X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/421b7b220ea958384617bd7d339f188bf601280e..0f0afa2a182178f8c70d50daf99f864372fe71f3:/youtube_dl/extractor/vimeo.py?ds=sidebyside diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index cedb548..a41178b 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 @@ -14,9 +15,13 @@ from ..compat import ( from ..utils import ( determine_ext, ExtractorError, + js_to_json, InAdvancePagedList, int_or_none, + merge_dicts, NO_DEFAULT, + parse_filesize, + qualities, RegexNotFoundError, sanitized_Request, smuggle_url, @@ -26,7 +31,6 @@ from ..utils import ( unsmuggle_url, urlencode_postdata, unescapeHTML, - parse_filesize, ) @@ -36,26 +40,35 @@ 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') @@ -96,23 +109,9 @@ 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', {}) @@ -129,6 +128,7 @@ 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') @@ -138,7 +138,7 @@ 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': @@ -151,6 +151,10 @@ 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, @@ -162,6 +166,15 @@ 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: @@ -171,17 +184,61 @@ 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.""" @@ -218,7 +275,7 @@ 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': r're:https?://(?:www\.)?vimeo\.com/user7108434', @@ -289,10 +346,13 @@ class VimeoIE(VimeoBaseInfoExtractor): 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/atencio', 'uploader_id': 'atencio', 'uploader': 'Peter Atencio', + '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', @@ -345,11 +405,13 @@ class VimeoIE(VimeoBaseInfoExtractor): 'url': 'https://vimeo.com/channels/tributes/6213729', 'info_dict': { 'id': '6213729', - 'ext': 'mov', + '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', @@ -375,6 +437,22 @@ 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 MarquÃnez Ferrándiz', + 'duration': 10, + }, + 'params': { + 'videopassword': 'youtube-dl', + }, + }, { 'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741', 'only_matching': True, @@ -401,6 +479,8 @@ 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 @@ -431,18 +511,22 @@ 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() @@ -455,11 +539,18 @@ 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 @@ -468,11 +559,12 @@ class VimeoIE(VimeoBaseInfoExtractor): request = sanitized_Request(url, headers=headers) try: 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(urlh.geturl()): - return self.url_result(urlh.geturl(), VimeoOndemandIE.ie_key()) + 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() @@ -511,7 +603,7 @@ 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') @@ -528,9 +620,11 @@ class VimeoIE(VimeoBaseInfoExtractor): # We try to find out to which variable is assigned the config dic 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) @@ -541,29 +635,33 @@ class VimeoIE(VimeoBaseInfoExtractor): if re.search(r'