X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9ed7fe4fe4c445eb7d9f3197bb300d0db8f1807a..d9bd2d488401884d00ee455f2a161a5011f7457d:/youtube_dl/extractor/dailymotion.py diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index ab7f3ae..040f0bd 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -1,21 +1,29 @@ # coding: utf-8 from __future__ import unicode_literals -import re -import json +import base64 +import functools +import hashlib import itertools +import json +import random +import re +import string from .common import InfoExtractor - -from ..compat import compat_str +from ..compat import compat_struct_pack from ..utils import ( - ExtractorError, determine_ext, + error_to_compat_str, + ExtractorError, int_or_none, + mimetype2ext, + OnDemandPagedList, parse_iso8601, sanitized_Request, str_to_int, unescapeHTML, + urlencode_postdata, ) @@ -37,7 +45,7 @@ class DailymotionBaseInfoExtractor(InfoExtractor): class DailymotionIE(DailymotionBaseInfoExtractor): - _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P[^/?_]+)' + _VALID_URL = r'(?i)https?://(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(?:(?:embed|swf|#)/)?video|swf)/(?P[^/?_]+)' IE_NAME = 'dailymotion' _FORMATS = [ @@ -48,59 +56,88 @@ class DailymotionIE(DailymotionBaseInfoExtractor): ('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', - 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News', - 'description': 'Several come bundled with the Steam Controller.', - 'thumbnail': 're:^https?:.*\.(?:jpg|png)$', - 'duration': 74, - 'timestamp': 1425657362, - 'upload_date': '20150306', - 'uploader': 'IGN', - 'uploader_id': 'xijv66', - 'age_limit': 0, - 'view_count': int, - 'comment_count': int, - } + _TESTS = [{ + 'url': 'http://www.dailymotion.com/video/x5kesuj_office-christmas-party-review-jason-bateman-olivia-munn-t-j-miller_news', + 'md5': '074b95bdee76b9e3654137aee9c79dfe', + 'info_dict': { + 'id': 'x5kesuj', + 'ext': 'mp4', + 'title': 'Office Christmas Party Review – Jason Bateman, Olivia Munn, T.J. Miller', + 'description': 'Office Christmas Party Review - Jason Bateman, Olivia Munn, T.J. Miller', + 'thumbnail': r're:^https?:.*\.(?:jpg|png)$', + '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', + '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, - } + '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, - } - ] + '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, + }] + + @staticmethod + def _extract_urls(webpage): + # Look for embedded Dailymotion player + matches = re.findall( + r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage) + return list(map(lambda m: unescapeHTML(m[1]), matches)) def _real_extract(self, url): video_id = self._match_id(url) @@ -110,24 +147,44 @@ class DailymotionIE(DailymotionBaseInfoExtractor): age_limit = self._rta_search(webpage) - description = self._og_search_description(webpage) or self._html_search_meta( + description = self._og_search_description( + webpage, default=None) or self._html_search_meta( 'description', webpage, 'description') - view_count = str_to_int(self._search_regex( - [r']+itemprop="interactionCount"[^>]+content="UserPlays:(\d+)"', - r'video_views_count[^>]+>\s+([\d\.,]+)'], - webpage, 'view count', fatal=False)) + 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', fatal=False)) + webpage, 'comment count', default=None)) player_v5 = self._search_regex( - [r'buildPlayer\(({.+?})\);', r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);'], + [r'buildPlayer\(({.+?})\);\n', # See https://github.com/rg3/youtube-dl/issues/7826 + r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);', + r'buildPlayer\(({.+?})\);', + r'var\s+config\s*=\s*({.+?});', + # New layout regex (see https://github.com/rg3/youtube-dl/issues/13580) + r'__PLAYER_CONFIG__\s*=\s*({.+?});'], webpage, 'player v5', default=None) if player_v5: player = self._parse_json(player_v5, video_id) metadata = player['metadata'] + if metadata.get('error', {}).get('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) + self._check_error(metadata) formats = [] @@ -139,21 +196,22 @@ class DailymotionIE(DailymotionBaseInfoExtractor): type_ = media.get('type') if type_ == 'application/vnd.lumberjack.manifest': continue - ext = determine_ext(media_url) - if type_ == 'application/x-mpegURL' or ext == 'm3u8': + ext = mimetype2ext(type_) or determine_ext(media_url) + if ext == 'm3u8': m3u8_formats = self._extract_m3u8_formats( - media_url, video_id, 'mp4', m3u8_id='hls', fatal=False) - if m3u8_formats: - formats.extend(m3u8_formats) - elif type_ == 'application/f4m' or ext == 'f4m': - f4m_formats = self._extract_f4m_formats( - media_url, video_id, preference=-1, f4m_id='hds', fatal=False) - if f4m_formats: - formats.extend(f4m_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': quality, + 'format_id': 'http-%s' % quality, + 'ext': ext, } m = re.search(r'H264-(?P\d+)x(?P\d+)', media_url) if m: @@ -172,11 +230,13 @@ class DailymotionIE(DailymotionBaseInfoExtractor): uploader_id = metadata.get('owner', {}).get('id') subtitles = {} - for subtitle_lang, subtitle in metadata.get('subtitles', {}).get('data', {}).items(): - subtitles[subtitle_lang] = [{ - 'ext': determine_ext(subtitle_url), - 'url': subtitle_url, - } for subtitle_url in subtitle.get('urls', [])] + 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, @@ -196,7 +256,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor): # vevo embed vevo_id = self._search_regex( - r'[\w]*)', + r'[\w]*)', webpage, 'vevo embed', default=None) if vevo_id: return self.url_result('vevo:%s' % vevo_id, 'Vevo') @@ -259,9 +319,14 @@ class DailymotionIE(DailymotionBaseInfoExtractor): } def _check_error(self, info): - if info.get('error') is not None: + 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, info['error']['title']), expected=True) + '%s said: %s' % (self.IE_NAME, title), expected=True) def _get_subtitles(self, video_id, webpage): try: @@ -269,7 +334,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor): '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)) + 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): @@ -281,56 +346,73 @@ class DailymotionIE(DailymotionBaseInfoExtractor): class DailymotionPlaylistIE(DailymotionBaseInfoExtractor): IE_NAME = 'dailymotion:playlist' - _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P.+?)/' - _MORE_PAGES_INDICATOR = r'(?s)
.*?x[0-9a-z]+)' _TESTS = [{ 'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q', 'info_dict': { 'title': 'SPORT', - 'id': 'xv4bw_nqtv_sport', + 'id': 'xv4bw', }, 'playlist_mincount': 20, }] - - def _extract_entries(self, id): - video_ids = set() - processed_urls = set() - for pagenum in itertools.count(1): - page_url = self._PAGE_TEMPLATE % (id, pagenum) - webpage, urlh = self._download_webpage_handle_no_ff( - page_url, id, 'Downloading page %s' % pagenum) - if urlh.geturl() in processed_urls: - self.report_warning('Stopped at duplicated page %s, which is the same as %s' % ( - page_url, urlh.geturl()), id) - break - - processed_urls.add(urlh.geturl()) - - for video_id in re.findall(r'data-xid="(.+?)"', webpage): - if video_id not in video_ids: - yield self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion') - video_ids.add(video_id) - - if re.search(self._MORE_PAGES_INDICATOR, webpage) is None: - break + _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') + playlist_id = self._match_id(url) webpage = self._download_webpage(url, playlist_id) - - return { - '_type': 'playlist', - 'id': playlist_id, - 'title': self._og_search_title(webpage), - 'entries': self._extract_entries(playlist_id), - } - - -class DailymotionUserIE(DailymotionPlaylistIE): + 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|#|video|playlist)/)(?:(?:old/)?user/)?(?P[^/]+)' + _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P[^/]+)' + _MORE_PAGES_INDICATOR = r'(?s)
.*?[^/?]+)' % _VALID_URL_PREFIX - _VALID_EMBED_URL = r'%s[^/]+/[^\'"]+' % _VALID_URL_PREFIX - - _TESTS = [{ - # From http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html - # Tested at FranceTvInfo_2 - 'url': 'http://api.dmcloud.net/embed/4e7343f894a6f677b10006b4/556e03339473995ee145930c?auth=1464865870-0-jyhsm84b-ead4c701fb750cf9367bf4447167a3db&autoplay=1', - 'only_matching': True, - }, { - # http://www.francetvinfo.fr/societe/larguez-les-amarres-le-cobaturage-se-developpe_980101.html - 'url': 'http://api.dmcloud.net/player/embed/4e7343f894a6f677b10006b4/559545469473996d31429f06?auth=1467430263-0-90tglw2l-a3a4b64ed41efe48d7fccad85b8b8fda&autoplay=1', - 'only_matching': True, - }] - - @classmethod - def _extract_dmcloud_url(self, webpage): - mobj = re.search(r']+src=[\'"](%s)[\'"]' % self._VALID_EMBED_URL, webpage) - if mobj: - return mobj.group(1) - - mobj = re.search( - r']+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % self._VALID_EMBED_URL, - webpage) - if mobj: - return mobj.group(1) - - def _real_extract(self, url): - video_id = self._match_id(url) - - webpage = self._download_webpage_no_ff(url, video_id) - - title = self._html_search_regex(r'([^>]+)', webpage, 'title') - - video_info = self._parse_json(self._search_regex( - r'var\s+info\s*=\s*([^;]+);', webpage, 'video info'), video_id) - - # TODO: parse ios_url, which is in fact a manifest - video_url = video_info['mp4_url'] - - return { - 'id': video_id, - 'url': video_url, - 'title': title, - 'thumbnail': video_info.get('thumbnail_url'), - }