X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fvimeo.py;h=50df79ca12d7082c9615bba0fdc979a5f7ed6578;hb=11bed5827dace09b5483b159476ce9f8c29d6078;hp=06b0bed41e68401a8667cbabdca0d9796ea8ca3d;hpb=2f15832f569834a37ac3ee6140a3b997407c04cd;p=youtube-dl diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index 06b0bed41..50df79ca1 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -6,7 +6,6 @@ import re import itertools from .common import InfoExtractor -from .subtitles import SubtitlesInfoExtractor from ..compat import ( compat_HTTPError, compat_urllib_parse, @@ -18,15 +17,19 @@ from ..utils import ( InAdvancePagedList, int_or_none, RegexNotFoundError, + smuggle_url, std_headers, + unified_strdate, unsmuggle_url, urlencode_postdata, + unescapeHTML, ) 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() @@ -35,23 +38,27 @@ class VimeoBaseInfoExtractor(InfoExtractor): raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True) return self.report_login() - login_url = 'https://vimeo.com/log_in' - webpage = self._download_webpage(login_url, None, False) - token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token') + webpage = self._download_webpage(self._LOGIN_URL, None, False) + token = self._extract_xsrft(webpage) data = urlencode_postdata({ + 'action': 'login', 'email': username, 'password': password, - 'action': 'login', 'service': 'vimeo', 'token': token, }) - login_request = compat_urllib_request.Request(login_url, data) + login_request = compat_urllib_request.Request(self._LOGIN_URL, data) login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - login_request.add_header('Cookie', 'xsrft=%s' % token) + login_request.add_header('Referer', self._LOGIN_URL) self._download_webpage(login_request, None, False, 'Wrong login info') + def _extract_xsrft(self, webpage): + return self._search_regex( + r'xsrft\s*[=:]\s*(?P["\'])(?P.+?)(?P=q)', + webpage, 'login token', group='xsrft') + -class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): +class VimeoIE(VimeoBaseInfoExtractor): """Information extractor for vimeo.com.""" # _VALID_URL matches Vimeo URLs @@ -139,6 +146,7 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): 'description': 'md5:8678b246399b070816b12313e8b4eb5c', 'uploader_id': 'atencio', 'uploader': 'Peter Atencio', + 'upload_date': '20130927', 'duration': 187, }, }, @@ -171,26 +179,39 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): }, ] + @staticmethod + def _extract_vimeo_url(url, webpage): + # Look for embedded (iframe) Vimeo player + mobj = re.search( + r']+?src=(["\'])(?P(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage) + if mobj: + player_url = unescapeHTML(mobj.group('url')) + surl = smuggle_url(player_url, {'Referer': url}) + return surl + # Look for embedded (swf embed) Vimeo player + mobj = re.search( + r']+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage) + if mobj: + return mobj.group(1) + def _verify_video_password(self, url, video_id, webpage): password = self._downloader.params.get('videopassword', None) if password is None: - raise ExtractorError('This video is protected by a password, use the --video-password option') - token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token') - data = compat_urllib_parse.urlencode({ + raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True) + token = self._extract_xsrft(webpage) + data = urlencode_postdata({ 'password': password, 'token': token, }) - # I didn't manage to use the password with https - if url.startswith('https'): - pass_url = url.replace('https', 'http') - else: - pass_url = url - password_request = compat_urllib_request.Request(pass_url + '/password', data) + if url.startswith('http://'): + # vimeo only supports https now, but the user can give an http url + url = url.replace('http://', 'https://') + password_request = compat_urllib_request.Request(url + '/password', data) password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - password_request.add_header('Cookie', 'xsrft=%s' % token) - self._download_webpage(password_request, video_id, - 'Verifying the password', - 'Wrong password') + password_request.add_header('Referer', url) + return self._download_webpage( + password_request, video_id, + 'Verifying the password', 'Wrong password') def _verify_player_video_password(self, url, video_id): password = self._downloader.params.get('videopassword', None) @@ -222,7 +243,9 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): video_id = mobj.group('id') orig_url = url if mobj.group('pro') or mobj.group('player'): - url = 'http://player.vimeo.com/video/' + video_id + url = 'https://player.vimeo.com/video/' + video_id + else: + url = 'https://vimeo.com/' + video_id # Retrieve video webpage to extract further information request = compat_urllib_request.Request(url, None, headers) @@ -244,6 +267,16 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): # and latter we extract those that are Vimeo specific. self.report_extraction(video_id) + vimeo_config = self._search_regex( + r'vimeo\.config\s*=\s*({.+?});', webpage, + 'vimeo config', default=None) + if vimeo_config: + seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {}) + if seed_status.get('state') == 'failed': + raise ExtractorError( + '%s returned error: %s' % (self.IE_NAME, seed_status['title']), + expected=True) + # Extract the config JSON try: try: @@ -266,9 +299,12 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage): raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option') - if re.search(']+?id="pw_form"', webpage) is not None: + if re.search(r']+?id="pw_form"', webpage) is not None: + if data and '_video_password_verified' in data: + raise ExtractorError('video password verification failed!') self._verify_video_password(url, video_id, webpage) - return self._real_extract(url) + return self._real_extract( + smuggle_url(url, {'_video_password_verified': 'verified'})) else: raise ExtractorError('Unable to extract info section', cause=e) @@ -314,9 +350,9 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): # Extract upload date video_upload_date = None - mobj = re.search(r'[^/?#]+)/?(?:$|[?#])' + _VALID_URL = r'https://vimeo\.com/channels/(?P[^/?#]+)/?(?:$|[?#])' _MORE_PAGES_INDICATOR = r']+?title="(.*?)"' _TESTS = [{ - 'url': 'http://vimeo.com/channels/tributes', + 'url': 'https://vimeo.com/channels/tributes', 'info_dict': { + 'id': 'tributes', 'title': 'Vimeo Tributes', }, 'playlist_mincount': 25, @@ -410,19 +446,50 @@ class VimeoChannelIE(InfoExtractor): return '%s/videos/page:%d/' % (base_url, pagenum) def _extract_list_title(self, webpage): - return self._html_search_regex(self._TITLE_RE, webpage, 'list title') + return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title') + + def _login_list_password(self, page_url, list_id, webpage): + login_form = self._search_regex( + r'(?s)]+?id="pw_form"(.*?)', + webpage, 'login form', default=None) + if not login_form: + return webpage + + password = self._downloader.params.get('videopassword', None) + 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 = self._extract_xsrft(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 = compat_urllib_request.Request(password_url, post) + password_request.add_header('Content-type', 'application/x-www-form-urlencoded') + self._set_cookie('vimeo.com', 'xsrft', token) + + return self._download_webpage( + password_request, list_id, + 'Verifying the password', 'Wrong password') def _extract_videos(self, list_id, base_url): video_ids = [] for pagenum in itertools.count(1): + page_url = self._page_url(base_url, pagenum) webpage = self._download_webpage( - self._page_url(base_url, pagenum), list_id, + page_url, list_id, 'Downloading page %s' % pagenum) + + if pagenum == 1: + webpage = self._login_list_password(page_url, list_id, webpage) + video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage)) if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None: break - entries = [self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo') + entries = [self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo') for video_id in video_ids] return {'_type': 'playlist', 'id': list_id, @@ -433,17 +500,18 @@ class VimeoChannelIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) channel_id = mobj.group('id') - return self._extract_videos(channel_id, 'http://vimeo.com/channels/%s' % channel_id) + return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id) class VimeoUserIE(VimeoChannelIE): IE_NAME = 'vimeo:user' - _VALID_URL = r'https?://vimeo\.com/(?![0-9]+(?:$|[?#/]))(?P[^/]+)(?:/videos|[#?]|$)' + _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P[^/]+)(?:/videos|[#?]|$)' _TITLE_RE = r']+?class="user">([^<>]+?)' _TESTS = [{ - 'url': 'http://vimeo.com/nkistudio/videos', + 'url': 'https://vimeo.com/nkistudio/videos', 'info_dict': { 'title': 'Nki', + 'id': 'nkistudio', }, 'playlist_mincount': 66, }] @@ -451,36 +519,48 @@ class VimeoUserIE(VimeoChannelIE): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) name = mobj.group('name') - return self._extract_videos(name, 'http://vimeo.com/%s' % name) + return self._extract_videos(name, 'https://vimeo.com/%s' % name) class VimeoAlbumIE(VimeoChannelIE): IE_NAME = 'vimeo:album' - _VALID_URL = r'https?://vimeo\.com/album/(?P\d+)' + _VALID_URL = r'https://vimeo\.com/album/(?P\d+)' _TITLE_RE = r'
\n\s*

(.*?)

' _TESTS = [{ - 'url': 'http://vimeo.com/album/2632481', + 'url': 'https://vimeo.com/album/2632481', 'info_dict': { + 'id': '2632481', 'title': 'Staff Favorites: November 2013', }, 'playlist_mincount': 13, + }, { + 'note': 'Password-protected album', + 'url': 'https://vimeo.com/album/3253534', + 'info_dict': { + 'title': 'test', + 'id': '3253534', + }, + 'playlist_count': 1, + 'params': { + 'videopassword': 'youtube-dl', + } }] def _page_url(self, base_url, pagenum): return '%s/page:%d/' % (base_url, pagenum) def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - album_id = mobj.group('id') - return self._extract_videos(album_id, 'http://vimeo.com/album/%s' % album_id) + album_id = self._match_id(url) + return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id) class VimeoGroupsIE(VimeoAlbumIE): IE_NAME = 'vimeo:group' - _VALID_URL = r'(?:https?://)?vimeo\.com/groups/(?P[^/]+)' + _VALID_URL = r'https://vimeo\.com/groups/(?P[^/]+)' _TESTS = [{ - 'url': 'http://vimeo.com/groups/rolexawards', + 'url': 'https://vimeo.com/groups/rolexawards', 'info_dict': { + 'id': 'rolexawards', 'title': 'Rolex Awards for Enterprise', }, 'playlist_mincount': 73, @@ -492,24 +572,25 @@ class VimeoGroupsIE(VimeoAlbumIE): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) name = mobj.group('name') - return self._extract_videos(name, 'http://vimeo.com/groups/%s' % name) + return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name) class VimeoReviewIE(InfoExtractor): IE_NAME = 'vimeo:review' IE_DESC = 'Review pages on vimeo' - _VALID_URL = r'https?://vimeo\.com/[^/]+/review/(?P[^/]+)' + _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P[^/]+)' _TESTS = [{ 'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d', - 'file': '75524534.mp4', 'md5': 'c507a72f780cacc12b2248bb4006d253', 'info_dict': { + 'id': '75524534', + 'ext': 'mp4', 'title': "DICK HARDWICK 'Comedian'", 'uploader': 'Richard Hardwick', } }, { 'note': 'video player needs Referer', - 'url': 'http://vimeo.com/user22258446/review/91613211/13f927e053', + 'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053', 'md5': '6295fdab8f4bf6a002d058b2c6dce276', 'info_dict': { 'id': '91613211', @@ -528,14 +609,14 @@ class VimeoReviewIE(InfoExtractor): return self.url_result(player_url, 'Vimeo', video_id) -class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE): +class VimeoWatchLaterIE(VimeoChannelIE): IE_NAME = 'vimeo:watchlater' IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)' - _VALID_URL = r'https?://vimeo\.com/home/watchlater|:vimeowatchlater' + _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater' + _TITLE = 'Watch Later' _LOGIN_REQUIRED = True - _TITLE_RE = r'href="/home/watchlater".*?>(.*?)<' _TESTS = [{ - 'url': 'http://vimeo.com/home/watchlater', + 'url': 'https://vimeo.com/watchlater', 'only_matching': True, }] @@ -551,17 +632,18 @@ class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE): return request def _real_extract(self, url): - return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater') + return self._extract_videos('watchlater', 'https://vimeo.com/watchlater') class VimeoLikesIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?vimeo\.com/user(?P[0-9]+)/likes/?(?:$|[?#]|sort:)' + _VALID_URL = r'https://(?:www\.)?vimeo\.com/user(?P[0-9]+)/likes/?(?:$|[?#]|sort:)' IE_NAME = 'vimeo:likes' IE_DESC = 'Vimeo user likes' _TEST = { 'url': 'https://vimeo.com/user755559/likes/', 'playlist_mincount': 293, "info_dict": { + 'id': 'user755559_likes', "description": "See all the videos urza likes", "title": 'Videos urza likes', }, @@ -582,8 +664,8 @@ class VimeoLikesIE(InfoExtractor): description = self._html_search_meta('description', webpage) def _get_page(idx): - page_url = '%s//vimeo.com/user%s/likes/page:%d/sort:date' % ( - self.http_scheme(), user_id, idx + 1) + page_url = 'https://vimeo.com/user%s/likes/page:%d/sort:date' % ( + user_id, idx + 1) webpage = self._download_webpage( page_url, user_id, note='Downloading page %d/%d' % (idx + 1, page_count))