compat_urlparse,
)
from ..utils import (
+ clean_html,
determine_ext,
+ dict_get,
ExtractorError,
js_to_json,
int_or_none,
merge_dicts,
- NO_DEFAULT,
OnDemandPagedList,
parse_filesize,
RegexNotFoundError,
sanitized_Request,
smuggle_url,
std_headers,
+ str_or_none,
try_get,
unified_timestamp,
unsmuggle_url,
urlencode_postdata,
+ urljoin,
unescapeHTML,
)
})
# 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)
- 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:
+ 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.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 {}
'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': 'https://vimeo.com' + tt['url'],
+ 'url': urljoin('https://vimeo.com', tt['url']),
}]
thumbnails = []
video_uploader_url = owner.get('url')
return {
- 'id': video_id,
+ '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,
(?:
(?:
www|
- (?P<player>player)
+ player
)
\.
)?
- vimeo(?P<pro>pro)?\.com/
+ vimeo(?:pro)?\.com/
(?!(?:channels|album|showcase)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/)
(?:.*?/)?
(?:
'id': '56015672',
'ext': 'mp4',
'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
- 'description': 'md5:509a9ad5c9bf97c60faee9203aca4479',
+ 'description': 'md5:2d3305bad981a06ff79f027f19865021',
'timestamp': 1355990239,
'upload_date': '20121220',
'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user7108434',
'duration': 10,
'license': 'by-sa',
},
+ 'params': {
+ 'format': 'best[protocol=https]',
+ },
},
{
'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
'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]',
},
},
{
'duration': 3610,
'description': None,
},
+ 'params': {
+ 'format': 'best[protocol=https]',
+ },
+ 'expected_warnings': ['Unable to download JSON metadata'],
},
{
'url': 'http://vimeo.com/68375962',
'description': 'md5:dca3ea23adb29ee387127bc4ddfce63f',
},
'params': {
+ 'format': 'best[protocol=https]',
'videopassword': 'youtube-dl',
},
},
'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',
'duration': 10,
},
'params': {
+ 'format': 'best[protocol=https]',
'videopassword': 'youtube-dl',
},
},
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')
+ raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
data = urlencode_postdata({
'password': base64.b64encode(password.encode()),
})
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'):
+ 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))
- elif mobj.group('player'):
+ if not url:
+ url = 'https://vimeo.com/' + video_id
+ elif is_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
- # Retrieve video webpage to extract further information
- 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(redirect_url):
- return self.url_result(redirect_url, VimeoOndemandIE.ie_key())
+ # 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()
cc_license = None
timestamp = None
+ video_description = None
# Extract the config JSON
try:
# Sometimes new react-based page is served instead of old one that require
# different config URL extraction approach (see
# 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')
- page_config = self._parse_json(vimeo_clip_page_config, video_id)
+ 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)
- config_json = self._download_webpage(config_url, video_id)
- config = json.loads(config_json)
+ 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
{'force_feature_id': True}), 'Vimeo')
# Extract video description
-
- video_description = self._html_search_regex(
- r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
- webpage, 'description', default=None)
+ if not video_description:
+ video_description = self._html_search_regex(
+
r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
+ 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',
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 upload date
return info_dict
-class VimeoOndemandIE(VimeoBaseInfoExtractor):
+class VimeoOndemandIE(VimeoIE):
IE_NAME = 'vimeo:ondemand'
- _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?P<id>[^/?#&]+)'
+ _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/([^/]+/)?(?P<id>[^/?#&]+)'
_TESTS = [{
# ondemand video not available via https://vimeo.com/id
'url': 'https://vimeo.com/ondemand/20704',
'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': '126682985',
+ 'id': '126584684',
'ext': 'mp4',
'title': 'Rävlock, rätt läte på rätt plats',
'uploader': 'Lindroth & Norin',
- 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user14430847',
- 'uploader_id': 'user14430847',
+ '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,
'only_matching': True,
}]
- def _real_extract(self, url):
- video_id = self._match_id(url)
- webpage = self._download_webpage(url, video_id)
- return self.url_result(
- # Some videos require Referer to be passed along with og:video:url
- # similarly to generic vimeo embeds (e.g.
- # https://vimeo.com/ondemand/36938/126682985).
- VimeoIE._smuggle_referrer(self._og_search_video_url(webpage), url),
- VimeoIE.ie_key())
-
class VimeoChannelIE(VimeoBaseInfoExtractor):
IE_NAME = 'vimeo:channel'
},
'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)
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(
- r'(?s)<form[^>]+?id="pw_form"(.*?)</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(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')
-
def _title_and_entries(self, list_id, base_url):
for pagenum in itertools.count(1):
page_url = self._page_url(base_url, pagenum)
'Downloading page %s' % pagenum)
if pagenum == 1:
- webpage = self._login_list_password(page_url, list_id, webpage)
yield self._extract_list_title(webpage)
# Try extracting href first since not all videos are available via
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<name>[^/]+)(?:/videos|[#?]|$)'
+ _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<id>[^/]+)(?:/videos|[#?]|$)'
_TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
_TESTS = [{
'url': 'https://vimeo.com/nkistudio/videos',
},
'playlist_mincount': 66,
}]
+ _BASE_URL_TEMPL = 'https://vimeo.com/%s'
- def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
- name = mobj.group('name')
- return self._extract_videos(name, 'https://vimeo.com/%s' % name)
-
-class VimeoAlbumIE(VimeoChannelIE):
+class VimeoAlbumIE(VimeoBaseInfoExtractor):
IE_NAME = 'vimeo:album'
_VALID_URL = r'https://vimeo\.com/(?:album|showcase)/(?P<id>\d+)(?:$|[?#]|/(?!video))'
_TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
def _fetch_page(self, album_id, authorizaion, hashed_pass, page):
api_page = page + 1
query = {
- 'fields': 'link',
+ 'fields': 'link,uri',
'page': api_page,
'per_page': self._PAGE_SIZE,
}
link = video.get('link')
if not link:
continue
- yield self.url_result(link, VimeoIE.ie_key(), VimeoIE._match_id(link))
+ 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)
webpage = self._download_webpage(url, album_id)
- webpage = self._login_list_password(url, album_id, webpage)
- api_config = self._extract_vimeo_config(webpage, album_id)['api']
+ viewer = self._parse_json(self._search_regex(
+ r'bootstrap_data\s*=\s*({.+?})</script>',
+ 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, api_config['jwt'],
- api_config.get('hashed_pass')), self._PAGE_SIZE)
- return self.playlist_result(entries, album_id, self._html_search_regex(
- r'<title>\s*(.+?)(?:\s+on Vimeo)?</title>', webpage, 'title', fatal=False))
+ 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(VimeoAlbumIE):
+class VimeoGroupsIE(VimeoChannelIE):
IE_NAME = 'vimeo:group'
- _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)'
+ _VALID_URL = r'https://vimeo\.com/groups/(?P<id>[^/]+)(?:/(?!videos?/\d+)|$)'
_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,
}]
-
- def _extract_list_title(self, webpage):
- return self._og_search_title(webpage, fatal=False)
-
- 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)
+ _BASE_URL_TEMPL = 'https://vimeo.com/groups/%s'
class VimeoReviewIE(VimeoBaseInfoExtractor):
'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',
'duration': 2773,
'thumbnail': r're:^https?://.*\.jpg$',
'uploader_id': 'user22258446',
- }
+ },
+ 'skip': 'video gone',
}, {
'note': 'Password protected',
'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde',
def _real_initialize(self):
self._login()
- def _get_config_url(self, webpage_url, video_id, video_password_verified=False):
- webpage = self._download_webpage(webpage_url, video_id)
- config_url = self._html_search_regex(
- r'data-config-url=(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
- 'config URL', default=None, group='url')
- if not config_url:
- 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 = 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(
- webpage_url, video_id, video_password_verified=True)
- return config_url
-
def _real_extract(self, url):
page_url, video_id = re.match(self._VALID_URL, url).groups()
- config_url = self._get_config_url(url, video_id)
+ 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, 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
projects / youtube-dl.git / blobdiff