X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/15b1d10671b48df598afd70e17ba21e9e64ac766..3712eb2135b6beeff3820f65bdfc6173fd1dbbab:/youtube_dl/extractor/arte.py?ds=inline diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index 219631b..ffc3218 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -1,16 +1,23 @@ -# encoding: utf-8 +# coding: utf-8 from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import ( + compat_parse_qs, + compat_str, + compat_urllib_parse_urlparse, +) from ..utils import ( + ExtractorError, find_xpath_attr, - unified_strdate, - get_element_by_id, get_element_by_attribute, int_or_none, + NO_DEFAULT, qualities, + try_get, + unified_strdate, ) # There are different sources of video in arte.tv, the extraction process @@ -19,7 +26,7 @@ from ..utils import ( class ArteTvIE(InfoExtractor): - _VALID_URL = r'http://videos\.arte\.tv/(?Pfr|de)/.*-(?P.*?)\.html' + _VALID_URL = r'https?://videos\.arte\.tv/(?Pfr|de|en|es)/.*-(?P.*?)\.html' IE_NAME = 'arte.tv' def _real_extract(self, url): @@ -37,7 +44,7 @@ class ArteTvIE(InfoExtractor): config_xml_url, video_id, note='Downloading configuration') formats = [{ - 'forma_id': q.attrib['quality'], + 'format_id': q.attrib['quality'], # The playpath starts at 'mp4:', if we don't manually # split the url, rtmpdump will incorrectly parse them 'url': q.text.split('mp4:', 1)[0], @@ -57,39 +64,39 @@ class ArteTvIE(InfoExtractor): } -class ArteTVPlus7IE(InfoExtractor): - IE_NAME = 'arte.tv:+7' - _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?Pfr|de)/(?:(?:sendungen|emissions)/)?(?P.*?)/(?P.*?)(\?.*)?' - +class ArteTVBaseIE(InfoExtractor): @classmethod def _extract_url_info(cls, url): mobj = re.match(cls._VALID_URL, url) lang = mobj.group('lang') - # This is not a real id, it can be for example AJT for the news - # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal - video_id = mobj.group('id') + query = compat_parse_qs(compat_urllib_parse_urlparse(url).query) + if 'vid' in query: + video_id = query['vid'][0] + else: + # This is not a real id, it can be for example AJT for the news + # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal + video_id = mobj.group('id') return video_id, lang - def _real_extract(self, url): - video_id, lang = self._extract_url_info(url) - webpage = self._download_webpage(url, video_id) - return self._extract_from_webpage(webpage, video_id, lang) - - def _extract_from_webpage(self, webpage, video_id, lang): - json_url = self._html_search_regex( - [r'arte_vp_url=["\'](.*?)["\']', r'data-url=["\']([^"]+)["\']'], - webpage, 'json vp url') - return self._extract_from_json_url(json_url, video_id, lang) - - def _extract_from_json_url(self, json_url, video_id, lang): + def _extract_from_json_url(self, json_url, video_id, lang, title=None): info = self._download_json(json_url, video_id) player_info = info['videoJsonPlayer'] + vsr = try_get(player_info, lambda x: x['VSR'], dict) + if not vsr: + error = None + if try_get(player_info, lambda x: x['custom_msg']['type']) == 'error': + error = try_get( + player_info, lambda x: x['custom_msg']['msg'], compat_str) + if not error: + error = 'Video %s is not available' % player_info.get('VID') or video_id + raise ExtractorError(error, expected=True) + upload_date_str = player_info.get('shootingDate') if not upload_date_str: - upload_date_str = player_info.get('VDA', '').split(' ')[0] + upload_date_str = (player_info.get('VRA') or player_info.get('VDA') or '').split(' ')[0] - title = player_info['VTI'].strip() + title = (player_info.get('VTI') or title or player_info['VID']).strip() subtitle = player_info.get('VSU', '').strip() if subtitle: title += ' - %s' % subtitle @@ -103,28 +110,60 @@ class ArteTVPlus7IE(InfoExtractor): } qfunc = qualities(['HQ', 'MQ', 'EQ', 'SQ']) + LANGS = { + 'fr': 'F', + 'de': 'A', + 'en': 'E[ANG]', + 'es': 'E[ESP]', + } + + langcode = LANGS.get(lang, lang) + formats = [] - for format_id, format_dict in player_info['VSR'].items(): + for format_id, format_dict in vsr.items(): f = dict(format_dict) versionCode = f.get('versionCode') + l = re.escape(langcode) + + # Language preference from most to least priority + # Reference: section 5.6.3 of + # http://www.arte.tv/sites/en/corporate/files/complete-technical-guidelines-arte-geie-v1-05.pdf + PREFERENCES = ( + # original version in requested language, without subtitles + r'VO{0}$'.format(l), + # original version in requested language, with partial subtitles in requested language + r'VO{0}-ST{0}$'.format(l), + # original version in requested language, with subtitles for the deaf and hard-of-hearing in requested language + r'VO{0}-STM{0}$'.format(l), + # non-original (dubbed) version in requested language, without subtitles + r'V{0}$'.format(l), + # non-original (dubbed) version in requested language, with subtitles partial subtitles in requested language + r'V{0}-ST{0}$'.format(l), + # non-original (dubbed) version in requested language, with subtitles for the deaf and hard-of-hearing in requested language + r'V{0}-STM{0}$'.format(l), + # original version in requested language, with partial subtitles in different language + r'VO{0}-ST(?!{0}).+?$'.format(l), + # original version in requested language, with subtitles for the deaf and hard-of-hearing in different language + r'VO{0}-STM(?!{0}).+?$'.format(l), + # original version in different language, with partial subtitles in requested language + r'VO(?:(?!{0}).+?)?-ST{0}$'.format(l), + # original version in different language, with subtitles for the deaf and hard-of-hearing in requested language + r'VO(?:(?!{0}).+?)?-STM{0}$'.format(l), + # original version in different language, without subtitles + r'VO(?:(?!{0}))?$'.format(l), + # original version in different language, with partial subtitles in different language + r'VO(?:(?!{0}).+?)?-ST(?!{0}).+?$'.format(l), + # original version in different language, with subtitles for the deaf and hard-of-hearing in different language + r'VO(?:(?!{0}).+?)?-STM(?!{0}).+?$'.format(l), + ) + + for pref, p in enumerate(PREFERENCES): + if re.match(p, versionCode): + lang_pref = len(PREFERENCES) - pref + break + else: + lang_pref = -1 - langcode = { - 'fr': 'F', - 'de': 'A', - }.get(lang, lang) - lang_rexs = [r'VO?%s' % langcode, r'VO?.-ST%s' % langcode] - lang_pref = ( - None if versionCode is None else ( - 10 if any(re.match(r, versionCode) for r in lang_rexs) - else -10)) - source_pref = 0 - if versionCode is not None: - # The original version with subtitles has lower relevance - if re.match(r'VO-ST(F|A)', versionCode): - source_pref -= 10 - # The version with sourds/mal subtitles has also lower relevance - elif re.match(r'VO?(F|A)-STM\1', versionCode): - source_pref -= 9 format = { 'format_id': format_id, 'preference': -10 if f.get('videoFormat') == 'M3U8' else None, @@ -133,8 +172,7 @@ class ArteTVPlus7IE(InfoExtractor): 'width': int_or_none(f.get('width')), 'height': int_or_none(f.get('height')), 'tbr': int_or_none(f.get('bitrate')), - 'quality': qfunc(f['quality']), - 'source_preference': source_pref, + 'quality': qfunc(f.get('quality')), } if f.get('mediaType') == 'rtmp': @@ -146,61 +184,148 @@ class ArteTVPlus7IE(InfoExtractor): formats.append(format) + self._check_formats(formats, video_id) self._sort_formats(formats) info_dict['formats'] = formats return info_dict +class ArteTVPlus7IE(ArteTVBaseIE): + IE_NAME = 'arte.tv:+7' + _VALID_URL = r'https?://(?:(?:www|sites)\.)?arte\.tv/(?:[^/]+/)?(?Pfr|de|en|es)/(?:videos/)?(?:[^/]+/)*(?P[^/?#&]+)' + + _TESTS = [{ + 'url': 'http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D', + 'only_matching': True, + }, { + 'url': 'http://sites.arte.tv/karambolage/de/video/karambolage-22', + 'only_matching': True, + }, { + 'url': 'http://www.arte.tv/de/videos/048696-000-A/der-kluge-bauch-unser-zweites-gehirn', + 'only_matching': True, + }] + + @classmethod + def suitable(cls, url): + return False if ArteTVPlaylistIE.suitable(url) else super(ArteTVPlus7IE, cls).suitable(url) + + def _real_extract(self, url): + video_id, lang = self._extract_url_info(url) + webpage = self._download_webpage(url, video_id) + return self._extract_from_webpage(webpage, video_id, lang) + + def _extract_from_webpage(self, webpage, video_id, lang): + patterns_templates = (r'arte_vp_url=["\'](.*?%s.*?)["\']', r'data-url=["\']([^"]+%s[^"]+)["\']') + ids = (video_id, '') + # some pages contain multiple videos (like + # http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D), + # so we first try to look for json URLs that contain the video id from + # the 'vid' parameter. + patterns = [t % re.escape(_id) for _id in ids for t in patterns_templates] + json_url = self._html_search_regex( + patterns, webpage, 'json vp url', default=None) + if not json_url: + def find_iframe_url(webpage, default=NO_DEFAULT): + return self._html_search_regex( + r']+src=(["\'])(?P.+\bjson_url=.+?)\1', + webpage, 'iframe url', group='url', default=default) + + iframe_url = find_iframe_url(webpage, None) + if not iframe_url: + embed_url = self._html_search_regex( + r'arte_vp_url_oembed=\'([^\']+?)\'', webpage, 'embed url', default=None) + if embed_url: + player = self._download_json( + embed_url, video_id, 'Downloading player page') + iframe_url = find_iframe_url(player['html']) + # en and es URLs produce react-based pages with different layout (e.g. + # http://www.arte.tv/guide/en/053330-002-A/carnival-italy?zone=world) + if not iframe_url: + program = self._search_regex( + r'program\s*:\s*({.+?["\']embed_html["\'].+?}),?\s*\n', + webpage, 'program', default=None) + if program: + embed_html = self._parse_json(program, video_id) + if embed_html: + iframe_url = find_iframe_url(embed_html['embed_html']) + if iframe_url: + json_url = compat_parse_qs( + compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0] + if json_url: + title = self._search_regex( + r']+title=(["\'])(?P.+?)\1', + webpage, 'title', default=None, group='title') + return self._extract_from_json_url(json_url, video_id, lang, title=title) + # Different kind of embed URL (e.g. + # http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium) + entries = [ + self.url_result(url) + for _, url in re.findall(r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1', webpage)] + return self.playlist_result(entries) + + # It also uses the arte_vp_url url from the webpage to extract the information class ArteTVCreativeIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:creative' - _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/(?:magazine?/)?(?P<id>[^?#]+)' + _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)' _TESTS = [{ - 'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design', + 'url': 'http://creative.arte.tv/fr/episode/osmosis-episode-1', 'info_dict': { - 'id': '72176', + 'id': '057405-001-A', 'ext': 'mp4', - 'title': 'Folge 2 - Corporate Design', - 'upload_date': '20131004', + 'title': 'OSMOSIS - N\'AYEZ PLUS PEUR D\'AIMER (1)', + 'upload_date': '20150716', }, }, { 'url': 'http://creative.arte.tv/fr/Monty-Python-Reunion', + 'playlist_count': 11, + 'add_ie': ['Youtube'], + }, { + 'url': 'http://creative.arte.tv/de/episode/agentur-amateur-4-der-erste-kunde', + 'only_matching': True, + }] + + +class ArteTVInfoIE(ArteTVPlus7IE): + IE_NAME = 'arte.tv:info' + _VALID_URL = r'https?://info\.arte\.tv/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)' + + _TESTS = [{ + 'url': 'http://info.arte.tv/fr/service-civique-un-cache-misere', 'info_dict': { - 'id': '160676', + 'id': '067528-000-A', 'ext': 'mp4', - 'title': 'Monty Python live (mostly)', - 'description': 'Événement ! Quarante-cinq ans après leurs premiers succès, les légendaires Monty Python remontent sur scène.\n', - 'upload_date': '20140805', - } + 'title': 'Service civique, un cache misère ?', + 'upload_date': '20160403', + }, }] class ArteTVFutureIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:future' - _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(thema|sujet)/.*?#article-anchor-(?P<id>\d+)' + _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)' - _TEST = { - 'url': 'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081', + _TESTS = [{ + 'url': 'http://future.arte.tv/fr/info-sciences/les-ecrevisses-aussi-sont-anxieuses', 'info_dict': { - 'id': '5201', + 'id': '050940-028-A', 'ext': 'mp4', - 'title': 'Les champignons au secours de la planète', - 'upload_date': '20131101', + 'title': 'Les écrevisses aussi peuvent être anxieuses', + 'upload_date': '20140902', }, - } - - def _real_extract(self, url): - anchor_id, lang = self._extract_url_info(url) - webpage = self._download_webpage(url, anchor_id) - row = get_element_by_id(anchor_id, webpage) - return self._extract_from_webpage(row, anchor_id, lang) + }, { + 'url': 'http://future.arte.tv/fr/la-science-est-elle-responsable', + 'only_matching': True, + }] class ArteTVDDCIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:ddc' - _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)' + _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>[^/?#&]+)' + + _TESTS = [] def _real_extract(self, url): video_id, lang = self._extract_url_info(url) @@ -218,9 +343,9 @@ class ArteTVDDCIE(ArteTVPlus7IE): class ArteTVConcertIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:concert' - _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>de|fr)/(?P<id>.+)' + _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)' - _TEST = { + _TESTS = [{ 'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde', 'md5': '9ea035b7bd69696b67aa2ccaaa218161', 'info_dict': { @@ -230,23 +355,119 @@ class ArteTVConcertIE(ArteTVPlus7IE): 'upload_date': '20140128', 'description': 'md5:486eb08f991552ade77439fe6d82c305', }, - } + }] + + +class ArteTVCinemaIE(ArteTVPlus7IE): + IE_NAME = 'arte.tv:cinema' + _VALID_URL = r'https?://cinema\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>.+)' + + _TESTS = [{ + 'url': 'http://cinema.arte.tv/fr/article/les-ailes-du-desir-de-julia-reck', + 'md5': 'a5b9dd5575a11d93daf0e3f404f45438', + 'info_dict': { + 'id': '062494-000-A', + 'ext': 'mp4', + 'title': 'Film lauréat du concours web - "Les ailes du désir" de Julia Reck', + 'upload_date': '20150807', + }, + }] + + +class ArteTVMagazineIE(ArteTVPlus7IE): + IE_NAME = 'arte.tv:magazine' + _VALID_URL = r'https?://(?:www\.)?arte\.tv/magazine/[^/]+/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)' + + _TESTS = [{ + # Embedded via <iframe src="http://www.arte.tv/arte_vp/index.php?json_url=..." + 'url': 'http://www.arte.tv/magazine/trepalium/fr/entretien-avec-le-realisateur-vincent-lannoo-trepalium', + 'md5': '2a9369bcccf847d1c741e51416299f25', + 'info_dict': { + 'id': '065965-000-A', + 'ext': 'mp4', + 'title': 'Trepalium - Extrait Ep.01', + 'upload_date': '20160121', + }, + }, { + # Embedded via <iframe src="http://www.arte.tv/guide/fr/embed/054813-004-A/medium" + 'url': 'http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium', + 'md5': 'fedc64fc7a946110fe311634e79782ca', + 'info_dict': { + 'id': '054813-004_PLUS7-F', + 'ext': 'mp4', + 'title': 'Trepalium (4/6)', + 'description': 'md5:10057003c34d54e95350be4f9b05cb40', + 'upload_date': '20160218', + }, + }, { + 'url': 'http://www.arte.tv/magazine/metropolis/de/frank-woeste-german-paris-metropolis', + 'only_matching': True, + }] class ArteTVEmbedIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:embed' _VALID_URL = r'''(?x) http://www\.arte\.tv - /playerv2/embed\.php\?json_url= + /(?:playerv2/embed|arte_vp/index)\.php\?json_url= (?P<json_url> http://arte\.tv/papi/tvguide/videos/stream/player/ (?P<lang>[^/]+)/(?P<id>[^/]+)[^&]* ) ''' + _TESTS = [] + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') lang = mobj.group('lang') json_url = mobj.group('json_url') return self._extract_from_json_url(json_url, video_id, lang) + + +class TheOperaPlatformIE(ArteTVPlus7IE): + IE_NAME = 'theoperaplatform' + _VALID_URL = r'https?://(?:www\.)?theoperaplatform\.eu/(?P<lang>fr|de|en|es)/(?P<id>[^/?#&]+)' + + _TESTS = [{ + 'url': 'http://www.theoperaplatform.eu/de/opera/verdi-otello', + 'md5': '970655901fa2e82e04c00b955e9afe7b', + 'info_dict': { + 'id': '060338-009-A', + 'ext': 'mp4', + 'title': 'Verdi - OTELLO', + 'upload_date': '20160927', + }, + }] + + +class ArteTVPlaylistIE(ArteTVBaseIE): + IE_NAME = 'arte.tv:playlist' + _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/[^#]*#collection/(?P<id>PL-\d+)' + + _TESTS = [{ + 'url': 'http://www.arte.tv/guide/de/plus7/?country=DE#collection/PL-013263/ARTETV', + 'info_dict': { + 'id': 'PL-013263', + 'title': 'Areva & Uramin', + 'description': 'md5:a1dc0312ce357c262259139cfd48c9bf', + }, + 'playlist_mincount': 6, + }, { + 'url': 'http://www.arte.tv/guide/de/playlists?country=DE#collection/PL-013190/ARTETV', + 'only_matching': True, + }] + + def _real_extract(self, url): + playlist_id, lang = self._extract_url_info(url) + collection = self._download_json( + 'https://api.arte.tv/api/player/v1/collectionData/%s/%s?source=videos' + % (lang, playlist_id), playlist_id) + title = collection.get('title') + description = collection.get('shortDescription') or collection.get('teaserText') + entries = [ + self._extract_from_json_url( + video['jsonUrl'], video.get('programId') or playlist_id, lang) + for video in collection['videos'] if video.get('jsonUrl')] + return self.playlist_result(entries, playlist_id, title, description)