dev/null +++ b/youtube_dl/extractor/tvnoe.py @@ -0,0 +1,49 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .jwplatform import JWPlatformBaseIE +from ..utils import ( + clean_html, + get_element_by_class, + js_to_json, +) + + +class TVNoeIE(JWPlatformBaseIE): + _VALID_URL = r'https?://(www\.)?tvnoe\.cz/video/(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://www.tvnoe.cz/video/10362', + 'md5': 'aee983f279aab96ec45ab6e2abb3c2ca', + 'info_dict': { + 'id': '10362', + 'ext': 'mp4', + 'series': 'Noční univerzita', + 'title': 'prof. Tomáš Halík, Th.D. - Návrat náboženství a střet civilizací', + 'description': 'md5:f337bae384e1a531a52c55ebc50fff41', + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + iframe_url = self._search_regex( + r'<iframe[^>]+src="([^"]+)"', webpage, 'iframe URL') + + ifs_page = self._download_webpage(iframe_url, video_id) + jwplayer_data = self._parse_json( + self._find_jwplayer_data(ifs_page), + video_id, transform_source=js_to_json) + info_dict = self._parse_jwplayer_data( + jwplayer_data, video_id, require_title=False, base_url=iframe_url) + + info_dict.update({ + 'id': video_id, + 'title': clean_html(get_element_by_class( + 'field-name-field-podnazev', webpage)), + 'description': clean_html(get_element_by_class( + 'field-name-body', webpage)), + 'series': clean_html(get_element_by_class('title', webpage)) + }) + + return info_dict diff --git a/youtube_dl/extractor/vimple.py b/youtube_dl/extractor/vimple.py index 92321d66e..7fd9b777b 100644 --- a/youtube_dl/extractor/vimple.py +++ b/youtube_dl/extractor/vimple.py @@ -28,23 +28,24 @@ class SprutoBaseIE(InfoExtractor): class VimpleIE(SprutoBaseIE): IE_DESC = 'Vimple - one-click video hosting' - _VALID_URL = r'https?://(?:player\.vimple\.ru/iframe|vimple\.ru)/(?P<id>[\da-f-]{32,36})' - _TESTS = [ - { - 'url': 'http://vimple.ru/c0f6b1687dcd4000a97ebe70068039cf', - 'md5': '2e750a330ed211d3fd41821c6ad9a279', - 'info_dict': { - 'id': 'c0f6b168-7dcd-4000-a97e-be70068039cf', - 'ext': 'mp4', - 'title': 'Sunset', - 'duration': 20, - 'thumbnail': 're:https?://.*?\.jpg', - }, - }, { - 'url': 'http://player.vimple.ru/iframe/52e1beec-1314-4a83-aeac-c61562eadbf9', - 'only_matching': True, - } - ] + _VALID_URL = r'https?://(?:player\.vimple\.(?:ru|co)/iframe|vimple\.(?:ru|co))/(?P<id>[\da-f-]{32,36})' + _TESTS = [{ + 'url': 'http://vimple.ru/c0f6b1687dcd4000a97ebe70068039cf', + 'md5': '2e750a330ed211d3fd41821c6ad9a279', + 'info_dict': { + 'id': 'c0f6b168-7dcd-4000-a97e-be70068039cf', + 'ext': 'mp4', + 'title': 'Sunset', + 'duration': 20, + 'thumbnail': 're:https?://.*?\.jpg', + }, + }, { + 'url': 'http://player.vimple.ru/iframe/52e1beec-1314-4a83-aeac-c61562eadbf9', + 'only_matching': True, + }, { + 'url': 'http://vimple.co/04506a053f124483b8fb05ed73899f19', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) diff --git a/youtube_dl/extractor/vodplatform.py b/youtube_dl/extractor/vodplatform.py index b49542b16..7bdd8b1dc 100644 --- a/youtube_dl/extractor/vodplatform.py +++ b/youtube_dl/extractor/vodplatform.py @@ -6,7 +6,7 @@ from ..utils import unescapeHTML class VODPlatformIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?vod-platform\.net/embed/(?P<id>[^/?#]+)' + _VALID_URL = r'https?://(?:www\.)?vod-platform\.net/[eE]mbed/(?P<id>[^/?#]+)' _TEST = { # from http://www.lbcgroup.tv/watch/chapter/29143/52844/%D8%A7%D9%84%D9%86%D8%B5%D8%B1%D8%A9-%D9%81%D9%8A-%D8%B6%D9%8A%D8%A7%D9%81%D8%A9-%D8%A7%D9%84%D9%80-cnn/ar 'url': 'http://vod-platform.net/embed/RufMcytHDolTH1MuKHY9Fw', diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py index b0679dfb7..d7a81ab8c 100644 --- a/youtube_dl/extractor/yahoo.py +++ b/youtube_dl/extractor/yahoo.py @@ -8,7 +8,6 @@ import re from .common import InfoExtractor, SearchInfoExtractor from ..compat import ( compat_urllib_parse, - compat_urllib_parse_urlencode, compat_urlparse, ) from ..utils import ( @@ -17,6 +16,7 @@ from ..utils import ( ExtractorError, int_or_none, mimetype2ext, + determine_ext, ) from .brightcove import BrightcoveNewIE @@ -39,7 +39,7 @@ class YahooIE(InfoExtractor): }, { 'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html', - 'md5': 'c3466d2b6d5dd6b9f41ba9ed04c24b23', + 'md5': '251af144a19ebc4a033e8ba91ac726bb', 'info_dict': { 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9', 'ext': 'mp4', @@ -50,7 +50,7 @@ class YahooIE(InfoExtractor): }, { 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed', - 'md5': '75ffabdb87c16d4ffe8c036dc4d1c136', + 'md5': '7993e572fac98e044588d0b5260f4352', 'info_dict': { 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb', 'ext': 'mp4', @@ -61,7 +61,7 @@ class YahooIE(InfoExtractor): }, { 'url': 'https://tw.news.yahoo.com/%E6%95%A2%E5%95%8F%E5%B8%82%E9%95%B7%20%E9%BB%83%E7%A7%80%E9%9C%9C%E6%89%B9%E8%B3%B4%E6%B8%85%E5%BE%B7%20%E9%9D%9E%E5%B8%B8%E9%AB%98%E5%82%B2-034024051.html', - 'md5': '9035d38f88b1782682a3e89f985be5bb', + 'md5': '45c024bad51e63e9b6f6fad7a43a8c23', 'info_dict': { 'id': 'cac903b3-fcf4-3c14-b632-643ab541712f', 'ext': 'mp4', @@ -72,10 +72,10 @@ class YahooIE(InfoExtractor): }, { 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html', - 'md5': '0b51660361f0e27c9789e7037ef76f4b', + 'md5': '71298482f7c64cbb7fa064e4553ff1c1', 'info_dict': { 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58', - 'ext': 'mp4', + 'ext': 'webm', 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder', 'description': 'md5:f66c890e1490f4910a9953c941dee944', 'duration': 97, @@ -98,7 +98,7 @@ class YahooIE(InfoExtractor): 'id': '154609075', }, 'playlist': [{ - 'md5': 'f8e336c6b66f503282e5f719641d6565', + 'md5': '000887d0dc609bc3a47c974151a40fb8', 'info_dict': { 'id': 'e624c4bc-3389-34de-9dfc-025f74943409', 'ext': 'mp4', @@ -107,7 +107,7 @@ class YahooIE(InfoExtractor): 'duration': 30, }, }, { - 'md5': '958bcb90b4d6df71c56312137ee1cd5a', + 'md5': '81bc74faf10750fe36e4542f9a184c66', 'info_dict': { 'id': '1fc8ada0-718e-3abe-a450-bf31f246d1a9', 'ext': 'mp4', @@ -139,7 +139,7 @@ class YahooIE(InfoExtractor): 'skip': 'Domain name in.lifestyle.yahoo.com gone', }, { 'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html', - 'md5': 'b17ac378b1134fa44370fb27db09a744', + 'md5': '2a9752f74cb898af5d1083ea9f661b58', 'info_dict': { 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1', 'ext': 'mp4', @@ -168,7 +168,7 @@ class YahooIE(InfoExtractor): }, { # Query result is embedded in webpage, but explicit request to video API fails with geo restriction 'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html', - 'md5': '1ddbf7c850777548438e5c4f147c7b8c', + 'md5': '4fbafb9c9b6f07aa8f870629f6671b35', 'info_dict': { 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504', 'ext': 'mp4', @@ -196,6 +196,7 @@ class YahooIE(InfoExtractor): 'description': 'Galactic', 'title': 'Dolla Diva (feat. Maggie Koerner)', }, + 'skip': 'redirect to https://www.yahoo.com/music', }, ] @@ -213,15 +214,7 @@ class YahooIE(InfoExtractor): entries = [] iframe_urls = re.findall(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage) for idx, iframe_url in enumerate(iframe_urls): - iframepage = self._download_webpage( - host + iframe_url, display_id, - note='Downloading iframe webpage for video #%d' % idx) - items_json = self._search_regex( - r'mediaItems: (\[.+?\])$', iframepage, 'items', flags=re.MULTILINE, default=None) - if items_json: - items = json.loads(items_json) - video_id = items[0]['id'] - entries.append(self._get_info(video_id, display_id, webpage)) + entries.append(self.url_result(host + iframe_url, 'Yahoo')) if entries: return self.playlist_result(entries, page_id) @@ -246,7 +239,9 @@ class YahooIE(InfoExtractor): if config: sapi = config.get('models', {}).get('applet_model', {}).get('data', {}).get('sapi') if sapi and 'query' in sapi: - return self._extract_info(display_id, sapi, webpage) + info = self._extract_info(display_id, sapi, webpage) + self._sort_formats(info['formats']) + return info items_json = self._search_regex( r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE, @@ -292,15 +287,17 @@ class YahooIE(InfoExtractor): formats = [] for s in info['streams']: + tbr = int_or_none(s.get('bitrate')) format_info = { 'width': int_or_none(s.get('width')), 'height': int_or_none(s.get('height')), - 'tbr': int_or_none(s.get('bitrate')), + 'tbr': tbr, } host = s['host'] path = s['path'] if host.startswith('rtmp'): + fmt = 'rtmp' format_info.update({ 'url': host, 'play_path': path, @@ -308,14 +305,18 @@ class YahooIE(InfoExtractor): }) else: if s.get('format') == 'm3u8_playlist': - format_info['protocol'] = 'm3u8_native' - format_info['ext'] = 'mp4' + fmt = 'hls' + format_info.update({ + 'protocol': 'm3u8_native', + 'ext': 'mp4', + }) + else: + fmt = format_info['ext'] = determine_ext(path) format_url = compat_urlparse.urljoin(host, path) format_info['url'] = format_url + format_info['format_id'] = fmt + ('-%d' % tbr if tbr else '') formats.append(format_info) - self._sort_formats(formats) - closed_captions = self._html_search_regex( r'"closedcaptions":(\[[^\]]+\])', webpage, 'closed captions', default='[]') @@ -346,17 +347,25 @@ class YahooIE(InfoExtractor): def _get_info(self, video_id, display_id, webpage): region = self._search_regex( r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"', - webpage, 'region', fatal=False, default='US') - data = compat_urllib_parse_urlencode({ - 'protocol': 'http', - 'region': region.upper(), - }) - query_url = ( - 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/' - '{id}?{data}'.format(id=video_id, data=data)) - query_result = self._download_json( - query_url, display_id, 'Downloading video info') - return self._extract_info(display_id, query_result, webpage) + webpage, 'region', fatal=False, default='US').upper() + formats = [] + info = {} + for fmt in ('webm', 'mp4'): + query_result = self._download_json( + 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/' + video_id, + display_id, 'Downloading %s video info' % fmt, query={ + 'protocol': 'http', + 'region': region, + 'format': fmt, + }) + info = self._extract_info(display_id, query_result, webpage) + formats.extend(info['formats']) + formats.extend(self._extract_m3u8_formats( + 'http://video.media.yql.yahoo.com/v1/hls/%s?region=%s' % (video_id, region), + video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) + self._sort_formats(formats) + info['formats'] = formats + return info class YahooSearchIE(SearchInfoExtractor): diff --git a/youtube_dl/extractor/youjizz.py b/youtube_dl/extractor/youjizz.py index 31e2f9263..b50f34e9b 100644 --- a/youtube_dl/extractor/youjizz.py +++ b/youtube_dl/extractor/youjizz.py @@ -1,21 +1,16 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor -from ..utils import ( - ExtractorError, -) class YouJizzIE(InfoExtractor): _VALID_URL = r'https?://(?:\w+\.)?youjizz\.com/videos/(?:[^/#?]+)?-(?P<id>[0-9]+)\.html(?:$|[?#])' _TESTS = [{ 'url': 'http://www.youjizz.com/videos/zeichentrick-1-2189178.html', - 'md5': '07e15fa469ba384c7693fd246905547c', + 'md5': '78fc1901148284c69af12640e01c6310', 'info_dict': { 'id': '2189178', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Zeichentrick 1', 'age_limit': 18, } @@ -27,38 +22,18 @@ class YouJizzIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) + # YouJizz's HTML5 player has invalid HTML + webpage = webpage.replace('"controls', '" controls') age_limit = self._rta_search(webpage) video_title = self._html_search_regex( r'<title>\s*(.*)\s*</title>', webpage, 'title') - embed_page_url = self._search_regex( - r'(https?://www.youjizz.com/videos/embed/[0-9]+)', - webpage, 'embed page') - webpage = self._download_webpage( - embed_page_url, video_id, note='downloading embed page') - - # Get the video URL - m_playlist = re.search(r'so.addVariable\("playlist", ?"(?P<playlist>.+?)"\);', webpage) - if m_playlist is not None: - playlist_url = m_playlist.group('playlist') - playlist_page = self._download_webpage(playlist_url, video_id, - 'Downloading playlist page') - m_levels = list(re.finditer(r'<level bitrate="(\d+?)" file="(.*?)"', playlist_page)) - if len(m_levels) == 0: - raise ExtractorError('Unable to extract video url') - videos = [(int(m.group(1)), m.group(2)) for m in m_levels] - (_, video_url) = sorted(videos)[0] - video_url = video_url.replace('%252F', '%2F') - else: - video_url = self._search_regex(r'so.addVariable\("file",encodeURIComponent\("(?P<source>[^"]+)"\)\);', - webpage, 'video URL') + info_dict = self._parse_html5_media_entries(url, webpage, video_id)[0] - return { + info_dict.update({ 'id': video_id, - 'url': video_url, 'title': video_title, - 'ext': 'flv', - 'format': 'flv', - 'player_url': embed_page_url, 'age_limit': age_limit, - } + }) + + return info_dict diff --git a/youtube_dl/extractor/youporn.py b/youtube_dl/extractor/youporn.py index 0df2d76ee..0265a64a7 100644 --- a/youtube_dl/extractor/youporn.py +++ b/youtube_dl/extractor/youporn.py @@ -35,7 +35,7 @@ class YouPornIE(InfoExtractor): 'age_limit': 18, }, }, { - # Anonymous User uploader + # Unknown uploader 'url': 'http://www.youporn.com/watch/561726/big-tits-awesome-brunette-on-amazing-webcam-show/?from=related3&al=2&from_id=561726&pos=4', 'info_dict': { 'id': '561726', @@ -44,7 +44,7 @@ class YouPornIE(InfoExtractor): 'title': 'Big Tits Awesome Brunette On amazing webcam show', 'description': 'http://sweetlivegirls.com Big Tits Awesome Brunette On amazing webcam show.mp4', 'thumbnail': 're:^https?://.*\.jpg$', - 'uploader': 'Anonymous User', + 'uploader': 'Unknown', 'upload_date': '20111125', 'average_rating': int, 'view_count': int, @@ -140,17 +140,17 @@ class YouPornIE(InfoExtractor): r'>All [Cc]omments? \(([\d,.]+)\)', webpage, 'comment count', fatal=False)) - def extract_tag_box(title): - tag_box = self._search_regex( - (r'<div[^>]+class=["\']tagBoxTitle["\'][^>]*>\s*%s\b.*?</div>\s*' - '<div[^>]+class=["\']tagBoxContent["\']>(.+?)</div>') % re.escape(title), - webpage, '%s tag box' % title, default=None) + def extract_tag_box(regex, title): + tag_box = self._search_regex(regex, webpage, title, default=None) if not tag_box: return [] return re.findall(r'<a[^>]+href=[^>]+>([^<]+)', tag_box) - categories = extract_tag_box('Category') - tags = extract_tag_box('Tags') + categories = extract_tag_box( + r'(?s)Categories:.*?</[^>]+>(.+?)</div>', 'categories') + tags = extract_tag_box( + r'(?s)Tags:.*?</div>\s*<div[^>]+class=["\']tagBoxContent["\'][^>]*>(.+?)</div>', + 'tags') return { 'id': video_id, diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index d5d5b7334..8fc26bd02 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -264,7 +264,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): ) )? # all until now is optional -> you can pass the naked ID ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID - (?!.*?&list=) # combined list/video URLs are handled by the playlist IE + (?!.*?\blist=) # combined list/video URLs are handled by the playlist IE (?(1).+)? # if we found the ID, everything can follow $""" _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' @@ -844,6 +844,24 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059) 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo', 'only_matching': True, + }, + { + # Rental video preview + 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg', + 'info_dict': { + 'id': 'uGpuVWrhIzE', + 'ext': 'mp4', + 'title': 'Piku - Trailer', + 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb', + 'upload_date': '20150811', + 'uploader': 'FlixMatrix', + 'uploader_id': 'FlixMatrixKaravan', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan', + 'license': 'Standard YouTube License', + }, + 'params': { + 'skip_download': True, + }, } ] @@ -1254,6 +1272,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # Convert to the same format returned by compat_parse_qs video_info = dict((k, [v]) for k, v in args.items()) add_dash_mpd(video_info) + # Rental video is not rented but preview is available (e.g. + # https://www.youtube.com/watch?v=yYr8q0y5Jfg, + # https://github.com/rg3/youtube-dl/issues/10532) + if not video_info and args.get('ypc_vid'): + return self.url_result( + args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid']) if args.get('livestream') == '1' or args.get('live_playback') == 1: is_live = True if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True): @@ -1754,11 +1778,14 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): _VALID_URL = r"""(?x)(?: (?:https?://)? (?:\w+\.)? - youtube\.com/ (?: - (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries) - \? (?:.*?[&;])*? (?:p|a|list)= - | p/ + youtube\.com/ + (?: + (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries) + \? (?:.*?[&;])*? (?:p|a|list)= + | p/ + )| + youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist= ) ( (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,} @@ -1841,6 +1868,31 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): 'id': 'UUXw-G3eDE9trcvY2sBMM_aA', }, 'playlist_mincout': 21, + }, { + # Playlist URL that does not actually serve a playlist + 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4', + 'info_dict': { + 'id': 'FqZTN594JQw', + 'ext': 'webm', + 'title': "Smiley's People 01 detective, Adventure Series, Action", + 'uploader': 'STREEM', + 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng', + 'upload_date': '20150526', + 'license': 'Standard YouTube License', + 'description': 'md5:507cdcb5a49ac0da37a920ece610be80', + 'categories': ['People & Blogs'], + 'tags': list, + 'like_count': int, + 'dislike_count': int, + }, + 'params': { + 'skip_download': True, + }, + 'add_ie': [YoutubeIE.ie_key()], + }, { + 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21', + 'only_matching': True, }] def _real_initialize(self): @@ -1901,9 +1953,20 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): playlist_title = self._html_search_regex( r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>', - page, 'title') + page, 'title', default=None) + + has_videos = True - return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title) + if not playlist_title: + try: + # Some playlist URLs don't actually serve a playlist (e.g. + # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4) + next(self._entries(page, playlist_id)) + except StopIteration: + has_videos = False + + return has_videos, self.playlist_result( + self._entries(page, playlist_id), playlist_id, playlist_title) def _check_download_just_video(self, url, playlist_id): # Check if it's a video-specific URL @@ -1912,9 +1975,11 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): video_id = query_dict['v'][0] if self._downloader.params.get('noplaylist'): self.to_screen('Downloading just video %s because of --no-playlist' % video_id) - return self.url_result(video_id, 'Youtube', video_id=video_id) + return video_id, self.url_result(video_id, 'Youtube', video_id=video_id) else: self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id)) + return video_id, None + return None, None def _real_extract(self, url): # Extract playlist id @@ -1923,7 +1988,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): raise ExtractorError('Invalid URL: %s' % url) playlist_id = mobj.group(1) or mobj.group(2) - video = self._check_download_just_video(url, playlist_id) + video_id, video = self._check_download_just_video(url, playlist_id) if video: return video @@ -1931,7 +1996,15 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): # Mixes require a custom extraction process return self._extract_mix(playlist_id) - return self._extract_playlist(playlist_id) + has_videos, playlist = self._extract_playlist(playlist_id) + if has_videos or not video_id: + return playlist + + # Some playlist URLs don't actually serve a playlist (see + # https://github.com/rg3/youtube-dl/issues/10537). + # Fallback to plain video extraction if there is a video id + # along with playlist id. + return self.url_result(video_id, 'Youtube', video_id=video_id) class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): @@ -2309,10 +2382,11 @@ class YoutubeWatchLaterIE(YoutubePlaylistIE): }] def _real_extract(self, url): - video = self._check_download_just_video(url, 'WL') + _, video = self._check_download_just_video(url, 'WL') if video: return video - return self._extract_playlist('WL') + _, playlist = self._extract_playlist('WL') + return playlist class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): diff --git a/youtube_dl/options.py b/youtube_dl/options.py index 5d62deef4..56f312f57 100644 --- a/youtube_dl/options.py +++ b/youtube_dl/options.py @@ -423,7 +423,15 @@ def parseOpts(overrideArguments=None): downloader.add_option( '--fragment-retries', dest='fragment_retries', metavar='RETRIES', default=10, - help='Number of retries for a fragment (default is %default), or "infinite" (DASH only)') + help='Number of retries for a fragment (default is %default), or "infinite" (DASH and hlsnative only)') + downloader.add_option( + '--skip-unavailable-fragments', + action='store_true', dest='skip_unavailable_fragments', default=True, + help='Skip unavailable fragments (DASH and hlsnative only)') + general.add_option( + '--abort-on-unavailable-fragment', + action='store_false', dest='skip_unavailable_fragments', + help='Abort downloading when some fragment is not available') downloader.add_option( '--buffer-size', dest='buffersize', metavar='SIZE', default='1024', diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 1091f17f3..ed199c4ad 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -2148,7 +2148,7 @@ def mimetype2ext(mt): return ext _, _, res = mt.rpartition('/') - res = res.lower() + res = res.split(';')[0].strip().lower() return { '3gpp': '3gp', @@ -2168,6 +2168,7 @@ def mimetype2ext(mt): 'f4m+xml': 'f4m', 'hds+xml': 'f4m', 'vnd.ms-sstr+xml': 'ism', + 'quicktime': 'mov', }.get(res, res) diff --git a/youtube_dl/version.py b/youtube_dl/version.py index ee30ca2ad..b2ea6dac6 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,3 +1,3 @@ from __future__ import unicode_literals -__version__ = '2016.08.28' +__version__ = '2016.09.04.1' |