X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fvimeo.py;h=88e462a4d02befd492f324709969a5791b4f31a2;hb=09670d5ba4f98e195837b9665205f0cfe3b0c80b;hp=1959d5e36f7e5cef060fbb47d44f82fae2f1e453;hpb=30965ac66a4e33b8e70d6c0cf5da1a746054d6cd;p=youtube-dl

diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py
index 1959d5e36..88e462a4d 100644
--- a/youtube_dl/extractor/vimeo.py
+++ b/youtube_dl/extractor/vimeo.py
@@ -6,7 +6,6 @@ import re
 import itertools
 
 from .common import InfoExtractor
-from .subtitles import SubtitlesInfoExtractor
 from ..compat import (
     compat_HTTPError,
     compat_urllib_parse,
@@ -20,14 +19,17 @@ from ..utils import (
     RegexNotFoundError,
     smuggle_url,
     std_headers,
+    unified_strdate,
     unsmuggle_url,
     urlencode_postdata,
+    unescapeHTML,
 )
 
 
 class VimeoBaseInfoExtractor(InfoExtractor):
     _NETRC_MACHINE = 'vimeo'
     _LOGIN_REQUIRED = False
+    _LOGIN_URL = 'https://vimeo.com/log_in'
 
     def _login(self):
         (username, password) = self._get_login_info()
@@ -36,23 +38,32 @@ class VimeoBaseInfoExtractor(InfoExtractor):
                 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')
+        webpage = self._download_webpage(self._LOGIN_URL, None, False)
+        token, vuid = self._extract_xsrft_and_vuid(webpage)
         data = urlencode_postdata({
+            'action': 'login',
             'email': username,
             'password': password,
-            'action': 'login',
             'service': 'vimeo',
             'token': token,
         })
-        login_request = compat_urllib_request.Request(login_url, data)
+        login_request = compat_urllib_request.Request(self._LOGIN_URL, data)
         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-        login_request.add_header('Cookie', 'xsrft=%s' % token)
+        login_request.add_header('Cookie', 'vuid=%s' % vuid)
+        login_request.add_header('Referer', self._LOGIN_URL)
         self._download_webpage(login_request, None, False, 'Wrong login info')
 
+    def _extract_xsrft_and_vuid(self, webpage):
+        xsrft = self._search_regex(
+            r'xsrft\s*[=:]\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
+            webpage, 'login token', group='xsrft')
+        vuid = self._search_regex(
+            r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
+            webpage, 'vuid', group='vuid')
+        return xsrft, vuid
+
 
-class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
+class VimeoIE(VimeoBaseInfoExtractor):
     """Information extractor for vimeo.com."""
 
     # _VALID_URL matches Vimeo URLs
@@ -74,12 +85,12 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
             '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,
+                'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
+                'description': 'md5:2d3305bad981a06ff79f027f19865021',
+                'upload_date': '20121220',
+                'uploader_id': 'user7108434',
+                'uploader': 'Filippo Valsorda',
+                'duration': 10,
             },
         },
         {
@@ -92,7 +103,7 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
                 'uploader_id': 'openstreetmapus',
                 'uploader': 'OpenStreetMap US',
                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
-                'description': 'md5:380943ec71b89736ff4bf27183233d09',
+                'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
                 'duration': 1595,
             },
         },
@@ -140,12 +151,12 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
                 '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',
@@ -172,23 +183,37 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
         },
     ]
 
+    @staticmethod
+    def _extract_vimeo_url(url, webpage):
+        # Look for embedded (iframe) Vimeo player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
+        if mobj:
+            player_url = unescapeHTML(mobj.group('url'))
+            surl = smuggle_url(player_url, {'Referer': url})
+            return surl
+        # Look for embedded (swf embed) Vimeo player
+        mobj = re.search(
+            r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
+        if mobj:
+            return mobj.group(1)
+
     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, vuid = self._extract_xsrft_and_vuid(webpage)
+        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)
+        password_request.add_header('Cookie', 'clip_test2=1; vuid=%s' % vuid)
+        password_request.add_header('Referer', url)
         return self._download_webpage(
             password_request, video_id,
             'Verifying the password', 'Wrong password')
@@ -223,7 +248,9 @@ 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)
@@ -245,11 +272,28 @@ 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:
                 config_url = self._html_search_regex(
-                    r' data-config-url="(.+?)"', webpage, 'config URL')
+                    r' data-config-url="(.+?)"', webpage,
+                    'config URL', default=None)
+                if not config_url:
+                    # New react-based page
+                    vimeo_clip_page_config = self._search_regex(
+                        r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
+                        'vimeo clip page config')
+                    config_url = self._parse_json(vimeo_clip_page_config, video_id)['player']['config_url']
                 config_json = self._download_webpage(config_url, video_id)
                 config = json.loads(config_json)
             except RegexNotFoundError:
@@ -318,9 +362,9 @@ 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'))
@@ -359,25 +403,29 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
                     'ext': codec_extension,
                     'url': video_url,
                     'format_id': format_id,
-                    'width': file_info.get('width'),
-                    'height': file_info.get('height'),
+                    'width': int_or_none(file_info.get('width')),
+                    'height': int_or_none(file_info.get('height')),
+                    'tbr': int_or_none(file_info.get('bitrate')),
                 })
         formats = []
+        m3u8_url = config_files.get('hls', {}).get('all')
+        if m3u8_url:
+            m3u8_formats = self._extract_m3u8_formats(
+                m3u8_url, video_id, 'mp4', 'm3u8_native', 0, 'hls', fatal=False)
+            if m3u8_formats:
+                formats.extend(m3u8_formats)
         for key in ('other', 'sd', 'hd'):
             formats += files[key]
-        if len(formats) == 0:
-            raise ExtractorError('No known codec found')
+        self._sort_formats(formats)
 
         subtitles = {}
         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,
@@ -393,18 +441,20 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
             'view_count': view_count,
             'like_count': like_count,
             'comment_count': comment_count,
-            'subtitles': video_subtitles,
+            'subtitles': subtitles,
         }
 
 
-class VimeoChannelIE(InfoExtractor):
+class VimeoChannelIE(VimeoBaseInfoExtractor):
     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 = None
     _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,
@@ -414,7 +464,7 @@ class VimeoChannelIE(InfoExtractor):
         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')
+        return self._TITLE or 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(
@@ -426,20 +476,17 @@ class VimeoChannelIE(InfoExtractor):
         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 = self._hidden_inputs(login_form)
+        token, vuid = self._extract_xsrft_and_vuid(webpage)
         fields['token'] = token
         fields['password'] = password
-        post = compat_urllib_parse.urlencode(fields)
+        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')
+        password_request.add_header('Cookie', 'vuid=%s' % vuid)
         self._set_cookie('vimeo.com', 'xsrft', token)
 
         return self._download_webpage(
@@ -461,7 +508,7 @@ class VimeoChannelIE(InfoExtractor):
             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,
@@ -472,17 +519,18 @@ 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]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
     _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,
     }]
@@ -490,16 +538,17 @@ 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,
@@ -521,15 +570,16 @@ class VimeoAlbumIE(VimeoChannelIE):
 
     def _real_extract(self, url):
         album_id = self._match_id(url)
-        return self._extract_videos(album_id, 'http://vimeo.com/album/%s' % album_id)
+        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,
@@ -541,13 +591,13 @@ 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',
         'md5': 'c507a72f780cacc12b2248bb4006d253',
@@ -559,7 +609,7 @@ class VimeoReviewIE(InfoExtractor):
         }
     }, {
         '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',
@@ -578,14 +628,14 @@ class VimeoReviewIE(InfoExtractor):
         return self.url_result(player_url, 'Vimeo', video_id)
 
 
-class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE):
+class VimeoWatchLaterIE(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'
+    _TITLE = 'Watch Later'
     _LOGIN_REQUIRED = True
-    _TITLE_RE = r'href="/home/watchlater".*?>(.*?)<'
     _TESTS = [{
-        'url': 'http://vimeo.com/home/watchlater',
+        'url': 'https://vimeo.com/watchlater',
         'only_matching': True,
     }]
 
@@ -601,17 +651,18 @@ class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE):
         return request
 
     def _real_extract(self, url):
-        return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater')
+        return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
 
 
 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',
         },
@@ -632,8 +683,8 @@ 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))