X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/99b46abf79266c2bf5eabc291dc4f366e61b5dde..4d055bc6eb5dbe9a42216ec3e659d8ff1f2a1256:/youtube_dl/extractor/vimeo.py diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index 7e854f3..ddf375c 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -1,12 +1,15 @@ -# encoding: utf-8 +# coding: utf-8 from __future__ import unicode_literals +import base64 +import functools import json import re import itertools from .common import InfoExtractor from ..compat import ( + compat_kwargs, compat_HTTPError, compat_str, compat_urlparse, @@ -14,19 +17,22 @@ from ..compat import ( from ..utils import ( determine_ext, ExtractorError, - InAdvancePagedList, + js_to_json, int_or_none, + merge_dicts, NO_DEFAULT, + OnDemandPagedList, + parse_filesize, + qualities, RegexNotFoundError, sanitized_Request, smuggle_url, std_headers, - unified_strdate, + try_get, + unified_timestamp, unsmuggle_url, urlencode_postdata, unescapeHTML, - parse_filesize, - try_get, ) @@ -36,26 +42,35 @@ class VimeoBaseInfoExtractor(InfoExtractor): _LOGIN_URL = 'https://vimeo.com/log_in' def _login(self): - (username, password) = self._get_login_info() + username, password = self._get_login_info() if username is None: if self._LOGIN_REQUIRED: raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True) return - self.report_login() - webpage = self._download_webpage(self._LOGIN_URL, None, False) + webpage = self._download_webpage( + self._LOGIN_URL, None, 'Downloading login page') token, vuid = self._extract_xsrft_and_vuid(webpage) - data = urlencode_postdata({ + data = { 'action': 'login', 'email': username, 'password': password, 'service': 'vimeo', 'token': token, - }) - login_request = sanitized_Request(self._LOGIN_URL, data) - login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - login_request.add_header('Referer', self._LOGIN_URL) + } self._set_vimeo_cookie('vuid', vuid) - self._download_webpage(login_request, None, False, 'Wrong login info') + try: + self._download_webpage( + self._LOGIN_URL, None, 'Logging in', + data=urlencode_postdata(data), headers={ + 'Content-Type': 'application/x-www-form-urlencoded', + 'Referer': self._LOGIN_URL, + }) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 418: + raise ExtractorError( + 'Unable to log in: bad username or password', + expected=True) + raise ExtractorError('Unable to log in') def _verify_video_password(self, url, video_id, webpage): password = self._downloader.params.get('videopassword') @@ -86,35 +101,29 @@ class VimeoBaseInfoExtractor(InfoExtractor): webpage, 'vuid', group='vuid') return xsrft, vuid + def _extract_vimeo_config(self, webpage, video_id, *args, **kwargs): + vimeo_config = self._search_regex( + r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', + webpage, 'vimeo config', *args, **compat_kwargs(kwargs)) + if vimeo_config: + return self._parse_json(vimeo_config, video_id) + def _set_vimeo_cookie(self, name, value): self._set_cookie('vimeo.com', name, value) def _vimeo_sort_formats(self, formats): # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps # at the same time without actual units specified. This lead to wrong sorting. - self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'format_id')) + self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'tbr', 'format_id')) def _parse_config(self, config, video_id): - # Extract title - video_title = config['video']['title'] - - # Extract uploader, uploader_url and uploader_id - video_uploader = config['video'].get('owner', {}).get('name') - video_uploader_url = config['video'].get('owner', {}).get('url') - video_uploader_id = video_uploader_url.split('/')[-1] if video_uploader_url else None - - # Extract video thumbnail - video_thumbnail = config['video'].get('thumbnail') - if video_thumbnail is None: - video_thumbs = config['video'].get('thumbs') - if video_thumbs and isinstance(video_thumbs, dict): - _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1] - - # Extract video duration - video_duration = int_or_none(config['video'].get('duration')) + video_data = config['video'] + video_title = video_data['title'] + live_event = video_data.get('live_event') or {} + is_live = live_event.get('status') == 'started' formats = [] - config_files = config['video'].get('files') or config['request'].get('files', {}) + config_files = video_data.get('files') or config['request'].get('files', {}) for f in config_files.get('progressive', []): video_url = f.get('url') if not video_url: @@ -127,10 +136,53 @@ class VimeoBaseInfoExtractor(InfoExtractor): 'fps': int_or_none(f.get('fps')), 'tbr': int_or_none(f.get('bitrate')), }) - m3u8_url = config_files.get('hls', {}).get('url') - if m3u8_url: - formats.extend(self._extract_m3u8_formats( - m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) + + # TODO: fix handling of 308 status code returned for live archive manifest requests + for files_type in ('hls', 'dash'): + for cdn_name, cdn_data in config_files.get(files_type, {}).get('cdns', {}).items(): + manifest_url = cdn_data.get('url') + if not manifest_url: + continue + format_id = '%s-%s' % (files_type, cdn_name) + if files_type == 'hls': + formats.extend(self._extract_m3u8_formats( + manifest_url, video_id, 'mp4', + 'm3u8' if is_live else 'm3u8_native', m3u8_id=format_id, + note='Downloading %s m3u8 information' % cdn_name, + fatal=False)) + elif files_type == 'dash': + mpd_pattern = r'/%s/(?:sep/)?video/' % video_id + mpd_manifest_urls = [] + if re.search(mpd_pattern, manifest_url): + for suffix, repl in (('', 'video'), ('_sep', 'sep/video')): + mpd_manifest_urls.append((format_id + suffix, re.sub( + mpd_pattern, '/%s/%s/' % (video_id, repl), manifest_url))) + else: + mpd_manifest_urls = [(format_id, manifest_url)] + for f_id, m_url in mpd_manifest_urls: + if 'json=1' in m_url: + real_m_url = (self._download_json(m_url, video_id, fatal=False) or {}).get('url') + if real_m_url: + m_url = real_m_url + mpd_formats = self._extract_mpd_formats( + m_url.replace('/master.json', '/master.mpd'), video_id, f_id, + 'Downloading %s MPD information' % cdn_name, + fatal=False) + for f in mpd_formats: + if f.get('vcodec') == 'none': + f['preference'] = -50 + elif f.get('acodec') == 'none': + f['preference'] = -40 + formats.extend(mpd_formats) + + live_archive = live_event.get('archive') or {} + live_archive_source_url = live_archive.get('source_url') + if live_archive_source_url and live_archive.get('status') == 'done': + formats.append({ + 'format_id': 'live-archive-source', + 'url': live_archive_source_url, + 'preference': 1, + }) subtitles = {} text_tracks = config['request'].get('text_tracks') @@ -141,17 +193,61 @@ class VimeoBaseInfoExtractor(InfoExtractor): 'url': 'https://vimeo.com' + tt['url'], }] + thumbnails = [] + if not is_live: + for key, thumb in video_data.get('thumbs', {}).items(): + thumbnails.append({ + 'id': key, + 'width': int_or_none(key), + 'url': thumb, + }) + thumbnail = video_data.get('thumbnail') + if thumbnail: + thumbnails.append({ + 'url': thumbnail, + }) + + owner = video_data.get('owner') or {} + video_uploader_url = owner.get('url') + return { - 'title': video_title, - 'uploader': video_uploader, - 'uploader_id': video_uploader_id, + 'title': self._live_title(video_title) if is_live else video_title, + 'uploader': owner.get('name'), + 'uploader_id': video_uploader_url.split('/')[-1] if video_uploader_url else None, 'uploader_url': video_uploader_url, - 'thumbnail': video_thumbnail, - 'duration': video_duration, + 'thumbnails': thumbnails, + 'duration': int_or_none(video_data.get('duration')), 'formats': formats, 'subtitles': subtitles, + 'is_live': is_live, } + def _extract_original_format(self, url, video_id): + download_data = self._download_json( + url, video_id, fatal=False, + query={'action': 'load_download_config'}, + headers={'X-Requested-With': 'XMLHttpRequest'}) + if download_data: + source_file = download_data.get('source_file') + if isinstance(source_file, dict): + download_url = source_file.get('download_url') + if download_url and not source_file.get('is_cold') and not source_file.get('is_defrosting'): + source_name = source_file.get('public_name', 'Original') + if self._is_valid_url(download_url, video_id, '%s video' % source_name): + ext = (try_get( + source_file, lambda x: x['extension'], + compat_str) or determine_ext( + download_url, None) or 'mp4').lower() + return { + 'url': download_url, + 'ext': ext, + 'width': int_or_none(source_file.get('width')), + 'height': int_or_none(source_file.get('height')), + 'filesize': parse_filesize(source_file.get('size')), + 'format_id': source_name, + 'preference': 1, + } + class VimeoIE(VimeoBaseInfoExtractor): """Information extractor for vimeo.com.""" @@ -167,7 +263,7 @@ class VimeoIE(VimeoBaseInfoExtractor): \. )? vimeo(?Ppro)?\.com/ - (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/) + (?!(?:channels|album|showcase)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/) (?:.*?/)? (?: (?: @@ -188,12 +284,14 @@ class VimeoIE(VimeoBaseInfoExtractor): 'id': '56015672', 'ext': 'mp4', 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", - 'description': 'md5:2d3305bad981a06ff79f027f19865021', + 'description': 'md5:509a9ad5c9bf97c60faee9203aca4479', + 'timestamp': 1355990239, 'upload_date': '20121220', - 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user7108434', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user7108434', 'uploader_id': 'user7108434', 'uploader': 'Filippo Valsorda', 'duration': 10, + 'license': 'by-sa', }, }, { @@ -203,7 +301,7 @@ class VimeoIE(VimeoBaseInfoExtractor): 'info_dict': { 'id': '68093876', 'ext': 'mp4', - 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/openstreetmapus', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/openstreetmapus', 'uploader_id': 'openstreetmapus', 'uploader': 'OpenStreetMap US', 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography', @@ -220,7 +318,7 @@ class VimeoIE(VimeoBaseInfoExtractor): 'ext': 'mp4', 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012', 'uploader': 'The BLN & Business of Software', - 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware', 'uploader_id': 'theblnbusinessofsoftware', 'duration': 3610, 'description': None, @@ -234,12 +332,13 @@ class VimeoIE(VimeoBaseInfoExtractor): 'id': '68375962', 'ext': 'mp4', 'title': 'youtube-dl password protected test video', + 'timestamp': 1371200155, 'upload_date': '20130614', - 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user18948128', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user18948128', 'uploader_id': 'user18948128', 'uploader': 'Jaime Marquínez Ferrándiz', 'duration': 10, - 'description': 'This is "youtube-dl password protected test video" by on Vimeo, the home for high quality videos and the people who love them.', + 'description': 'md5:dca3ea23adb29ee387127bc4ddfce63f', }, 'params': { 'videopassword': 'youtube-dl', @@ -253,12 +352,16 @@ class VimeoIE(VimeoBaseInfoExtractor): 'ext': 'mp4', 'title': 'Key & Peele: Terrorist Interrogation', 'description': 'md5:8678b246399b070816b12313e8b4eb5c', - 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/atencio', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/atencio', 'uploader_id': 'atencio', 'uploader': 'Peter Atencio', - 'upload_date': '20130927', + 'channel_id': 'keypeele', + 'channel_url': r're:https?://(?:www\.)?vimeo\.com/channels/keypeele', + 'timestamp': 1380339469, + 'upload_date': '20130928', 'duration': 187, }, + 'expected_warnings': ['Unable to download JSON metadata'], }, { 'url': 'http://vimeo.com/76979871', @@ -268,8 +371,9 @@ class VimeoIE(VimeoBaseInfoExtractor): 'ext': 'mp4', 'title': 'The New Vimeo Player (You Know, For Videos)', 'description': 'md5:2ec900bf97c3f389378a96aee11260ea', + 'timestamp': 1381846109, 'upload_date': '20131015', - 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/staff', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/staff', 'uploader_id': 'staff', 'uploader': 'Vimeo Staff', 'duration': 62, @@ -284,21 +388,22 @@ class VimeoIE(VimeoBaseInfoExtractor): 'ext': 'mp4', 'title': 'Pier Solar OUYA Official Trailer', 'uploader': 'Tulio Gonçalves', - 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user28849593', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user28849593', 'uploader_id': 'user28849593', }, }, { # contains original format 'url': 'https://vimeo.com/33951933', - 'md5': '2d9f5475e0537f013d0073e812ab89e6', + 'md5': '53c688fa95a55bf4b7293d37a89c5c53', 'info_dict': { 'id': '33951933', 'ext': 'mp4', 'title': 'FOX CLASSICS - Forever Classic ID - A Full Minute', 'uploader': 'The DMCI', - 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/dmci', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/dmci', 'uploader_id': 'dmci', + 'timestamp': 1324343742, 'upload_date': '20111220', 'description': 'md5:ae23671e82d05415868f7ad1aec21147', }, @@ -312,8 +417,11 @@ class VimeoIE(VimeoBaseInfoExtractor): 'ext': 'mp4', 'title': 'Vimeo Tribute: The Shining', 'uploader': 'Casey Donahue', - 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/caseydonahue', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/caseydonahue', 'uploader_id': 'caseydonahue', + 'channel_url': r're:https?://(?:www\.)?vimeo\.com/channels/tributes', + 'channel_id': 'tributes', + 'timestamp': 1250886430, 'upload_date': '20090821', 'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6', }, @@ -322,6 +430,38 @@ class VimeoIE(VimeoBaseInfoExtractor): }, 'expected_warnings': ['Unable to download JSON metadata'], }, + { + # redirects to ondemand extractor and should be passed through it + # for successful extraction + 'url': 'https://vimeo.com/73445910', + 'info_dict': { + 'id': '73445910', + 'ext': 'mp4', + 'title': 'The Reluctant Revolutionary', + 'uploader': '10Ft Films', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/tenfootfilms', + 'uploader_id': 'tenfootfilms', + }, + 'params': { + 'skip_download': True, + }, + }, + { + 'url': 'http://player.vimeo.com/video/68375962', + 'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7', + 'info_dict': { + 'id': '68375962', + 'ext': 'mp4', + 'title': 'youtube-dl password protected test video', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user18948128', + 'uploader_id': 'user18948128', + 'uploader': 'Jaime Marquínez Ferrándiz', + 'duration': 10, + }, + 'params': { + 'videopassword': 'youtube-dl', + }, + }, { 'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741', 'only_matching': True, @@ -348,40 +488,54 @@ class VimeoIE(VimeoBaseInfoExtractor): 'url': 'https://vimeo.com/160743502/abd0e13fb4', 'only_matching': True, } + # https://gettingthingsdone.com/workflowmap/ + # vimeo embed with check-password page protected by Referer header ] @staticmethod - def _extract_vimeo_url(url, webpage): + def _smuggle_referrer(url, referrer_url): + return smuggle_url(url, {'http_headers': {'Referer': referrer_url}}) + + @staticmethod + def _extract_urls(url, webpage): + urls = [] # Look for embedded (iframe) Vimeo player - mobj = re.search( - r']+?src=(["\'])(?P(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage) - if mobj: - player_url = unescapeHTML(mobj.group('url')) - surl = smuggle_url(player_url, {'http_headers': {'Referer': url}}) - return surl - # Look for embedded (swf embed) Vimeo player - mobj = re.search( - r']+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage) - if mobj: - return mobj.group(1) - # Look more for non-standard embedded Vimeo player - mobj = re.search( - r']+src=(?P[\'"])(?P(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)(?P=q1)', webpage) - if mobj: - return mobj.group('url') - - def _verify_player_video_password(self, url, video_id): + for mobj in re.finditer( + r']+?src=(["\'])(?P(?:https?:)?//player\.vimeo\.com/video/\d+.*?)\1', + webpage): + urls.append(VimeoIE._smuggle_referrer(unescapeHTML(mobj.group('url')), url)) + PLAIN_EMBED_RE = ( + # Look for embedded (swf embed) Vimeo player + r']+?src=(["\'])(?P(?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)\1', + # Look more for non-standard embedded Vimeo player + r']+src=(["\'])(?P(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)\1', + ) + for embed_re in PLAIN_EMBED_RE: + for mobj in re.finditer(embed_re, webpage): + urls.append(mobj.group('url')) + return urls + + @staticmethod + def _extract_url(url, webpage): + urls = VimeoIE._extract_urls(url, webpage) + return urls[0] if urls else None + + def _verify_player_video_password(self, url, video_id, headers): password = self._downloader.params.get('videopassword') if password is None: raise ExtractorError('This video is protected by a password, use the --video-password option') - data = urlencode_postdata({'password': password}) - pass_url = url + '/check-password' - password_request = sanitized_Request(pass_url, data) - password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - password_request.add_header('Referer', url) - return self._download_json( - password_request, video_id, - 'Verifying the password', 'Wrong password') + data = urlencode_postdata({ + 'password': base64.b64encode(password.encode()), + }) + headers = merge_dicts(headers, { + 'Content-Type': 'application/x-www-form-urlencoded', + }) + checked = self._download_json( + url + '/check-password', video_id, + 'Verifying the password', data=data, headers=headers) + if checked is False: + raise ExtractorError('Wrong video password', expected=True) + return checked def _real_initialize(self): self._login() @@ -394,11 +548,18 @@ class VimeoIE(VimeoBaseInfoExtractor): if 'Referer' not in headers: headers['Referer'] = url + channel_id = self._search_regex( + r'vimeo\.com/channels/([^/]+)', url, 'channel id', default=None) + # Extract ID from URL mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') orig_url = url - if mobj.group('pro') or mobj.group('player'): + if mobj.group('pro'): + # some videos require portfolio_id to be present in player url + # https://github.com/ytdl-org/youtube-dl/issues/20070 + url = self._extract_url(url, self._download_webpage(url, video_id)) + elif mobj.group('player'): url = 'https://player.vimeo.com/video/' + video_id elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')): url = 'https://vimeo.com/' + video_id @@ -406,7 +567,13 @@ class VimeoIE(VimeoBaseInfoExtractor): # Retrieve video webpage to extract further information request = sanitized_Request(url, headers=headers) try: - webpage = self._download_webpage(request, video_id) + webpage, urlh = self._download_webpage_handle(request, video_id) + redirect_url = compat_str(urlh.geturl()) + # Some URLs redirect to ondemand can't be extracted with + # this extractor right away thus should be passed through + # ondemand extractor (e.g. https://vimeo.com/73445910) + if VimeoOndemandIE.suitable(redirect_url): + return self.url_result(redirect_url, VimeoOndemandIE.ie_key()) except ExtractorError as ee: if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403: errmsg = ee.cause.read() @@ -423,16 +590,17 @@ class VimeoIE(VimeoBaseInfoExtractor): # and latter we extract those that are Vimeo specific. self.report_extraction(video_id) - vimeo_config = self._search_regex( - r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage, - 'vimeo config', default=None) + vimeo_config = self._extract_vimeo_config(webpage, video_id, default=None) if vimeo_config: - seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {}) + seed_status = vimeo_config.get('seed_status', {}) if seed_status.get('state') == 'failed': raise ExtractorError( '%s said: %s' % (self.IE_NAME, seed_status['title']), expected=True) + cc_license = None + timestamp = None + # Extract the config JSON try: try: @@ -442,22 +610,28 @@ class VimeoIE(VimeoBaseInfoExtractor): if not config_url: # Sometimes new react-based page is served instead of old one that require # different config URL extraction approach (see - # https://github.com/rg3/youtube-dl/pull/7209) + # https://github.com/ytdl-org/youtube-dl/pull/7209) vimeo_clip_page_config = self._search_regex( r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage, 'vimeo clip page config') - config_url = self._parse_json( - vimeo_clip_page_config, video_id)['player']['config_url'] + page_config = self._parse_json(vimeo_clip_page_config, video_id) + config_url = page_config['player']['config_url'] + cc_license = page_config.get('cc_license') + timestamp = try_get( + page_config, lambda x: x['clip']['uploaded_on'], + compat_str) config_json = self._download_webpage(config_url, video_id) config = json.loads(config_json) except RegexNotFoundError: # For pro videos or player.vimeo.com urls # We try to find out to which variable is assigned the config dic - m_variable_name = re.search('(\w)\.video\.id', webpage) + m_variable_name = re.search(r'(\w)\.video\.id', webpage) if m_variable_name is not None: - config_re = r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1)) + config_re = [r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))] else: config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});'] + config_re.append(r'\bvar\s+r\s*=\s*({.+?})\s*;') + config_re.append(r'\bconfig\s*=\s*({.+?})\s*;') config = self._search_regex(config_re, webpage, 'info section', flags=re.DOTALL) config = json.loads(config) @@ -468,29 +642,33 @@ class VimeoIE(VimeoBaseInfoExtractor): if re.search(r']+?id="pw_form"', webpage) is not None: if '_video_password_verified' in data: raise ExtractorError('video password verification failed!') - self._verify_video_password(url, video_id, webpage) + self._verify_video_password(redirect_url, video_id, webpage) return self._real_extract( - smuggle_url(url, {'_video_password_verified': 'verified'})) + smuggle_url(redirect_url, {'_video_password_verified': 'verified'})) else: raise ExtractorError('Unable to extract info section', cause=e) else: if config.get('view') == 4: - config = self._verify_player_video_password(url, video_id) + config = self._verify_player_video_password(redirect_url, video_id, headers) + + vod = config.get('video', {}).get('vod', {}) def is_rented(): if '>You rented this title.<' in webpage: return True if config.get('user', {}).get('purchased'): return True - label = try_get( - config, lambda x: x['video']['vod']['purchase_options'][0]['label_string'], compat_str) - if label and label.startswith('You rented this'): - return True + for purchase_option in vod.get('purchase_options', []): + if purchase_option.get('purchased'): + return True + label = purchase_option.get('label_string') + if label and (label.startswith('You rented this') or label.endswith(' remaining')): + return True return False - if is_rented(): - feature_id = config.get('video', {}).get('vod', {}).get('feature_id') + if is_rented() and vod.get('is_trailer'): + feature_id = vod.get('feature_id') if feature_id and not data.get('force_feature_id', False): return self.url_result(smuggle_url( 'https://player.vimeo.com/player/%s' % feature_id, @@ -516,10 +694,10 @@ class VimeoIE(VimeoBaseInfoExtractor): self._downloader.report_warning('Cannot find video description') # Extract upload date - video_upload_date = None - mobj = re.search(r']+datetime="([^"]+)"', webpage) - if mobj is not None: - video_upload_date = unified_strdate(mobj.group(1)) + if not timestamp: + timestamp = self._search_regex( + r']+datetime="([^"]+)"', webpage, + 'timestamp', default=None) try: view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count')) @@ -532,40 +710,40 @@ class VimeoIE(VimeoBaseInfoExtractor): comment_count = None formats = [] - download_request = sanitized_Request('https://vimeo.com/%s?action=load_download_config' % video_id, headers={ - 'X-Requested-With': 'XMLHttpRequest'}) - download_data = self._download_json(download_request, video_id, fatal=False) - if download_data: - source_file = download_data.get('source_file') - if isinstance(source_file, dict): - download_url = source_file.get('download_url') - if download_url and not source_file.get('is_cold') and not source_file.get('is_defrosting'): - source_name = source_file.get('public_name', 'Original') - if self._is_valid_url(download_url, video_id, '%s video' % source_name): - ext = source_file.get('extension', determine_ext(download_url)).lower() - formats.append({ - 'url': download_url, - 'ext': ext, - 'width': int_or_none(source_file.get('width')), - 'height': int_or_none(source_file.get('height')), - 'filesize': parse_filesize(source_file.get('size')), - 'format_id': source_name, - 'preference': 1, - }) - info_dict = self._parse_config(config, video_id) - formats.extend(info_dict['formats']) + source_format = self._extract_original_format( + 'https://vimeo.com/' + video_id, video_id) + if source_format: + formats.append(source_format) + + info_dict_config = self._parse_config(config, video_id) + formats.extend(info_dict_config['formats']) self._vimeo_sort_formats(formats) - info_dict.update({ + + json_ld = self._search_json_ld(webpage, video_id, default={}) + + if not cc_license: + cc_license = self._search_regex( + r']+rel=["\']license["\'][^>]+href=(["\'])(?P(?:(?!\1).)+)\1', + webpage, 'license', default=None, group='license') + + channel_url = 'https://vimeo.com/channels/%s' % channel_id if channel_id else None + + info_dict = { 'id': video_id, 'formats': formats, - 'upload_date': video_upload_date, + 'timestamp': unified_timestamp(timestamp), 'description': video_description, 'webpage_url': url, 'view_count': view_count, 'like_count': like_count, 'comment_count': comment_count, - }) + 'license': cc_license, + 'channel_id': channel_id, + 'channel_url': channel_url, + } + + info_dict = merge_dicts(info_dict, info_dict_config, json_ld) return info_dict @@ -582,9 +760,26 @@ class VimeoOndemandIE(VimeoBaseInfoExtractor): 'ext': 'mp4', 'title': 'המעבדה - במאי יותם פלדמן', 'uploader': 'גם סרטים', - 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/gumfilms', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/gumfilms', 'uploader_id': 'gumfilms', }, + 'params': { + 'format': 'best[protocol=https]', + }, + }, { + # requires Referer to be passed along with og:video:url + 'url': 'https://vimeo.com/ondemand/36938/126682985', + 'info_dict': { + 'id': '126682985', + 'ext': 'mp4', + 'title': 'Rävlock, rätt läte på rätt plats', + 'uploader': 'Lindroth & Norin', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user14430847', + 'uploader_id': 'user14430847', + }, + 'params': { + 'skip_download': True, + }, }, { 'url': 'https://vimeo.com/ondemand/nazmaalik', 'only_matching': True, @@ -599,7 +794,12 @@ class VimeoOndemandIE(VimeoBaseInfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - return self.url_result(self._og_search_video_url(webpage), VimeoIE.ie_key()) + return self.url_result( + # Some videos require Referer to be passed along with og:video:url + # similarly to generic vimeo embeds (e.g. + # https://vimeo.com/ondemand/36938/126682985). + VimeoIE._smuggle_referrer(self._og_search_video_url(webpage), url), + VimeoIE.ie_key()) class VimeoChannelIE(VimeoBaseInfoExtractor): @@ -621,7 +821,8 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): return '%s/videos/page:%d/' % (base_url, pagenum) def _extract_list_title(self, webpage): - return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title') + return self._TITLE or self._html_search_regex( + self._TITLE_RE, webpage, 'list title', fatal=False) def _login_list_password(self, page_url, list_id, webpage): login_form = self._search_regex( @@ -664,12 +865,12 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): # Try extracting href first since not all videos are available via # short https://vimeo.com/id URL (e.g. https://vimeo.com/channels/tributes/6213729) clips = re.findall( - r'id="clip_(\d+)"[^>]*>\s*]+href="(/(?:[^/]+/)*\1)', webpage) + r'id="clip_(\d+)"[^>]*>\s*]+href="(/(?:[^/]+/)*\1)(?:[^>]+\btitle="([^"]+)")?', webpage) if clips: - for video_id, video_url in clips: + for video_id, video_url, video_title in clips: yield self.url_result( compat_urlparse.urljoin(base_url, video_url), - VimeoIE.ie_key(), video_id=video_id) + VimeoIE.ie_key(), video_id=video_id, video_title=video_title) # More relaxed fallback else: for video_id in re.findall(r'id=["\']clip_(\d+)', webpage): @@ -712,7 +913,7 @@ class VimeoUserIE(VimeoChannelIE): class VimeoAlbumIE(VimeoChannelIE): IE_NAME = 'vimeo:album' - _VALID_URL = r'https://vimeo\.com/album/(?P\d+)(?:$|[?#]|/(?!video))' + _VALID_URL = r'https://vimeo\.com/(?:album|showcase)/(?P\d+)(?:$|[?#]|/(?!video))' _TITLE_RE = r'
\n\s*

(.*?)

' _TESTS = [{ 'url': 'https://vimeo.com/album/2632481', @@ -732,21 +933,39 @@ class VimeoAlbumIE(VimeoChannelIE): 'params': { 'videopassword': 'youtube-dl', } - }, { - 'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail', - 'only_matching': True, - }, { - # TODO: respect page number - 'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail', - 'only_matching': True, }] - - def _page_url(self, base_url, pagenum): - return '%s/page:%d/' % (base_url, pagenum) + _PAGE_SIZE = 100 + + def _fetch_page(self, album_id, authorizaion, hashed_pass, page): + api_page = page + 1 + query = { + 'fields': 'link', + 'page': api_page, + 'per_page': self._PAGE_SIZE, + } + if hashed_pass: + query['_hashed_pass'] = hashed_pass + videos = self._download_json( + 'https://api.vimeo.com/albums/%s/videos' % album_id, + album_id, 'Downloading page %d' % api_page, query=query, headers={ + 'Authorization': 'jwt ' + authorizaion, + })['data'] + for video in videos: + link = video.get('link') + if not link: + continue + yield self.url_result(link, VimeoIE.ie_key(), VimeoIE._match_id(link)) def _real_extract(self, url): album_id = self._match_id(url) - return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id) + webpage = self._download_webpage(url, album_id) + webpage = self._login_list_password(url, album_id, webpage) + api_config = self._extract_vimeo_config(webpage, album_id)['api'] + entries = OnDemandPagedList(functools.partial( + self._fetch_page, album_id, api_config['jwt'], + api_config.get('hashed_pass')), self._PAGE_SIZE) + return self.playlist_result(entries, album_id, self._html_search_regex( + r'\s*(.+?)(?:\s+on Vimeo)?', webpage, 'title', fatal=False)) class VimeoGroupsIE(VimeoAlbumIE): @@ -762,7 +981,7 @@ class VimeoGroupsIE(VimeoAlbumIE): }] def _extract_list_title(self, webpage): - return self._og_search_title(webpage) + return self._og_search_title(webpage, fatal=False) def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -773,7 +992,7 @@ class VimeoGroupsIE(VimeoAlbumIE): class VimeoReviewIE(VimeoBaseInfoExtractor): IE_NAME = 'vimeo:review' IE_DESC = 'Review pages on vimeo' - _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P[^/]+)' + _VALID_URL = r'(?Phttps://vimeo\.com/[^/]+/review/(?P[^/]+)/[0-9a-f]{10})' _TESTS = [{ 'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d', 'md5': 'c507a72f780cacc12b2248bb4006d253', @@ -794,7 +1013,7 @@ class VimeoReviewIE(VimeoBaseInfoExtractor): 'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn', 'uploader': 'DevWeek Events', 'duration': 2773, - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'uploader_id': 'user22258446', } }, { @@ -810,6 +1029,7 @@ class VimeoReviewIE(VimeoBaseInfoExtractor): 'params': { 'videopassword': 'holygrail', }, + 'skip': 'video gone', }] def _real_initialize(self): @@ -818,8 +1038,14 @@ class VimeoReviewIE(VimeoBaseInfoExtractor): def _get_config_url(self, webpage_url, video_id, video_password_verified=False): webpage = self._download_webpage(webpage_url, video_id) config_url = self._html_search_regex( - r'data-config-url="([^"]+)"', webpage, 'config URL', - default=NO_DEFAULT if video_password_verified else None) + r'data-config-url=(["\'])(?P(?:(?!\1).)+)\1', webpage, + 'config URL', default=None, group='url') + if not config_url: + data = self._parse_json(self._search_regex( + r'window\s*=\s*_extend\(window,\s*({.+?})\);', webpage, 'data', + default=NO_DEFAULT if video_password_verified else '{}'), video_id) + config = data.get('vimeo_esi', {}).get('config', {}) + config_url = config.get('configUrl') or try_get(config, lambda x: x['clipData']['configUrl']) if config_url is None: self._verify_video_password(webpage_url, video_id, webpage) config_url = self._get_config_url( @@ -827,10 +1053,13 @@ class VimeoReviewIE(VimeoBaseInfoExtractor): return config_url def _real_extract(self, url): - video_id = self._match_id(url) + page_url, video_id = re.match(self._VALID_URL, url).groups() config_url = self._get_config_url(url, video_id) config = self._download_json(config_url, video_id) info_dict = self._parse_config(config, video_id) + source_format = self._extract_original_format(page_url, video_id) + if source_format: + info_dict['formats'].append(source_format) self._vimeo_sort_formats(info_dict['formats']) info_dict['id'] = video_id return info_dict @@ -862,57 +1091,118 @@ class VimeoWatchLaterIE(VimeoChannelIE): return self._extract_videos('watchlater', 'https://vimeo.com/watchlater') -class VimeoLikesIE(InfoExtractor): - _VALID_URL = r'https://(?:www\.)?vimeo\.com/user(?P[0-9]+)/likes/?(?:$|[?#]|sort:)' +class VimeoLikesIE(VimeoChannelIE): + _VALID_URL = r'https://(?:www\.)?vimeo\.com/(?P[^/]+)/likes/?(?:$|[?#]|sort:)' IE_NAME = 'vimeo:likes' IE_DESC = 'Vimeo user likes' - _TEST = { + _TESTS = [{ 'url': 'https://vimeo.com/user755559/likes/', 'playlist_mincount': 293, 'info_dict': { - 'id': 'user755559_likes', - 'description': 'See all the videos urza likes', - 'title': 'Videos urza likes', + 'id': 'user755559', + 'title': 'urza’s Likes', }, - } + }, { + 'url': 'https://vimeo.com/stormlapse/likes', + 'only_matching': True, + }] + + def _page_url(self, base_url, pagenum): + return '%s/page:%d/' % (base_url, pagenum) def _real_extract(self, url): user_id = self._match_id(url) - webpage = self._download_webpage(url, user_id) - page_count = self._int( - self._search_regex( - r'''(?x)
  • - .*?
    • \s* - ''', webpage, 'page count'), - 'page count', fatal=True) - PAGE_SIZE = 12 - title = self._html_search_regex( - r'(?s)

    (.+?)

    ', webpage, 'title', fatal=False) - description = self._html_search_meta('description', webpage) - - def _get_page(idx): - page_url = 'https://vimeo.com/user%s/likes/page:%d/sort:date' % ( - user_id, idx + 1) - webpage = self._download_webpage( - page_url, user_id, - note='Downloading page %d/%d' % (idx + 1, page_count)) - video_list = self._search_regex( - r'(?s)
      ]*>(.*?)
    ', - webpage, 'video content') - paths = re.findall( - r']*>\s*\d+)' + + def _call_api(self, video_id, access_token, path='', query=None): + return self._download_json( + 'https://api.vhx.tv/videos/' + video_id + path, video_id, headers={ + 'Authorization': 'Bearer ' + access_token, + }, query=query) + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + credentials = self._parse_json(self._search_regex( + r'(?s)credentials\s*:\s*({.+?}),', webpage, + 'config'), video_id, js_to_json) + access_token = credentials['access_token'] + + query = {} + for k, v in credentials.items(): + if k in ('authorization', 'authUserToken', 'ticket') and v and v != 'undefined': + if k == 'authUserToken': + query['auth_user_token'] = v + else: + query[k] = v + files = self._call_api(video_id, access_token, '/files', query) + + formats = [] + for f in files: + href = try_get(f, lambda x: x['_links']['source']['href']) + if not href: + continue + method = f.get('method') + if method == 'hls': + formats.extend(self._extract_m3u8_formats( + href, video_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) + elif method == 'dash': + formats.extend(self._extract_mpd_formats( + href, video_id, mpd_id='dash', fatal=False)) + else: + fmt = { + 'filesize': int_or_none(try_get(f, lambda x: x['size']['bytes'])), + 'format_id': 'http', + 'preference': 1, + 'url': href, + 'vcodec': f.get('codec'), } + quality = f.get('quality') + if quality: + fmt.update({ + 'format_id': 'http-' + quality, + 'height': int_or_none(self._search_regex(r'(\d+)p', quality, 'height', default=None)), + }) + formats.append(fmt) + self._sort_formats(formats) + + video_data = self._call_api(video_id, access_token) + title = video_data.get('title') or video_data['name'] - pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE) + subtitles = {} + for subtitle in try_get(video_data, lambda x: x['tracks']['subtitles'], list) or []: + lang = subtitle.get('srclang') or subtitle.get('label') + for _link in subtitle.get('_links', {}).values(): + href = _link.get('href') + if not href: + continue + subtitles.setdefault(lang, []).append({ + 'url': href, + }) + + q = qualities(['small', 'medium', 'large', 'source']) + thumbnails = [] + for thumbnail_id, thumbnail_url in video_data.get('thumbnail', {}).items(): + thumbnails.append({ + 'id': thumbnail_id, + 'url': thumbnail_url, + 'preference': q(thumbnail_id), + }) return { - '_type': 'playlist', - 'id': 'user%s_likes' % user_id, + 'id': video_id, 'title': title, - 'description': description, - 'entries': pl, + 'description': video_data.get('description'), + 'duration': int_or_none(try_get(video_data, lambda x: x['duration']['seconds'])), + 'formats': formats, + 'subtitles': subtitles, + 'thumbnails': thumbnails, + 'timestamp': unified_timestamp(video_data.get('created_at')), + 'view_count': int_or_none(video_data.get('plays_count')), }