X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/b238854ce845f3796daac74edab2e8a373e8ba1a..38bb9b1b0a044cabaf5691553815e334cd2e9213:/youtube_dl/extractor/canalplus.py diff --git a/youtube_dl/extractor/canalplus.py b/youtube_dl/extractor/canalplus.py index 7cdcd83..699b4f7 100644 --- a/youtube_dl/extractor/canalplus.py +++ b/youtube_dl/extractor/canalplus.py @@ -1,53 +1,138 @@ # encoding: utf-8 +from __future__ import unicode_literals + import re from .common import InfoExtractor -from ..utils import unified_strdate +from ..utils import ( + ExtractorError, + HEADRequest, + unified_strdate, + url_basename, + qualities, +) 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'https?://(?:www\.(?Pcanalplus\.fr|piwiplus\.fr|d8\.tv|itele\.fr)/.*?/(?P.*)|player\.canalplus\.fr/#/(?P[0-9]+))' + _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/%s/%s' + _SITE_ID_MAP = { + 'canalplus.fr': 'cplus', + 'piwiplus.fr': 'teletoon', + 'd8.tv': 'd8', + 'itele.fr': 'itele', + } + + _TESTS = [{ + 'url': 'http://www.canalplus.fr/c-emissions/pid1830-c-zapping.html?vid=1263092', + 'md5': 'b3481d7ca972f61e37420798d0a9d934', + 'info_dict': { + 'id': '1263092', + 'ext': 'flv', + 'title': 'Le Zapping - 13/05/15', + 'description': 'md5:09738c0d06be4b5d06a0940edb0da73f', + 'upload_date': '20150513', }, - u'params': { - u'skip_download': True, + }, { + 'url': 'http://www.piwiplus.fr/videos-piwi/pid1405-le-labyrinthe-boing-super-ranger.html?vid=1108190', + 'info_dict': { + 'id': '1108190', + 'ext': 'flv', + 'title': 'Le labyrinthe - Boing super ranger', + 'description': 'md5:4cea7a37153be42c1ba2c1d3064376ff', + 'upload_date': '20140724', }, - } + 'skip': 'Only works from France', + }, { + 'url': 'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html', + 'info_dict': { + 'id': '966289', + 'ext': 'flv', + 'title': 'Campagne intime - Documentaire exceptionnel', + 'description': 'md5:d2643b799fb190846ae09c61e59a859f', + 'upload_date': '20131108', + }, + 'skip': 'videos get deleted after a while', + }, { + 'url': 'http://www.itele.fr/france/video/aubervilliers-un-lycee-en-colere-111559', + 'md5': 'f3a46edcdf28006598ffaf5b30e6a2d4', + 'info_dict': { + 'id': '1213714', + 'ext': 'flv', + 'title': 'Aubervilliers : un lycée en colère - Le 11/02/2015 à 06h45', + 'description': 'md5:8216206ec53426ea6321321f3b3c16db', + 'upload_date': '20150211', + }, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.groupdict().get('id') + + site_id = self._SITE_ID_MAP[mobj.group('site') or 'canal'] + + # Beware, some subclasses do not define an id group + display_id = url_basename(mobj.group('path')) + 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 - doc = self._download_xml(info_url,video_id, - u'Downloading video info') + webpage = self._download_webpage(url, display_id) + video_id = self._search_regex( + r']+?videoId="(\d+)"', webpage, 'video id') + + info_url = self._VIDEO_INFO_TEMPLATE % (site_id, video_id) + doc = self._download_xml(info_url, video_id, 'Downloading video XML') - self.report_extraction(video_id) 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), - } + infos = video_info.find('INFOS') + + preference = qualities(['MOBILE', 'BAS_DEBIT', 'HAUT_DEBIT', 'HD', 'HLS', 'HDS']) + + fmt_url = next(iter(media.find('VIDEOS'))).text + 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 fmt in media.find('VIDEOS'): + format_url = fmt.text + if not format_url: + continue + format_id = fmt.tag + if format_id == 'HLS': + hls_formats = self._extract_m3u8_formats(format_url, video_id, 'flv') + for fmt in hls_formats: + fmt['preference'] = preference(format_id) + formats.extend(hls_formats) + elif format_id == 'HDS': + hds_formats = self._extract_f4m_formats(format_url + '?hdcore=2.11.3', video_id) + for fmt in hds_formats: + fmt['preference'] = preference(format_id) + formats.extend(hds_formats) + else: + formats.append({ + 'url': format_url, + 'format_id': format_id, + 'preference': preference(format_id), + }) + self._sort_formats(formats) + + return { + 'id': video_id, + 'display_id': display_id, + 'title': '%s - %s' % (infos.find('TITRAGE/TITRE').text, + infos.find('TITRAGE/SOUS_TITRE').text), + '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), + 'like_count': int(infos.find('NB_LIKES').text), + 'comment_count': int(infos.find('NB_COMMENTS').text), + 'formats': formats, + }