# coding: utf-8
from __future__ import unicode_literals
+import base64
import json
import re
import itertools
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,
unsmuggle_url,
urlencode_postdata,
unescapeHTML,
- parse_filesize,
)
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', {})
'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')
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':
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,
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:
'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."""
'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',
'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',
'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,
'url': 'https://vimeo.com/160743502/abd0e13fb4',
'only_matching': True,
}
+ # https://gettingthingsdone.com/workflowmap/
+ # vimeo embed with check-password page protected by Referer header
]
@staticmethod
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()
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
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')
# 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)
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', {})
def is_rented():
if '>You rented this title.<' in webpage:
return True
if config.get('user', {}).get('purchased'):
return True
- label = try_get(
- config, lambda x: x['video']['vod']['purchase_options'][0]['label_string'], compat_str)
- if label and label.startswith('You rented this'):
- return True
+ for purchase_option in vod.get('purchase_options', []):
+ if purchase_option.get('purchased'):
+ return True
+ label = purchase_option.get('label_string')
+ if label and (label.startswith('You rented this') or label.endswith(' remaining')):
+ return True
return False
- if is_rented():
- feature_id = config.get('video', {}).get('vod', {}).get('feature_id')
+ if is_rented() and vod.get('is_trailer'):
+ feature_id = vod.get('feature_id')
if feature_id and not data.get('force_feature_id', False):
return self.url_result(smuggle_url(
'https://player.vimeo.com/player/%s' % feature_id,
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'])
r'<link[^>]+rel=["\']license["\'][^>]+href=(["\'])(?P<license>(?:(?!\1).)+)\1',
webpage, 'license', default=None, group='license')
+ channel_url = 'https://vimeo.com/channels/%s' % channel_id if channel_id else None
+
info_dict = {
'id': video_id,
'formats': formats,
'like_count': like_count,
'comment_count': comment_count,
'license': cc_license,
+ 'channel_id': channel_id,
+ 'channel_url': channel_url,
}
info_dict = merge_dicts(info_dict, info_dict_config, json_ld)
class VimeoReviewIE(VimeoBaseInfoExtractor):
IE_NAME = 'vimeo:review'
IE_DESC = 'Review pages on vimeo'
- _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
+ _VALID_URL = r'(?P<url>https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)/[0-9a-f]{10})'
_TESTS = [{
'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
'md5': 'c507a72f780cacc12b2248bb4006d253',
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(
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
'description': description,
'entries': pl,
}
+
+
+class VHXEmbedIE(InfoExtractor):
+ IE_NAME = 'vhx:embed'
+ _VALID_URL = r'https?://embed\.vhx\.tv/videos/(?P<id>\d+)'
+
+ def _call_api(self, video_id, access_token, path='', query=None):
+ return self._download_json(
+ 'https://api.vhx.tv/videos/' + video_id + path, video_id, headers={
+ 'Authorization': 'Bearer ' + access_token,
+ }, query=query)
+
+ def _real_extract(self, url):
+ video_id = self._match_id(url)
+ webpage = self._download_webpage(url, video_id)
+ credentials = self._parse_json(self._search_regex(
+ r'(?s)credentials\s*:\s*({.+?}),', webpage,
+ 'config'), video_id, js_to_json)
+ access_token = credentials['access_token']
+
+ query = {}
+ for k, v in credentials.items():
+ if k in ('authorization', 'authUserToken', 'ticket') and v and v != 'undefined':
+ if k == 'authUserToken':
+ query['auth_user_token'] = v
+ else:
+ query[k] = v
+ files = self._call_api(video_id, access_token, '/files', query)
+
+ formats = []
+ for f in files:
+ href = try_get(f, lambda x: x['_links']['source']['href'])
+ if not href:
+ continue
+ method = f.get('method')
+ if method == 'hls':
+ formats.extend(self._extract_m3u8_formats(
+ href, video_id, 'mp4', 'm3u8_native',
+ m3u8_id='hls', fatal=False))
+ elif method == 'dash':
+ formats.extend(self._extract_mpd_formats(
+ href, video_id, mpd_id='dash', fatal=False))
+ else:
+ fmt = {
+ 'filesize': int_or_none(try_get(f, lambda x: x['size']['bytes'])),
+ 'format_id': 'http',
+ 'preference': 1,
+ 'url': href,
+ 'vcodec': f.get('codec'),
+ }
+ quality = f.get('quality')
+ if quality:
+ fmt.update({
+ 'format_id': 'http-' + quality,
+ 'height': int_or_none(self._search_regex(r'(\d+)p', quality, 'height', default=None)),
+ })
+ formats.append(fmt)
+ self._sort_formats(formats)
+
+ video_data = self._call_api(video_id, access_token)
+ title = video_data.get('title') or video_data['name']
+
+ subtitles = {}
+ for subtitle in try_get(video_data, lambda x: x['tracks']['subtitles'], list) or []:
+ lang = subtitle.get('srclang') or subtitle.get('label')
+ for _link in subtitle.get('_links', {}).values():
+ href = _link.get('href')
+ if not href:
+ continue
+ subtitles.setdefault(lang, []).append({
+ 'url': href,
+ })
+
+ q = qualities(['small', 'medium', 'large', 'source'])
+ thumbnails = []
+ for thumbnail_id, thumbnail_url in video_data.get('thumbnail', {}).items():
+ thumbnails.append({
+ 'id': thumbnail_id,
+ 'url': thumbnail_url,
+ 'preference': q(thumbnail_id),
+ })
+
+ return {
+ 'id': video_id,
+ 'title': title,
+ 'description': video_data.get('description'),
+ 'duration': int_or_none(try_get(video_data, lambda x: x['duration']['seconds'])),
+ 'formats': formats,
+ 'subtitles': subtitles,
+ 'thumbnails': thumbnails,
+ 'timestamp': unified_timestamp(video_data.get('created_at')),
+ 'view_count': int_or_none(video_data.get('plays_count')),
+ }
Rapsys Git / youtubedl / blobdiff