diff options
47 files changed, 878 insertions, 280 deletions
@@ -173,6 +173,10 @@ which means you can modify it, redistribute it or use it however you like. expected filesize (experimental) --hls-prefer-native Use the native HLS downloader instead of ffmpeg (experimental) + --hls-use-mpegts Use the mpegts container for HLS videos, + allowing to play the video while + downloading (some players may not be able + to play it) --external-downloader COMMAND Use the specified external downloader. Currently supports aria2c,axel,curl,httpie,wget diff --git a/docs/supportedsites.md b/docs/supportedsites.md index eb68c23b5..61be9990d 100644 --- a/docs/supportedsites.md +++ b/docs/supportedsites.md @@ -55,6 +55,7 @@ - **audiomack** - **audiomack:album** - **Azubu** + - **AzubuLive** - **BaiduVideo**: 百度视频 - **bambuser** - **bambuser:channel** @@ -133,6 +134,8 @@ - **DailymotionCloud** - **daum.net** - **daum.net:clip** + - **daum.net:playlist** + - **daum.net:user** - **DBTV** - **DCN** - **dcn:live** @@ -315,6 +318,7 @@ - **mailru**: Видео@Mail.Ru - **MakerTV** - **Malemotion** + - **MatchTV** - **MDR**: MDR.DE and KiKA - **media.ccc.de** - **metacafe** @@ -507,6 +511,7 @@ - **Sapo**: SAPO Vídeos - **savefrom.net** - **SBS**: sbs.com.au + - **schooltv** - **SciVee** - **screen.yahoo:search**: Yahoo screen search - **Screencast** diff --git a/test/test_YoutubeDL.py b/test/test_YoutubeDL.py index b53cfbe78..73910eaec 100644 --- a/test/test_YoutubeDL.py +++ b/test/test_YoutubeDL.py @@ -14,6 +14,7 @@ from test.helper import FakeYDL, assertRegexpMatches from youtube_dl import YoutubeDL from youtube_dl.compat import compat_str, compat_urllib_error from youtube_dl.extractor import YoutubeIE +from youtube_dl.extractor.common import InfoExtractor from youtube_dl.postprocessor.common import PostProcessor from youtube_dl.utils import ExtractorError, match_filter_func @@ -646,6 +647,42 @@ class TestYoutubeDL(unittest.TestCase): ydl = YDL() self.assertRaises(compat_urllib_error.URLError, ydl.urlopen, 'file:///etc/passwd') + def test_do_not_override_ie_key_in_url_transparent(self): + ydl = YDL() + + class Foo1IE(InfoExtractor): + _VALID_URL = r'foo1:' + + def _real_extract(self, url): + return { + '_type': 'url_transparent', + 'url': 'foo2:', + 'ie_key': 'Foo2', + } + + class Foo2IE(InfoExtractor): + _VALID_URL = r'foo2:' + + def _real_extract(self, url): + return { + '_type': 'url', + 'url': 'foo3:', + 'ie_key': 'Foo3', + } + + class Foo3IE(InfoExtractor): + _VALID_URL = r'foo3:' + + def _real_extract(self, url): + return _make_result([{'url': TEST_URL}]) + + ydl.add_info_extractor(Foo1IE(ydl)) + ydl.add_info_extractor(Foo2IE(ydl)) + ydl.add_info_extractor(Foo3IE(ydl)) + ydl.extract_info('foo1:') + downloaded = ydl.downloaded_info_dicts[0] + self.assertEqual(downloaded['url'], TEST_URL) + if __name__ == '__main__': unittest.main() diff --git a/test/test_all_urls.py b/test/test_all_urls.py index a0c11e6c1..f5af184e6 100644 --- a/test/test_all_urls.py +++ b/test/test_all_urls.py @@ -56,7 +56,7 @@ class TestAllURLsMatching(unittest.TestCase): assertChannel('https://www.youtube.com/channel/HCtnHdj3df7iM/videos') def test_youtube_user_matching(self): - self.assertMatch('www.youtube.com/NASAgovVideo/videos', ['youtube:user']) + self.assertMatch('http://www.youtube.com/NASAgovVideo/videos', ['youtube:user']) def test_youtube_feeds(self): self.assertMatch('https://www.youtube.com/feed/watch_later', ['youtube:watchlater']) diff --git a/test/test_subtitles.py b/test/test_subtitles.py index 9ed9fe622..9a695c4e8 100644 --- a/test/test_subtitles.py +++ b/test/test_subtitles.py @@ -21,7 +21,7 @@ from youtube_dl.extractor import ( NPOIE, ComedyCentralIE, NRKTVIE, - RaiIE, + RaiTVIE, VikiIE, ThePlatformIE, ThePlatformFeedIE, @@ -260,7 +260,7 @@ class TestNRKSubtitles(BaseTestSubtitles): class TestRaiSubtitles(BaseTestSubtitles): url = 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html' - IE = RaiIE + IE = RaiTVIE def test_allsubtitles(self): self.DL.params['writesubtitles'] = True diff --git a/test/test_youtube_lists.py b/test/test_youtube_lists.py index 26aadb34f..47df0f348 100644 --- a/test/test_youtube_lists.py +++ b/test/test_youtube_lists.py @@ -34,7 +34,7 @@ class TestYoutubeLists(unittest.TestCase): ie = YoutubePlaylistIE(dl) # TODO find a > 100 (paginating?) videos course result = ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8') - entries = result['entries'] + entries = list(result['entries']) self.assertEqual(YoutubeIE().extract_id(entries[0]['url']), 'j9WZyLZCBzs') self.assertEqual(len(entries), 25) self.assertEqual(YoutubeIE().extract_id(entries[-1]['url']), 'rYefUsYuEp0') diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index e1bd40843..50228bb32 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -263,7 +263,7 @@ class YoutubeDL(object): the downloader (see youtube_dl/downloader/common.py): nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test, noresizebuffer, retries, continuedl, noprogress, consoletitle, - xattr_set_filesize, external_downloader_args. + xattr_set_filesize, external_downloader_args, hls_use_mpegts. The following options are used by the post processors: prefer_ffmpeg: If True, use ffmpeg instead of avconv if both are available, @@ -707,7 +707,6 @@ class YoutubeDL(object): It will also download the videos if 'download'. Returns the resolved ie_result. """ - result_type = ie_result.get('_type', 'video') if result_type in ('url', 'url_transparent'): @@ -736,7 +735,7 @@ class YoutubeDL(object): force_properties = dict( (k, v) for k, v in ie_result.items() if v is not None) - for f in ('_type', 'url'): + for f in ('_type', 'url', 'ie_key'): if f in force_properties: del force_properties[f] new_result = info.copy() diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index 9f131f5db..f5f064241 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -369,6 +369,7 @@ def _real_main(argv=None): 'no_color': opts.no_color, 'ffmpeg_location': opts.ffmpeg_location, 'hls_prefer_native': opts.hls_prefer_native, + 'hls_use_mpegts': opts.hls_use_mpegts, 'external_downloader_args': external_downloader_args, 'postprocessor_args': postprocessor_args, 'cn_verification_proxy': opts.cn_verification_proxy, diff --git a/youtube_dl/downloader/common.py b/youtube_dl/downloader/common.py index fc7521598..de815612c 100644 --- a/youtube_dl/downloader/common.py +++ b/youtube_dl/downloader/common.py @@ -45,6 +45,7 @@ class FileDownloader(object): (experimental) external_downloader_args: A list of additional command-line arguments for the external downloader. + hls_use_mpegts: Use the mpegts container for HLS videos. Subclasses of this one must re-define the real_download method. """ diff --git a/youtube_dl/downloader/f4m.py b/youtube_dl/downloader/f4m.py index aaf0c49c8..fc9642905 100644 --- a/youtube_dl/downloader/f4m.py +++ b/youtube_dl/downloader/f4m.py @@ -273,15 +273,21 @@ class F4mFD(FragmentFD): return fragments_list def _parse_bootstrap_node(self, node, base_url): - if node.text is None: + # Sometimes non empty inline bootstrap info can be specified along + # with bootstrap url attribute (e.g. dummy inline bootstrap info + # contains whitespace characters in [1]). We will prefer bootstrap + # url over inline bootstrap info when present. + # 1. http://live-1-1.rutube.ru/stream/1024/HDS/SD/C2NKsS85HQNckgn5HdEmOQ/1454167650/S-s604419906/move/four/dirs/upper/1024-576p.f4m + bootstrap_url = node.get('url') + if bootstrap_url: bootstrap_url = compat_urlparse.urljoin( - base_url, node.attrib['url']) + base_url, bootstrap_url) boot_info = self._get_bootstrap_from_url(bootstrap_url) else: bootstrap_url = None bootstrap = base64.b64decode(node.text.encode('ascii')) boot_info = read_bootstrap_info(bootstrap) - return (boot_info, bootstrap_url) + return boot_info, bootstrap_url def real_download(self, filename, info_dict): man_url = info_dict['url'] @@ -316,7 +322,8 @@ class F4mFD(FragmentFD): metadata = None fragments_list = build_fragments_list(boot_info) - if self.params.get('test', False): + test = self.params.get('test', False) + if test: # We only download the first fragment fragments_list = fragments_list[:1] total_frags = len(fragments_list) @@ -326,6 +333,7 @@ class F4mFD(FragmentFD): ctx = { 'filename': filename, 'total_frags': total_frags, + 'live': live, } self._prepare_frag_download(ctx) @@ -380,7 +388,7 @@ class F4mFD(FragmentFD): else: raise - if not fragments_list and live and bootstrap_url: + if not fragments_list and not test and live and bootstrap_url: fragments_list = self._update_live_fragments(bootstrap_url, frag_i) total_frags += len(fragments_list) if fragments_list and (fragments_list[0][1] > frag_i + 1): diff --git a/youtube_dl/downloader/fragment.py b/youtube_dl/downloader/fragment.py index 0c9113d0f..8b96eceb9 100644 --- a/youtube_dl/downloader/fragment.py +++ b/youtube_dl/downloader/fragment.py @@ -26,7 +26,11 @@ class FragmentFD(FileDownloader): self._start_frag_download(ctx) def _prepare_frag_download(self, ctx): - self.to_screen('[%s] Total fragments: %d' % (self.FD_NAME, ctx['total_frags'])) + if 'live' not in ctx: + ctx['live'] = False + self.to_screen( + '[%s] Total fragments: %s' + % (self.FD_NAME, ctx['total_frags'] if not ctx['live'] else 'unknown (live)')) self.report_destination(ctx['filename']) dl = HttpQuietDownloader( self.ydl, @@ -74,14 +78,14 @@ class FragmentFD(FileDownloader): if s['status'] not in ('downloading', 'finished'): return - frag_total_bytes = s.get('total_bytes') or 0 - - estimated_size = ( - (ctx['complete_frags_downloaded_bytes'] + frag_total_bytes) / - (state['frag_index'] + 1) * total_frags) time_now = time.time() - state['total_bytes_estimate'] = estimated_size state['elapsed'] = time_now - start + frag_total_bytes = s.get('total_bytes') or 0 + if not ctx['live']: + estimated_size = ( + (ctx['complete_frags_downloaded_bytes'] + frag_total_bytes) / + (state['frag_index'] + 1) * total_frags) + state['total_bytes_estimate'] = estimated_size if s['status'] == 'finished': state['frag_index'] += 1 @@ -91,9 +95,10 @@ class FragmentFD(FileDownloader): else: frag_downloaded_bytes = s['downloaded_bytes'] state['downloaded_bytes'] += frag_downloaded_bytes - ctx['prev_frag_downloaded_bytes'] - state['eta'] = self.calc_eta( - start, time_now, estimated_size, - state['downloaded_bytes']) + if not ctx['live']: + state['eta'] = self.calc_eta( + start, time_now, estimated_size, + state['downloaded_bytes']) state['speed'] = s.get('speed') ctx['prev_frag_downloaded_bytes'] = frag_downloaded_bytes self._hook_progress(state) diff --git a/youtube_dl/downloader/hls.py b/youtube_dl/downloader/hls.py index 10b83c6b2..cb34dc4ab 100644 --- a/youtube_dl/downloader/hls.py +++ b/youtube_dl/downloader/hls.py @@ -39,7 +39,11 @@ class HlsFD(FileDownloader): '-headers', ''.join('%s: %s\r\n' % (key, val) for key, val in headers.items())] - args += ['-i', url, '-f', 'mp4', '-c', 'copy', '-bsf:a', 'aac_adtstoasc'] + args += ['-i', url, '-c', 'copy'] + if self.params.get('hls_use_mpegts', False): + args += ['-f', 'mpegts'] + else: + args += ['-f', 'mp4', '-bsf:a', 'aac_adtstoasc'] args = [encodeArgument(opt) for opt in args] args.append(encodeFilename(ffpp._ffmpeg_filename_argument(tmpfilename), True)) diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 532be7e4c..e61a88de7 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -50,7 +50,7 @@ from .atresplayer import AtresPlayerIE from .atttechchannel import ATTTechChannelIE from .audimedia import AudiMediaIE from .audiomack import AudiomackIE, AudiomackAlbumIE -from .azubu import AzubuIE +from .azubu import AzubuIE, AzubuLiveIE from .baidu import BaiduVideoIE from .bambuser import BambuserIE, BambuserChannelIE from .bandcamp import BandcampIE, BandcampAlbumIE @@ -142,6 +142,8 @@ from .dailymotion import ( from .daum import ( DaumIE, DaumClipIE, + DaumPlaylistIE, + DaumUserIE, ) from .dbtv import DBTVIE from .dcn import ( @@ -372,6 +374,7 @@ from .macgamestore import MacGameStoreIE from .mailru import MailRuIE from .makertv import MakerTVIE from .malemotion import MalemotionIE +from .matchtv import MatchTVIE from .mdr import MDRIE from .metacafe import MetacafeIE from .metacritic import MetacriticIE @@ -482,6 +485,7 @@ from .npo import ( NPOLiveIE, NPORadioIE, NPORadioFragmentIE, + SchoolTVIE, VPROIE, WNLIE ) diff --git a/youtube_dl/extractor/acast.py b/youtube_dl/extractor/acast.py index be7913bc7..92eee8119 100644 --- a/youtube_dl/extractor/acast.py +++ b/youtube_dl/extractor/acast.py @@ -8,11 +8,7 @@ from ..compat import compat_str from ..utils import int_or_none -class ACastBaseIE(InfoExtractor): - _API_BASE_URL = 'https://www.acast.com/api/' - - -class ACastIE(ACastBaseIE): +class ACastIE(InfoExtractor): IE_NAME = 'acast' _VALID_URL = r'https?://(?:www\.)?acast\.com/(?P<channel>[^/]+)/(?P<id>[^/#?]+)' _TEST = { @@ -23,14 +19,19 @@ class ACastIE(ACastBaseIE): 'ext': 'mp3', 'title': '"Where Are You?": Taipei 101, Taiwan', 'timestamp': 1196172000000, - 'description': 'md5:0c5d8201dfea2b93218ea986c91eee6e', + 'description': 'md5:a0b4ef3634e63866b542e5b1199a1a0e', 'duration': 211, } } def _real_extract(self, url): channel, display_id = re.match(self._VALID_URL, url).groups() - cast_data = self._download_json(self._API_BASE_URL + 'channels/%s/acasts/%s/playback' % (channel, display_id), display_id) + + embed_page = self._download_webpage( + re.sub('(?:www\.)?acast\.com', 'embedcdn.acast.com', url), display_id) + cast_data = self._parse_json(self._search_regex( + r'window\[\'acast/queries\'\]\s*=\s*([^;]+);', embed_page, 'acast data'), + display_id)['GetAcast/%s/%s' % (channel, display_id)] return { 'id': compat_str(cast_data['id']), @@ -44,7 +45,7 @@ class ACastIE(ACastBaseIE): } -class ACastChannelIE(ACastBaseIE): +class ACastChannelIE(InfoExtractor): IE_NAME = 'acast:channel' _VALID_URL = r'https?://(?:www\.)?acast\.com/(?P<id>[^/#?]+)' _TEST = { @@ -56,6 +57,7 @@ class ACastChannelIE(ACastBaseIE): }, 'playlist_mincount': 20, } + _API_BASE_URL = 'https://www.acast.com/api/' @classmethod def suitable(cls, url): diff --git a/youtube_dl/extractor/allocine.py b/youtube_dl/extractor/allocine.py index 7d65b8193..190bc2cc8 100644 --- a/youtube_dl/extractor/allocine.py +++ b/youtube_dl/extractor/allocine.py @@ -8,6 +8,8 @@ from .common import InfoExtractor from ..compat import compat_str from ..utils import ( qualities, + unescapeHTML, + xpath_element, ) @@ -31,7 +33,7 @@ class AllocineIE(InfoExtractor): 'id': '19540403', 'ext': 'mp4', 'title': 'Planes 2 Bande-annonce VF', - 'description': 'md5:eeaffe7c2d634525e21159b93acf3b1e', + 'description': 'Regardez la bande annonce du film Planes 2 (Planes 2 Bande-annonce VF). Planes 2, un film de Roberts Gannaway', 'thumbnail': 're:http://.*\.jpg', }, }, { @@ -41,7 +43,7 @@ class AllocineIE(InfoExtractor): 'id': '19544709', 'ext': 'mp4', 'title': 'Dragons 2 - Bande annonce finale VF', - 'description': 'md5:71742e3a74b0d692c7fce0dd2017a4ac', + 'description': 'md5:601d15393ac40f249648ef000720e7e3', 'thumbnail': 're:http://.*\.jpg', }, }, { @@ -59,14 +61,18 @@ class AllocineIE(InfoExtractor): if typ == 'film': video_id = self._search_regex(r'href="/video/player_gen_cmedia=([0-9]+).+"', webpage, 'video id') else: - player = self._search_regex(r'data-player=\'([^\']+)\'>', webpage, 'data player') - - player_data = json.loads(player) - video_id = compat_str(player_data['refMedia']) + player = self._search_regex(r'data-player=\'([^\']+)\'>', webpage, 'data player', default=None) + if player: + player_data = json.loads(player) + video_id = compat_str(player_data['refMedia']) + else: + model = self._search_regex(r'data-model="([^"]+)">', webpage, 'data model') + model_data = self._parse_json(unescapeHTML(model), display_id) + video_id = compat_str(model_data['id']) xml = self._download_xml('http://www.allocine.fr/ws/AcVisiondataV4.ashx?media=%s' % video_id, display_id) - video = xml.find('.//AcVisionVideo').attrib + video = xpath_element(xml, './/AcVisionVideo').attrib quality = qualities(['ld', 'md', 'hd']) formats = [] diff --git a/youtube_dl/extractor/azubu.py b/youtube_dl/extractor/azubu.py index 0961d339f..011edf128 100644 --- a/youtube_dl/extractor/azubu.py +++ b/youtube_dl/extractor/azubu.py @@ -3,7 +3,11 @@ from __future__ import unicode_literals import json from .common import InfoExtractor -from ..utils import float_or_none +from ..utils import ( + ExtractorError, + float_or_none, + sanitized_Request, +) class AzubuIE(InfoExtractor): @@ -91,3 +95,37 @@ class AzubuIE(InfoExtractor): 'view_count': view_count, 'formats': formats, } + + +class AzubuLiveIE(InfoExtractor): + _VALID_URL = r'http://www.azubu.tv/(?P<id>[^/]+)$' + + _TEST = { + 'url': 'http://www.azubu.tv/MarsTVMDLen', + 'only_matching': True, + } + + def _real_extract(self, url): + user = self._match_id(url) + + info = self._download_json( + 'http://api.azubu.tv/public/modules/last-video/{0}/info'.format(user), + user)['data'] + if info['type'] != 'STREAM': + raise ExtractorError('{0} is not streaming live'.format(user), expected=True) + + req = sanitized_Request( + 'https://edge-elb.api.brightcove.com/playback/v1/accounts/3361910549001/videos/ref:' + info['reference_id']) + req.add_header('Accept', 'application/json;pk=BCpkADawqM1gvI0oGWg8dxQHlgT8HkdE2LnAlWAZkOlznO39bSZX726u4JqnDsK3MDXcO01JxXK2tZtJbgQChxgaFzEVdHRjaDoxaOu8hHOO8NYhwdxw9BzvgkvLUlpbDNUuDoc4E4wxDToV') + bc_info = self._download_json(req, user) + m3u8_url = next(source['src'] for source in bc_info['sources'] if source['container'] == 'M2TS') + formats = self._extract_m3u8_formats(m3u8_url, user, ext='mp4') + + return { + 'id': info['id'], + 'title': self._live_title(info['title']), + 'uploader_id': user, + 'formats': formats, + 'is_live': True, + 'thumbnail': bc_info['poster'], + } diff --git a/youtube_dl/extractor/bbc.py b/youtube_dl/extractor/bbc.py index 210ce568b..6ddee686c 100644 --- a/youtube_dl/extractor/bbc.py +++ b/youtube_dl/extractor/bbc.py @@ -194,6 +194,19 @@ class BBCCoUkIE(InfoExtractor): 'skip_download': True, }, }, { + # compact player (https://github.com/rg3/youtube-dl/issues/8147) + 'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player', + 'info_dict': { + 'id': 'p028bfkj', + 'ext': 'flv', + 'title': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews', + 'description': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews', + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + }, { 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4', 'only_matching': True, }, { @@ -485,7 +498,8 @@ class BBCCoUkIE(InfoExtractor): (r'<h2[^>]+id="parent-title"[^>]*>(.+?)</h2>', r'<div[^>]+class="info"[^>]*>\s*<h1>(.+?)</h1>'), webpage, 'title') description = self._search_regex( - r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>', + (r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>', + r'<div[^>]+class="info_+synopsis"[^>]*>([^<]+)</div>'), webpage, 'description', default=None) if not description: description = self._html_search_meta('description', webpage) diff --git a/youtube_dl/extractor/bpb.py b/youtube_dl/extractor/bpb.py index 510813f76..c28e72927 100644 --- a/youtube_dl/extractor/bpb.py +++ b/youtube_dl/extractor/bpb.py @@ -1,7 +1,13 @@ # coding: utf-8 from __future__ import unicode_literals +import re + from .common import InfoExtractor +from ..utils import ( + js_to_json, + determine_ext, +) class BpbIE(InfoExtractor): @@ -10,7 +16,8 @@ class BpbIE(InfoExtractor): _TEST = { 'url': 'http://www.bpb.de/mediathek/297/joachim-gauck-zu-1989-und-die-erinnerung-an-die-ddr', - 'md5': '0792086e8e2bfbac9cdf27835d5f2093', + # md5 fails in Python 2.6 due to buggy server response and wrong handling of urllib2 + 'md5': 'c4f84c8a8044ca9ff68bb8441d300b3f', 'info_dict': { 'id': '297', 'ext': 'mp4', @@ -25,13 +32,26 @@ class BpbIE(InfoExtractor): title = self._html_search_regex( r'<h2 class="white">(.*?)</h2>', webpage, 'title') - video_url = self._html_search_regex( - r'(http://film\.bpb\.de/player/dokument_[0-9]+\.mp4)', - webpage, 'video URL') + video_info_dicts = re.findall( + r"({\s*src:\s*'http://film\.bpb\.de/[^}]+})", webpage) + + formats = [] + for video_info in video_info_dicts: + video_info = self._parse_json(video_info, video_id, transform_source=js_to_json) + quality = video_info['quality'] + video_url = video_info['src'] + formats.append({ + 'url': video_url, + 'preference': 10 if quality == 'high' else 0, + 'format_note': quality, + 'format_id': '%s-%s' % (quality, determine_ext(video_url)), + }) + + self._sort_formats(formats) return { 'id': video_id, - 'url': video_url, + 'formats': formats, 'title': title, 'description': self._og_search_description(webpage), } diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 33290fd74..199a04d1c 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -828,7 +828,7 @@ class InfoExtractor(object): for f in formats: # Automatically determine tbr when missing based on abr and vbr (improves # formats sorting in some cases) - if 'tbr' not in f and 'abr' in f and 'vbr' in f: + if 'tbr' not in f and f.get('abr') is not None and f.get('vbr') is not None: f['tbr'] = f['abr'] + f['vbr'] def _formats_key(f): @@ -1330,6 +1330,83 @@ class InfoExtractor(object): }) return entries + def _download_dash_manifest(self, dash_manifest_url, video_id, fatal=True): + return self._download_xml( + dash_manifest_url, video_id, + note='Downloading DASH manifest', + errnote='Could not download DASH manifest', + fatal=fatal) + + def _extract_dash_manifest_formats(self, dash_manifest_url, video_id, fatal=True, namespace=None, formats_dict={}): + dash_doc = self._download_dash_manifest(dash_manifest_url, video_id, fatal) + if dash_doc is False: + return [] + + return self._parse_dash_manifest( + dash_doc, namespace=namespace, formats_dict=formats_dict) + + def _parse_dash_manifest(self, dash_doc, namespace=None, formats_dict={}): + def _add_ns(path): + return self._xpath_ns(path, namespace) + + formats = [] + for a in dash_doc.findall('.//' + _add_ns('AdaptationSet')): + mime_type = a.attrib.get('mimeType') + for r in a.findall(_add_ns('Representation')): + mime_type = r.attrib.get('mimeType') or mime_type + url_el = r.find(_add_ns('BaseURL')) + if mime_type == 'text/vtt': + # TODO implement WebVTT downloading + pass + elif mime_type.startswith('audio/') or mime_type.startswith('video/'): + segment_list = r.find(_add_ns('SegmentList')) + format_id = r.attrib['id'] + video_url = url_el.text if url_el is not None else None + filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el is not None else None) + f = { + 'format_id': format_id, + 'url': video_url, + 'width': int_or_none(r.attrib.get('width')), + 'height': int_or_none(r.attrib.get('height')), + 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000), + 'asr': int_or_none(r.attrib.get('audioSamplingRate')), + 'filesize': filesize, + 'fps': int_or_none(r.attrib.get('frameRate')), + } + if segment_list is not None: + initialization_url = segment_list.find(_add_ns('Initialization')).attrib['sourceURL'] + f.update({ + 'initialization_url': initialization_url, + 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall(_add_ns('SegmentURL'))], + 'protocol': 'http_dash_segments', + }) + if not f.get('url'): + f['url'] = initialization_url + try: + existing_format = next( + fo for fo in formats + if fo['format_id'] == format_id) + except StopIteration: + full_info = formats_dict.get(format_id, {}).copy() + full_info.update(f) + codecs = r.attrib.get('codecs') + if codecs: + if mime_type.startswith('video/'): + vcodec, acodec = codecs, 'none' + else: # mime_type.startswith('audio/') + vcodec, acodec = 'none', codecs + + full_info.update({ + 'vcodec': vcodec, + 'acodec': acodec, + }) + formats.append(full_info) + else: + existing_format.update(f) + else: + self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type) + return formats + def _live_title(self, name): """ Generate the title for a live video """ now = datetime.datetime.now() diff --git a/youtube_dl/extractor/cspan.py b/youtube_dl/extractor/cspan.py index b78edf729..b8b9d058d 100644 --- a/youtube_dl/extractor/cspan.py +++ b/youtube_dl/extractor/cspan.py @@ -113,7 +113,7 @@ class CSpanIE(InfoExtractor): 'tbr': int_or_none(get_text_attr(quality, 'bitrate')), }) if not formats: - path = get_text_attr(f, 'path') + path = unescapeHTML(get_text_attr(f, 'path')) if not path: continue formats = self._extract_m3u8_formats( diff --git a/youtube_dl/extractor/daum.py b/youtube_dl/extractor/daum.py index f08f57157..c84c51058 100644 --- a/youtube_dl/extractor/daum.py +++ b/youtube_dl/extractor/daum.py @@ -2,17 +2,26 @@ from __future__ import unicode_literals +import re +import itertools + from .common import InfoExtractor -from ..compat import compat_urllib_parse +from ..compat import ( + compat_parse_qs, + compat_urllib_parse, + compat_urllib_parse_unquote, + compat_urlparse, +) from ..utils import ( int_or_none, str_to_int, xpath_text, + unescapeHTML, ) class DaumIE(InfoExtractor): - _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/v/(?P<id>[^?#&]+)' + _VALID_URL = r'https?://(?:(?:m\.)?tvpot\.daum\.net/v/|videofarm\.daum\.net/controller/player/VodPlayer\.swf\?vid=)(?P<id>[^?#&]+)' IE_NAME = 'daum.net' _TESTS = [{ @@ -23,25 +32,57 @@ class DaumIE(InfoExtractor): 'title': '마크 헌트 vs 안토니오 실바', 'description': 'Mark Hunt vs Antonio Silva', 'upload_date': '20131217', + 'thumbnail': 're:^https?://.*\.(?:jpg|png)', 'duration': 2117, 'view_count': int, 'comment_count': int, }, }, { + 'url': 'http://m.tvpot.daum.net/v/65139429', + 'info_dict': { + 'id': '65139429', + 'ext': 'mp4', + 'title': '1297회, \'아빠 아들로 태어나길 잘 했어\' 민수, 감동의 눈물[아빠 어디가] 20150118', + 'description': 'md5:79794514261164ff27e36a21ad229fc5', + 'upload_date': '20150604', + 'thumbnail': 're:^https?://.*\.(?:jpg|png)', + 'duration': 154, + 'view_count': int, + 'comment_count': int, + }, + }, { 'url': 'http://tvpot.daum.net/v/07dXWRka62Y%24', 'only_matching': True, + }, { + 'url': 'http://videofarm.daum.net/controller/player/VodPlayer.swf?vid=vwIpVpCQsT8%24&ref=', + 'info_dict': { + 'id': 'vwIpVpCQsT8$', + 'ext': 'flv', + 'title': '01-Korean War ( Trouble on the horizon )', + 'description': '\nKorean War 01\nTrouble on the horizon\n전쟁의 먹구름', + 'upload_date': '20080223', + 'thumbnail': 're:^https?://.*\.(?:jpg|png)', + 'duration': 249, + 'view_count': int, + 'comment_count': int, + }, }] def _real_extract(self, url): - video_id = self._match_id(url) + video_id = compat_urllib_parse_unquote(self._match_id(url)) query = compat_urllib_parse.urlencode({'vid': video_id}) - info = self._download_xml( - 'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id, - 'Downloading video info') movie_data = self._download_json( 'http://videofarm.daum.net/controller/api/closed/v1_2/IntegratedMovieData.json?' + query, video_id, 'Downloading video formats info') + # For urls like http://m.tvpot.daum.net/v/65139429, where the video_id is really a clipid + if not movie_data.get('output_list', {}).get('output_list') and re.match(r'^\d+$', video_id): + return self.url_result('http://tvpot.daum.net/clip/ClipView.do?clipid=%s' % video_id) + + info = self._download_xml( + 'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id, + 'Downloading video info') + formats = [] for format_el in movie_data['output_list']['output_list']: profile = format_el['profile'] @@ -76,8 +117,9 @@ class DaumIE(InfoExtractor): class DaumClipIE(InfoExtractor): - _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/(?:clip/ClipView.do|mypot/View.do)\?.*?clipid=(?P<id>\d+)' + _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/(?:clip/ClipView.(?:do|tv)|mypot/View.do)\?.*?clipid=(?P<id>\d+)' IE_NAME = 'daum.net:clip' + _URL_TEMPLATE = 'http://tvpot.daum.net/clip/ClipView.do?clipid=%s' _TESTS = [{ 'url': 'http://tvpot.daum.net/clip/ClipView.do?clipid=52554690', @@ -87,11 +129,19 @@ class DaumClipIE(InfoExtractor): 'title': 'DOTA 2GETHER 시즌2 6회 - 2부', 'description': 'DOTA 2GETHER 시즌2 6회 - 2부', 'upload_date': '20130831', + 'thumbnail': 're:^https?://.*\.(?:jpg|png)', 'duration': 3868, 'view_count': int, }, + }, { + 'url': 'http://m.tvpot.daum.net/clip/ClipView.tv?clipid=54999425', + 'only_matching': True, }] + @classmethod + def suitable(cls, url): + return False if DaumPlaylistIE.suitable(url) or DaumUserIE.suitable(url) else super(DaumClipIE, cls).suitable(url) + def _real_extract(self, url): video_id = self._match_id(url) clip_info = self._download_json( @@ -102,7 +152,7 @@ class DaumClipIE(InfoExtractor): '_type': 'url_transparent', 'id': video_id, 'url': 'http://tvpot.daum.net/v/%s' % clip_info['vid'], - 'title': clip_info['title'], + 'title': unescapeHTML(clip_info['title']), 'thumbnail': clip_info.get('thumb_url'), 'description': clip_info.get('contents'), 'duration': int_or_none(clip_info.get('duration')), @@ -110,3 +160,139 @@ class DaumClipIE(InfoExtractor): 'view_count': int_or_none(clip_info.get('play_count')), 'ie_key': 'Daum', } + + +class DaumListIE(InfoExtractor): + def _get_entries(self, list_id, list_id_type): + name = None + entries = [] + for pagenum in itertools.count(1): + list_info = self._download_json( + 'http://tvpot.daum.net/mypot/json/GetClipInfo.do?size=48&init=true&order=date&page=%d&%s=%s' % ( + pagenum, list_id_type, list_id), list_id, 'Downloading list info - %s' % pagenum) + + entries.extend([ + self.url_result( + 'http://tvpot.daum.net/v/%s' % clip['vid']) + for clip in list_info['clip_list'] + ]) + + if not name: + name = list_info.get('playlist_bean', {}).get('name') or \ + list_info.get('potInfo', {}).get('name') + + if not list_info.get('has_more'): + break + + return name, entries + + def _check_clip(self, url, list_id): + query_dict = compat_parse_qs(compat_urlparse.urlparse(url).query) + if 'clipid' in query_dict: + clip_id = query_dict['clipid'][0] + if self._downloader.params.get('noplaylist'): + self.to_screen('Downloading just video %s because of --no-playlist' % clip_id) + return self.url_result(DaumClipIE._URL_TEMPLATE % clip_id, 'DaumClip') + else: + self.to_screen('Downloading playlist %s - add --no-playlist to just download video' % list_id) + + +class DaumPlaylistIE(DaumListIE): + _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/mypot/(?:View\.do|Top\.tv)\?.*?playlistid=(?P<id>[0-9]+)' + IE_NAME = 'daum.net:playlist' + _URL_TEMPLATE = 'http://tvpot.daum.net/mypot/View.do?playlistid=%s' + + _TESTS = [{ + 'note': 'Playlist url with clipid', + 'url': 'http://tvpot.daum.net/mypot/View.do?playlistid=6213966&clipid=73806844', + 'info_dict': { + 'id': '6213966', + 'title': 'Woorissica Official', + }, + 'playlist_mincount': 181 + }, { + 'note': 'Playlist url with clipid - noplaylist', + 'url': 'http://tvpot.daum.net/mypot/View.do?playlistid=6213966&clipid=73806844', + 'info_dict': { + 'id': '73806844', + 'ext': 'mp4', + 'title': '151017 Airport', + 'upload_date': '20160117', + }, + 'params': { + 'noplaylist': True, + 'skip_download': True, + } + }] + + @classmethod + def suitable(cls, url): + return False if DaumUserIE.suitable(url) else super(DaumPlaylistIE, cls).suitable(url) + + def _real_extract(self, url): + list_id = self._match_id(url) + + clip_result = self._check_clip(url, list_id) + if clip_result: + return clip_result + + name, entries = self._get_entries(list_id, 'playlistid') + + return self.playlist_result(entries, list_id, name) + + +class DaumUserIE(DaumListIE): + _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/mypot/(?:View|Top)\.(?:do|tv)\?.*?ownerid=(?P<id>[0-9a-zA-Z]+)' + IE_NAME = 'daum.net:user' + + _TESTS = [{ + 'url': 'http://tvpot.daum.net/mypot/View.do?ownerid=o2scDLIVbHc0', + 'info_dict': { + 'id': 'o2scDLIVbHc0', + 'title': '마이 리틀 텔레비전', + }, + 'playlist_mincount': 213 + }, { + 'url': 'http://tvpot.daum.net/mypot/View.do?ownerid=o2scDLIVbHc0&clipid=73801156', + 'info_dict': { + 'id': '73801156', + 'ext': 'mp4', + 'title': '[미공개] 김구라, 오만석이 부릅니다 \'오케피\' - 마이 리틀 텔레비전 20160116', + 'upload_date': '20160117', + 'description': 'md5:5e91d2d6747f53575badd24bd62b9f36' + }, + 'params': { + 'noplaylist': True, + 'skip_download': True, + } + }, { + 'note': 'Playlist url has ownerid and playlistid, playlistid takes precedence', + 'url': 'http://tvpot.daum.net/mypot/View.do?ownerid=o2scDLIVbHc0&playlistid=6196631', + 'info_dict': { + 'id': '6196631', + 'title': '마이 리틀 텔레비전 - 20160109', + }, + 'playlist_count': 11 + }, { + 'url': 'http://tvpot.daum.net/mypot/Top.do?ownerid=o2scDLIVbHc0', + 'only_matching': True, + }, { + 'url': 'http://m.tvpot.daum.net/mypot/Top.tv?ownerid=45x1okb1If50&playlistid=3569733', + 'only_matching': True, + }] + + def _real_extract(self, url): + list_id = self._match_id(url) + + clip_result = self._check_clip(url, list_id) + if clip_result: + return clip_result + + query_dict = compat_parse_qs(compat_urlparse.urlparse(url).query) + if 'playlistid' in query_dict: + playlist_id = query_dict['playlistid'][0] + return self.url_result(DaumPlaylistIE._URL_TEMPLATE % playlist_id, 'DaumPlaylist') + + name, entries = self._get_entries(list_id, 'ownerid') + + return self.playlist_result(entries, list_id, name) diff --git a/youtube_dl/extractor/espn.py b/youtube_dl/extractor/espn.py index 3762d8748..db4b263bc 100644 --- a/youtube_dl/extractor/espn.py +++ b/youtube_dl/extractor/espn.py @@ -53,8 +53,8 @@ class ESPNIE(InfoExtractor): webpage = self._download_webpage(url, video_id) video_id = self._search_regex( - r'class="video-play-button"[^>]+data-id="(\d+)', - webpage, 'video id') + r'class=(["\']).*?video-play-button.*?\1[^>]+data-id=["\'](?P<id>\d+)', + webpage, 'video id', group='id') cms = 'espn' if 'data-source="intl"' in webpage: diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py index cb5dd57fb..b6d1180f0 100644 --- a/youtube_dl/extractor/facebook.py +++ b/youtube_dl/extractor/facebook.py @@ -6,9 +6,11 @@ import socket from .common import InfoExtractor from ..compat import ( + compat_etree_fromstring, compat_http_client, compat_urllib_error, compat_urllib_parse_unquote, + compat_urllib_parse_unquote_plus, ) from ..utils import ( error_to_compat_str, @@ -44,6 +46,9 @@ class FacebookIE(InfoExtractor): _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1' _NETRC_MACHINE = 'facebook' IE_NAME = 'facebook' + + _CHROME_USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/48.0.2564.97 Safari/537.36' + _TESTS = [{ 'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf', 'md5': '6a40d33c0eccbb1af76cf0485a052659', @@ -66,6 +71,16 @@ class FacebookIE(InfoExtractor): 'title' ] }, { + 'note': 'Video with DASH manifest', + 'url': 'https://www.facebook.com/video.php?v=957955867617029', + 'md5': '54706e4db4f5ad58fbad82dde1f1213f', + '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', + }, + }, { 'url': 'https://www.facebook.com/video.php?v=10204634152394104', 'only_matching': True, }, { @@ -147,13 +162,36 @@ class FacebookIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - url = 'https://www.facebook.com/video/video.php?v=%s' % video_id - webpage = self._download_webpage(url, video_id) + req = sanitized_Request('https://www.facebook.com/video/video.php?v=%s' % video_id) + req.add_header('User-Agent', self._CHROME_USER_AGENT) + webpage = self._download_webpage(req, video_id) + + video_data = None BEFORE = '{swf.addParam(param[0], param[1]);});\n' AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});' m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage) - if not m: + if m: + data = dict(json.loads(m.group(1))) + params_raw = compat_urllib_parse_unquote(data['params']) + video_data = json.loads(params_raw)['video_data'] + + def video_data_list2dict(video_data): + ret = {} + for item in video_data: + format_id = item['stream_type'] + ret.setdefault(format_id, []).append(item) + return ret + + if not video_data: + server_js_data = self._parse_json(self._search_regex( + r'handleServerJS\(({.+})\);', webpage, 'server js data'), video_id) + for item in server_js_data['instances']: + if item[1][0] == 'VideoConfig': + video_data = video_data_list2dict(item[2][0]['videoData']) + break + + if not video_data: m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*"><div>(.*?)</div>', webpage) if m_msg is not None: raise ExtractorError( @@ -161,12 +199,9 @@ class FacebookIE(InfoExtractor): expected=True) else: raise ExtractorError('Cannot parse data') - data = dict(json.loads(m.group(1))) - params_raw = compat_urllib_parse_unquote(data['params']) - params = json.loads(params_raw) formats = [] - for format_id, f in params['video_data'].items(): + for format_id, f in video_data.items(): if not f or not isinstance(f, list): continue for quality in ('sd', 'hd'): @@ -178,9 +213,16 @@ class FacebookIE(InfoExtractor): 'url': src, 'preference': -10 if format_id == 'progressive' else 0, }) + dash_manifest = f[0].get('dash_manifest') + if dash_manifest: + formats.extend(self._parse_dash_manifest( + compat_etree_fromstring(compat_urllib_parse_unquote_plus(dash_manifest)), + namespace='urn:mpeg:dash:schema:mpd:2011')) if not formats: raise ExtractorError('Cannot find video formats') + self._sort_formats(formats) + video_title = self._html_search_regex( r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage, 'title', default=None) diff --git a/youtube_dl/extractor/gamekings.py b/youtube_dl/extractor/gamekings.py index 027f55eb2..f6b9046f9 100644 --- a/youtube_dl/extractor/gamekings.py +++ b/youtube_dl/extractor/gamekings.py @@ -6,24 +6,29 @@ from ..utils import ( xpath_text, xpath_with_ns, ) +from .youtube import YoutubeIE class GamekingsIE(InfoExtractor): - _VALID_URL = r'http://www\.gamekings\.tv/(?:videos|nieuws)/(?P<id>[^/]+)' + _VALID_URL = r'http://www\.gamekings\.nl/(?:videos|nieuws)/(?P<id>[^/]+)' _TESTS = [{ - 'url': 'http://www.gamekings.tv/videos/phoenix-wright-ace-attorney-dual-destinies-review/', - # MD5 is flaky, seems to change regularly - # 'md5': '2f32b1f7b80fdc5cb616efb4f387f8a3', + # YouTube embed video + 'url': 'http://www.gamekings.nl/videos/phoenix-wright-ace-attorney-dual-destinies-review/', + 'md5': '5208d3a17adeaef829a7861887cb9029', 'info_dict': { - 'id': 'phoenix-wright-ace-attorney-dual-destinies-review', + 'id': 'HkSQKetlGOU', 'ext': 'mp4', - 'title': 'Phoenix Wright: Ace Attorney \u2013 Dual Destinies Review', - 'description': 'md5:36fd701e57e8c15ac8682a2374c99731', + 'title': 'Phoenix Wright: Ace Attorney - Dual Destinies Review', + 'description': 'md5:db88c0e7f47e9ea50df3271b9dc72e1d', 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader_id': 'UCJugRGo4STYMeFr5RoOShtQ', + 'uploader': 'Gamekings Vault', + 'upload_date': '20151123', }, + 'add_ie': ['Youtube'], }, { # vimeo video - 'url': 'http://www.gamekings.tv/videos/the-legend-of-zelda-majoras-mask/', + 'url': 'http://www.gamekings.nl/videos/the-legend-of-zelda-majoras-mask/', 'md5': '12bf04dfd238e70058046937657ea68d', 'info_dict': { 'id': 'the-legend-of-zelda-majoras-mask', @@ -33,7 +38,7 @@ class GamekingsIE(InfoExtractor): 'thumbnail': 're:^https?://.*\.jpg$', }, }, { - 'url': 'http://www.gamekings.tv/nieuws/gamekings-extra-shelly-en-david-bereiden-zich-voor-op-de-livestream/', + 'url': 'http://www.gamekings.nl/nieuws/gamekings-extra-shelly-en-david-bereiden-zich-voor-op-de-livestream/', 'only_matching': True, }] @@ -43,7 +48,11 @@ class GamekingsIE(InfoExtractor): webpage = self._download_webpage(url, video_id) playlist_id = self._search_regex( - r'gogoVideo\(\s*\d+\s*,\s*"([^"]+)', webpage, 'playlist id') + r'gogoVideo\([^,]+,\s*"([^"]+)', webpage, 'playlist id') + + # Check if a YouTube embed is used + if YoutubeIE.suitable(playlist_id): + return self.url_result(playlist_id, ie='Youtube') playlist = self._download_xml( 'http://www.gamekings.tv/wp-content/themes/gk2010/rss_playlist.php?id=%s' % playlist_id, diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 26d3698c8..b18e734c4 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -1819,6 +1819,17 @@ class GenericIE(InfoExtractor): if digiteka_url: return self.url_result(self._proto_relative_url(digiteka_url), DigitekaIE.ie_key()) + # Look for Limelight embeds + mobj = re.search(r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})', webpage) + if mobj: + lm = { + 'Media': 'media', + 'Channel': 'channel', + 'ChannelList': 'channel_list', + } + return self.url_result('limelight:%s:%s' % ( + lm[mobj.group(1)], mobj.group(2)), 'Limelight%s' % mobj.group(1), mobj.group(2)) + # Look for AdobeTVVideo embeds mobj = re.search( r'<iframe[^>]+src=[\'"]((?:https?:)?//video\.tv\.adobe\.com/v/\d+[^"]+)[\'"]', diff --git a/youtube_dl/extractor/kickstarter.py b/youtube_dl/extractor/kickstarter.py index 1d391e69f..9f1ade2e4 100644 --- a/youtube_dl/extractor/kickstarter.py +++ b/youtube_dl/extractor/kickstarter.py @@ -2,12 +2,13 @@ from __future__ import unicode_literals from .common import InfoExtractor +from ..utils import smuggle_url class KickStarterIE(InfoExtractor): _VALID_URL = r'https?://www\.kickstarter\.com/projects/(?P<id>[^/]*)/.*' _TESTS = [{ - 'url': 'https://www.kickstarter.com/projects/1404461844/intersection-the-story-of-josh-grant?ref=home_location', + 'url': 'https://www.kickstarter.com/projects/1404461844/intersection-the-story-of-josh-grant/description', 'md5': 'c81addca81327ffa66c642b5d8b08cab', 'info_dict': { 'id': '1404461844', @@ -27,7 +28,8 @@ class KickStarterIE(InfoExtractor): 'uploader_id': 'pebble', 'uploader': 'Pebble Technology', 'title': 'Pebble iOS Notifications', - } + }, + 'add_ie': ['Vimeo'], }, { 'url': 'https://www.kickstarter.com/projects/1420158244/power-drive-2000/widget/video.html', 'info_dict': { @@ -43,7 +45,7 @@ class KickStarterIE(InfoExtractor): webpage = self._download_webpage(url, video_id) title = self._html_search_regex( - r'<title>\s*(.*?)(?:\s*— Kickstarter)?\s*</title>', + r'<title>\s*(.*?)(?:\s*—\s*Kickstarter)?\s*</title>', webpage, 'title') video_url = self._search_regex( r'data-video-url="(.*?)"', @@ -52,7 +54,7 @@ class KickStarterIE(InfoExtractor): return { '_type': 'url_transparent', 'ie_key': 'Generic', - 'url': url, + 'url': smuggle_url(url, {'to_generic': True}), 'title': title, } diff --git a/youtube_dl/extractor/letv.py b/youtube_dl/extractor/letv.py index 08bdae8a2..9665ece89 100644 --- a/youtube_dl/extractor/letv.py +++ b/youtube_dl/extractor/letv.py @@ -5,11 +5,13 @@ import datetime import re import time import base64 +import hashlib from .common import InfoExtractor from ..compat import ( compat_urllib_parse, compat_ord, + compat_str, ) from ..utils import ( determine_ext, @@ -258,6 +260,7 @@ class LetvCloudIE(InfoExtractor): }, }, { 'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=ec93197892&pu=2c7cd40209&auto_play=1&gpcflag=1&width=640&height=360', + 'md5': 'e03d9cc8d9c13191e1caf277e42dbd31', 'info_dict': { 'id': 'p7jnfw5hw9_ec93197892', 'ext': 'mp4', @@ -265,6 +268,7 @@ class LetvCloudIE(InfoExtractor): }, }, { 'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=187060b6fd', + 'md5': 'cb988699a776b22d4a41b9d43acfb3ac', 'info_dict': { 'id': 'p7jnfw5hw9_187060b6fd', 'ext': 'mp4', @@ -272,21 +276,37 @@ class LetvCloudIE(InfoExtractor): }, }] - def _real_extract(self, url): - uu_mobj = re.search('uu=([\w]+)', url) - vu_mobj = re.search('vu=([\w]+)', url) - - if not uu_mobj or not vu_mobj: - raise ExtractorError('Invalid URL: %s' % url, expected=True) - - uu = uu_mobj.group(1) - vu = vu_mobj.group(1) - media_id = uu + '_' + vu - - play_json_req = sanitized_Request( - 'http://api.letvcloud.com/gpc.php?cf=html5&sign=signxxxxx&ver=2.2&format=json&' + - 'uu=' + uu + '&vu=' + vu) - play_json = self._download_json(play_json_req, media_id, 'Downloading playJson data') + @staticmethod + def sign_data(obj): + if obj['cf'] == 'flash': + salt = '2f9d6924b33a165a6d8b5d3d42f4f987' + items = ['cf', 'format', 'ran', 'uu', 'ver', 'vu'] + elif obj['cf'] == 'html5': + salt = 'fbeh5player12c43eccf2bec3300344' + items = ['cf', 'ran', 'uu', 'bver', 'vu'] + input_data = ''.join([item + obj[item] for item in items]) + salt + obj['sign'] = hashlib.md5(input_data.encode('utf-8')).hexdigest() + + def _get_formats(self, cf, uu, vu, media_id): + def get_play_json(cf, timestamp): + data = { + 'cf': cf, + 'ver': '2.2', + 'bver': 'firefox44.0', + 'format': 'json', + 'uu': uu, + 'vu': vu, + 'ran': compat_str(timestamp), + } + self.sign_data(data) + return self._download_json( + 'http://api.letvcloud.com/gpc.php?' + compat_urllib_parse.urlencode(data), + media_id, 'Downloading playJson data for type %s' % cf) + + play_json = get_play_json(cf, time.time()) + # The server time may be different from local time + if play_json.get('code') == 10071: + play_json = get_play_json(cf, play_json['timestamp']) if not play_json.get('data'): if play_json.get('message'): @@ -312,6 +332,21 @@ class LetvCloudIE(InfoExtractor): 'width': int_or_none(play_url.get('vwidth')), 'height': int_or_none(play_url.get('vheight')), }) + + return formats + + def _real_extract(self, url): + uu_mobj = re.search('uu=([\w]+)', url) + vu_mobj = re.search('vu=([\w]+)', url) + + if not uu_mobj or not vu_mobj: + raise ExtractorError('Invalid URL: %s' % url, expected=True) + + uu = uu_mobj.group(1) + vu = vu_mobj.group(1) + media_id = uu + '_' + vu + + formats = self._get_formats('flash', uu, vu, media_id) + self._get_formats('html5', uu, vu, media_id) self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/limelight.py b/youtube_dl/extractor/limelight.py index fb03dd527..1a0625ac3 100644 --- a/youtube_dl/extractor/limelight.py +++ b/youtube_dl/extractor/limelight.py @@ -40,7 +40,8 @@ class LimelightBaseIE(InfoExtractor): if not stream_url: continue if '.f4m' in stream_url: - formats.extend(self._extract_f4m_formats(stream_url, video_id)) + formats.extend(self._extract_f4m_formats( + stream_url, video_id, fatal=False)) else: fmt = { 'url': stream_url, @@ -72,8 +73,8 @@ class LimelightBaseIE(InfoExtractor): format_id = mobile_url.get('targetMediaPlatform') if determine_ext(media_url) == 'm3u8': formats.extend(self._extract_m3u8_formats( - media_url, video_id, 'mp4', entry_protocol='m3u8_native', - preference=-1, m3u8_id=format_id)) + media_url, video_id, 'mp4', 'm3u8_native', + m3u8_id=format_id, fatal=False)) else: formats.append({ 'url': media_url, diff --git a/youtube_dl/extractor/matchtv.py b/youtube_dl/extractor/matchtv.py new file mode 100644 index 000000000..28e0dfe63 --- /dev/null +++ b/youtube_dl/extractor/matchtv.py @@ -0,0 +1,55 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import random + +from .common import InfoExtractor +from ..compat import compat_urllib_parse +from ..utils import ( + sanitized_Request, + xpath_text, +) + + +class MatchTVIE(InfoExtractor): + _VALID_URL = r'https?://matchtv\.ru/?#live-player' + _TEST = { + 'url': 'http://matchtv.ru/#live-player', + 'info_dict': { + 'id': 'matchtv-live', + 'ext': 'flv', + 'title': 're:^Матч ТВ - Прямой эфир \d{4}-\d{2}-\d{2} \d{2}:\d{2}$', + 'is_live': True, + }, + 'params': { + 'skip_download': True, + }, + } + + def _real_extract(self, url): + video_id = 'matchtv-live' + request = sanitized_Request( + 'http://player.matchtv.ntvplus.tv/player/smil?%s' % compat_urllib_parse.urlencode({ + 'ts': '', + 'quality': 'SD', + 'contentId': '561d2c0df7159b37178b4567', + 'sign': '', + 'includeHighlights': '0', + 'userId': '', + 'sessionId': random.randint(1, 1000000000), + 'contentType': 'channel', + 'timeShift': '0', + 'platform': 'portal', + }), + headers={ + 'Referer': 'http://player.matchtv.ntvplus.tv/embed-player/NTVEmbedPlayer.swf', + }) + video_url = self._download_json(request, video_id)['data']['videoUrl'] + f4m_url = xpath_text(self._download_xml(video_url, video_id), './to') + formats = self._extract_f4m_formats(f4m_url, video_id) + return { + 'id': video_id, + 'title': self._live_title('Матч ТВ - Прямой эфир'), + 'is_live': True, + 'formats': formats, + } diff --git a/youtube_dl/extractor/nba.py b/youtube_dl/extractor/nba.py index 9d26030d3..a071378b6 100644 --- a/youtube_dl/extractor/nba.py +++ b/youtube_dl/extractor/nba.py @@ -18,13 +18,17 @@ class NBAIE(InfoExtractor): 'md5': '9e7729d3010a9c71506fd1248f74e4f4', 'info_dict': { 'id': '0021200253-okc-bkn-recap', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Thunder vs. Nets', 'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.', 'duration': 181, 'timestamp': 1354638466, 'upload_date': '20121204', }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, { 'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/', 'only_matching': True, @@ -68,7 +72,7 @@ class NBAIE(InfoExtractor): if video_url.startswith('/'): continue if video_url.endswith('.m3u8'): - formats.extend(self._extract_m3u8_formats(video_url, video_id, m3u8_id='hls', fatal=False)) + formats.extend(self._extract_m3u8_formats(video_url, video_id, ext='mp4', m3u8_id='hls', fatal=False)) elif video_url.endswith('.f4m'): formats.extend(self._extract_f4m_formats(video_url + '?hdcore=3.4.1.1', video_id, f4m_id='hds', fatal=False)) else: diff --git a/youtube_dl/extractor/nbc.py b/youtube_dl/extractor/nbc.py index 1dd54c2f1..18d01f423 100644 --- a/youtube_dl/extractor/nbc.py +++ b/youtube_dl/extractor/nbc.py @@ -19,32 +19,39 @@ class NBCIE(InfoExtractor): _TESTS = [ { 'url': 'http://www.nbc.com/the-tonight-show/segments/112966', - # md5 checksum is not stable 'info_dict': { - 'id': 'c9xnCo0YPOPH', - 'ext': 'flv', + 'id': '112966', + 'ext': 'mp4', 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s', 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.', }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, { 'url': 'http://www.nbc.com/the-tonight-show/episodes/176', 'info_dict': { - 'id': 'XwU9KZkp98TH', + 'id': '176', 'ext': 'flv', 'title': 'Ricky Gervais, Steven Van Zandt, ILoveMakonnen', 'description': 'A brand new episode of The Tonight Show welcomes Ricky Gervais, Steven Van Zandt and ILoveMakonnen.', }, - 'skip': 'Only works from US', + 'skip': '404 Not Found', }, { 'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821', 'info_dict': { - 'id': '8iUuyzWDdYUZ', - 'ext': 'flv', + 'id': '2832821', + 'ext': 'mp4', 'title': 'Star Wars Teaser', 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442', }, + 'params': { + # m3u8 download + 'skip_download': True, + }, 'skip': 'Only works from US', }, { @@ -66,7 +73,11 @@ class NBCIE(InfoExtractor): webpage, 'theplatform url').replace('_no_endcard', '').replace('\\/', '/'))) if theplatform_url.startswith('//'): theplatform_url = 'http:' + theplatform_url - return self.url_result(smuggle_url(theplatform_url, {'source_url': url})) + return { + '_type': 'url_transparent', + 'url': smuggle_url(theplatform_url, {'source_url': url}), + 'id': video_id, + } class NBCSportsVPlayerIE(InfoExtractor): diff --git a/youtube_dl/extractor/npo.py b/youtube_dl/extractor/npo.py index eb12fb810..87f5675c7 100644 --- a/youtube_dl/extractor/npo.py +++ b/youtube_dl/extractor/npo.py @@ -189,7 +189,7 @@ class NPOIE(NPOBaseIE): if not video_url: continue if format_id == 'adaptive': - formats.extend(self._extract_m3u8_formats(video_url, video_id)) + formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4')) else: formats.append({ 'url': video_url, @@ -406,6 +406,38 @@ class NPORadioFragmentIE(InfoExtractor): } +class SchoolTVIE(InfoExtractor): + IE_NAME = 'schooltv' + _VALID_URL = r'https?://(?:www\.)?schooltv\.nl/video/(?P<id>[^/?#&]+)' + + _TEST = { + 'url': 'http://www.schooltv.nl/video/ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam/', + 'info_dict': { + 'id': 'WO_NTR_429477', + 'display_id': 'ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam', + 'title': 'Ademhaling: De hele dag haal je adem. Maar wat gebeurt er dan eigenlijk in je lichaam?', + 'ext': 'mp4', + 'description': 'md5:abfa0ff690adb73fd0297fd033aaa631' + }, + 'params': { + # Skip because of m3u8 download + 'skip_download': True + } + } + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + video_id = self._search_regex( + r'data-mid=(["\'])(?P<id>.+?)\1', webpage, 'video_id', group='id') + return { + '_type': 'url_transparent', + 'ie_key': 'NPO', + 'url': 'npo:%s' % video_id, + 'display_id': display_id + } + + class VPROIE(NPOIE): IE_NAME = 'vpro' _VALID_URL = r'https?://(?:www\.)?(?:tegenlicht\.)?vpro\.nl/(?:[^/]+/){2,}(?P<id>[^/]+)\.html' diff --git a/youtube_dl/extractor/nrk.py b/youtube_dl/extractor/nrk.py index 6ff13050d..a126f5054 100644 --- a/youtube_dl/extractor/nrk.py +++ b/youtube_dl/extractor/nrk.py @@ -133,26 +133,32 @@ class NRKTVIE(InfoExtractor): _TESTS = [ { 'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014', - 'md5': 'adf2c5454fa2bf032f47a9f8fb351342', 'info_dict': { 'id': 'MUHH48000314', - 'ext': 'flv', + 'ext': 'mp4', 'title': '20 spørsmål', 'description': 'md5:bdea103bc35494c143c6a9acdd84887a', 'upload_date': '20140523', 'duration': 1741.52, }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, { 'url': 'https://tv.nrk.no/program/mdfp15000514', - 'md5': '383650ece2b25ecec996ad7b5bb2a384', 'info_dict': { 'id': 'mdfp15000514', - 'ext': 'flv', - 'title': 'Kunnskapskanalen: Grunnlovsjubiléet - Stor ståhei for ingenting', + 'ext': 'mp4', + 'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting', 'description': 'md5:654c12511f035aed1e42bdf5db3b206a', 'upload_date': '20140524', - 'duration': 4605.0, + 'duration': 4605.08, + }, + 'params': { + # m3u8 download + 'skip_download': True, }, }, { diff --git a/youtube_dl/extractor/screenwavemedia.py b/youtube_dl/extractor/screenwavemedia.py index 05f93904c..e5d62a139 100644 --- a/youtube_dl/extractor/screenwavemedia.py +++ b/youtube_dl/extractor/screenwavemedia.py @@ -71,7 +71,7 @@ class ScreenwaveMediaIE(InfoExtractor): formats = [] for source in sources: if source['type'] == 'hls': - formats.extend(self._extract_m3u8_formats(source['file'], video_id)) + formats.extend(self._extract_m3u8_formats(source['file'], video_id, ext='mp4')) else: file_ = source.get('file') if not file_: @@ -107,7 +107,11 @@ class TeamFourIE(InfoExtractor): 'upload_date': '20130401', 'description': 'Check out this and more on our website: http://teamfourstar.com\nTFS Store: http://sharkrobot.com/team-four-star\nFollow on Twitter: http://twitter.com/teamfourstar\nLike on FB: http://facebook.com/teamfourstar', 'title': 'A Moment With TFS Episode 4', - } + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, } def _real_extract(self, url): diff --git a/youtube_dl/extractor/senateisvp.py b/youtube_dl/extractor/senateisvp.py index 474ebb49b..990ea0fa8 100644 --- a/youtube_dl/extractor/senateisvp.py +++ b/youtube_dl/extractor/senateisvp.py @@ -53,17 +53,25 @@ class SenateISVPIE(InfoExtractor): 'url': 'http://www.senate.gov/isvp/?comm=judiciary&type=live&stt=&filename=judiciary031715&auto_play=false&wmode=transparent&poster=http%3A%2F%2Fwww.judiciary.senate.gov%2Fthemes%2Fjudiciary%2Fimages%2Fvideo-poster-flash-fit.png', 'info_dict': { 'id': 'judiciary031715', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Integrated Senate Video Player', 'thumbnail': 're:^https?://.*\.(?:jpg|png)$', - } + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, { 'url': 'http://www.senate.gov/isvp/?type=live&comm=commerce&filename=commerce011514.mp4&auto_play=false', 'info_dict': { 'id': 'commerce011514', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Integrated Senate Video Player' - } + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, { 'url': 'http://www.senate.gov/isvp/?type=arch&comm=intel&filename=intel090613&hc_location=ufi', # checksum differs each time diff --git a/youtube_dl/extractor/tv2.py b/youtube_dl/extractor/tv2.py index fa338b936..1457e524e 100644 --- a/youtube_dl/extractor/tv2.py +++ b/youtube_dl/extractor/tv2.py @@ -17,18 +17,21 @@ class TV2IE(InfoExtractor): _VALID_URL = 'http://(?:www\.)?tv2\.no/v/(?P<id>\d+)' _TEST = { 'url': 'http://www.tv2.no/v/916509/', - 'md5': '9cb9e3410b18b515d71892f27856e9b1', 'info_dict': { 'id': '916509', - 'ext': 'flv', - 'title': 'Se Gryttens hyllest av Steven Gerrard', + 'ext': 'mp4', + 'title': 'Se Frode Gryttens hyllest av Steven Gerrard', 'description': 'TV 2 Sportens huspoet tar avskjed med Liverpools kaptein Steven Gerrard.', 'timestamp': 1431715610, 'upload_date': '20150515', 'duration': 156.967, 'view_count': int, 'categories': list, - } + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, } def _real_extract(self, url): diff --git a/youtube_dl/extractor/vgtv.py b/youtube_dl/extractor/vgtv.py index 86ba70ed9..14e945d49 100644 --- a/youtube_dl/extractor/vgtv.py +++ b/youtube_dl/extractor/vgtv.py @@ -86,10 +86,9 @@ class VGTVIE(XstreamIE): { # streamType: wasLive 'url': 'http://www.vgtv.no/#!/live/113063/direkte-v75-fra-solvalla', - 'md5': '458f4841239dab414343b50e5af8869c', 'info_dict': { 'id': '113063', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'V75 fra Solvalla 30.05.15', 'description': 'md5:b3743425765355855f88e096acc93231', 'thumbnail': 're:^https?://.*\.jpg', @@ -98,6 +97,10 @@ class VGTVIE(XstreamIE): 'upload_date': '20150530', 'view_count': int, }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, { 'url': 'http://www.aftenposten.no/webtv/#!/video/21039/trailer-sweatshop-i-can-t-take-any-more', diff --git a/youtube_dl/extractor/vidzi.py b/youtube_dl/extractor/vidzi.py index 2ba9f31df..7c6e98026 100644 --- a/youtube_dl/extractor/vidzi.py +++ b/youtube_dl/extractor/vidzi.py @@ -2,6 +2,7 @@ from __future__ import unicode_literals from .common import InfoExtractor +from ..utils import smuggle_url class VidziIE(InfoExtractor): @@ -13,6 +14,11 @@ class VidziIE(InfoExtractor): 'id': 'cghql9yq6emu', 'ext': 'mp4', 'title': 'youtube-dl test video 1\\\\2\'3/4<5\\\\6ä7↭', + 'uploader': 'vidzi.tv', + }, + 'params': { + # m3u8 download + 'skip_download': True, }, } @@ -20,19 +26,14 @@ class VidziIE(InfoExtractor): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - video_host = self._html_search_regex( - r'id=\'vplayer\'><img src="http://(.*?)/i', webpage, - 'video host') - video_hash = self._html_search_regex( - r'\|([a-z0-9]+)\|hls\|type', webpage, 'video_hash') - ext = self._html_search_regex( - r'\|tracks\|([a-z0-9]+)\|', webpage, 'video ext') - video_url = 'http://' + video_host + '/' + video_hash + '/v.' + ext title = self._html_search_regex( r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title') + # Vidzi now uses jwplayer, which can be handled by GenericIE return { + '_type': 'url_transparent', 'id': video_id, 'title': title, - 'url': video_url, + 'url': smuggle_url(url, {'to_generic': True}), + 'ie_key': 'Generic', } diff --git a/youtube_dl/extractor/viidea.py b/youtube_dl/extractor/viidea.py index 525e303d4..315984bf9 100644 --- a/youtube_dl/extractor/viidea.py +++ b/youtube_dl/extractor/viidea.py @@ -45,6 +45,10 @@ class ViideaIE(InfoExtractor): 'upload_date': '20130627', 'duration': 565, }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, { # video with invalid direct format links (HTTP 403) 'url': 'http://videolectures.net/russir2010_filippova_nlp/', diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 90557fa61..11014865a 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -321,7 +321,7 @@ class VKIE(InfoExtractor): class VKUserVideosIE(InfoExtractor): IE_NAME = 'vk:uservideos' IE_DESC = "VK - User's Videos" - _VALID_URL = r'https?://vk\.com/videos(?P<id>-?[0-9]+)$' + _VALID_URL = r'https?://vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)' _TEMPLATE_URL = 'https://vk.com/videos' _TESTS = [{ 'url': 'http://vk.com/videos205387401', @@ -333,6 +333,9 @@ class VKUserVideosIE(InfoExtractor): }, { 'url': 'http://vk.com/videos-77521', 'only_matching': True, + }, { + 'url': 'http://vk.com/videos-97664626?section=all', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/xuite.py b/youtube_dl/extractor/xuite.py index 8bbac54e2..2466410fa 100644 --- a/youtube_dl/extractor/xuite.py +++ b/youtube_dl/extractor/xuite.py @@ -34,19 +34,20 @@ class XuiteIE(InfoExtractor): }, }, { # Video with only one format - 'url': 'http://vlog.xuite.net/play/TkRZNjhULTM0NDE2MjkuZmx2', - 'md5': 'c45737fc8ac5dc8ac2f92ecbcecf505e', + 'url': 'http://vlog.xuite.net/play/WUxxR2xCLTI1OTI1MDk5LmZsdg==', + 'md5': '21f7b39c009b5a4615b4463df6eb7a46', 'info_dict': { - 'id': '3441629', + 'id': '25925099', 'ext': 'mp4', - 'title': '孫燕姿 - 眼淚成詩', + 'title': 'BigBuckBunny_320x180', 'thumbnail': 're:^https?://.*\.jpg$', - 'duration': 217.399, - 'timestamp': 1299383640, - 'upload_date': '20110306', - 'uploader': 'Valen', - 'uploader_id': '10400126', - 'categories': ['影視娛樂'], + 'duration': 596.458, + 'timestamp': 1454242500, + 'upload_date': '20160131', + 'uploader': 'yan12125', + 'uploader_id': '12158353', + 'categories': ['個人短片'], + 'description': 'http://download.blender.org/peach/bigbuckbunny_movies/BigBuckBunny_320x180.mp4', }, }, { # Video with two formats diff --git a/youtube_dl/extractor/youporn.py b/youtube_dl/extractor/youporn.py index dd724085a..b29baafc4 100644 --- a/youtube_dl/extractor/youporn.py +++ b/youtube_dl/extractor/youporn.py @@ -114,15 +114,13 @@ class YouPornIE(InfoExtractor): formats.append(f) self._sort_formats(formats) - description = self._html_search_regex( - r'(?s)<div[^>]+class=["\']video-description["\'][^>]*>(.+?)</div>', - webpage, 'description', default=None) + description = self._og_search_description(webpage, default=None) thumbnail = self._search_regex( r'(?:imageurl\s*=|poster\s*:)\s*(["\'])(?P<thumbnail>.+?)\1', webpage, 'thumbnail', fatal=False, group='thumbnail') uploader = self._html_search_regex( - r'(?s)<div[^>]+class=["\']videoInfoBy["\'][^>]*>\s*By:\s*</div>(.+?)</(?:a|div)>', + r'(?s)<div[^>]+class=["\']videoInfoBy(?:\s+[^"\']+)?["\'][^>]*>\s*By:\s*</div>(.+?)</(?:a|div)>', webpage, 'uploader', fatal=False) upload_date = unified_strdate(self._html_search_regex( r'(?s)<div[^>]+class=["\']videoInfoTime["\'][^>]*>(.+?)</div>', diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 92b9f3ae4..a7f8c968e 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -181,7 +181,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor): return -class YoutubeEntryListBaseInfoExtractor(InfoExtractor): +class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor): # Extract entries from page with "Load more" button def _entries(self, page, playlist_id): more_widget_html = content_html = page @@ -233,7 +233,7 @@ class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): def _process_page(self, content): - for playlist_id in re.findall(r'href="/?playlist\?list=(.+?)"', content): + for playlist_id in orderedSet(re.findall(r'href="/?playlist\?list=([0-9A-Za-z-_]{10,})"', content)): yield self.url_result( 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist') @@ -316,55 +316,55 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10}, # DASH mp4 video - '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40}, - '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40}, - '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40}, - '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40}, - '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40}, - '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559) - '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40}, - '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40}, - '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'fps': 60, 'preference': -40}, - '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'fps': 60, 'preference': -40}, - '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40}, + '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559) + '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40}, + '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40}, + '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Dash mp4 audio - '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'}, - '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'}, - '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'}, + '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'}, + '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'}, + '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'}, # Dash webm - '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, - '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, - '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, - '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, - '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, - '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, - '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40}, - '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40}, - '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40}, - '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40}, - '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40}, - '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40}, - '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40}, - '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40}, - '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40}, + '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40}, + '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug) - '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40}, - '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'fps': 60, 'preference': -40}, - '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'fps': 60, 'preference': -40}, - '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'fps': 60, 'preference': -40}, - '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40}, - '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'fps': 60, 'preference': -40}, + '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40}, + '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40}, + '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40}, + '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40}, # Dash webm audio - '171': {'ext': 'webm', 'acodec': 'vorbis', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50}, - '172': {'ext': 'webm', 'acodec': 'vorbis', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50}, + '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50}, + '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50}, # Dash webm audio with opus inside - '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50}, - '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50}, - '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50}, + '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50}, + '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50}, + '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50}, # RTMP (unnamed) '_rtmp': {'protocol': 'rtmp'}, @@ -1035,73 +1035,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor): url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.') - def _parse_dash_manifest( - self, video_id, dash_manifest_url, player_url, age_gate, fatal=True): - def decrypt_sig(mobj): - s = mobj.group(1) - dec_s = self._decrypt_signature(s, video_id, player_url, age_gate) - return '/signature/%s' % dec_s - dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url) - dash_doc = self._download_xml( - dash_manifest_url, video_id, - note='Downloading DASH manifest', - errnote='Could not download DASH manifest', - fatal=fatal) - - if dash_doc is False: - return [] - - formats = [] - for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'): - mime_type = a.attrib.get('mimeType') - for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'): - url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL') - if url_el is None: - continue - if mime_type == 'text/vtt': - # TODO implement WebVTT downloading - pass - elif mime_type.startswith('audio/') or mime_type.startswith('video/'): - segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList') - format_id = r.attrib['id'] - video_url = url_el.text - filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength')) - f = { - 'format_id': format_id, - 'url': video_url, - 'width': int_or_none(r.attrib.get('width')), - 'height': int_or_none(r.attrib.get('height')), - 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000), - 'asr': int_or_none(r.attrib.get('audioSamplingRate')), - 'filesize': filesize, - 'fps': int_or_none(r.attrib.get('frameRate')), - } - if segment_list is not None: - f.update({ - 'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'], - 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')], - 'protocol': 'http_dash_segments', - }) - try: - existing_format = next( - fo for fo in formats - if fo['format_id'] == format_id) - except StopIteration: - full_info = self._formats.get(format_id, {}).copy() - full_info.update(f) - codecs = r.attrib.get('codecs') - if codecs: - if full_info.get('acodec') == 'none': - full_info['vcodec'] = codecs - elif full_info.get('vcodec') == 'none': - full_info['acodec'] = codecs - formats.append(full_info) - else: - existing_format.update(f) - else: - self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type) - return formats - def _real_extract(self, url): url, smuggled_data = unsmuggle_url(url, {}) @@ -1533,8 +1466,16 @@ class YoutubeIE(YoutubeBaseInfoExtractor): for dash_manifest_url in dash_mpds: dash_formats = {} try: - for df in self._parse_dash_manifest( - video_id, dash_manifest_url, player_url, age_gate, dash_mpd_fatal): + def decrypt_sig(mobj): + s = mobj.group(1) + dec_s = self._decrypt_signature(s, video_id, player_url, age_gate) + return '/signature/%s' % dec_s + + dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url) + + for df in self._extract_dash_manifest_formats( + dash_manifest_url, video_id, fatal=dash_mpd_fatal, + namespace='urn:mpeg:DASH:schema:MPD:2011', formats_dict=self._formats): # Do not overwrite DASH format found in some previous DASH manifest if df['format_id'] not in dash_formats: dash_formats[df['format_id']] = df @@ -1602,7 +1543,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): } -class YoutubePlaylistIE(YoutubeBaseInfoExtractor, YoutubePlaylistBaseInfoExtractor): +class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): IE_DESC = 'YouTube.com playlists' _VALID_URL = r"""(?x)(?: (?:https?://)? @@ -1846,7 +1787,7 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): class YoutubeUserIE(YoutubeChannelIE): IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)' - _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)' + _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)' _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos' IE_NAME = 'youtube:user' diff --git a/youtube_dl/options.py b/youtube_dl/options.py index 433245f00..2137dfb3f 100644 --- a/youtube_dl/options.py +++ b/youtube_dl/options.py @@ -416,6 +416,11 @@ def parseOpts(overrideArguments=None): dest='hls_prefer_native', action='store_true', help='Use the native HLS downloader instead of ffmpeg (experimental)') downloader.add_option( + '--hls-use-mpegts', + dest='hls_use_mpegts', action='store_true', + help='Use the mpegts container for HLS videos, allowing to play the ' + 'video while downloading (some players may not be able to play it)') + downloader.add_option( '--external-downloader', dest='external_downloader', metavar='COMMAND', help='Use the specified external downloader. ' diff --git a/youtube_dl/postprocessor/ffmpeg.py b/youtube_dl/postprocessor/ffmpeg.py index 48f86e21c..d3d9d4f1d 100644 --- a/youtube_dl/postprocessor/ffmpeg.py +++ b/youtube_dl/postprocessor/ffmpeg.py @@ -483,6 +483,7 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor): self._downloader.to_screen('[ffmpeg] There aren\'t any subtitles to convert') return [], info self._downloader.to_screen('[ffmpeg] Converting subtitles') + sub_filenames = [] for lang, sub in subs.items(): ext = sub['ext'] if ext == new_ext: @@ -490,6 +491,8 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor): '[ffmpeg] Subtitle file for %s is already in the requested' 'format' % new_ext) continue + old_file = subtitles_filename(filename, lang, ext) + sub_filenames.append(old_file) new_file = subtitles_filename(filename, lang, new_ext) if ext == 'dfxp' or ext == 'ttml': @@ -497,7 +500,7 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor): 'You have requested to convert dfxp (TTML) subtitles into another format, ' 'which results in style information loss') - dfxp_file = subtitles_filename(filename, lang, ext) + dfxp_file = old_file srt_file = subtitles_filename(filename, lang, 'srt') with io.open(dfxp_file, 'rt', encoding='utf-8') as f: @@ -515,9 +518,7 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor): if new_ext == 'srt': continue - self.run_ffmpeg( - subtitles_filename(filename, lang, ext), - new_file, ['-f', new_format]) + self.run_ffmpeg(old_file, new_file, ['-f', new_format]) with io.open(new_file, 'rt', encoding='utf-8') as f: subs[lang] = { @@ -525,4 +526,4 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor): 'data': f.read(), } - return [], info + return sub_filenames, info diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index c63b61598..18dbe28bb 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -2017,20 +2017,27 @@ def dfxp2srt(dfxp_data): 'ttaf1': 'http://www.w3.org/2006/10/ttaf1', }) - def parse_node(node): - str_or_empty = functools.partial(str_or_none, default='') + class TTMLPElementParser: + out = '' - out = str_or_empty(node.text) + def start(self, tag, attrib): + if tag in (_x('ttml:br'), _x('ttaf1:br'), 'br'): + self.out += '\n' - for child in node: - if child.tag in (_x('ttml:br'), _x('ttaf1:br'), 'br'): - out += '\n' + str_or_empty(child.tail) - elif child.tag in (_x('ttml:span'), _x('ttaf1:span'), 'span'): - out += str_or_empty(parse_node(child)) - else: - out += str_or_empty(xml.etree.ElementTree.tostring(child)) + def end(self, tag): + pass - return out + def data(self, data): + self.out += data + + def close(self): + return self.out.strip() + + def parse_node(node): + target = TTMLPElementParser() + parser = xml.etree.ElementTree.XMLParser(target=target) + parser.feed(xml.etree.ElementTree.tostring(node)) + return parser.close() dfxp = compat_etree_fromstring(dfxp_data.encode('utf-8')) out = [] diff --git a/youtube_dl/version.py b/youtube_dl/version.py index 4ac7f9e93..6da42c5a5 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,3 +1,3 @@ from __future__ import unicode_literals -__version__ = '2016.01.27' +__version__ = '2016.02.01' |