X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9f2b33881274af98a9145c533a1d295fad71521a..c8c68895a8818fe160e81ce9c6f8baea38fe3051:/youtube_dl/extractor/vimeo.py diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index 3049dff..8cd611e 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -1,30 +1,40 @@ -# 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, ) from ..utils import ( + clean_html, determine_ext, - encode_dict, + dict_get, ExtractorError, - InAdvancePagedList, + js_to_json, int_or_none, + merge_dicts, + OnDemandPagedList, + parse_filesize, RegexNotFoundError, sanitized_Request, smuggle_url, std_headers, - unified_strdate, + str_or_none, + try_get, + unified_timestamp, unsmuggle_url, urlencode_postdata, + urljoin, unescapeHTML, - parse_filesize, ) @@ -34,26 +44,55 @@ 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(encode_dict({ + 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) + 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') + if password is None: + raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True) + token, vuid = self._extract_xsrft_and_vuid(webpage) + data = urlencode_postdata({ + 'password': password, + 'token': token, + }) + if url.startswith('http://'): + # vimeo only supports https now, but the user can give an http url + url = url.replace('http://', 'https://') + password_request = sanitized_Request(url + '/password', data) + password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') + password_request.add_header('Referer', url) self._set_vimeo_cookie('vuid', vuid) - self._download_webpage(login_request, None, False, 'Wrong login info') + return self._download_webpage( + password_request, video_id, + 'Verifying the password', 'Wrong password') def _extract_xsrft_and_vuid(self, webpage): xsrft = self._search_regex( @@ -64,24 +103,181 @@ 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', 'tbr', 'format_id')) + + def _parse_config(self, config, video_id): + 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 = 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: + continue + formats.append({ + 'url': video_url, + 'format_id': 'http-%s' % f.get('quality'), + 'width': int_or_none(f.get('width')), + 'height': int_or_none(f.get('height')), + 'fps': int_or_none(f.get('fps')), + 'tbr': int_or_none(f.get('bitrate')), + }) + + # 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') + if text_tracks: + for tt in text_tracks: + subtitles[tt['lang']] = [{ + 'ext': 'vtt', + 'url': urljoin('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 { + 'id': str_or_none(video_data.get('id')) or video_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, + '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.""" # _VALID_URL matches Vimeo URLs _VALID_URL = r'''(?x) - https?:// - (?:(?:www|(?Pplayer))\.)? - vimeo(?Ppro)?\.com/ - (?!channels/[^/?#]+/?(?:$|[?#])|album/) - (?:.*?/)? - (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)? - (?:videos?/)? - (?P[0-9]+) - /?(?:[?&].*)?(?:[#].*)?$''' + https?:// + (?: + (?: + www| + player + ) + \. + )? + vimeo(?:pro)?\.com/ + (?!(?:channels|album|showcase)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/) + (?:.*?/)? + (?: + (?: + play_redirect_hls| + moogaloop\.swf)\?clip_id= + )? + (?:videos?/)? + (?P[0-9]+) + (?:/[\da-f]+)? + /?(?:[?&].*)?(?:[#].*)?$ + ''' IE_NAME = 'vimeo' _TESTS = [ { @@ -92,10 +288,16 @@ class VimeoIE(VimeoBaseInfoExtractor): 'ext': 'mp4', 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", 'description': 'md5:2d3305bad981a06ff79f027f19865021', + 'timestamp': 1355990239, 'upload_date': '20121220', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user7108434', 'uploader_id': 'user7108434', 'uploader': 'Filippo Valsorda', 'duration': 10, + 'license': 'by-sa', + }, + 'params': { + 'format': 'best[protocol=https]', }, }, { @@ -105,11 +307,17 @@ class VimeoIE(VimeoBaseInfoExtractor): 'info_dict': { 'id': '68093876', 'ext': 'mp4', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/openstreetmapus', 'uploader_id': 'openstreetmapus', 'uploader': 'OpenStreetMap US', 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography', - 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30', + 'description': 'md5:2c362968038d4499f4d79f88458590c1', 'duration': 1595, + 'upload_date': '20130610', + 'timestamp': 1370893156, + }, + 'params': { + 'format': 'best[protocol=https]', }, }, { @@ -121,10 +329,15 @@ 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': r're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware', 'uploader_id': 'theblnbusinessofsoftware', 'duration': 3610, 'description': None, }, + 'params': { + 'format': 'best[protocol=https]', + }, + 'expected_warnings': ['Unable to download JSON metadata'], }, { 'url': 'http://vimeo.com/68375962', @@ -134,31 +347,37 @@ class VimeoIE(VimeoBaseInfoExtractor): 'id': '68375962', 'ext': 'mp4', 'title': 'youtube-dl password protected test video', + 'timestamp': 1371200155, 'upload_date': '20130614', + '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 Jaime Marquínez Ferrándiz on Vimeo, the home for high quality videos and the people\u2026', + 'description': 'md5:dca3ea23adb29ee387127bc4ddfce63f', }, 'params': { + 'format': 'best[protocol=https]', 'videopassword': 'youtube-dl', }, }, { 'url': 'http://vimeo.com/channels/keypeele/75629013', 'md5': '2f86a05afe9d7abc0b9126d229bbe15d', - 'note': 'Video is freely available via original URL ' - 'and protected with password when accessed via http://vimeo.com/75629013', 'info_dict': { 'id': '75629013', 'ext': 'mp4', 'title': 'Key & Peele: Terrorist Interrogation', 'description': 'md5:8678b246399b070816b12313e8b4eb5c', + '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', @@ -168,7 +387,9 @@ class VimeoIE(VimeoBaseInfoExtractor): 'ext': 'mp4', 'title': 'The New Vimeo Player (You Know, For Videos)', 'description': 'md5:2ec900bf97c3f389378a96aee11260ea', + 'timestamp': 1381846109, 'upload_date': '20131015', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/staff', 'uploader_id': 'staff', 'uploader': 'Vimeo Staff', 'duration': 62, @@ -183,6 +404,7 @@ class VimeoIE(VimeoBaseInfoExtractor): 'ext': 'mp4', 'title': 'Pier Solar OUYA Official Trailer', 'uploader': 'Tulio Gonçalves', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user28849593', 'uploader_id': 'user28849593', }, }, @@ -195,11 +417,76 @@ class VimeoIE(VimeoBaseInfoExtractor): 'ext': 'mp4', 'title': 'FOX CLASSICS - Forever Classic ID - A Full Minute', 'uploader': 'The DMCI', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/dmci', 'uploader_id': 'dmci', + 'timestamp': 1324343742, 'upload_date': '20111220', 'description': 'md5:ae23671e82d05415868f7ad1aec21147', }, }, + { + # only available via https://vimeo.com/channels/tributes/6213729 and + # not via https://vimeo.com/6213729 + 'url': 'https://vimeo.com/channels/tributes/6213729', + 'info_dict': { + 'id': '6213729', + 'ext': 'mp4', + 'title': 'Vimeo Tribute: The Shining', + 'uploader': 'Casey Donahue', + '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', + }, + 'params': { + 'skip_download': True, + }, + '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', + 'description': 'md5:0fa704e05b04f91f40b7f3ca2e801384', + 'upload_date': '20130830', + 'timestamp': 1377853339, + }, + 'params': { + 'skip_download': True, + }, + 'expected_warnings': ['Unable to download JSON metadata'], + }, + { + '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': { + 'format': 'best[protocol=https]', + 'videopassword': 'youtube-dl', + }, + }, + { + 'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741', + 'only_matching': True, + }, { 'url': 'https://vimeo.com/109815029', 'note': 'Video not completely processed, "failed" seed status', @@ -209,86 +496,103 @@ class VimeoIE(VimeoBaseInfoExtractor): 'url': 'https://vimeo.com/groups/travelhd/videos/22439234', 'only_matching': True, }, + { + 'url': 'https://vimeo.com/album/2632481/video/79010983', + 'only_matching': True, + }, { # source file returns 403: Forbidden 'url': 'https://vimeo.com/7809605', 'only_matching': True, }, + { + '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) + 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 - def _verify_video_password(self, url, video_id, webpage): - password = self._downloader.params.get('videopassword') - if password is None: - raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True) - token, vuid = self._extract_xsrft_and_vuid(webpage) - data = urlencode_postdata(encode_dict({ - 'password': password, - 'token': token, - })) - if url.startswith('http://'): - # vimeo only supports https now, but the user can give an http url - url = url.replace('http://', 'https://') - password_request = sanitized_Request(url + '/password', data) - password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - password_request.add_header('Referer', url) - self._set_vimeo_cookie('vuid', vuid) - return self._download_webpage( - password_request, video_id, - 'Verifying the password', 'Wrong password') + @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): + 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(encode_dict({'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') - return self._download_json( - password_request, video_id, - 'Verifying the password', - 'Wrong password') + raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True) + 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() def _real_extract(self, url): url, data = unsmuggle_url(url, {}) - headers = std_headers + headers = std_headers.copy() if 'http_headers' in data: - headers = headers.copy() headers.update(data['http_headers']) 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') + video_id = self._match_id(url) orig_url = url - if mobj.group('pro') or mobj.group('player'): + is_pro = 'vimeopro.com/' in url + is_player = '://player.vimeo.com/video/' in url + if is_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)) + if not url: + url = 'https://vimeo.com/' + video_id + elif is_player: url = 'https://player.vimeo.com/video/' + video_id - else: + elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')): url = 'https://vimeo.com/' + video_id - # Retrieve video webpage to extract further information - request = sanitized_Request(url, None, headers) try: - webpage = self._download_webpage(request, video_id) + # Retrieve video webpage to extract further information + webpage, urlh = self._download_webpage_handle( + url, video_id, headers=headers) + redirect_url = urlh.geturl() except ExtractorError as ee: if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403: errmsg = ee.cause.read() @@ -305,16 +609,18 @@ 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 + video_description = None + # Extract the config JSON try: try: @@ -324,22 +630,28 @@ 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) - 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'] - config_json = self._download_webpage(config_url, video_id) - config = json.loads(config_json) + # https://github.com/ytdl-org/youtube-dl/pull/7209) + page_config = self._parse_json(self._search_regex( + r'vimeo\.(?:clip|vod_title)_page_config\s*=\s*({.+?});', + webpage, '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) + video_description = clean_html(dict_get( + page_config, ('description', 'description_html_escaped'))) + config = self._download_json(config_url, video_id) 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) @@ -350,46 +662,47 @@ 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) - - if '>You rented this title.<' in webpage: - feature_id = config.get('video', {}).get('vod', {}).get('feature_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 + 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() 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, {'force_feature_id': True}), 'Vimeo') - # Extract title - video_title = config['video']['title'] - - # Extract uploader and uploader_id - video_uploader = config['video']['owner']['name'] - video_uploader_id = config['video']['owner']['url'].split('/')[-1] if config['video']['owner']['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 description - - video_description = self._html_search_regex( - r'(?s)]*>(.*?)', - webpage, 'description', default=None) + if not video_description: + video_description = self._html_search_regex( + r'(?s)]*>(.*?)', + webpage, 'description', default=None) if not video_description: video_description = self._html_search_meta( 'description', webpage, default=None) - if not video_description and mobj.group('pro'): + if not video_description and is_pro: orig_webpage = self._download_webpage( orig_url, video_id, note='Downloading webpage for description', @@ -397,17 +710,14 @@ class VimeoIE(VimeoBaseInfoExtractor): if orig_webpage: video_description = self._html_search_meta( 'description', orig_webpage, default=None) - if not video_description and not mobj.group('player'): + if not video_description and not is_player: self._downloader.report_warning('Cannot find video description') - # Extract video duration - video_duration = int_or_none(config['video'].get('duration')) - # 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')) @@ -420,73 +730,94 @@ 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, - }) - config_files = config['video'].get('files') or config['request'].get('files', {}) - for f in config_files.get('progressive', []): - video_url = f.get('url') - if not video_url: - continue - formats.append({ - 'url': video_url, - 'format_id': 'http-%s' % f.get('quality'), - 'width': int_or_none(f.get('width')), - 'height': int_or_none(f.get('height')), - '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)) - # 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')) - subtitles = {} - text_tracks = config['request'].get('text_tracks') - if text_tracks: - for tt in text_tracks: - subtitles[tt['lang']] = [{ - 'ext': 'vtt', - 'url': 'https://vimeo.com' + tt['url'], - }] + source_format = self._extract_original_format( + 'https://vimeo.com/' + video_id, video_id) + if source_format: + formats.append(source_format) - return { - 'id': video_id, - 'uploader': video_uploader, - 'uploader_id': video_uploader_id, - 'upload_date': video_upload_date, - 'title': video_title, - 'thumbnail': video_thumbnail, - 'description': video_description, - 'duration': video_duration, + info_dict_config = self._parse_config(config, video_id) + formats.extend(info_dict_config['formats']) + self._vimeo_sort_formats(formats) + + 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 = { 'formats': formats, + 'timestamp': unified_timestamp(timestamp), + 'description': video_description, 'webpage_url': url, 'view_count': view_count, 'like_count': like_count, 'comment_count': comment_count, - 'subtitles': subtitles, + '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 + + +class VimeoOndemandIE(VimeoIE): + IE_NAME = 'vimeo:ondemand' + _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/([^/]+/)?(?P[^/?#&]+)' + _TESTS = [{ + # ondemand video not available via https://vimeo.com/id + 'url': 'https://vimeo.com/ondemand/20704', + 'md5': 'c424deda8c7f73c1dfb3edd7630e2f35', + 'info_dict': { + 'id': '105442900', + 'ext': 'mp4', + 'title': 'המעבדה - במאי יותם פלדמן', + 'uploader': 'גם סרטים', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/gumfilms', + 'uploader_id': 'gumfilms', + 'description': 'md5:4c027c965e439de4baab621e48b60791', + 'upload_date': '20140906', + 'timestamp': 1410032453, + }, + 'params': { + 'format': 'best[protocol=https]', + }, + 'expected_warnings': ['Unable to download JSON metadata'], + }, { + # requires Referer to be passed along with og:video:url + 'url': 'https://vimeo.com/ondemand/36938/126682985', + 'info_dict': { + 'id': '126584684', + 'ext': 'mp4', + 'title': 'Rävlock, rätt läte på rätt plats', + 'uploader': 'Lindroth & Norin', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/lindrothnorin', + 'uploader_id': 'lindrothnorin', + 'description': 'md5:c3c46a90529612c8279fb6af803fc0df', + 'upload_date': '20150502', + 'timestamp': 1430586422, + }, + 'params': { + 'skip_download': True, + }, + 'expected_warnings': ['Unable to download JSON metadata'], + }, { + 'url': 'https://vimeo.com/ondemand/nazmaalik', + 'only_matching': True, + }, { + 'url': 'https://vimeo.com/ondemand/141692381', + 'only_matching': True, + }, { + 'url': 'https://vimeo.com/ondemand/thelastcolony/150274832', + 'only_matching': True, + }] + class VimeoChannelIE(VimeoBaseInfoExtractor): IE_NAME = 'vimeo:channel' @@ -502,39 +833,14 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): }, 'playlist_mincount': 25, }] + _BASE_URL_TEMPL = 'https://vimeo.com/channels/%s' def _page_url(self, base_url, pagenum): 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') - - def _login_list_password(self, page_url, list_id, webpage): - login_form = self._search_regex( - r'(?s)]+?id="pw_form"(.*?)', - webpage, 'login form', default=None) - if not login_form: - return webpage - - password = self._downloader.params.get('videopassword') - if password is None: - raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True) - fields = self._hidden_inputs(login_form) - token, vuid = self._extract_xsrft_and_vuid(webpage) - fields['token'] = token - fields['password'] = password - post = urlencode_postdata(encode_dict(fields)) - password_path = self._search_regex( - r'action="([^"]+)"', login_form, 'password URL') - password_url = compat_urlparse.urljoin(page_url, password_path) - password_request = sanitized_Request(password_url, post) - password_request.add_header('Content-type', 'application/x-www-form-urlencoded') - self._set_vimeo_cookie('vuid', vuid) - self._set_vimeo_cookie('xsrft', token) - - return self._download_webpage( - password_request, list_id, - 'Verifying the password', 'Wrong password') + return self._TITLE or self._html_search_regex( + self._TITLE_RE, webpage, 'list title', fatal=False) def _title_and_entries(self, list_id, base_url): for pagenum in itertools.count(1): @@ -544,11 +850,23 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): 'Downloading page %s' % pagenum) if pagenum == 1: - webpage = self._login_list_password(page_url, list_id, webpage) yield self._extract_list_title(webpage) - for video_id in re.findall(r'id="clip_(\d+?)"', webpage): - yield self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo') + # 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)(?:[^>]+\btitle="([^"]+)")?', webpage) + if 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, video_title=video_title) + # More relaxed fallback + else: + for video_id in re.findall(r'id=["\']clip_(\d+)', webpage): + yield self.url_result( + 'https://vimeo.com/%s' % video_id, + VimeoIE.ie_key(), video_id=video_id) if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None: break @@ -559,14 +877,13 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): return self.playlist_result(title_and_entries, list_id, list_title) def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - channel_id = mobj.group('id') - return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id) + channel_id = self._match_id(url) + return self._extract_videos(channel_id, self._BASE_URL_TEMPL % channel_id) class VimeoUserIE(VimeoChannelIE): IE_NAME = 'vimeo:user' - _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P[^/]+)(?:/videos|[#?]|$)' + _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P[^/]+)(?:/videos|[#?]|$)' _TITLE_RE = r']+?class="user">([^<>]+?)' _TESTS = [{ 'url': 'https://vimeo.com/nkistudio/videos', @@ -576,16 +893,12 @@ class VimeoUserIE(VimeoChannelIE): }, 'playlist_mincount': 66, }] - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - name = mobj.group('name') - return self._extract_videos(name, 'https://vimeo.com/%s' % name) + _BASE_URL_TEMPL = 'https://vimeo.com/%s' -class VimeoAlbumIE(VimeoChannelIE): +class VimeoAlbumIE(VimeoBaseInfoExtractor): IE_NAME = 'vimeo:album' - _VALID_URL = r'https://vimeo\.com/album/(?P\d+)' + _VALID_URL = r'https://vimeo\.com/(?:album|showcase)/(?P\d+)(?:$|[?#]|/(?!video))' _TITLE_RE = r'
\n\s*

(.*?)

' _TESTS = [{ 'url': 'https://vimeo.com/album/2632481', @@ -606,40 +919,86 @@ class VimeoAlbumIE(VimeoChannelIE): 'videopassword': 'youtube-dl', } }] - - 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,uri', + '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 + uri = video.get('uri') + video_id = self._search_regex(r'/videos/(\d+)', uri, 'video_id', default=None) if uri else None + yield self.url_result(link, VimeoIE.ie_key(), video_id) def _real_extract(self, url): album_id = self._match_id(url) - return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id) - - -class VimeoGroupsIE(VimeoAlbumIE): + webpage = self._download_webpage(url, album_id) + viewer = self._parse_json(self._search_regex( + r'bootstrap_data\s*=\s*({.+?})', + webpage, 'bootstrap data'), album_id)['viewer'] + jwt = viewer['jwt'] + album = self._download_json( + 'https://api.vimeo.com/albums/' + album_id, + album_id, headers={'Authorization': 'jwt ' + jwt}, + query={'fields': 'description,name,privacy'}) + hashed_pass = None + if try_get(album, lambda x: x['privacy']['view']) == 'password': + password = self._downloader.params.get('videopassword') + if not password: + raise ExtractorError( + 'This album is protected by a password, use the --video-password option', + expected=True) + self._set_vimeo_cookie('vuid', viewer['vuid']) + try: + hashed_pass = self._download_json( + 'https://vimeo.com/showcase/%s/auth' % album_id, + album_id, 'Verifying the password', data=urlencode_postdata({ + 'password': password, + 'token': viewer['xsrft'], + }), headers={ + 'X-Requested-With': 'XMLHttpRequest', + })['hashed_pass'] + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401: + raise ExtractorError('Wrong password', expected=True) + raise + entries = OnDemandPagedList(functools.partial( + self._fetch_page, album_id, jwt, hashed_pass), self._PAGE_SIZE) + return self.playlist_result( + entries, album_id, album.get('name'), album.get('description')) + + +class VimeoGroupsIE(VimeoChannelIE): IE_NAME = 'vimeo:group' - _VALID_URL = r'https://vimeo\.com/groups/(?P[^/]+)(?:/(?!videos?/\d+)|$)' + _VALID_URL = r'https://vimeo\.com/groups/(?P[^/]+)(?:/(?!videos?/\d+)|$)' _TESTS = [{ - 'url': 'https://vimeo.com/groups/rolexawards', + 'url': 'https://vimeo.com/groups/kattykay', 'info_dict': { - 'id': 'rolexawards', - 'title': 'Rolex Awards for Enterprise', + 'id': 'kattykay', + 'title': 'Katty Kay', }, - 'playlist_mincount': 73, + 'playlist_mincount': 27, }] + _BASE_URL_TEMPL = 'https://vimeo.com/groups/%s' - def _extract_list_title(self, webpage): - return self._og_search_title(webpage) - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - name = mobj.group('name') - return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name) - - -class VimeoReviewIE(InfoExtractor): +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', @@ -648,7 +1007,10 @@ class VimeoReviewIE(InfoExtractor): 'ext': 'mp4', 'title': "DICK HARDWICK 'Comedian'", 'uploader': 'Richard Hardwick', - } + 'uploader_id': 'user21297594', + 'description': "Comedian Dick Hardwick's five minute demo filmed in front of a live theater audience.\nEdit by Doug Mattocks", + }, + 'expected_warnings': ['Unable to download JSON metadata'], }, { 'note': 'video player needs Referer', 'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053', @@ -659,15 +1021,44 @@ class VimeoReviewIE(InfoExtractor): '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', + }, + 'skip': 'video gone', + }, { + 'note': 'Password protected', + 'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde', + 'info_dict': { + 'id': '138823582', + 'ext': 'mp4', + 'title': 'EFFICIENT PICKUP MASTERCLASS MODULE 1', + 'uploader': 'TMB', + 'uploader_id': 'user37284429', + }, + 'params': { + 'videopassword': 'holygrail', + }, + 'skip': 'video gone', }] + def _real_initialize(self): + self._login() + def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - player_url = 'https://player.vimeo.com/player/' + video_id - return self.url_result(player_url, 'Vimeo', video_id) + page_url, video_id = re.match(self._VALID_URL, url).groups() + clip_data = self._download_json( + page_url.replace('/review/', '/review/data/'), + video_id)['clipData'] + config_url = clip_data['configUrl'] + config = self._download_json(config_url, video_id) + info_dict = self._parse_config(config, video_id) + source_format = self._extract_original_format( + page_url + '/action', video_id) + if source_format: + info_dict['formats'].append(source_format) + self._vimeo_sort_formats(info_dict['formats']) + info_dict['description'] = clean_html(clip_data.get('description')) + return info_dict class VimeoWatchLaterIE(VimeoChannelIE): @@ -696,57 +1087,41 @@ 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 _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + config_url = self._parse_json(self._search_regex( + r'window\.OTTData\s*=\s*({.+})', webpage, + 'ott data'), video_id, js_to_json)['config_url'] + config = self._download_json(config_url, video_id) + info = self._parse_config(config, video_id) + self._vimeo_sort_formats(info['formats']) + return info