+
+class VimeoBaseInfoExtractor(InfoExtractor):
+ _NETRC_MACHINE = 'vimeo'
+ _LOGIN_REQUIRED = False
+ _LOGIN_URL = 'https://vimeo.com/log_in'
+
+ def _login(self):
+ 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
+ webpage = self._download_webpage(
+ self._LOGIN_URL, None, 'Downloading login page')
+ token, vuid = self._extract_xsrft_and_vuid(webpage)
+ data = {
+ 'action': 'login',
+ 'email': username,
+ 'password': password,
+ 'service': 'vimeo',
+ 'token': token,
+ }
+ 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)
+ return self._download_webpage(
+ password_request, video_id,
+ 'Verifying the password', 'Wrong password')
+
+ def _extract_xsrft_and_vuid(self, webpage):
+ xsrft = self._search_regex(
+ r'(?:(?P<q1>["\'])xsrft(?P=q1)\s*:|xsrft\s*[=:])\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
+ webpage, 'login token', group='xsrft')
+ vuid = self._search_regex(
+ r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
+ webpage, 'vuid', group='vuid')
+ return xsrft, vuid
+
+ 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']
+ # 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'))
+
+ 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')),
+ })
+
+ 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_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:
+ 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)
+
+ 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'],
+ }]
+
+ return {
+ 'title': video_title,
+ 'uploader': video_uploader,
+ 'uploader_id': video_uploader_id,
+ 'uploader_url': video_uploader_url,
+ 'thumbnail': video_thumbnail,
+ 'duration': video_duration,
+ 'formats': formats,
+ 'subtitles': subtitles,
+ }
+
+
+class VimeoIE(VimeoBaseInfoExtractor):
Rapsys Git / youtubedl / blobdiff