+
+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 _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
+ sep_pattern = r'/sep/video/'
+ 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)
+ sep_manifest_urls = []
+ if re.search(sep_pattern, manifest_url):
+ for suffix, repl in (('', 'video'), ('_sep', 'sep/video')):
+ sep_manifest_urls.append((format_id + suffix, re.sub(
+ sep_pattern, '/%s/' % repl, manifest_url)))
+ else:
+ sep_manifest_urls = [(format_id, manifest_url)]
+ for f_id, m_url in sep_manifest_urls:
+ if files_type == 'hls':
+ formats.extend(self._extract_m3u8_formats(
+ m_url, video_id, 'mp4',
+ 'm3u8' if is_live else 'm3u8_native', m3u8_id=f_id,
+ note='Downloading %s m3u8 information' % cdn_name,
+ fatal=False))
+ elif files_type == 'dash':
+ 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)
+ 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,
+ })
+
+ for f in formats:
+ if f.get('vcodec') == 'none':
+ f['preference'] = -50
+ elif f.get('acodec') == 'none':
+ f['preference'] = -40
+
+ 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):
Rapsys Git / youtubedl / blobdiff