X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/89bb9e23b34f43c6563524917f7dffb9841a972e..51057713322a0d2ff542520876a4c675f26c57ac:/youtube_dl/extractor/vimeo.py diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index 5e15f06..b5b44a7 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -1,6 +1,7 @@ # coding: utf-8 from __future__ import unicode_literals +import base64 import json import re import itertools @@ -15,7 +16,6 @@ from ..utils import ( determine_ext, ExtractorError, js_to_json, - InAdvancePagedList, int_or_none, merge_dicts, NO_DEFAULT, @@ -108,23 +108,9 @@ class VimeoBaseInfoExtractor(InfoExtractor): def _parse_config(self, config, video_id): video_data = config['video'] - # Extract title video_title = video_data['title'] - - # Extract uploader, uploader_url and uploader_id - video_uploader = video_data.get('owner', {}).get('name') - video_uploader_url = video_data.get('owner', {}).get('url') - video_uploader_id = video_uploader_url.split('/')[-1] if video_uploader_url else None - - # Extract video thumbnail - video_thumbnail = video_data.get('thumbnail') - if video_thumbnail is None: - video_thumbs = video_data.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(video_data.get('duration')) + 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', {}) @@ -141,6 +127,7 @@ class VimeoBaseInfoExtractor(InfoExtractor): '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') @@ -150,7 +137,7 @@ class VimeoBaseInfoExtractor(InfoExtractor): if files_type == 'hls': formats.extend(self._extract_m3u8_formats( manifest_url, video_id, 'mp4', - 'm3u8_native', m3u8_id=format_id, + 'm3u8' if is_live else 'm3u8_native', m3u8_id=format_id, note='Downloading %s m3u8 information' % cdn_name, fatal=False)) elif files_type == 'dash': @@ -163,6 +150,10 @@ class VimeoBaseInfoExtractor(InfoExtractor): 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, @@ -174,6 +165,15 @@ class VimeoBaseInfoExtractor(InfoExtractor): 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: @@ -183,17 +183,61 @@ 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.""" @@ -392,6 +436,22 @@ class VimeoIE(VimeoBaseInfoExtractor): '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, @@ -418,6 +478,8 @@ 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 @@ -448,18 +510,22 @@ class VimeoIE(VimeoBaseInfoExtractor): 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({'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() @@ -479,7 +545,11 @@ class VimeoIE(VimeoBaseInfoExtractor): 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 @@ -532,7 +602,7 @@ 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') @@ -572,7 +642,7 @@ class VimeoIE(VimeoBaseInfoExtractor): cause=e) else: if config.get('view') == 4: - config = self._verify_player_video_password(redirect_url, video_id) + config = self._verify_player_video_password(redirect_url, video_id, headers) vod = config.get('video', {}).get('vod', {}) @@ -632,29 +702,11 @@ 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 = (try_get( - source_file, lambda x: x['extension'], - compat_str) or determine_ext( - download_url, None) or 'mp4').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, - }) + + 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']) @@ -761,7 +813,8 @@ 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( @@ -902,7 +955,7 @@ 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) @@ -913,7 +966,7 @@ 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', @@ -965,7 +1018,8 @@ class VimeoReviewIE(VimeoBaseInfoExtractor): 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_url = data.get('vimeo_esi', {}).get('config', {}).get('configUrl') + 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( @@ -973,10 +1027,13 @@ 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 @@ -1008,7 +1065,7 @@ class VimeoWatchLaterIE(VimeoChannelIE): return self._extract_videos('watchlater', 'https://vimeo.com/watchlater') -class VimeoLikesIE(InfoExtractor): +class VimeoLikesIE(VimeoChannelIE): _VALID_URL = r'https://(?:www\.)?vimeo\.com/(?P[^/]+)/likes/?(?:$|[?#]|sort:)' IE_NAME = 'vimeo:likes' IE_DESC = 'Vimeo user likes' @@ -1016,55 +1073,20 @@ class VimeoLikesIE(InfoExtractor): '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', default=1), - '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/%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*