X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/d1a5eac72c0b1116e8efbbf50766d8d00678f5c8..779bc665c512f2802f1436a30b6b09ee7ad83e02:/youtube_dl/extractor/vimeo.py diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index 1936755..2558555 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -6,20 +6,50 @@ 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): +class VimeoBaseInfoExtractor(InfoExtractor): + _NETRC_MACHINE = 'vimeo' + _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 = 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, 'Wrong login info') + + +class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): """Information extractor for vimeo.com.""" # _VALID_URL matches Vimeo URLs @@ -32,92 +62,106 @@ class VimeoIE(InfoExtractor): (?:videos?/)? (?P[0-9]+) /?(?:[?&].*)?(?:[#].*)?$''' - _NETRC_MACHINE = 'vimeo' IE_NAME = 'vimeo' _TESTS = [ { 'url': 'http://vimeo.com/56015672#at=0', - 'file': '56015672.mp4', 'md5': '8879b6cc097e987f02484baf890129e5', 'info_dict': { - "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", + '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', - 'file': '68093876.mp4', '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', - 'file': '54469442.mp4', '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', 'uploader': 'The BLN & Business of Software', 'uploader_id': 'theblnbusinessofsoftware', + 'duration': 3610, }, }, { 'url': 'http://vimeo.com/68375962', - 'file': '68375962.mp4', '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/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 _login(self): - (username, password) = self._get_login_info() - if username is None: - 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') - data = compat_urllib_parse.urlencode({'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, 'Wrong login info') + @classmethod + def suitable(cls, url): + if VimeoChannelIE.suitable(url): + # Otherwise channel urls like http://vimeo.com/channels/31259 would + # match + return False + else: + return super(VimeoIE, cls).suitable(url) 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({'password': password, - 'token': 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, @@ -157,7 +201,18 @@ class VimeoIE(InfoExtractor): # 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, @@ -206,13 +261,16 @@ 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), 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 @@ -221,6 +279,9 @@ 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'[^/]+)' + _VALID_URL = r'(?:https?://)?vimeo\.com/channels/(?P[^/]+)/?(\?.*)?$' _MORE_PAGES_INDICATOR = r']+?title="(.*?)"' @@ -305,7 +379,7 @@ class VimeoChannelIE(InfoExtractor): video_ids = [] for pagenum in itertools.count(1): webpage = self._download_webpage( - self._page_url(base_url, pagenum) ,list_id, + 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: @@ -321,13 +395,13 @@ class VimeoChannelIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - channel_id = mobj.group('id') + 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/(?P[^/]+)(?:/videos|[#?]|$)' + _VALID_URL = r'(?:https?://)?vimeo\.com/(?P[^/]+)(?:/videos|[#?]|$)' _TITLE_RE = r']+?class="user">([^<>]+?)' @classmethod @@ -344,7 +418,7 @@ class VimeoUserIE(VimeoChannelIE): 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*

(.*?)

' def _page_url(self, base_url, pagenum): @@ -358,7 +432,7 @@ class VimeoAlbumIE(VimeoChannelIE): class VimeoGroupsIE(VimeoAlbumIE): IE_NAME = 'vimeo:group' - _VALID_URL = r'(?:https?://)?vimeo.\com/groups/(?P[^/]+)' + _VALID_URL = r'(?:https?://)?vimeo\.com/groups/(?P[^/]+)' def _extract_list_title(self, webpage): return self._og_search_title(webpage) @@ -372,7 +446,7 @@ class VimeoGroupsIE(VimeoAlbumIE): 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[^/]+)' _TEST = { 'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d', 'file': '75524534.mp4', @@ -388,3 +462,25 @@ class VimeoReviewIE(InfoExtractor): 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".*?>(.*?)<' + + 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')