diff options
Diffstat (limited to 'hypervideo_dl/extractor/vimeo.py')
| -rw-r--r-- | hypervideo_dl/extractor/vimeo.py | 252 |
1 files changed, 144 insertions, 108 deletions
diff --git a/hypervideo_dl/extractor/vimeo.py b/hypervideo_dl/extractor/vimeo.py index 4f025a5..516b76d 100644 --- a/hypervideo_dl/extractor/vimeo.py +++ b/hypervideo_dl/extractor/vimeo.py @@ -1,14 +1,11 @@ -# coding: utf-8 -from __future__ import unicode_literals - import base64 import functools import re import itertools +import urllib.error from .common import InfoExtractor from ..compat import ( - compat_kwargs, compat_HTTPError, compat_str, compat_urlparse, @@ -34,7 +31,6 @@ from ..utils import ( unsmuggle_url, urlencode_postdata, urljoin, - unescapeHTML, urlhandle_detect_ext, ) @@ -44,6 +40,18 @@ class VimeoBaseInfoExtractor(InfoExtractor): _LOGIN_REQUIRED = False _LOGIN_URL = 'https://vimeo.com/log_in' + @staticmethod + def _smuggle_referrer(url, referrer_url): + return smuggle_url(url, {'http_headers': {'Referer': referrer_url}}) + + def _unsmuggle_headers(self, url): + """@returns (url, smuggled_data, headers)""" + url, data = unsmuggle_url(url, {}) + headers = self.get_param('http_headers').copy() + if 'http_headers' in data: + headers.update(data['http_headers']) + return url, data, headers + def _perform_login(self, username, password): webpage = self._download_webpage( self._LOGIN_URL, None, 'Downloading login page') @@ -109,21 +117,16 @@ class VimeoBaseInfoExtractor(InfoExtractor): 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)) + webpage, 'vimeo config', *args, **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): - # Note: Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps - # at the same time without actual units specified. - self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source')) - def _parse_config(self, config, video_id): video_data = config['video'] - video_title = video_data['title'] + video_title = video_data.get('title') live_event = video_data.get('live_event') or {} is_live = live_event.get('status') == 'started' request = config.get('request') or {} @@ -235,6 +238,9 @@ class VimeoBaseInfoExtractor(InfoExtractor): 'formats': formats, 'subtitles': subtitles, 'is_live': is_live, + # Note: Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps + # at the same time without actual units specified. + '_format_sort_fields': ('quality', 'res', 'fps', 'hdr:12', 'source'), } def _extract_original_format(self, url, video_id, unlisted_hash=None): @@ -306,7 +312,7 @@ class VimeoIE(VimeoBaseInfoExtractor): ) \. )? - vimeo(?:pro)?\.com/ + vimeo\.com/ (?!(?:channels|album|showcase)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/) (?:[^/]+/)*? (?: @@ -320,6 +326,14 @@ class VimeoIE(VimeoBaseInfoExtractor): /?(?:[?&].*)?(?:[#].*)?$ ''' IE_NAME = 'vimeo' + _EMBED_REGEX = [ + # iframe + r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/\d+.*?)\1', + # Embedded (swf embed) Vimeo player + r'<embed[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)\1', + # Non-standard embedded Vimeo player + r'<video[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)\1', + ] _TESTS = [ { 'url': 'http://vimeo.com/56015672#at=0', @@ -343,31 +357,6 @@ class VimeoIE(VimeoBaseInfoExtractor): 'skip': 'No longer available' }, { - 'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876', - 'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82', - 'note': 'Vimeo Pro video (#1197)', - 'info_dict': { - 'id': '68093876', - 'ext': 'mp4', - 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/openstreetmapus', - 'uploader_id': 'openstreetmapus', - 'uploader': 'OpenStreetMap US', - 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography', - 'description': 'md5:2c362968038d4499f4d79f88458590c1', - 'duration': 1595, - 'upload_date': '20130610', - 'timestamp': 1370893156, - 'license': 'by', - 'thumbnail': 'https://i.vimeocdn.com/video/440260469-19b0d92fca3bd84066623b53f1eb8aaa3980c6c809e2d67b6b39ab7b4a77a344-d_960', - 'view_count': int, - 'comment_count': int, - 'like_count': int, - }, - 'params': { - 'format': 'best[protocol=https]', - }, - }, - { 'url': 'http://player.vimeo.com/video/54469442', 'md5': 'b3e7f4d2cbb53bd7dc3bb6ff4ed5cfbd', 'note': 'Videos that embed the url in the player page', @@ -721,33 +710,14 @@ class VimeoIE(VimeoBaseInfoExtractor): # vimeo embed with check-password page protected by Referer header ] - @staticmethod - def _smuggle_referrer(url, referrer_url): - return smuggle_url(url, {'http_headers': {'Referer': referrer_url}}) - - @staticmethod - def _extract_urls(url, webpage): - urls = [] - # Look for embedded (iframe) Vimeo player - for mobj in re.finditer( - r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/\d+.*?)\1', - webpage): - urls.append(VimeoIE._smuggle_referrer(unescapeHTML(mobj.group('url')), url)) - PLAIN_EMBED_RE = ( - # Look for embedded (swf embed) Vimeo player - r'<embed[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)\1', - # Look more for non-standard embedded Vimeo player - r'<video[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)\1', - ) - for embed_re in PLAIN_EMBED_RE: - for mobj in re.finditer(embed_re, webpage): - urls.append(mobj.group('url')) - return urls + @classmethod + def _extract_embed_urls(cls, url, webpage): + for embed_url in super()._extract_embed_urls(url, webpage): + yield cls._smuggle_referrer(embed_url, url) - @staticmethod - def _extract_url(url, webpage): - urls = VimeoIE._extract_urls(url, webpage) - return urls[0] if urls else None + @classmethod + def _extract_url(cls, url, webpage): + return next(cls._extract_embed_urls(url, webpage), None) def _verify_player_video_password(self, url, video_id, headers): password = self._get_video_password() @@ -758,8 +728,8 @@ class VimeoIE(VimeoBaseInfoExtractor): 'Content-Type': 'application/x-www-form-urlencoded', }) checked = self._download_json( - url + '/check-password', video_id, - 'Verifying the password', data=data, headers=headers) + f'{compat_urlparse.urlsplit(url)._replace(query=None).geturl()}/check-password', + video_id, 'Verifying the password', data=data, headers=headers) if checked is False: raise ExtractorError('Wrong video password', expected=True) return checked @@ -780,7 +750,6 @@ class VimeoIE(VimeoBaseInfoExtractor): }) info = self._parse_config(self._download_json( video['config_url'], video_id), video_id) - self._vimeo_sort_formats(info['formats']) get_timestamp = lambda x: parse_iso8601(video.get(x + '_time')) info.update({ 'description': video.get('description'), @@ -834,10 +803,7 @@ class VimeoIE(VimeoBaseInfoExtractor): raise def _real_extract(self, url): - url, data = unsmuggle_url(url, {}) - headers = self.get_param('http_headers').copy() - if 'http_headers' in data: - headers.update(data['http_headers']) + url, data, headers = self._unsmuggle_headers(url) if 'Referer' not in headers: headers['Referer'] = url @@ -847,15 +813,7 @@ class VimeoIE(VimeoBaseInfoExtractor): if unlisted_hash: return self._extract_from_api(video_id, unlisted_hash) - orig_url = url - is_pro = 'vimeopro.com/' 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)) - if not url: - url = 'https://vimeo.com/' + video_id - elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')): + if any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')): url = 'https://vimeo.com/' + video_id self._try_album_password(url) @@ -877,13 +835,11 @@ class VimeoIE(VimeoBaseInfoExtractor): if '://player.vimeo.com/video/' in url: config = self._parse_json(self._search_regex( - r'\bconfig\s*=\s*({.+?})\s*;', webpage, 'info section'), video_id) + r'\b(?:playerC|c)onfig\s*=\s*({.+?})\s*;', webpage, 'info section'), video_id) if config.get('view') == 4: config = self._verify_player_video_password( redirect_url, video_id, headers) - info = self._parse_config(config, video_id) - self._vimeo_sort_formats(info['formats']) - return info + return self._parse_config(config, video_id) if re.search(r'<form[^>]+?id="pw_form"', webpage): video_password = self._get_video_password() @@ -959,14 +915,6 @@ class VimeoIE(VimeoBaseInfoExtractor): video_description = self._html_search_meta( ['description', 'og:description', 'twitter:description'], webpage, default=None) - if not video_description and is_pro: - orig_webpage = self._download_webpage( - orig_url, video_id, - note='Downloading webpage for description', - fatal=False) - if orig_webpage: - video_description = self._html_search_meta( - 'description', orig_webpage, default=None) if not video_description: self.report_warning('Cannot find video description') @@ -988,7 +936,7 @@ class VimeoIE(VimeoBaseInfoExtractor): info_dict_config = self._parse_config(config, video_id) formats.extend(info_dict_config['formats']) - self._vimeo_sort_formats(formats) + info_dict['_format_sort_fields'] = info_dict_config['_format_sort_fields'] json_ld = self._search_json_ld(webpage, video_id, default={}) @@ -1011,7 +959,7 @@ class VimeoIE(VimeoBaseInfoExtractor): return merge_dicts(info_dict, info_dict_config, json_ld) -class VimeoOndemandIE(VimeoIE): +class VimeoOndemandIE(VimeoIE): # XXX: Do not subclass from concrete IE IE_NAME = 'vimeo:ondemand' _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?:[^/]+/)?(?P<id>[^/?#&]+)' _TESTS = [{ @@ -1136,9 +1084,9 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): return self._extract_videos(channel_id, self._BASE_URL_TEMPL % channel_id) -class VimeoUserIE(VimeoChannelIE): +class VimeoUserIE(VimeoChannelIE): # XXX: Do not subclass from concrete IE IE_NAME = 'vimeo:user' - _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<id>[^/]+)(?:/videos|[#?]|$)' + _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', @@ -1147,6 +1095,9 @@ class VimeoUserIE(VimeoChannelIE): 'id': 'nkistudio', }, 'playlist_mincount': 66, + }, { + 'url': 'https://vimeo.com/nkistudio/', + 'only_matching': True, }] _BASE_URL_TEMPL = 'https://vimeo.com/%s' @@ -1243,7 +1194,7 @@ class VimeoAlbumIE(VimeoBaseInfoExtractor): entries, album_id, album.get('name'), album.get('description')) -class VimeoGroupsIE(VimeoChannelIE): +class VimeoGroupsIE(VimeoChannelIE): # XXX: Do not subclass from concrete IE IE_NAME = 'vimeo:group' _VALID_URL = r'https://vimeo\.com/groups/(?P<id>[^/]+)(?:/(?!videos?/\d+)|$)' _TESTS = [{ @@ -1330,14 +1281,13 @@ class VimeoReviewIE(VimeoBaseInfoExtractor): page_url + '/action', video_id) if source_format: info_dict['formats'].append(source_format) - self._vimeo_sort_formats(info_dict['formats']) info_dict['description'] = clean_html(clip_data.get('description')) return info_dict -class VimeoWatchLaterIE(VimeoChannelIE): +class VimeoWatchLaterIE(VimeoChannelIE): # XXX: Do not subclass from concrete IE IE_NAME = 'vimeo:watchlater' - IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)' + IE_DESC = 'Vimeo watch later list, ":vimeowatchlater" keyword (requires authentication)' _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater' _TITLE = 'Watch Later' _LOGIN_REQUIRED = True @@ -1358,7 +1308,7 @@ class VimeoWatchLaterIE(VimeoChannelIE): return self._extract_videos('watchlater', 'https://vimeo.com/watchlater') -class VimeoLikesIE(VimeoChannelIE): +class VimeoLikesIE(VimeoChannelIE): # XXX: Do not subclass from concrete IE _VALID_URL = r'https://(?:www\.)?vimeo\.com/(?P<id>[^/]+)/likes/?(?:$|[?#]|sort:)' IE_NAME = 'vimeo:likes' IE_DESC = 'Vimeo user likes' @@ -1385,21 +1335,107 @@ class VimeoLikesIE(VimeoChannelIE): class VHXEmbedIE(VimeoBaseInfoExtractor): IE_NAME = 'vhx:embed' _VALID_URL = r'https?://embed\.vhx\.tv/videos/(?P<id>\d+)' + _EMBED_REGEX = [r'<iframe[^>]+src="(?P<url>https?://embed\.vhx\.tv/videos/\d+[^"]*)"'] - @staticmethod - def _extract_url(webpage): - mobj = re.search( - r'<iframe[^>]+src="(https?://embed\.vhx\.tv/videos/\d+[^"]*)"', webpage) - return unescapeHTML(mobj.group(1)) if mobj else None + @classmethod + def _extract_embed_urls(cls, url, webpage): + for embed_url in super()._extract_embed_urls(url, webpage): + yield cls._smuggle_referrer(embed_url, url) def _real_extract(self, url): video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) + url, _, headers = self._unsmuggle_headers(url) + webpage = self._download_webpage(url, video_id, headers=headers) config_url = self._parse_json(self._search_regex( r'window\.OTTData\s*=\s*({.+})', webpage, 'ott data'), video_id, js_to_json)['config_url'] config = self._download_json(config_url, video_id) info = self._parse_config(config, video_id) info['id'] = video_id - self._vimeo_sort_formats(info['formats']) return info + + +class VimeoProIE(VimeoBaseInfoExtractor): + IE_NAME = 'vimeo:pro' + _VALID_URL = r'https?://(?:www\.)?vimeopro\.com/[^/?#]+/(?P<slug>[^/?#]+)(?:(?:/videos?/(?P<id>[0-9]+)))?' + _TESTS = [{ + # Vimeo URL derived from video_id + 'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876', + 'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82', + 'note': 'Vimeo Pro video (#1197)', + 'info_dict': { + 'id': '68093876', + 'ext': 'mp4', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/openstreetmapus', + 'uploader_id': 'openstreetmapus', + 'uploader': 'OpenStreetMap US', + 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography', + 'description': 'md5:2c362968038d4499f4d79f88458590c1', + 'duration': 1595, + 'upload_date': '20130610', + 'timestamp': 1370893156, + 'license': 'by', + 'thumbnail': 'https://i.vimeocdn.com/video/440260469-19b0d92fca3bd84066623b53f1eb8aaa3980c6c809e2d67b6b39ab7b4a77a344-d_960', + 'view_count': int, + 'comment_count': int, + 'like_count': int, + 'tags': 'count:1', + }, + 'params': { + 'format': 'best[protocol=https]', + }, + }, { + # password-protected VimeoPro page with Vimeo player embed + 'url': 'https://vimeopro.com/cadfem/simulation-conference-mechanische-systeme-in-perfektion', + 'info_dict': { + 'id': '764543723', + 'ext': 'mp4', + 'title': 'Mechanische Systeme in Perfektion: Realität erfassen, Innovation treiben', + 'thumbnail': 'https://i.vimeocdn.com/video/1543784598-a1a750494a485e601110136b9fe11e28c2131942452b3a5d30391cb3800ca8fd-d_1280', + 'description': 'md5:2a9d195cd1b0f6f79827107dc88c2420', + 'uploader': 'CADFEM', + 'uploader_id': 'cadfem', + 'uploader_url': 'https://vimeo.com/cadfem', + 'duration': 12505, + 'chapters': 'count:10', + }, + 'params': { + 'videopassword': 'Conference2022', + 'skip_download': True, + }, + }] + + def _real_extract(self, url): + display_id, video_id = self._match_valid_url(url).group('slug', 'id') + if video_id: + display_id = video_id + webpage = self._download_webpage(url, display_id) + + password_form = self._search_regex( + r'(?is)<form[^>]+?method=["\']post["\'][^>]*>(.+?password.+?)</form>', + webpage, 'password form', default=None) + if password_form: + try: + webpage = self._download_webpage(url, display_id, data=urlencode_postdata({ + 'password': self._get_video_password(), + **self._hidden_inputs(password_form), + }), note='Logging in with video password') + except ExtractorError as e: + if isinstance(e.cause, urllib.error.HTTPError) and e.cause.code == 418: + raise ExtractorError('Wrong video password', expected=True) + raise + + description = None + # even if we have video_id, some videos require player URL with portfolio_id query param + # https://github.com/ytdl-org/youtube-dl/issues/20070 + vimeo_url = VimeoIE._extract_url(url, webpage) + if vimeo_url: + description = self._html_search_meta('description', webpage, default=None) + elif video_id: + vimeo_url = f'https://vimeo.com/{video_id}' + else: + raise ExtractorError( + 'No Vimeo embed or video ID could be found in VimeoPro page', expected=True) + + return self.url_result(vimeo_url, VimeoIE, video_id, url_transparent=True, + description=description) |