]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/vimeo.py
Prepare for release.
[youtubedl] / youtube_dl / extractor / vimeo.py
index 11c7d7e817f1f0839604311534dd918b5b5e4fee..8f540f5780570d06fa10e695555026c537b7c0f0 100644 (file)
@@ -4,21 +4,24 @@ from __future__ import unicode_literals
 import json
 import re
 import itertools
+import hashlib
 
 from .common import InfoExtractor
-from .subtitles import SubtitlesInfoExtractor
-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,
     unsmuggle_url,
     urlencode_postdata,
-    int_or_none,
 )
 
 
@@ -49,14 +52,15 @@ 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
     _VALID_URL = r'''(?x)
-        (?P<proto>(?:https?:)?//)?
+        https?://
         (?:(?:www|(?P<player>player))\.)?
         vimeo(?P<pro>pro)?\.com/
+        (?!channels/[^/?#]+/?(?:$|[?#])|album/)
         (?:.*?/)?
         (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)?
         (?:videos?/)?
@@ -88,6 +92,7 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
                 'uploader_id': 'openstreetmapus',
                 'uploader': 'OpenStreetMap US',
                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
+                'description': 'md5:380943ec71b89736ff4bf27183233d09',
                 'duration': 1595,
             },
         },
@@ -102,6 +107,7 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
                 'uploader': 'The BLN & Business of Software',
                 'uploader_id': 'theblnbusinessofsoftware',
                 'duration': 3610,
+                'description': None,
             },
         },
         {
@@ -116,6 +122,7 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
                 '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',
@@ -151,21 +158,24 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
                 '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',
+            },
+        },
     ]
 
-    @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')
+            raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
         data = compat_urllib_parse.urlencode({
             'password': password,
@@ -179,9 +189,9 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
         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,
-                               '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)
@@ -205,13 +215,21 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
         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)
         video_id = mobj.group('id')
+        orig_url = url
         if mobj.group('pro') or mobj.group('player'):
             url = 'http://player.vimeo.com/video/' + video_id
 
+        password = self._downloader.params.get('videopassword', None)
+        if password:
+            headers['Cookie'] = '%s_password=%s' % (
+                video_id, hashlib.md5(password.encode('utf-8')).hexdigest())
+
         # Retrieve video webpage to extract further information
         request = compat_urllib_request.Request(url, None, headers)
         try:
@@ -244,19 +262,22 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
                 # 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))
+                    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)
+                                            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('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)
@@ -279,18 +300,23 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
                 _, 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("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
-            if err.args[0] == 'we should not get here!':
-                pass
-            else:
-                raise
+
+        video_description = self._html_search_regex(
+            r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
+            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"))
@@ -351,12 +377,10 @@ 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': 'http://vimeo.com' + tt['url'],
+                }]
 
         return {
             'id': video_id,
@@ -372,15 +396,23 @@ 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',
+        'info_dict': {
+            'id': 'tributes',
+            'title': 'Vimeo Tributes',
+        },
+        'playlist_mincount': 25,
+    }]
 
     def _page_url(self, base_url, pagenum):
         return '%s/videos/page:%d/' % (base_url, pagenum)
@@ -388,12 +420,47 @@ 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: \'(.*?)\'', webpage, 'login token')
+        fields['token'] = token
+        fields['password'] = password
+        post = compat_urllib_parse.urlencode(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
@@ -414,14 +481,16 @@ class VimeoChannelIE(InfoExtractor):
 
 class VimeoUserIE(VimeoChannelIE):
     IE_NAME = 'vimeo:user'
-    _VALID_URL = r'(?:https?://)?vimeo\.com/(?P<name>[^/]+)(?:/videos|[#?]|$)'
+    _VALID_URL = r'https?://vimeo\.com/(?![0-9]+(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
-
-    @classmethod
-    def suitable(cls, url):
-        if VimeoChannelIE.suitable(url) or VimeoIE.suitable(url) or VimeoAlbumIE.suitable(url) or VimeoGroupsIE.suitable(url):
-            return False
-        return super(VimeoUserIE, cls).suitable(url)
+    _TESTS = [{
+        'url': 'http://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)
@@ -431,21 +500,47 @@ class VimeoUserIE(VimeoChannelIE):
 
 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',
+        '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')
+        album_id = self._match_id(url)
         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<name>[^/]+)'
+    _TESTS = [{
+        'url': 'http://vimeo.com/groups/rolexawards',
+        'info_dict': {
+            'id': 'rolexawards',
+            'title': 'Rolex Awards for Enterprise',
+        },
+        'playlist_mincount': 73,
+    }]
 
     def _extract_list_title(self, webpage):
         return self._og_search_title(webpage)
@@ -459,16 +554,29 @@ class VimeoGroupsIE(VimeoAlbumIE):
 class VimeoReviewIE(InfoExtractor):
     IE_NAME = 'vimeo:review'
     IE_DESC = 'Review pages on vimeo'
-    _VALID_URL = r'(?:https?://)?vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
-    _TEST = {
+    _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',
+        'md5': '6295fdab8f4bf6a002d058b2c6dce276',
+        'info_dict': {
+            'id': '91613211',
+            'ext': 'mp4',
+            'title': 're:(?i)^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)
@@ -483,6 +591,10 @@ class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE):
     _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()
@@ -497,3 +609,59 @@ class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE):
 
     def _real_extract(self, url):
         return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater')
+
+
+class VimeoLikesIE(InfoExtractor):
+    _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',
+        },
+    }
+
+    def _real_extract(self, url):
+        user_id = self._match_id(url)
+        webpage = self._download_webpage(url, user_id)
+        page_count = self._int(
+            self._search_regex(
+                r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
+                    .*?</a></li>\s*<li\s+class="pagination_next">
+                ''', webpage, 'page count'),
+            'page count', fatal=True)
+        PAGE_SIZE = 12
+        title = self._html_search_regex(
+            r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
+        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)
+            webpage = self._download_webpage(
+                page_url, user_id,
+                note='Downloading page %d/%d' % (idx + 1, page_count))
+            video_list = self._search_regex(
+                r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
+                webpage, 'video content')
+            paths = re.findall(
+                r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
+            for path in paths:
+                yield {
+                    '_type': 'url',
+                    'url': compat_urlparse.urljoin(page_url, path),
+                }
+
+        pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
+
+        return {
+            '_type': 'playlist',
+            'id': 'user%s_likes' % user_id,
+            'title': title,
+            'description': description,
+            'entries': pl,
+        }