X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/af478477605bdf3f5d57562035885cfee905f379..d2e1a98478a93d5d191b83a9aa545df270ebf323:/youtube_dl/extractor/vk.py

diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py
index a293b88..7dea8c5 100644
--- a/youtube_dl/extractor/vk.py
+++ b/youtube_dl/extractor/vk.py
@@ -5,18 +5,22 @@ import re
 import json
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_str,
+    compat_urllib_parse,
+    compat_urllib_request,
+)
 from ..utils import (
     ExtractorError,
-    compat_urllib_request,
-    compat_urllib_parse,
-    compat_str,
+    orderedSet,
     unescapeHTML,
+    unified_strdate,
 )
 
 
 class VKIE(InfoExtractor):
     IE_NAME = 'vk.com'
-    _VALID_URL = r'https?://vk\.com/(?:videos.*?\?.*?z=)?video(?P<id>.*?)(?:\?|%2F|$)'
+    _VALID_URL = r'https?://(?:m\.)?vk\.com/(?:video_ext\.php\?.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+)|(?:.+?\?.*?z=)?video(?P<videoid>[^s].*?)(?:\?|%2F|$))'
     _NETRC_MACHINE = 'vk'
 
     _TESTS = [
@@ -27,22 +31,39 @@ class VKIE(InfoExtractor):
                 'id': '162222515',
                 'ext': 'flv',
                 'title': 'ProtivoGunz - Хуёвая песня',
-                'uploader': 'Noize MC',
+                'uploader': 're:Noize MC.*',
                 'duration': 195,
+                'upload_date': '20120212',
             },
         },
         {
-            'url': 'http://vk.com/video4643923_163339118',
-            'md5': 'f79bccb5cd182b1f43502ca5685b2b36',
+            'url': 'http://vk.com/video205387401_165548505',
+            'md5': '6c0aeb2e90396ba97035b9cbde548700',
+            'info_dict': {
+                'id': '165548505',
+                'ext': 'mp4',
+                'uploader': 'Tom Cruise',
+                'title': 'No name',
+                'duration': 9,
+                'upload_date': '20130721'
+            }
+        },
+        {
+            'note': 'Embedded video',
+            'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
+            'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
             'info_dict': {
-                'id': '163339118',
+                'id': '162925554',
                 'ext': 'mp4',
-                'uploader': 'Elvira Dzhonik',
-                'title': 'Dream Theater - Hollow Years Live at Budokan 720*',
-                'duration': 558,
+                'uploader': 'Vladimir Gavrin',
+                'title': 'Lin Dan',
+                'duration': 101,
+                'upload_date': '20120730',
             }
         },
         {
+            # VIDEO NOW REMOVED
+            # please update if you find a video whose URL follows the same pattern
             'url': 'http://vk.com/video-8871596_164049491',
             'md5': 'a590bcaf3d543576c9bd162812387666',
             'note': 'Only available for registered users',
@@ -50,11 +71,43 @@ class VKIE(InfoExtractor):
                 'id': '164049491',
                 'ext': 'mp4',
                 'uploader': 'Триллеры',
-                'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]\u00a0',
+                'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
                 'duration': 8352,
+                'upload_date': '20121218'
             },
             'skip': 'Requires vk account credentials',
-        }
+        },
+        {
+            'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
+            'md5': '4d7a5ef8cf114dfa09577e57b2993202',
+            'info_dict': {
+                'id': '168067957',
+                'ext': 'mp4',
+                'uploader': 'Киномания - лучшее из мира кино',
+                'title': ' ',
+                'duration': 7291,
+                'upload_date': '20140328',
+            },
+            'skip': 'Requires vk account credentials',
+        },
+        {
+            'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
+            'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
+            'note': 'ivi.ru embed',
+            'info_dict': {
+                'id': '60690',
+                'ext': 'mp4',
+                'title': 'Книга Илая',
+                'duration': 6771,
+                'upload_date': '20140626',
+            },
+            'skip': 'Only works from Russia',
+        },
+        {
+            # removed video, just testing that we match the pattern
+            'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
+            'only_matching': True,
+        },
     ]
 
     def _login(self):
@@ -71,7 +124,7 @@ class VKIE(InfoExtractor):
         }
 
         request = compat_urllib_request.Request('https://login.vk.com/?act=login',
-            compat_urllib_parse.urlencode(login_form).encode('utf-8'))
+                                                compat_urllib_parse.urlencode(login_form).encode('utf-8'))
         login_page = self._download_webpage(request, None, note='Logging in as %s' % username)
 
         if re.search(r'onLoginFailed', login_page):
@@ -82,22 +135,62 @@ class VKIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = mobj.group('videoid')
+
+        if not video_id:
+            video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
 
         info_url = 'http://vk.com/al_video.php?act=show&al=1&video=%s' % video_id
         info_page = self._download_webpage(info_url, video_id)
 
-        if re.search(r'<!>Please log in or <', info_page):
-            raise ExtractorError('This video is only available for registered users, '
-                'use --username and --password options to provide account credentials.', expected=True)
+        ERRORS = {
+            r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
+            'Video %s has been removed from public access due to rightholder complaint.',
+
+            r'<!>Please log in or <':
+            'Video %s is only available for registered users, '
+            'use --username and --password options to provide account credentials.',
+
+            r'<!>Unknown error':
+            'Video %s does not exist.'
+        }
+
+        for error_re, error_msg in ERRORS.items():
+            if re.search(error_re, info_page):
+                raise ExtractorError(error_msg % video_id, expected=True)
 
         m_yt = re.search(r'src="(http://www.youtube.com/.*?)"', info_page)
         if m_yt is not None:
-            self.to_screen(u'Youtube video detected')
+            self.to_screen('Youtube video detected')
             return self.url_result(m_yt.group(1), 'Youtube')
+
+        m_rutube = re.search(
+            r'\ssrc="((?:https?:)?//rutube\.ru\\?/video\\?/embed(?:.*?))\\?"', info_page)
+        if m_rutube is not None:
+            self.to_screen('rutube video detected')
+            rutube_url = self._proto_relative_url(
+                m_rutube.group(1).replace('\\', ''))
+            return self.url_result(rutube_url)
+
+        m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.*?});', info_page)
+        if m_opts:
+            m_opts_url = re.search(r"url\s*:\s*'([^']+)", m_opts.group(1))
+            if m_opts_url:
+                opts_url = m_opts_url.group(1)
+                if opts_url.startswith('//'):
+                    opts_url = 'http:' + opts_url
+                return self.url_result(opts_url)
+
         data_json = self._search_regex(r'var vars = ({.*?});', info_page, 'vars')
         data = json.loads(data_json)
 
+        # Extract upload date
+        upload_date = None
+        mobj = re.search(r'id="mv_date_wrap".*?Added ([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page)
+        if mobj is not None:
+            mobj.group(1) + ' ' + mobj.group(2)
+            upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2))
+
         formats = [{
             'format_id': k,
             'url': v,
@@ -112,5 +205,31 @@ class VKIE(InfoExtractor):
             'title': unescapeHTML(data['md_title']),
             'thumbnail': data.get('jpg'),
             'uploader': data.get('md_author'),
-            'duration': data.get('duration')
+            'duration': data.get('duration'),
+            'upload_date': upload_date,
         }
+
+
+class VKUserVideosIE(InfoExtractor):
+    IE_NAME = 'vk.com:user-videos'
+    IE_DESC = 'vk.com:All of a user\'s videos'
+    _VALID_URL = r'https?://vk\.com/videos(?P<id>[0-9]+)(?:m\?.*)?'
+    _TEMPLATE_URL = 'https://vk.com/videos'
+    _TEST = {
+        'url': 'http://vk.com/videos205387401',
+        'info_dict': {
+            'id': '205387401',
+        },
+        'playlist_mincount': 4,
+    }
+
+    def _real_extract(self, url):
+        page_id = self._match_id(url)
+        page = self._download_webpage(url, page_id)
+        video_ids = orderedSet(
+            m.group(1) for m in re.finditer(r'href="/video([0-9_]+)"', page))
+        url_entries = [
+            self.url_result(
+                'http://vk.com/video' + video_id, 'VK', video_id=video_id)
+            for video_id in video_ids]
+        return self.playlist_result(url_entries, page_id)