X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fdailymotion.py;h=e87690f9d288103ea222e1c216786b42e89364de;hb=38db46794f3ccfef09094db9b411e55acd4c1a3d;hp=0b47d652a678c992a0ed08a02becd75e0de7af6a;hpb=b27c856fbc1d36537a6caaaffe18e07ccc2bf4cf;p=youtube-dl diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index 0b47d652a..e87690f9d 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -1,28 +1,80 @@ import re import json +import itertools from .common import InfoExtractor +from .subtitles import SubtitlesInfoExtractor + from ..utils import ( compat_urllib_request, + compat_str, + get_element_by_attribute, + get_element_by_id, + orderedSet, ExtractorError, - 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') + request.add_header('Cookie', 'ff=off') + return request + +class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): """Information Extractor for Dailymotion""" - _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^/]+)' + _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?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\"" + + _FORMATS = [ + (u'stream_h264_ld_url', u'ld'), + (u'stream_h264_url', u'standard'), + (u'stream_h264_hq_url', u'hq'), + (u'stream_h264_hd_url', u'hd'), + (u'stream_h264_hd1080_url', u'hd180'), + ] + + _TESTS = [ + { + 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"Amphora Alex and Van .", + u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\"" + } + }, + # Vevo video + { + u'url': u'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi', + u'file': u'USUV71301934.mp4', + u'info_dict': { + u'title': u'Roar (Official)', + u'uploader': u'Katy Perry', + u'upload_date': u'20130905', + }, + u'params': { + u'skip_download': True, + }, + u'skip': u'VEVO is only available in some countries', + }, + # age-restricted video + { + u'url': u'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband', + u'file': u'xyh2zz.mp4', + u'md5': u'0d667a7b9cebecc3c89ee93099c4159d', + u'info_dict': { + u'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]', + u'uploader': 'HotWaves1012', + u'age_limit': 18, + } + } - } + ] def _real_extract(self, url): # Extract id and simplified title from URL @@ -30,27 +82,29 @@ class DailymotionIE(InfoExtractor): video_id = mobj.group(1).split('_')[0].split('?')[0] - video_extension = 'mp4' + url = 'http://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) + # 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(u'Vevo video detected: %s' % vevo_id) + return self.url_result(u'vevo:%s' % vevo_id, ie='Vevo') - mobj = re.search(r'', 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".*?>([^<]+?)([0-9]{2})-([0-9]{2})-([0-9]{4})', webpage) @@ -60,27 +114,115 @@ class DailymotionIE(InfoExtractor): embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id embed_page = self._download_webpage(embed_url, video_id, u'Downloading embed page') - info = self._search_regex(r'var info = ({.*?}),', embed_page, 'video info') + info = self._search_regex(r'var info = ({.*?}),$', embed_page, + 'video info', flags=re.MULTILINE) info = json.loads(info) - - # TODO: support choosing qualities - - for key in ['stream_h264_hd1080_url','stream_h264_hd_url', - 'stream_h264_hq_url','stream_h264_url', - 'stream_h264_ld_url']: - if info.get(key):#key in info and info[key]: - max_quality = key - self.to_screen(u'Using %s' % key) - break - else: + 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 = 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(u'Unable to extract video URL') - video_url = info[max_quality] - return [{ + # subtitles + video_subtitles = self.extract_subtitles(video_id, webpage) + if self._downloader.params.get('listsubtitles', False): + self._list_available_subtitles(video_id, webpage) + return + + return { 'id': video_id, - 'url': video_url, + 'formats': formats, 'uploader': video_uploader, 'upload_date': video_upload_date, - 'title': video_title, - 'ext': video_extension, - }] + 'title': self._og_search_title(webpage), + 'subtitles': video_subtitles, + 'thumbnail': info['thumbnail_url'], + 'age_limit': age_limit, + } + + def _get_available_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(u'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'], l['url']) for l in info['list']) + return sub_lang_list + self._downloader.report_warning(u'video doesn\'t have subtitles') + return {} + + +class DailymotionPlaylistIE(DailymotionBaseInfoExtractor): + IE_NAME = u'dailymotion:playlist' + _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P.+?)/' + _MORE_PAGES_INDICATOR = r'' + _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s' + + def _extract_entries(self, id): + video_ids = [] + for pagenum in itertools.count(1): + request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum)) + webpage = self._download_webpage(request, + id, u'Downloading page %s' % pagenum) + + playlist_el = get_element_by_attribute(u'class', u'video_list', webpage) + video_ids.extend(re.findall(r'data-id="(.+?)"', playlist_el)) + + if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None: + break + return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion') + for video_id in orderedSet(video_ids)] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + playlist_id = mobj.group('id') + webpage = self._download_webpage(url, playlist_id) + + return {'_type': 'playlist', + 'id': playlist_id, + 'title': get_element_by_id(u'playlist_name', webpage), + 'entries': self._extract_entries(playlist_id), + } + + +class DailymotionUserIE(DailymotionPlaylistIE): + IE_NAME = u'dailymotion:user' + _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/user/(?P[^/]+)' + _MORE_PAGES_INDICATOR = r'' + _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + user = mobj.group('user') + webpage = self._download_webpage(url, user) + full_user = self._html_search_regex( + r'(.*?)