From: Timendum Date: Tue, 14 Mar 2017 15:11:09 +0000 (+0100) Subject: [rai] Fix extraction X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=commitdiff_plain;h=51342717cddafde83dbf39f2212be40a196a577a;p=youtube-dl [rai] Fix extraction --- diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 980333a11..d9e8d53ac 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -802,7 +802,7 @@ from .radiojavan import RadioJavanIE from .radiobremen import RadioBremenIE from .radiofrance import RadioFranceIE from .rai import ( - RaiTVIE, + RaiPlayIE, RaiIE, ) from .rbmaradio import RBMARadioIE diff --git a/youtube_dl/extractor/rai.py b/youtube_dl/extractor/rai.py index 41afbd9af..b67e94f88 100644 --- a/youtube_dl/extractor/rai.py +++ b/youtube_dl/extractor/rai.py @@ -3,8 +3,8 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..compat import compat_urlparse from ..utils import ( - determine_ext, ExtractorError, + determine_ext, find_xpath_attr, fix_xml_ampersands, int_or_none, @@ -55,181 +55,157 @@ class RaiBaseIE(InfoExtractor): return formats - def _extract_from_content_id(self, content_id, base_url): - media = self._download_json( - 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id, - content_id, 'Downloading video JSON') - thumbnails = [] - for image_type in ('image', 'image_medium', 'image_300'): - thumbnail_url = media.get(image_type) - if thumbnail_url: - thumbnails.append({ - 'url': compat_urlparse.urljoin(base_url, thumbnail_url), - }) +class RaiPlayIE(RaiBaseIE): + _VALID_URL = r'https?://(?:www\.)?raiplay\.it/.+?-(?P[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})\.html' + _TESTS = [{ + 'url': 'http://www.raiplay.it/video/2016/10/La-Casa-Bianca-e06118bb-59a9-4636-b914-498e4cfd2c66.html?source=twitter', + 'md5': '340aa3b7afb54bfd14a8c11786450d76', + 'info_dict': { + 'id': 'e06118bb-59a9-4636-b914-498e4cfd2c66', + 'ext': 'mp4', + 'title': 'La Casa Bianca', + 'thumbnail': r're:^https?://.*\.jpg$', + 'uploader': r're:^Rai.+', + 'description': 're:^[A-Za-z]+' + } + }, { + 'url': 'http://www.raiplay.it/video/2016/11/gazebotraindesi-efebe701-969c-4593-92f3-285f0d1ce750.html?', + 'md5': 'ed4da3d70ccf8129a33ab16b34d20ab8', + 'info_dict': { + 'id': 'efebe701-969c-4593-92f3-285f0d1ce750', + 'ext': 'mp4', + 'title': 'Gazebo - #gazebotraindesi', + 'thumbnail': r're:^https?://.*\.png$', + 'uploader': r're:^Rai.+', + 'description': r're:^[A-Za-z]+' + } + }, { + 'url': 'http://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html', + 'md5': '8970abf8caf8aef4696e7b1f2adfc696', + 'info_dict': { + 'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391', + 'ext': 'mp4', + 'title': 'Report - Report del 07/04/2014', + 'thumbnail': r're:^https?://.*\.jpg$', + 'uploader': r're:^Rai.+', + 'description': r're:^[A-Za-z]+' + } + }] + _RESOLUTION = '600x400' - formats = [] - media_type = media['type'] - if 'Audio' in media_type: - formats.append({ - 'format_id': media.get('formatoAudio'), - 'url': media['audioUrl'], - 'ext': media.get('formatoAudio'), - }) - elif 'Video' in media_type: - formats.extend(self._extract_relinker_formats(media['mediaUri'], content_id)) - self._sort_formats(formats) - else: - raise ExtractorError('not a media file') + def _real_extract(self, url): + video_id = self._match_id(url) - subtitles = {} - captions = media.get('subtitlesUrl') - if captions: - STL_EXT = '.stl' - SRT_EXT = '.srt' - if captions.endswith(STL_EXT): - captions = captions[:-len(STL_EXT)] + SRT_EXT - subtitles['it'] = [{ - 'ext': 'srt', - 'url': captions, - }] + # remove query and fragment part from url + canonical_url = compat_urlparse.urljoin(url, compat_urlparse.urlparse(url).path) + webpage = self._download_webpage(canonical_url, video_id) - return { - 'id': content_id, - 'title': media['name'], - 'description': media.get('desc'), - 'thumbnails': thumbnails, - 'uploader': media.get('author'), - 'upload_date': unified_strdate(media.get('date')), - 'duration': parse_duration(media.get('length')), - 'formats': formats, - 'subtitles': subtitles, - } + media = self._download_json('%s?json' % canonical_url, + video_id, 'Downloading video JSON') + thumbnails = [] + if 'images' in media: + for _, value in media.get('images').items(): + if value: + thumbnails.append({ + 'url': value.replace('[RESOLUTION]', self._RESOLUTION) + }) -class RaiTVIE(RaiBaseIE): - _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/(?:[^/]+/)+(?:media|ondemand)/.+?-(?P[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html' - _TESTS = [ - { - 'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html', - 'md5': '8970abf8caf8aef4696e7b1f2adfc696', - 'info_dict': { - 'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391', - 'ext': 'mp4', - 'title': 'Report del 07/04/2014', - 'description': 'md5:f27c544694cacb46a078db84ec35d2d9', - 'upload_date': '20140407', - 'duration': 6160, - 'thumbnail': r're:^https?://.*\.jpg$', - } - }, - { - # no m3u8 stream - 'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html', - # HDS download, MD5 is unstable - 'info_dict': { - 'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9', - 'ext': 'flv', - 'title': 'TG PRIMO TEMPO', - 'upload_date': '20140612', - 'duration': 1758, - 'thumbnail': r're:^https?://.*\.jpg$', - }, - 'skip': 'Geo-restricted to Italy', - }, - { - 'url': 'http://www.rainews.it/dl/rainews/media/state-of-the-net-Antonella-La-Carpia-regole-virali-7aafdea9-0e5d-49d5-88a6-7e65da67ae13.html', - 'md5': '35cf7c229f22eeef43e48b5cf923bef0', - 'info_dict': { - 'id': '7aafdea9-0e5d-49d5-88a6-7e65da67ae13', - 'ext': 'mp4', - 'title': 'State of the Net, Antonella La Carpia: regole virali', - 'description': 'md5:b0ba04a324126903e3da7763272ae63c', - 'upload_date': '20140613', - }, - 'skip': 'Error 404', - }, - { - 'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-b4a49761-e0cc-4b14-8736-2729f6f73132-tg2.html', - 'info_dict': { - 'id': 'b4a49761-e0cc-4b14-8736-2729f6f73132', - 'ext': 'mp4', - 'title': 'Alluvione in Sardegna e dissesto idrogeologico', - 'description': 'Edizione delle ore 20:30 ', - }, - 'skip': 'invalid urls', - }, - { - 'url': 'http://www.ilcandidato.rai.it/dl/ray/media/Il-Candidato---Primo-episodio-Le-Primarie-28e5525a-b495-45e8-a7c3-bc48ba45d2b6.html', - 'md5': 'e57493e1cb8bc7c564663f363b171847', - 'info_dict': { - 'id': '28e5525a-b495-45e8-a7c3-bc48ba45d2b6', - 'ext': 'mp4', - 'title': 'Il Candidato - Primo episodio: "Le Primarie"', - 'description': 'md5:364b604f7db50594678f483353164fb8', - 'upload_date': '20140923', - 'duration': 386, - 'thumbnail': r're:^https?://.*\.jpg$', - } - }, - ] + if 'video' not in media: + raise ExtractorError('No video found') - def _real_extract(self, url): - video_id = self._match_id(url) + video = media.get('video') + duration = parse_duration(video.get('duration')), + formats = self._extract_relinker_formats(video.get('contentUrl'), video_id) + self._sort_formats(formats) - return self._extract_from_content_id(video_id, url) + return { + 'id': video_id, + 'title': self._og_search_title(webpage).replace(' - video - RaiPlay', ''), + 'description': self._og_search_description(webpage), + 'uploader': media.get('channel'), + 'duration': duration, + 'thumbnails': thumbnails, + 'formats': formats + } class RaiIE(RaiBaseIE): - _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/.+?-(?P[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html' - _TESTS = [ - { - 'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html', - 'md5': '2dd727e61114e1ee9c47f0da6914e178', - 'info_dict': { - 'id': '59d69d28-6bb6-409d-a4b5-ed44096560af', - 'ext': 'mp4', - 'title': 'Il pacco', - 'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a', - 'upload_date': '20141221', - }, + _VALID_URL = r'https?://.+\.(?:rai|rainews)\.it/dl/.+?-(?P[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html' + _TESTS = [{ + # subdomain test case + 'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html', + 'info_dict': { + 'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9', + 'ext': 'mp4', + 'title': 'TG PRIMO TEMPO', + 'upload_date': '20140612', + 'duration': 1758, + 'thumbnail': r're:^https?://.*\.jpg$' + } + }, { + # rainews test case + 'url': 'http://www.rainews.it/dl/rainews/media/Weekend-al-cinema-da-Hollywood-arriva-il-thriller-di-Tate-Taylor-La-ragazza-del-treno-1632c009-c843-4836-bb65-80c33084a64b.html', + 'info_dict': { + 'id': '1632c009-c843-4836-bb65-80c33084a64b', + 'ext': 'mp4', + 'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor \"La ragazza del treno\" ', + 'upload_date': '20161103', + 'thumbnail': r're:^https?://.*\.png$', + 'description': r're:^[A-Za-z]+' + } + }, { + # with media information + 'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html', + 'md5': '11959b4e44fa74de47011b5799490adf', + 'info_dict': { + 'id': 'efb17665-691c-45d5-a60c-5301333cbb0c', + 'ext': 'mp4', + 'title': 'TG1 ore 20:00 del 03/11/2016', + 'thumbnail': r're:^https?://.*\.jpg$', + 'upload_date': '20161103', + 'description': r're:^[A-Za-z]+' + } + }, { + # drawMediaRaiTV test case + 'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html', + 'md5': '2dd727e61114e1ee9c47f0da6914e178', + 'info_dict': { + 'id': '59d69d28-6bb6-409d-a4b5-ed44096560af', + 'ext': 'mp4', + 'title': 'Il pacco', + 'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a', + 'upload_date': '20141221', }, - { - # Direct relinker URL - 'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews', - # HDS live stream, MD5 is unstable - 'info_dict': { - 'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc', - 'ext': 'flv', - 'title': 'EuroNews', - }, - 'skip': 'Geo-restricted to Italy', + }, { + # Direct relinker URL + 'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews', + # HDS live stream, MD5 is unstable + 'info_dict': { + 'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc', + 'ext': 'flv', + 'title': 'EuroNews', }, - { - # Embedded content item ID - 'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined', - 'md5': '84c1135ce960e8822ae63cec34441d63', - 'info_dict': { - 'id': '0960e765-62c8-474a-ac4b-7eb3e2be39c8', - 'ext': 'mp4', - 'title': 'TG1 ore 20:00 del 02/07/2016', - 'upload_date': '20160702', - }, + }, { + # Embedded content item ID + 'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined', + 'info_dict': { + 'id': 'd80d4b70-3812-4501-a888-92edec729f00', + 'ext': 'mp4', + 'title': r're:TG1 ore \d{2}:\d{2} del \d{2}/\d{2}/\d{4}', + 'upload_date': r're:\d{8}', + 'description': r're:.+', }, - { - 'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html', - # HDS live stream, MD5 is unstable - 'info_dict': { - 'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9', - 'ext': 'flv', - 'title': 'La diretta di Rainews24', - }, + }, { + 'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html', + # HDS live stream, MD5 is unstable + 'info_dict': { + 'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9', + 'ext': 'mp4', + 'title': 'La diretta di Rainews24', }, - ] - - @classmethod - def suitable(cls, url): - return False if RaiTVIE.suitable(url) else super(RaiIE, cls).suitable(url) + }] def _real_extract(self, url): video_id = self._match_id(url) @@ -250,6 +226,12 @@ class RaiIE(RaiBaseIE): if content_item_id: return self._extract_from_content_id(content_item_id, url) + try: + return self._extract_from_content_id(video_id, url) + except ExtractorError: + # no media data, only direct relinker + pass + relinker_url = compat_urlparse.urljoin(url, self._search_regex( r'(?:var\s+videoURL|mediaInfo\.mediaUri)\s*=\s*(?P[\'"])(?P(https?:)?//mediapolis\.rai\.it/relinker/relinkerServlet\.htm\?cont=\d+)(?P=q1)', webpage, 'relinker URL', group='url')) @@ -265,3 +247,54 @@ class RaiIE(RaiBaseIE): 'title': title, 'formats': formats, } + + def _extract_from_content_id(self, content_id, url): + media = self._download_json( + 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id, + content_id, 'Downloading video JSON') + + thumbnails = [] + for image_type in ('image', 'image_medium', 'image_300'): + thumbnail_url = media.get(image_type) + if thumbnail_url: + thumbnails.append({ + 'url': compat_urlparse.urljoin(url, thumbnail_url), + }) + + formats = [] + media_type = media['type'] + if 'Audio' in media_type: + formats.append({ + 'format_id': media.get('formatoAudio'), + 'url': media['audioUrl'], + 'ext': media.get('formatoAudio'), + }) + elif 'Video' in media_type: + formats.extend(self._extract_relinker_formats(media['mediaUri'], content_id)) + self._sort_formats(formats) + else: + raise ExtractorError('not a media file') + + subtitles = {} + captions = media.get('subtitlesUrl') + if captions: + STL_EXT = '.stl' + SRT_EXT = '.srt' + if captions.endswith(STL_EXT): + captions = captions[:-len(STL_EXT)] + SRT_EXT + subtitles['it'] = [{ + 'ext': 'srt', + 'url': captions, + }] + + return { + 'id': content_id, + 'title': media['name'], + 'description': media.get('desc'), + 'thumbnails': thumbnails, + 'uploader': media.get('author'), + 'upload_date': unified_strdate(media.get('date')), + 'duration': parse_duration(media.get('length')), + 'formats': formats, + 'subtitles': subtitles, + }