X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fvimeo.py;h=bc01d7fbf583eb44ccc708c7fb00b4b113f82bc0;hb=cc7fec5818254f4679896823c7de9d17f50201ca;hp=ad2f75d6bb2075aef326d8b06839edb9aaf2f3ee;hpb=55b3e45bbab3af5132d45c8f3f8f19fae5f5f1d9;p=youtube-dl diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index ad2f75d6b..bc01d7fbf 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -1,127 +1,225 @@ +# encoding: utf-8 +from __future__ import unicode_literals + import json import re import itertools from .common import InfoExtractor +from .subtitles import SubtitlesInfoExtractor from ..utils import ( + compat_HTTPError, compat_urllib_parse, compat_urllib_request, - clean_html, get_element_by_attribute, ExtractorError, RegexNotFoundError, std_headers, unsmuggle_url, + urlencode_postdata, + int_or_none, ) -class VimeoIE(InfoExtractor): - """Information extractor for vimeo.com.""" - # _VALID_URL matches Vimeo URLs - _VALID_URL = r'(?Phttps?://)?(?:(?:www|player)\.)?vimeo(?Ppro)?\.com/(?:(?:(?:groups|album)/[^/]+)|(?:.*?)/)?(?Pplay_redirect_hls\?clip_id=)?(?:videos?/)?(?P[0-9]+)/?(?:[?].*)?$' +class VimeoBaseInfoExtractor(InfoExtractor): _NETRC_MACHINE = 'vimeo' - IE_NAME = u'vimeo' - _TESTS = [ - { - u'url': u'http://vimeo.com/56015672', - u'file': u'56015672.mp4', - u'md5': u'ae7a1d8b183758a0506b0622f37dfa14', - u'info_dict': { - u"upload_date": u"20121220", - u"description": u"This is a test case for youtube-dl.\nFor more information, see github.com/rg3/youtube-dl\nTest chars: \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", - u"uploader_id": u"user7108434", - u"uploader": u"Filippo Valsorda", - u"title": u"youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", - }, - }, - { - u'url': u'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876', - u'file': u'68093876.mp4', - u'md5': u'3b5ca6aa22b60dfeeadf50b72e44ed82', - u'note': u'Vimeo Pro video (#1197)', - u'info_dict': { - u'uploader_id': u'openstreetmapus', - u'uploader': u'OpenStreetMap US', - u'title': u'Andy Allan - Putting the Carto into OpenStreetMap Cartography', - }, - }, - { - u'url': u'http://player.vimeo.com/video/54469442', - u'file': u'54469442.mp4', - u'md5': u'619b811a4417aa4abe78dc653becf511', - u'note': u'Videos that embed the url in the player page', - u'info_dict': { - u'title': u'Kathy Sierra: Building the minimum Badass User, Business of Software', - u'uploader': u'The BLN & Business of Software', - }, - } - ] + _LOGIN_REQUIRED = False 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 self.report_login() login_url = 'https://vimeo.com/log_in' webpage = self._download_webpage(login_url, None, False) - token = re.search(r'xsrft: \'(.*?)\'', webpage).group(1) - data = compat_urllib_parse.urlencode({'email': username, - 'password': password, - 'action': 'login', - 'service': 'vimeo', - 'token': token, - }) + token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token') + data = urlencode_postdata({ + 'email': username, + 'password': password, + 'action': 'login', + 'service': 'vimeo', + 'token': token, + }) login_request = compat_urllib_request.Request(login_url, data) login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') login_request.add_header('Cookie', 'xsrft=%s' % token) - self._download_webpage(login_request, None, False, u'Wrong login info') + self._download_webpage(login_request, None, False, 'Wrong login info') + + +class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): + """Information extractor for vimeo.com.""" + + # _VALID_URL matches Vimeo URLs + _VALID_URL = r'''(?x) + (?P(?:https?:)?//)? + (?:(?:www|(?Pplayer))\.)? + vimeo(?Ppro)?\.com/ + (?!channels/[^/?#]+/?(?:$|[?#])|album/) + (?:.*?/)? + (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)? + (?:videos?/)? + (?P[0-9]+) + /?(?:[?&].*)?(?:[#].*)?$''' + IE_NAME = 'vimeo' + _TESTS = [ + { + 'url': 'http://vimeo.com/56015672#at=0', + 'md5': '8879b6cc097e987f02484baf890129e5', + 'info_dict': { + 'id': '56015672', + 'ext': 'mp4', + "upload_date": "20121220", + "description": "This is a test case for youtube-dl.\nFor more information, see github.com/rg3/youtube-dl\nTest chars: \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", + "uploader_id": "user7108434", + "uploader": "Filippo Valsorda", + "title": "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550", + "duration": 10, + }, + }, + { + '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_id': 'openstreetmapus', + 'uploader': 'OpenStreetMap US', + 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography', + 'duration': 1595, + }, + }, + { + 'url': 'http://player.vimeo.com/video/54469442', + 'md5': '619b811a4417aa4abe78dc653becf511', + 'note': 'Videos that embed the url in the player page', + 'info_dict': { + 'id': '54469442', + 'ext': 'mp4', + 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012', + 'uploader': 'The BLN & Business of Software', + 'uploader_id': 'theblnbusinessofsoftware', + 'duration': 3610, + }, + }, + { + 'url': 'http://vimeo.com/68375962', + 'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7', + 'note': 'Video protected with password', + 'info_dict': { + 'id': '68375962', + 'ext': 'mp4', + 'title': 'youtube-dl password protected test video', + 'upload_date': '20130614', + 'uploader_id': 'user18948128', + 'uploader': 'Jaime Marquínez Ferrándiz', + 'duration': 10, + }, + 'params': { + 'videopassword': 'youtube-dl', + }, + }, + { + 'url': 'http://vimeo.com/channels/keypeele/75629013', + 'md5': '2f86a05afe9d7abc0b9126d229bbe15d', + 'note': 'Video is freely available via original URL ' + 'and protected with password when accessed via http://vimeo.com/75629013', + 'info_dict': { + 'id': '75629013', + 'ext': 'mp4', + 'title': 'Key & Peele: Terrorist Interrogation', + 'description': 'md5:8678b246399b070816b12313e8b4eb5c', + 'uploader_id': 'atencio', + 'uploader': 'Peter Atencio', + 'duration': 187, + }, + }, + { + 'url': 'http://vimeo.com/76979871', + 'md5': '3363dd6ffebe3784d56f4132317fd446', + 'note': 'Video with subtitles', + 'info_dict': { + 'id': '76979871', + 'ext': 'mp4', + 'title': 'The New Vimeo Player (You Know, For Videos)', + 'description': 'md5:2ec900bf97c3f389378a96aee11260ea', + 'upload_date': '20131015', + 'uploader_id': 'staff', + 'uploader': 'Vimeo Staff', + 'duration': 62, + } + }, + ] def _verify_video_password(self, url, video_id, webpage): password = self._downloader.params.get('videopassword', None) if password is None: - raise ExtractorError(u'This video is protected by a password, use the --video-password option') - token = re.search(r'xsrft: \'(.*?)\'', webpage).group(1) - data = compat_urllib_parse.urlencode({'password': password, - 'token': token}) + 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({ + 'password': password, + 'token': token, + }) # I didn't manage to use the password with https if url.startswith('https'): - pass_url = url.replace('https','http') + pass_url = url.replace('https', 'http') else: pass_url = url - password_request = compat_urllib_request.Request(pass_url+'/password', data) + password_request = compat_urllib_request.Request(pass_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, - u'Verifying the password', - u'Wrong password') + 'Verifying the password', + 'Wrong password') + + def _verify_player_video_password(self, url, video_id): + 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') + data = compat_urllib_parse.urlencode({'password': password}) + pass_url = url + '/check-password' + password_request = compat_urllib_request.Request(pass_url, data) + password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') + return self._download_json( + password_request, video_id, + 'Verifying the password', + 'Wrong password') def _real_initialize(self): self._login() - def _real_extract(self, url, new_video=True): + def _real_extract(self, url): url, data = unsmuggle_url(url) headers = std_headers if data is not None: headers = headers.copy() headers.update(data) + if 'Referer' not in headers: + headers['Referer'] = url # Extract ID from URL mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - video_id = mobj.group('id') - if not mobj.group('proto'): - url = 'https://' + url - elif mobj.group('pro'): + if mobj.group('pro') or mobj.group('player'): url = 'http://player.vimeo.com/video/' + video_id - elif mobj.group('direct_link'): - url = 'https://vimeo.com/' + video_id # Retrieve video webpage to extract further information request = compat_urllib_request.Request(url, None, headers) - webpage = self._download_webpage(request, video_id) + try: + webpage = self._download_webpage(request, video_id) + except ExtractorError as ee: + if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403: + errmsg = ee.cause.read() + if b'Because of its privacy settings, this video cannot be played here' in errmsg: + raise ExtractorError( + 'Cannot download embed-only video without embedding ' + 'URL. Please call youtube-dl with the URL of the page ' + 'that embeds this video.', + expected=True) + raise # Now we begin extracting as much information as we can from what we # retrieved. First we extract the information common to all extractors, @@ -130,25 +228,35 @@ class VimeoIE(InfoExtractor): # Extract the config JSON try: - config_url = self._html_search_regex( - r' data-config-url="(.+?)"', webpage, u'config URL') - config_json = self._download_webpage(config_url, video_id) - config = json.loads(config_json) - except RegexNotFoundError: - # For pro videos or player.vimeo.com urls - config = self._search_regex([r' = {config:({.+?}),assets:', r'c=({.+?);'], - webpage, u'info section', flags=re.DOTALL) - config = json.loads(config) + try: + config_url = self._html_search_regex( + r' data-config-url="(.+?)"', webpage, 'config URL') + config_json = self._download_webpage(config_url, video_id) + config = json.loads(config_json) + except RegexNotFoundError: + # For pro videos or player.vimeo.com urls + # We try to find out to which variable is assigned the config dic + m_variable_name = re.search('(\w)\.video\.id', webpage) + if m_variable_name is not None: + config_re = r'%s=({.+?});' % re.escape(m_variable_name.group(1)) + else: + config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});'] + config = self._search_regex(config_re, webpage, 'info section', + flags=re.DOTALL) + config = json.loads(config) except Exception as e: if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage): - raise ExtractorError(u'The author has restricted the access to this video, try with the "--referer" option') + raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option') - if re.search('If so please provide the correct password.', webpage): + if re.search(']+?id="pw_form"', webpage) is not None: self._verify_video_password(url, video_id, webpage) return self._real_extract(url) else: - raise ExtractorError(u'Unable to extract info section', + raise ExtractorError('Unable to extract info section', cause=e) + else: + if config.get('view') == 4: + config = self._verify_player_video_password(url, video_id) # Extract title video_title = config["video"]["title"] @@ -160,13 +268,16 @@ class VimeoIE(InfoExtractor): # Extract video thumbnail video_thumbnail = config["video"].get("thumbnail") if video_thumbnail is None: - _, video_thumbnail = sorted((int(width), t_url) for (width, t_url) in config["video"]["thumbs"].items())[-1] + video_thumbs = config["video"].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 description video_description = None try: - video_description = get_element_by_attribute("itemprop", "description", webpage) - if video_description: video_description = clean_html(video_description) + video_description = get_element_by_attribute("class", "description_wrapper", webpage) + if video_description: + video_description = clean_html(video_description) except AssertionError as err: # On some pages like (http://player.vimeo.com/video/54469442) the # html tags are not closed, python 2.6 cannot handle it @@ -175,12 +286,25 @@ class VimeoIE(InfoExtractor): else: raise + # Extract video duration + video_duration = int_or_none(config["video"].get("duration")) + # Extract upload date video_upload_date = None mobj = re.search(r'[^/]+)' + IE_NAME = 'vimeo:channel' + _VALID_URL = r'https?://vimeo\.com/channels/(?P[^/?#]+)/?(?:$|[?#])' _MORE_PAGES_INDICATOR = r']+?title="(.*?)"' + _TESTS = [{ + 'url': 'http://vimeo.com/channels/tributes', + 'info_dict': { + 'title': 'Vimeo Tributes', + }, + 'playlist_mincount': 25, + }] - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - channel_id = mobj.group('id') - video_ids = [] + def _page_url(self, base_url, pagenum): + 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') + + def _extract_videos(self, list_id, base_url): + video_ids = [] for pagenum in itertools.count(1): - webpage = self._download_webpage('http://vimeo.com/channels/%s/videos/page:%d' % (channel_id, pagenum), - channel_id, u'Downloading page %s' % pagenum) + webpage = self._download_webpage( + self._page_url(base_url, pagenum), list_id, + 'Downloading page %s' % pagenum) 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') for video_id in video_ids] - channel_title = self._html_search_regex(r'(.*?)' % channel_id, - webpage, u'channel title') return {'_type': 'playlist', - 'id': channel_id, - 'title': channel_title, + 'id': list_id, + 'title': self._extract_list_title(webpage), 'entries': entries, } + + 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) + + +class VimeoUserIE(VimeoChannelIE): + IE_NAME = 'vimeo:user' + _VALID_URL = r'https?://vimeo\.com/(?![0-9]+(?:$|[?#/]))(?P[^/]+)(?:/videos|[#?]|$)' + _TITLE_RE = r']+?class="user">([^<>]+?)' + _TESTS = [{ + 'url': 'http://vimeo.com/nkistudio/videos', + 'info_dict': { + 'title': 'Nki', + }, + 'playlist_mincount': 66, + }] + + 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) + + +class VimeoAlbumIE(VimeoChannelIE): + IE_NAME = 'vimeo:album' + _VALID_URL = r'https?://vimeo\.com/album/(?P\d+)' + _TITLE_RE = r'
\n\s*

(.*?)

' + _TESTS = [{ + 'url': 'http://vimeo.com/album/2632481', + 'info_dict': { + 'title': 'Staff Favorites: November 2013', + }, + 'playlist_mincount': 13, + }] + + 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) + + +class VimeoGroupsIE(VimeoAlbumIE): + IE_NAME = 'vimeo:group' + _VALID_URL = r'(?:https?://)?vimeo\.com/groups/(?P[^/]+)' + _TESTS = [{ + 'url': 'http://vimeo.com/groups/rolexawards', + 'info_dict': { + 'title': 'Rolex Awards for Enterprise', + }, + 'playlist_mincount': 73, + }] + + def _extract_list_title(self, webpage): + return self._og_search_title(webpage) + + 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) + + +class VimeoReviewIE(InfoExtractor): + IE_NAME = 'vimeo:review' + IE_DESC = 'Review pages on vimeo' + _VALID_URL = r'https?://vimeo\.com/[^/]+/review/(?P[^/]+)' + _TESTS = [{ + 'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d', + 'file': '75524534.mp4', + 'md5': 'c507a72f780cacc12b2248bb4006d253', + 'info_dict': { + 'title': "DICK HARDWICK 'Comedian'", + 'uploader': 'Richard Hardwick', + } + }, { + 'note': 'video player needs Referer', + 'url': 'http://vimeo.com/user22258446/review/91613211/13f927e053', + 'md5': '6295fdab8f4bf6a002d058b2c6dce276', + 'info_dict': { + 'id': '91613211', + 'ext': 'mp4', + 'title': 'Death by dogma versus assembling agile - Sander Hoogendoorn', + 'uploader': 'DevWeek Events', + 'duration': 2773, + 'thumbnail': 're:^https?://.*\.jpg$', + } + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + player_url = 'https://player.vimeo.com/player/' + video_id + return self.url_result(player_url, 'Vimeo', video_id) + + +class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE): + IE_NAME = 'vimeo:watchlater' + IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)' + _VALID_URL = r'https?://vimeo\.com/home/watchlater|:vimeowatchlater' + _LOGIN_REQUIRED = True + _TITLE_RE = r'href="/home/watchlater".*?>(.*?)<' + _TESTS = [{ + 'url': 'http://vimeo.com/home/watchlater', + 'only_matching': True, + }] + + def _real_initialize(self): + self._login() + + def _page_url(self, base_url, pagenum): + url = '%s/page:%d/' % (base_url, pagenum) + request = compat_urllib_request.Request(url) + # Set the header to get a partial html page with the ids, + # the normal page doesn't contain them. + request.add_header('X-Requested-With', 'XMLHttpRequest') + return request + + def _real_extract(self, url): + return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater')