X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/03342304420e5daeb428ffdcc7bbd2bbfecfa61a..5a9d00a5190582368d533f12db7638e735d7f7c8:/youtube_dl/extractor/ard.py diff --git a/youtube_dl/extractor/ard.py b/youtube_dl/extractor/ard.py index 9fb8491..ef73d5a 100644 --- a/youtube_dl/extractor/ard.py +++ b/youtube_dl/extractor/ard.py @@ -5,22 +5,23 @@ import re from .common import InfoExtractor from .generic import GenericIE +from ..compat import compat_str from ..utils import ( determine_ext, ExtractorError, - get_element_by_attribute, qualities, int_or_none, parse_duration, unified_strdate, xpath_text, + update_url_query, ) from ..compat import compat_etree_fromstring class ARDMediathekIE(InfoExtractor): IE_NAME = 'ARD:mediathek' - _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?' + _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.(?:daserste|rbb-online)\.de)/(?:.*/)(?P[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?' _TESTS = [{ 'url': 'http://www.ardmediathek.de/tv/Dokumentation-und-Reportage/Ich-liebe-das-Leben-trotzdem/rbb-Fernsehen/Video?documentId=29582122&bcastId=3822114', @@ -35,6 +36,7 @@ class ARDMediathekIE(InfoExtractor): # m3u8 download 'skip_download': True, }, + 'skip': 'HTTP Error 404: Not Found', }, { 'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916', 'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e', @@ -45,6 +47,7 @@ class ARDMediathekIE(InfoExtractor): 'description': 'md5:196392e79876d0ac94c94e8cdb2875f1', 'duration': 5252, }, + 'skip': 'HTTP Error 404: Not Found', }, { # audio 'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086', @@ -56,9 +59,22 @@ class ARDMediathekIE(InfoExtractor): 'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef', 'duration': 3240, }, + 'skip': 'HTTP Error 404: Not Found', }, { 'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht', 'only_matching': True, + }, { + # audio + 'url': 'http://mediathek.rbb-online.de/radio/Hörspiel/Vor-dem-Fest/kulturradio/Audio?documentId=30796318&topRessort=radio&bcastId=9839158', + 'md5': '4e8f00631aac0395fee17368ac0e9867', + 'info_dict': { + 'id': '30796318', + 'ext': 'mp3', + 'title': 'Vor dem Fest', + 'description': 'md5:c0c1c8048514deaed2a73b3a60eecacb', + 'duration': 3287, + }, + 'skip': 'Video is no longer available', }] def _extract_media_info(self, media_info_url, webpage, video_id): @@ -78,12 +94,13 @@ class ARDMediathekIE(InfoExtractor): duration = int_or_none(media_info.get('_duration')) thumbnail = media_info.get('_previewImage') + is_live = media_info.get('_isLive') is True subtitles = {} subtitle_url = media_info.get('_subtitleUrl') if subtitle_url: subtitles['de'] = [{ - 'ext': 'srt', + 'ext': 'ttml', 'url': subtitle_url, }] @@ -91,6 +108,7 @@ class ARDMediathekIE(InfoExtractor): 'id': video_id, 'duration': duration, 'thumbnail': thumbnail, + 'is_live': is_live, 'formats': formats, 'subtitles': subtitles, } @@ -109,16 +127,21 @@ class ARDMediathekIE(InfoExtractor): quality = stream.get('_quality') server = stream.get('_server') for stream_url in stream_urls: + if not isinstance(stream_url, compat_str) or '//' not in stream_url: + continue ext = determine_ext(stream_url) if quality != 'auto' and ext in ('f4m', 'm3u8'): continue if ext == 'f4m': formats.extend(self._extract_f4m_formats( - stream_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', - video_id, preference=-1, f4m_id='hds', fatal=False)) + update_url_query(stream_url, { + 'hdcore': '3.1.1', + 'plugin': 'aasp-3.1.1.69.124' + }), + video_id, f4m_id='hds', fatal=False)) elif ext == 'm3u8': formats.extend(self._extract_m3u8_formats( - stream_url, video_id, 'mp4', preference=1, m3u8_id='hls', fatal=False)) + stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False)) else: if server and server.startswith('rtmp'): f = { @@ -126,13 +149,11 @@ class ARDMediathekIE(InfoExtractor): 'play_path': stream_url, 'format_id': 'a%s-rtmp-%s' % (num, quality), } - elif stream_url.startswith('http'): + else: f = { 'url': stream_url, 'format_id': 'a%s-%s-%s' % (num, ext, quality) } - else: - continue m = re.search(r'_(?P\d+)x(?P\d+)\.mp4$', stream_url) if m: f.update({ @@ -148,19 +169,25 @@ class ARDMediathekIE(InfoExtractor): # determine video id from url m = re.match(self._VALID_URL, url) + document_id = None + numid = re.search(r'documentId=([0-9]+)', url) if numid: - video_id = numid.group(1) + document_id = video_id = numid.group(1) else: video_id = m.group('video_id') webpage = self._download_webpage(url, video_id) - if '>Der gewünschte Beitrag ist nicht mehr verfügbar.<' in webpage: - raise ExtractorError('Video %s is no longer available' % video_id, expected=True) + ERRORS = ( + ('>Leider liegt eine Störung vor.', 'Video %s is unavailable'), + ('>Der gewünschte Beitrag ist nicht mehr verfügbar.<', + 'Video %s is no longer available'), + ) - if 'Diese Sendung ist für Jugendliche unter 12 Jahren nicht geeignet. Der Clip ist deshalb nur von 20 bis 6 Uhr verfügbar.' in webpage: - raise ExtractorError('This program is only suitable for those aged 12 and older. Video %s is therefore only available between 20 pm and 6 am.' % video_id, expected=True) + for pattern, message in ERRORS: + if pattern in webpage: + raise ExtractorError(message % video_id, expected=True) if re.search(r'[\?&]rss($|[=&])', url): doc = compat_etree_fromstring(webpage.encode('utf-8')) @@ -169,7 +196,7 @@ class ARDMediathekIE(InfoExtractor): title = self._html_search_regex( [r'(.*?)', - r'', + r'', r'

(.*?)

'], webpage, 'title') description = self._html_search_meta( @@ -206,12 +233,16 @@ class ARDMediathekIE(InfoExtractor): 'formats': formats, } else: # request JSON file + if not document_id: + video_id = self._search_regex( + r'/play/(?:config|media)/(\d+)', webpage, 'media id') info = self._extract_media_info( - 'http://www.ardmediathek.de/play/media/%s' % video_id, webpage, video_id) + 'http://www.ardmediathek.de/play/media/%s' % video_id, + webpage, video_id) info.update({ 'id': video_id, - 'title': title, + 'title': self._live_title(title) if info.get('is_live') else title, 'description': description, 'thumbnail': thumbnail, }) @@ -220,7 +251,7 @@ class ARDMediathekIE(InfoExtractor): class ARDIE(InfoExtractor): - _VALID_URL = '(?Phttps?://(www\.)?daserste\.de/[^?#]+/videos/(?P[^/?#]+)-(?P[0-9]+))\.html' + _VALID_URL = r'(?Phttps?://(www\.)?daserste\.de/[^?#]+/videos/(?P[^/?#]+)-(?P[0-9]+))\.html' _TEST = { 'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html', 'md5': 'd216c3a86493f9322545e045ddc3eb35', @@ -231,8 +262,9 @@ class ARDIE(InfoExtractor): 'duration': 2600, 'title': 'Die Story im Ersten: Mission unter falscher Flagge', 'upload_date': '20140804', - 'thumbnail': 're:^https?://.*\.jpg$', - } + 'thumbnail': r're:^https?://.*\.jpg$', + }, + 'skip': 'HTTP Error 404: Not Found', } def _real_extract(self, url): @@ -274,41 +306,3 @@ class ARDIE(InfoExtractor): 'upload_date': upload_date, 'thumbnail': thumbnail, } - - -class SportschauIE(ARDMediathekIE): - IE_NAME = 'Sportschau' - _VALID_URL = r'(?Phttps?://(?:www\.)?sportschau\.de/(?:[^/]+/)+video(?P[^/#?]+))\.html' - _TESTS = [{ - 'url': 'http://www.sportschau.de/tourdefrance/videoseppeltkokainhatnichtsmitklassischemdopingzutun100.html', - 'info_dict': { - 'id': 'seppeltkokainhatnichtsmitklassischemdopingzutun100', - 'ext': 'mp4', - 'title': 'Seppelt: "Kokain hat nichts mit klassischem Doping zu tun"', - 'thumbnail': 're:^https?://.*\.jpg$', - 'description': 'Der ARD-Doping Experte Hajo Seppelt gibt seine Einschätzung zum ersten Dopingfall der diesjährigen Tour de France um den Italiener Luca Paolini ab.', - }, - 'params': { - # m3u8 download - 'skip_download': True, - }, - }] - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - base_url = mobj.group('baseurl') - - webpage = self._download_webpage(url, video_id) - title = get_element_by_attribute('class', 'headline', webpage) - description = self._html_search_meta('description', webpage, 'description') - - info = self._extract_media_info( - base_url + '-mc_defaultQuality-h.json', webpage, video_id) - - info.update({ - 'title': title, - 'description': description, - }) - - return info