X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/a316b1d93c357b5edf19d5e3100526a94191c029..139d14b198add5b26a11ed5dca022feaa6581d98:/youtube_dl/extractor/vimeo.py diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index 512e06e..f300c7c 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -1,128 +1,295 @@ +# encoding: utf-8 +from __future__ import unicode_literals + import json import re import itertools from .common import InfoExtractor -from ..utils import ( +from ..compat import ( + compat_HTTPError, compat_urllib_parse, compat_urllib_request, - - clean_html, - get_element_by_attribute, + compat_urlparse, +) +from ..utils import ( ExtractorError, + InAdvancePagedList, + int_or_none, + RegexNotFoundError, + smuggle_url, std_headers, + unified_strdate, + unsmuggle_url, + urlencode_postdata, ) -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'8879b6cc097e987f02484baf890129e5', - 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', - }, - }, - ] + _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): + """Information extractor for vimeo.com.""" + + # _VALID_URL matches Vimeo URLs + _VALID_URL = r'''(?x) + 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', + 'description': 'md5:380943ec71b89736ff4bf27183233d09', + '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, + 'description': None, + }, + }, + { + '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, + 'description': 'This is "youtube-dl password protected test video" by Jaime Marquínez Ferrándiz on Vimeo, the home for high quality videos and the people who love them.', + }, + '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', + 'upload_date': '20130927', + '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, + } + }, + { + # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/ + 'url': 'https://player.vimeo.com/video/98044508', + 'note': 'The js code contains assignments to the same variable as the config', + 'info_dict': { + 'id': '98044508', + 'ext': 'mp4', + 'title': 'Pier Solar OUYA Official Trailer', + 'uploader': 'Tulio Gonçalves', + 'uploader_id': 'user28849593', + }, + }, + ] 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}) - # 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) + raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True) + token = self._search_regex(r'xsrft[\s=:"\']+([^"\']+)', webpage, 'login token') + data = urlencode_postdata({ + 'password': password, + 'token': token, + }) + 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, - u'Verifying the password', - u'Wrong password') + 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) + 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'): - url = 'http://player.vimeo.com/video/' + video_id - elif mobj.group('direct_link'): + orig_url = url + if mobj.group('pro') or mobj.group('player'): + 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, std_headers) - webpage = self._download_webpage(request, video_id) + request = compat_urllib_request.Request(url, None, headers) + 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, # 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: - config = webpage.split(' = {config:')[1].split(',assets:')[0] - config = json.loads(config) - except: + 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(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(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"] @@ -132,18 +299,49 @@ class VimeoIE(InfoExtractor): video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None # Extract video thumbnail - video_thumbnail = config["video"]["thumbnail"] + video_thumbnail = config["video"].get("thumbnail") + if video_thumbnail is None: + 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 = get_element_by_attribute("itemprop", "description", webpage) - if video_description: video_description = clean_html(video_description) - else: video_description = u'' + + video_description = self._html_search_regex( + r'(?s)]*>(.*?)', + 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'): + 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 and not mobj.group('player'): + self._downloader.report_warning('Cannot find video description') + + # Extract video duration + video_duration = int_or_none(config["video"].get("duration")) # Extract upload date video_upload_date = None - mobj = re.search(r' 0: - video_quality = files[quality][0][2] - video_codec = files[quality][0][0] - video_extension = files[quality][0][1] - self.to_screen(u'%s: Downloading %s file at %s quality' % (video_id, video_codec.upper(), video_quality)) - break - else: - raise ExtractorError(u'No known codec found') + file_info = {} + if video_url is None: + video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \ + % (video_id, sig, timestamp, quality, codec_name.upper()) + + files[key].append({ + 'ext': codec_extension, + 'url': video_url, + 'format_id': format_id, + 'width': file_info.get('width'), + 'height': file_info.get('height'), + }) + formats = [] + for key in ('other', 'sd', 'hd'): + formats += files[key] + if len(formats) == 0: + raise ExtractorError('No known codec found') - video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \ - %(video_id, sig, timestamp, video_quality, video_codec.upper()) + 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'], + }] - return [{ - 'id': video_id, - 'url': video_url, + return { + 'id': video_id, 'uploader': video_uploader, 'uploader_id': video_uploader_id, - 'upload_date': video_upload_date, - 'title': video_title, - 'ext': video_extension, - 'thumbnail': video_thumbnail, - 'description': video_description, - }] + 'upload_date': video_upload_date, + 'title': video_title, + 'thumbnail': video_thumbnail, + 'description': video_description, + 'duration': video_duration, + 'formats': formats, + 'webpage_url': url, + 'view_count': view_count, + 'like_count': like_count, + 'comment_count': comment_count, + 'subtitles': subtitles, + } class VimeoChannelIE(InfoExtractor): - IE_NAME = u'vimeo:channel' - _VALID_URL = r'(?:https?://)?vimeo.\com/channels/(?P[^/]+)' + IE_NAME = 'vimeo:channel' + _VALID_URL = r'https://vimeo\.com/channels/(?P[^/?#]+)/?(?:$|[?#])' _MORE_PAGES_INDICATOR = r']+?title="(.*?)"' + _TESTS = [{ + 'url': 'https://vimeo.com/channels/tributes', + 'info_dict': { + 'id': 'tributes', + '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 _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 = dict(re.findall(r'''(?x)(.*?)' % 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, 'https://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': 'https://vimeo.com/nkistudio/videos', + 'info_dict': { + 'title': 'Nki', + 'id': 'nkistudio', + }, + 'playlist_mincount': 66, + }] + + 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): + IE_NAME = 'vimeo:album' + _VALID_URL = r'https://vimeo\.com/album/(?P\d+)' + _TITLE_RE = r'