X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9a117f94b4bfe84cfe1d904d5132aefcf41511c9..fd1446fb8834d066905da673eef608c5030392ba:/youtube_dl/extractor/dailymotion.py diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index 3297a85..70aa433 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -1,86 +1,253 @@ +# coding: utf-8 +from __future__ import unicode_literals + import re +import json +import itertools from .common import InfoExtractor -from ..utils import ( - compat_urllib_request, - compat_urllib_parse, +from ..compat import ( + compat_str, + compat_urllib_request, +) +from ..utils import ( ExtractorError, + int_or_none, + orderedSet, + str_to_int, unescapeHTML, ) -class DailymotionIE(InfoExtractor): + +class DailymotionBaseInfoExtractor(InfoExtractor): + @staticmethod + def _build_request(url): + """Build a request with the family filter disabled""" + request = compat_urllib_request.Request(url) + request.add_header('Cookie', 'family_filter=off; ff=off') + return request + + +class DailymotionIE(DailymotionBaseInfoExtractor): """Information Extractor for Dailymotion""" - _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^/]+)' - IE_NAME = u'dailymotion' - _TEST = { - u'url': u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech', - u'file': u'x33vw9.mp4', - u'md5': u'392c4b85a60a90dc4792da41ce3144eb', - u'info_dict': { - u"uploader": u"Alex and Van .", - u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\"" + _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P[^/?_]+)' + IE_NAME = 'dailymotion' + + _FORMATS = [ + ('stream_h264_ld_url', 'ld'), + ('stream_h264_url', 'standard'), + ('stream_h264_hq_url', 'hq'), + ('stream_h264_hd_url', 'hd'), + ('stream_h264_hd1080_url', 'hd180'), + ] + + _TESTS = [ + { + 'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames', + 'md5': '2137c41a8e78554bb09225b8eb322406', + 'info_dict': { + 'id': 'x2iuewm', + 'ext': 'mp4', + 'uploader': 'IGN', + 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News', + 'upload_date': '20150306', + } + }, + # Vevo video + { + 'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi', + 'info_dict': { + 'title': 'Roar (Official)', + 'id': 'USUV71301934', + 'ext': 'mp4', + 'uploader': 'Katy Perry', + 'upload_date': '20130905', + }, + 'params': { + 'skip_download': True, + }, + 'skip': 'VEVO is only available in some countries', + }, + # age-restricted video + { + 'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband', + 'md5': '0d667a7b9cebecc3c89ee93099c4159d', + 'info_dict': { + 'id': 'xyh2zz', + 'ext': 'mp4', + 'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]', + 'uploader': 'HotWaves1012', + 'age_limit': 18, + } } - } + ] def _real_extract(self, url): - # Extract id and simplified title from URL - mobj = re.match(self._VALID_URL, url) - - video_id = mobj.group(1).split('_')[0].split('?')[0] - - video_extension = 'mp4' + video_id = self._match_id(url) + url = 'https://www.dailymotion.com/video/%s' % video_id # Retrieve video webpage to extract further information - request = compat_urllib_request.Request(url) - request.add_header('Cookie', 'family_filter=off') + request = self._build_request(url) webpage = self._download_webpage(request, video_id) # Extract URL, uploader and title from webpage self.report_extraction(video_id) - mobj = re.search(r'\s*var flashvars = (.*)', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract media URL') - flashvars = compat_urllib_parse.unquote(mobj.group(1)) - - for key in ['hd1080URL', 'hd720URL', 'hqURL', 'sdURL', 'ldURL', 'video_url']: - if key in flashvars: - max_quality = key - self.to_screen(u'Using %s' % key) - break - else: - raise ExtractorError(u'Unable to extract video URL') - mobj = re.search(r'"' + max_quality + r'":"(.+?)"', flashvars) - if mobj is None: - raise ExtractorError(u'Unable to extract video URL') + # It may just embed a vevo video: + m_vevo = re.search( + r'[\w]*)', + webpage) + if m_vevo is not None: + vevo_id = m_vevo.group('id') + self.to_screen('Vevo video detected: %s' % vevo_id) + return self.url_result('vevo:%s' % vevo_id, ie='Vevo') - video_url = compat_urllib_parse.unquote(mobj.group(1)).replace('\\/', '/') + age_limit = self._rta_search(webpage) - # TODO: support choosing qualities + video_upload_date = None + mobj = re.search(r'', webpage) + if mobj is not None: + video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3) + + embed_url = 'https://www.dailymotion.com/embed/video/%s' % video_id + embed_request = self._build_request(embed_url) + embed_page = self._download_webpage( + embed_request, video_id, 'Downloading embed page') + info = self._search_regex(r'var info = ({.*?}),$', embed_page, + 'video info', flags=re.MULTILINE) + info = json.loads(info) + if info.get('error') is not None: + msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title'] + raise ExtractorError(msg, expected=True) + + formats = [] + for (key, format_id) in self._FORMATS: + video_url = info.get(key) + if video_url is not None: + m_size = re.search(r'H264-(\d+)x(\d+)', video_url) + if m_size is not None: + width, height = map(int_or_none, (m_size.group(1), m_size.group(2))) + else: + width, height = None, None + formats.append({ + 'url': video_url, + 'ext': 'mp4', + 'format_id': format_id, + 'width': width, + 'height': height, + }) + if not formats: + raise ExtractorError('Unable to extract video URL') + + # subtitles + video_subtitles = self.extract_subtitles(video_id, webpage) + + view_count = str_to_int(self._search_regex( + r'video_views_count[^>]+>\s+([\d\.,]+)', + webpage, 'view count', fatal=False)) + + title = self._og_search_title(webpage, default=None) + if title is None: + title = self._html_search_regex( + r'(?s)]*>(.*?)', webpage, + 'title') + + return { + 'id': video_id, + 'formats': formats, + 'uploader': info['owner.screenname'], + 'upload_date': video_upload_date, + 'title': title, + 'subtitles': video_subtitles, + 'thumbnail': info['thumbnail_url'], + 'age_limit': age_limit, + 'view_count': view_count, + } + + def _get_subtitles(self, video_id, webpage): + try: + sub_list = self._download_webpage( + 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id, + video_id, note=False) + except ExtractorError as err: + self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err)) + return {} + info = json.loads(sub_list) + if (info['total'] > 0): + sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list']) + return sub_lang_list + self._downloader.report_warning('video doesn\'t have subtitles') + return {} + + +class DailymotionPlaylistIE(DailymotionBaseInfoExtractor): + IE_NAME = 'dailymotion:playlist' + _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P.+?)/' + _MORE_PAGES_INDICATOR = r'(?s)
.*?', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = unescapeHTML(mobj.group('title')) - video_uploader = None - video_uploader = self._search_regex([r'(?im)[^<]+?]+?>([^<]+?)', - # Looking for official user - r'<(?:span|a) .*?rel="author".*?>([^<]+?)[^/]+)$' + _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s' + _TESTS = [{ + 'url': 'https://www.dailymotion.com/user/nqtv', + 'info_dict': { + 'id': 'nqtv', + 'title': 'Rémi Gaillard', + }, + 'playlist_mincount': 100, + }] - video_upload_date = None - mobj = re.search(r'
([0-9]{2})-([0-9]{2})-([0-9]{4})
', webpage) - if mobj is not None: - video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1) - - return [{ - 'id': video_id, - 'url': video_url, - 'uploader': video_uploader, - 'upload_date': video_upload_date, - 'title': video_title, - 'ext': video_extension, - }] + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + user = mobj.group('user') + webpage = self._download_webpage( + 'https://www.dailymotion.com/user/%s' % user, user) + full_user = unescapeHTML(self._html_search_regex( + r'' % re.escape(user), + webpage, 'user')) + + return { + '_type': 'playlist', + 'id': user, + 'title': full_user, + 'entries': self._extract_entries(user), + }