X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/782e271dfe61ed8d5dbf100286ad876068ea585c..ed1c1d2a80ce1ad325853e6c59a1fbaf2156adcc:/youtube_dl/extractor/canalplus.py?ds=sidebyside diff --git a/youtube_dl/extractor/canalplus.py b/youtube_dl/extractor/canalplus.py index 1db9b24..d8bf073 100644 --- a/youtube_dl/extractor/canalplus.py +++ b/youtube_dl/extractor/canalplus.py @@ -1,54 +1,191 @@ -# encoding: utf-8 +# coding: utf-8 +from __future__ import unicode_literals + import re -import xml.etree.ElementTree from .common import InfoExtractor -from ..utils import unified_strdate +from ..compat import compat_urllib_parse_urlparse +from ..utils import ( + dict_get, + # ExtractorError, + # HEADRequest, + int_or_none, + qualities, + remove_end, + unified_strdate, +) + class CanalplusIE(InfoExtractor): - _VALID_URL = r'https?://(www\.canalplus\.fr/.*?/(?P.*)|player\.canalplus\.fr/#/(?P\d+))' - _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/cplus/%s' - IE_NAME = u'canalplus.fr' - - _TEST = { - u'url': u'http://www.canalplus.fr/c-infos-documentaires/pid1830-c-zapping.html?vid=922470', - u'file': u'922470.flv', - u'info_dict': { - u'title': u'Zapping - 26/08/13', - u'description': u'Le meilleur de toutes les chaînes, tous les jours.\nEmission du 26 août 2013', - u'upload_date': u'20130826', + IE_DESC = 'canalplus.fr, piwiplus.fr and d8.tv' + _VALID_URL = r'''(?x) + https?:// + (?: + (?: + (?:(?:www|m)\.)?canalplus\.fr| + (?:www\.)?piwiplus\.fr| + (?:www\.)?d8\.tv| + (?:www\.)?c8\.fr| + (?:www\.)?d17\.tv| + (?:(?:football|www)\.)?cstar\.fr| + (?:www\.)?itele\.fr + )/(?:(?:[^/]+/)*(?P[^/?#&]+))?(?:\?.*\bvid=(?P\d+))?| + player\.canalplus\.fr/#/(?P\d+) + ) + + ''' + _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/%s/%s?format=json' + _SITE_ID_MAP = { + 'canalplus': 'cplus', + 'piwiplus': 'teletoon', + 'd8': 'd8', + 'c8': 'd8', + 'd17': 'd17', + 'cstar': 'd17', + 'itele': 'itele', + } + + # Only works for direct mp4 URLs + _GEO_COUNTRIES = ['FR'] + + _TESTS = [{ + 'url': 'http://www.canalplus.fr/c-emissions/pid1830-c-zapping.html?vid=1192814', + 'info_dict': { + 'id': '1405510', + 'display_id': 'pid1830-c-zapping', + 'ext': 'mp4', + 'title': 'Zapping - 02/07/2016', + 'description': 'Le meilleur de toutes les chaînes, tous les jours', + 'upload_date': '20160702', }, - u'params': { - u'skip_download': True, + }, { + # geo restricted, bypassed + 'url': 'http://www.piwiplus.fr/videos-piwi/pid1405-le-labyrinthe-boing-super-ranger.html?vid=1108190', + 'info_dict': { + 'id': '1108190', + 'display_id': 'pid1405-le-labyrinthe-boing-super-ranger', + 'ext': 'mp4', + 'title': 'BOING SUPER RANGER - Ep : Le labyrinthe', + 'description': 'md5:4cea7a37153be42c1ba2c1d3064376ff', + 'upload_date': '20140724', }, - } + 'expected_warnings': ['HTTP Error 403: Forbidden'], + }, { + # geo restricted, bypassed + 'url': 'http://www.c8.fr/c8-divertissement/ms-touche-pas-a-mon-poste/pid6318-videos-integrales.html?vid=1443684', + 'md5': 'bb6f9f343296ab7ebd88c97b660ecf8d', + 'info_dict': { + 'id': '1443684', + 'display_id': 'pid6318-videos-integrales', + 'ext': 'mp4', + 'title': 'Guess my iep ! - TPMP - 07/04/2017', + 'description': 'md5:6f005933f6e06760a9236d9b3b5f17fa', + 'upload_date': '20170407', + }, + 'expected_warnings': ['HTTP Error 403: Forbidden'], + }, { + 'url': 'http://www.itele.fr/chroniques/invite-michael-darmon/rachida-dati-nicolas-sarkozy-est-le-plus-en-phase-avec-les-inquietudes-des-francais-171510', + 'info_dict': { + 'id': '1420176', + 'display_id': 'rachida-dati-nicolas-sarkozy-est-le-plus-en-phase-avec-les-inquietudes-des-francais-171510', + 'ext': 'mp4', + 'title': 'L\'invité de Michaël Darmon du 14/10/2016 - ', + 'description': 'Chaque matin du lundi au vendredi, Michaël Darmon reçoit un invité politique à 8h25.', + 'upload_date': '20161014', + }, + }, { + 'url': 'http://football.cstar.fr/cstar-minisite-foot/pid7566-feminines-videos.html?vid=1416769', + 'info_dict': { + 'id': '1416769', + 'display_id': 'pid7566-feminines-videos', + 'ext': 'mp4', + 'title': 'France - Albanie : les temps forts de la soirée - 20/09/2016', + 'description': 'md5:c3f30f2aaac294c1c969b3294de6904e', + 'upload_date': '20160921', + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://m.canalplus.fr/?vid=1398231', + 'only_matching': True, + }, { + 'url': 'http://www.d17.tv/emissions/pid8303-lolywood.html?vid=1397061', + 'only_matching': True, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - if video_id is None: - webpage = self._download_webpage(url, mobj.group('path')) - video_id = self._search_regex(r'videoId = "(\d+)";', webpage, u'video id') - info_url = self._VIDEO_INFO_TEMPLATE % video_id - info_page = self._download_webpage(info_url,video_id, - u'Downloading video info') - - self.report_extraction(video_id) - doc = xml.etree.ElementTree.fromstring(info_page.encode('utf-8')) - video_info = [video for video in doc if video.find('ID').text == video_id][0] - infos = video_info.find('INFOS') - media = video_info.find('MEDIA') - formats = [media.find('VIDEOS/%s' % format) - for format in ['BAS_DEBIT', 'HAUT_DEBIT', 'HD']] - video_url = [format.text for format in formats if format is not None][-1] - - return {'id': video_id, - 'title': u'%s - %s' % (infos.find('TITRAGE/TITRE').text, - infos.find('TITRAGE/SOUS_TITRE').text), - 'url': video_url, - 'ext': 'flv', - 'upload_date': unified_strdate(infos.find('PUBLICATION/DATE').text), - 'thumbnail': media.find('IMAGES/GRAND').text, - 'description': infos.find('DESCRIPTION').text, - 'view_count': int(infos.find('NB_VUES').text), - } + + site_id = self._SITE_ID_MAP[compat_urllib_parse_urlparse(url).netloc.rsplit('.', 2)[-2]] + + # Beware, some subclasses do not define an id group + display_id = remove_end(dict_get(mobj.groupdict(), ('display_id', 'id', 'vid')), '.html') + + webpage = self._download_webpage(url, display_id) + video_id = self._search_regex( + [r']+?videoId=(["\'])(?P\d+)', + r'id=["\']canal_video_player(?P\d+)', + r'data-video=["\'](?P\d+)'], + webpage, 'video id', default=mobj.group('vid'), group='id') + + info_url = self._VIDEO_INFO_TEMPLATE % (site_id, video_id) + video_data = self._download_json(info_url, video_id, 'Downloading video JSON') + + if isinstance(video_data, list): + video_data = [video for video in video_data if video.get('ID') == video_id][0] + media = video_data['MEDIA'] + infos = video_data['INFOS'] + + preference = qualities(['MOBILE', 'BAS_DEBIT', 'HAUT_DEBIT', 'HD']) + + # _, fmt_url = next(iter(media['VIDEOS'].items())) + # if '/geo' in fmt_url.lower(): + # response = self._request_webpage( + # HEADRequest(fmt_url), video_id, + # 'Checking if the video is georestricted') + # if '/blocage' in response.geturl(): + # raise ExtractorError( + # 'The video is not available in your country', + # expected=True) + + formats = [] + for format_id, format_url in media['VIDEOS'].items(): + if not format_url: + continue + if format_id == 'HLS': + formats.extend(self._extract_m3u8_formats( + format_url, video_id, 'mp4', 'm3u8_native', m3u8_id=format_id, fatal=False)) + elif format_id == 'HDS': + formats.extend(self._extract_f4m_formats( + format_url + '?hdcore=2.11.3', video_id, f4m_id=format_id, fatal=False)) + else: + formats.append({ + # the secret extracted ya function in http://player.canalplus.fr/common/js/canalPlayer.js + 'url': format_url + '?secret=pqzerjlsmdkjfoiuerhsdlfknaes', + 'format_id': format_id, + 'preference': preference(format_id), + }) + self._sort_formats(formats) + + thumbnails = [{ + 'id': image_id, + 'url': image_url, + } for image_id, image_url in media.get('images', {}).items()] + + titrage = infos['TITRAGE'] + + return { + 'id': video_id, + 'display_id': display_id, + 'title': '%s - %s' % (titrage['TITRE'], + titrage['SOUS_TITRE']), + 'upload_date': unified_strdate(infos.get('PUBLICATION', {}).get('DATE')), + 'thumbnails': thumbnails, + 'description': infos.get('DESCRIPTION'), + 'duration': int_or_none(infos.get('DURATION')), + 'view_count': int_or_none(infos.get('NB_VUES')), + 'like_count': int_or_none(infos.get('NB_LIKES')), + 'comment_count': int_or_none(infos.get('NB_COMMENTS')), + 'formats': formats, + }