]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/vimeo.py
Update changelog.
[youtubedl] / youtube_dl / extractor / vimeo.py
index 06b0bed41e68401a8667cbabdca0d9796ea8ca3d..f300c7ca40eeba75f872bbd6cd79c3f0720e0955 100644 (file)
@@ -6,7 +6,6 @@ import re
 import itertools
 
 from .common import InfoExtractor
-from .subtitles import SubtitlesInfoExtractor
 from ..compat import (
     compat_HTTPError,
     compat_urllib_parse,
@@ -18,7 +17,9 @@ from ..utils import (
     InAdvancePagedList,
     int_or_none,
     RegexNotFoundError,
+    smuggle_url,
     std_headers,
+    unified_strdate,
     unsmuggle_url,
     urlencode_postdata,
 )
@@ -37,7 +38,7 @@ class VimeoBaseInfoExtractor(InfoExtractor):
         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')
+        token = self._search_regex(r'xsrft":"(.*?)"', webpage, 'login token')
         data = urlencode_postdata({
             'email': username,
             'password': password,
@@ -51,7 +52,7 @@ class VimeoBaseInfoExtractor(InfoExtractor):
         self._download_webpage(login_request, None, False, 'Wrong login info')
 
 
-class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
+class VimeoIE(VimeoBaseInfoExtractor):
     """Information extractor for vimeo.com."""
 
     # _VALID_URL matches Vimeo URLs
@@ -139,6 +140,7 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
                 'uploader_id': 'atencio',
                 'uploader': 'Peter Atencio',
+                'upload_date': '20130927',
                 'duration': 187,
             },
         },
@@ -174,23 +176,21 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
     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._search_regex(r'xsrft[\s=:"\']+([^"\']+)', webpage, 'login token')
+        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')
+        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 +222,9 @@ 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 +246,16 @@ 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 +278,12 @@ 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('<form[^>]+?id="pw_form"', webpage) is not None:
+            if re.search(r'<form[^>]+?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 +329,9 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
 
         # Extract upload date
         video_upload_date = None
-        mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
+        mobj = re.search(r'<time[^>]+datetime="([^"]+)"', webpage)
         if mobj is not None:
-            video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
+            video_upload_date = unified_strdate(mobj.group(1))
 
         try:
             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
@@ -368,12 +383,10 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
         text_tracks = config['request'].get('text_tracks')
         if text_tracks:
             for tt in text_tracks:
-                subtitles[tt['lang']] = 'http://vimeo.com' + tt['url']
-
-        video_subtitles = self.extract_subtitles(video_id, subtitles)
-        if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(video_id, subtitles)
-            return
+                subtitles[tt['lang']] = [{
+                    'ext': 'vtt',
+                    'url': 'https://vimeo.com' + tt['url'],
+                }]
 
         return {
             'id': video_id,
@@ -389,18 +402,19 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
             'view_count': view_count,
             'like_count': like_count,
             'comment_count': comment_count,
-            'subtitles': video_subtitles,
+            'subtitles': subtitles,
         }
 
 
 class VimeoChannelIE(InfoExtractor):
     IE_NAME = 'vimeo:channel'
-    _VALID_URL = r'https?://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
+    _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
     _TESTS = [{
-        'url': 'http://vimeo.com/channels/tributes',
+        'url': 'https://vimeo.com/channels/tributes',
         'info_dict': {
+            'id': 'tributes',
             'title': 'Vimeo Tributes',
         },
         'playlist_mincount': 25,
@@ -412,17 +426,52 @@ class VimeoChannelIE(InfoExtractor):
     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)<form[^>]+?id="pw_form"(.*?)</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)<input\s+
+            type="hidden"\s+
+            name="([^"]+)"\s+
+            value="([^"]*)"
+            ''', login_form))
+        token = self._search_regex(r'xsrft[\s=:"\']+([^"\']+)', webpage, 'login token')
+        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 +482,18 @@ 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]+(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
+    _VALID_URL = r'https://vimeo\.com/(?![0-9]+(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
     _TESTS = [{
-        'url': 'http://vimeo.com/nkistudio/videos',
+        'url': 'https://vimeo.com/nkistudio/videos',
         'info_dict': {
             'title': 'Nki',
+            'id': 'nkistudio',
         },
         'playlist_mincount': 66,
     }]
@@ -451,36 +501,48 @@ 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<id>\d+)'
+    _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)'
     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
     _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<name>[^/]+)'
+    _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)'
     _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 +554,25 @@ 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<id>[^/]+)'
+    _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
     _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',
@@ -531,11 +594,11 @@ class VimeoReviewIE(InfoExtractor):
 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'
+    _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',
+        'url': 'https://vimeo.com/home/watchlater',
         'only_matching': True,
     }]
 
@@ -555,13 +618,14 @@ class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE):
 
 
 class VimeoLikesIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
+    _VALID_URL = r'https://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|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 +646,8 @@ 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))