From 6e6b9f600f2f447604f6108fb6486b73cc25def1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Thu, 2 Jun 2016 01:10:23 +0700 Subject: [PATCH] [arte] Add support for playlists and rework tests (Closes #9632) --- youtube_dl/extractor/arte.py | 173 ++++++++++++++++++----------- youtube_dl/extractor/extractors.py | 1 + 2 files changed, 110 insertions(+), 64 deletions(-) diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index e37fdae13..f40532929 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -61,10 +61,7 @@ class ArteTvIE(InfoExtractor): } -class ArteTVPlus7IE(InfoExtractor): - IE_NAME = 'arte.tv:+7' - _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?Pfr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P[^/]+)/(?P[^/?#&]+)' - +class ArteTVBaseIE(InfoExtractor): @classmethod def _extract_url_info(cls, url): mobj = re.match(cls._VALID_URL, url) @@ -78,60 +75,6 @@ class ArteTVPlus7IE(InfoExtractor): 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): - 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) - embed_url = self._search_regex( - r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1', - webpage, 'embed url', group='url') - return self.url_result(embed_url) - 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'] @@ -235,6 +178,74 @@ class ArteTVPlus7IE(InfoExtractor): return info_dict +class ArteTVPlus7IE(ArteTVBaseIE): + IE_NAME = 'arte.tv:+7' + _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&]+)' + + _TESTS = [{ + 'url': 'http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D', + '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'<iframe[^>]+src=(["\'])(?P<url>.+\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'<h3[^>]+title=(["\'])(?P<title>.+?)\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) + embed_url = self._search_regex( + r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1', + webpage, 'embed url', group='url') + return self.url_result(embed_url) + + # It also uses the arte_vp_url url from the webpage to extract the information class ArteTVCreativeIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:creative' @@ -267,7 +278,7 @@ class ArteTVInfoIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:info' _VALID_URL = r'https?://info\.arte\.tv/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)' - _TEST = { + _TESTS = [{ 'url': 'http://info.arte.tv/fr/service-civique-un-cache-misere', 'info_dict': { 'id': '067528-000-A', @@ -275,7 +286,7 @@ class ArteTVInfoIE(ArteTVPlus7IE): 'title': 'Service civique, un cache misère ?', 'upload_date': '20160403', }, - } + }] class ArteTVFutureIE(ArteTVPlus7IE): @@ -300,6 +311,8 @@ class ArteTVDDCIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:ddc' _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) if lang == 'folge': @@ -318,7 +331,7 @@ class ArteTVConcertIE(ArteTVPlus7IE): IE_NAME = 'arte.tv:concert' _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': { @@ -328,14 +341,14 @@ 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>.+)' - _TEST = { + _TESTS = [{ 'url': 'http://cinema.arte.tv/de/node/38291', 'md5': '6b275511a5107c60bacbeeda368c3aa1', 'info_dict': { @@ -345,7 +358,7 @@ class ArteTVCinemaIE(ArteTVPlus7IE): 'upload_date': '20160122', 'description': 'md5:7f749bbb77d800ef2be11d54529b96bc', }, - } + }] class ArteTVMagazineIE(ArteTVPlus7IE): @@ -390,9 +403,41 @@ class ArteTVEmbedIE(ArteTVPlus7IE): ) ''' + _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 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', + }, + '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) diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index dd4b2b838..dc21cfed9 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -56,6 +56,7 @@ from .arte import ( ArteTVDDCIE, ArteTVMagazineIE, ArteTVEmbedIE, + ArteTVPlaylistIE, ) from .atresplayer import AtresPlayerIE from .atttechchannel import ATTTechChannelIE