X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/467f1716f3b46ffc038848b9b08a833a7811be09..4d055bc6eb5dbe9a42216ec3e659d8ff1f2a1256:/youtube_dl/extractor/dailymotion.py diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index fa8c630..7459719 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -1,110 +1,512 @@ -import re -import json +# coding: utf-8 +from __future__ import unicode_literals + +import base64 +import functools +import hashlib import itertools +import json +import random +import re +import string from .common import InfoExtractor +from ..compat import compat_struct_pack from ..utils import ( - compat_urllib_request, - get_element_by_attribute, - get_element_by_id, - + determine_ext, + error_to_compat_str, ExtractorError, + int_or_none, + mimetype2ext, + OnDemandPagedList, + parse_iso8601, + sanitized_Request, + str_to_int, + try_get, + unescapeHTML, + update_url_query, + url_or_none, + urlencode_postdata, ) -class DailymotionIE(InfoExtractor): - """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\"" - } - } + +class DailymotionBaseInfoExtractor(InfoExtractor): + @staticmethod + def _build_request(url): + """Build a request with the family filter disabled""" + request = sanitized_Request(url) + request.add_header('Cookie', 'family_filter=off; ff=off') + return request + + def _download_webpage_handle_no_ff(self, url, *args, **kwargs): + request = self._build_request(url) + return self._download_webpage_handle(request, *args, **kwargs) + + def _download_webpage_no_ff(self, url, *args, **kwargs): + request = self._build_request(url) + return self._download_webpage(request, *args, **kwargs) + + +class DailymotionIE(DailymotionBaseInfoExtractor): + _VALID_URL = r'''(?ix) + https?:// + (?: + (?:(?:www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(?:(?:embed|swf|\#)/)?video|swf)| + (?:www\.)?lequipe\.fr/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': 'http://www.dailymotion.com/video/x5kesuj_office-christmas-party-review-jason-bateman-olivia-munn-t-j-miller_news', + 'md5': '074b95bdee76b9e3654137aee9c79dfe', + 'info_dict': { + 'id': 'x5kesuj', + 'ext': 'mp4', + 'title': 'Office Christmas Party Review – Jason Bateman, Olivia Munn, T.J. Miller', + 'description': 'Office Christmas Party Review - Jason Bateman, Olivia Munn, T.J. Miller', + 'thumbnail': r're:^https?:.*\.(?:jpg|png)$', + 'duration': 187, + 'timestamp': 1493651285, + 'upload_date': '20170501', + 'uploader': 'Deadline', + 'uploader_id': 'x1xm8ri', + 'age_limit': 0, + }, + }, { + 'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames', + 'md5': '2137c41a8e78554bb09225b8eb322406', + 'info_dict': { + 'id': 'x2iuewm', + 'ext': 'mp4', + 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News', + 'description': 'Several come bundled with the Steam Controller.', + 'thumbnail': r're:^https?:.*\.(?:jpg|png)$', + 'duration': 74, + 'timestamp': 1425657362, + 'upload_date': '20150306', + 'uploader': 'IGN', + 'uploader_id': 'xijv66', + 'age_limit': 0, + 'view_count': int, + }, + 'skip': 'video gone', + }, { + # 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, + }, + 'skip': 'video gone', + }, { + # geo-restricted, player v5 + 'url': 'http://www.dailymotion.com/video/xhza0o', + 'only_matching': True, + }, { + # with subtitles + 'url': 'http://www.dailymotion.com/video/x20su5f_the-power-of-nightmares-1-the-rise-of-the-politics-of-fear-bbc-2004_news', + 'only_matching': True, + }, { + 'url': 'http://www.dailymotion.com/swf/video/x3n92nf', + 'only_matching': True, + }, { + 'url': 'http://www.dailymotion.com/swf/x3ss1m_funny-magic-trick-barry-and-stuart_fun', + 'only_matching': True, + }, { + 'url': 'https://www.lequipe.fr/video/x791mem', + 'only_matching': True, + }, { + 'url': 'https://www.lequipe.fr/video/k7MtHciueyTcrFtFKA2', + 'only_matching': True, + }] + + @staticmethod + def _extract_urls(webpage): + urls = [] + # Look for embedded Dailymotion player + # https://developer.dailymotion.com/player#player-parameters + for mobj in re.finditer( + r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage): + urls.append(unescapeHTML(mobj.group('url'))) + for mobj in re.finditer( + r'(?s)DM\.player\([^,]+,\s*{.*?video[\'"]?\s*:\s*["\']?(?P[0-9a-zA-Z]+).+?}\s*\);', webpage): + urls.append('https://www.dailymotion.com/embed/video/' + mobj.group('id')) + return urls def _real_extract(self, url): - # Extract id and simplified title from URL - mobj = re.match(self._VALID_URL, url) + video_id = self._match_id(url) - video_id = mobj.group(1).split('_')[0].split('?')[0] + webpage = self._download_webpage_no_ff( + 'https://www.dailymotion.com/video/%s' % video_id, video_id) - video_extension = 'mp4' + age_limit = self._rta_search(webpage) - # Retrieve video webpage to extract further information - request = compat_urllib_request.Request(url) - request.add_header('Cookie', 'family_filter=off') - webpage = self._download_webpage(request, video_id) + description = self._og_search_description( + webpage, default=None) or self._html_search_meta( + 'description', webpage, 'description') - # Extract URL, uploader and title from webpage - self.report_extraction(video_id) + view_count_str = self._search_regex( + (r']+itemprop="interactionCount"[^>]+content="UserPlays:([\s\d,.]+)"', + r'video_views_count[^>]+>\s+([\s\d\,.]+)'), + webpage, 'view count', default=None) + if view_count_str: + view_count_str = re.sub(r'\s', '', view_count_str) + view_count = str_to_int(view_count_str) + comment_count = int_or_none(self._search_regex( + r']+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"', + webpage, 'comment count', default=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) - if mobj is not None: - video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1) + if try_get(metadata, lambda x: x['error']['type']) == 'password_protected': + password = self._downloader.params.get('videopassword') + if password: + r = int(metadata['id'][1:], 36) + us64e = lambda x: base64.urlsafe_b64encode(x).decode().strip('=') + t = ''.join(random.choice(string.ascii_letters) for i in range(10)) + n = us64e(compat_struct_pack('I', r)) + i = us64e(hashlib.md5(('%s%d%s' % (password, r, t)).encode()).digest()) + metadata = self._download_json( + 'http://www.dailymotion.com/player/metadata/video/p' + i + t + n, video_id) - 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 = json.loads(info) + self._check_error(metadata) - # TODO: support choosing qualities + formats = [] + for quality, media_list in metadata['qualities'].items(): + for media in media_list: + media_url = media.get('url') + if not media_url: + continue + type_ = media.get('type') + if type_ == 'application/vnd.lumberjack.manifest': + continue + ext = mimetype2ext(type_) or determine_ext(media_url) + if ext == 'm3u8': + m3u8_formats = self._extract_m3u8_formats( + media_url, video_id, 'mp4', preference=-1, + m3u8_id='hls', fatal=False) + for f in m3u8_formats: + f['url'] = f['url'].split('#')[0] + formats.append(f) + elif ext == 'f4m': + formats.extend(self._extract_f4m_formats( + media_url, video_id, preference=-1, f4m_id='hds', fatal=False)) + else: + f = { + 'url': media_url, + 'format_id': 'http-%s' % quality, + 'ext': ext, + } + m = re.search(r'H264-(?P\d+)x(?P\d+)', media_url) + if m: + f.update({ + 'width': int(m.group('width')), + 'height': int(m.group('height')), + }) + formats.append(f) + self._sort_formats(formats) - 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: - raise ExtractorError(u'Unable to extract video URL') - video_url = info[max_quality] + title = metadata['title'] + duration = int_or_none(metadata.get('duration')) + timestamp = int_or_none(metadata.get('created_time')) + thumbnail = metadata.get('poster_url') + uploader = metadata.get('owner', {}).get('screenname') + uploader_id = metadata.get('owner', {}).get('id') + + subtitles = {} + subtitles_data = metadata.get('subtitles', {}).get('data', {}) + if subtitles_data and isinstance(subtitles_data, dict): + for subtitle_lang, subtitle in subtitles_data.items(): + subtitles[subtitle_lang] = [{ + 'ext': determine_ext(subtitle_url), + 'url': subtitle_url, + } for subtitle_url in subtitle.get('urls', [])] + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'timestamp': timestamp, + 'uploader': uploader, + 'uploader_id': uploader_id, + 'age_limit': age_limit, + 'view_count': view_count, + 'comment_count': comment_count, + 'formats': formats, + 'subtitles': subtitles, + } + + # vevo embed + vevo_id = self._search_regex( + r'[\w]*)', + webpage, 'vevo embed', default=None) + if vevo_id: + return self.url_result('vevo:%s' % vevo_id, 'Vevo') + + # fallback old player + embed_page = self._download_webpage_no_ff( + 'https://www.dailymotion.com/embed/video/%s' % video_id, + video_id, 'Downloading embed page') - return [{ - 'id': video_id, - 'url': video_url, - 'uploader': video_uploader, - 'upload_date': video_upload_date, - 'title': self._og_search_title(webpage), - 'ext': video_extension, - 'thumbnail': info['thumbnail_url'] - }] + timestamp = parse_iso8601(self._html_search_meta( + 'video:release_date', webpage, 'upload date')) + info = self._parse_json( + self._search_regex( + r'var info = ({.*?}),$', embed_page, + 'video info', flags=re.MULTILINE), + video_id) -class DailymotionPlaylistIE(InfoExtractor): - _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P.+?)/' - _MORE_PAGES_INDICATOR = r'' + self._check_error(info) + + 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, + }) + self._sort_formats(formats) + + # subtitles + video_subtitles = self.extract_subtitles(video_id, webpage) + + 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'], + 'timestamp': timestamp, + 'title': title, + 'description': description, + 'subtitles': video_subtitles, + 'thumbnail': info['thumbnail_url'], + 'age_limit': age_limit, + 'view_count': view_count, + 'duration': info['duration'] + } + + def _check_error(self, info): + error = info.get('error') + if error: + title = error.get('title') or error['message'] + # See https://developer.dailymotion.com/api#access-error + if error.get('code') == 'DM007': + self.raise_geo_restricted(msg=title) + raise ExtractorError( + '%s said: %s' % (self.IE_NAME, title), expected=True) + + 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' % error_to_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/(?Px[0-9a-z]+)' + _TESTS = [{ + 'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q', + 'info_dict': { + 'title': 'SPORT', + 'id': 'xv4bw', + }, + 'playlist_mincount': 20, + }] + _PAGE_SIZE = 100 + + def _fetch_page(self, playlist_id, authorizaion, page): + page += 1 + videos = self._download_json( + 'https://graphql.api.dailymotion.com', + playlist_id, 'Downloading page %d' % page, + data=json.dumps({ + 'query': '''{ + collection(xid: "%s") { + videos(first: %d, page: %d) { + pageInfo { + hasNextPage + nextPage + } + edges { + node { + xid + url + } + } + } + } +}''' % (playlist_id, self._PAGE_SIZE, page) + }).encode(), headers={ + 'Authorization': authorizaion, + 'Origin': 'https://www.dailymotion.com', + })['data']['collection']['videos'] + for edge in videos['edges']: + node = edge['node'] + yield self.url_result( + node['url'], DailymotionIE.ie_key(), node['xid']) def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - playlist_id = mobj.group('id') - video_ids = [] + playlist_id = self._match_id(url) + webpage = self._download_webpage(url, playlist_id) + api = self._parse_json(self._search_regex( + r'__PLAYER_CONFIG__\s*=\s*({.+?});', + webpage, 'player config'), playlist_id)['context']['api'] + auth = self._download_json( + api.get('auth_url', 'https://graphql.api.dailymotion.com/oauth/token'), + playlist_id, data=urlencode_postdata({ + 'client_id': api.get('client_id', 'f1a362d288c1b98099c7'), + 'client_secret': api.get('client_secret', 'eea605b96e01c796ff369935357eca920c5da4c5'), + 'grant_type': 'client_credentials', + })) + authorizaion = '%s %s' % (auth.get('token_type', 'Bearer'), auth['access_token']) + entries = OnDemandPagedList(functools.partial( + self._fetch_page, playlist_id, authorizaion), self._PAGE_SIZE) + return self.playlist_result( + entries, playlist_id, + self._og_search_title(webpage)) + +class DailymotionUserIE(DailymotionBaseInfoExtractor): + IE_NAME = 'dailymotion:user' + _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P[^/]+)' + _MORE_PAGES_INDICATOR = r'(?s)
.*?' % re.escape(user), + webpage, 'user')) + + return { + '_type': 'playlist', + 'id': user, + 'title': full_user, + 'entries': self._extract_entries(user), + }