Rebuild with only Python 3 support.

[youtubedl] / youtube_dl / extractor / vk.py

diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py
index 7dea8c59d2a30673b93ae181bab128b8ef0a8b58..3ee66e23e22b1d5350148d5cb21c2f55e4885f75 100644 (file)
--- a/youtube_dl/extractor/vk.py
+++ b/youtube_dl/extractor/vk.py
@@ -3,26 +3,96 @@ from __future__ import unicode_literals
 
 import re
 import json
 
 import re
 import json

+import sys

 
 from .common import InfoExtractor
 from ..compat import (
     compat_str,
 
 from .common import InfoExtractor
 from ..compat import (
     compat_str,

-    compat_urllib_parse,
-    compat_urllib_request,
+    compat_urlparse,

 )
 from ..utils import (
 )
 from ..utils import (

+    clean_html,

     ExtractorError,
     ExtractorError,

+    get_element_by_class,
+    int_or_none,

     orderedSet,
     orderedSet,

+    parse_duration,
+    remove_start,
+    str_to_int,

     unescapeHTML,
     unified_strdate,
     unescapeHTML,
     unified_strdate,

+    urlencode_postdata,

 )
 )

+from .vimeo import VimeoIE
+from .pladform import PladformIE

 
 
 
 

-class VKIE(InfoExtractor):
-    IE_NAME = 'vk.com'
-    _VALID_URL = r'https?://(?:m\.)?vk\.com/(?:video_ext\.php\?.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+)|(?:.+?\?.*?z=)?video(?P<videoid>[^s].*?)(?:\?|%2F|$))'
+class VKBaseIE(InfoExtractor):

     _NETRC_MACHINE = 'vk'
 
     _NETRC_MACHINE = 'vk'
 

+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None:
+            return
+
+        login_page, url_handle = self._download_webpage_handle(
+            'https://vk.com', None, 'Downloading login page')
+
+        login_form = self._hidden_inputs(login_page)
+
+        login_form.update({
+            'email': username.encode('cp1251'),
+            'pass': password.encode('cp1251'),
+        })
+
+        # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
+        # and expects the first one to be set rather than second (see
+        # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201).
+        # As of RFC6265 the newer one cookie should be set into cookie store
+        # what actually happens.
+        # We will workaround this VK issue by resetting the remixlhk cookie to
+        # the first one manually.
+        cookies = url_handle.headers.get('Set-Cookie')
+        if cookies:
+            if sys.version_info[0] >= 3:
+                cookies = cookies.encode('iso-8859-1')
+            cookies = cookies.decode('utf-8')
+            remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies)
+            if remixlhk:
+                value, domain = remixlhk.groups()
+                self._set_cookie(domain, 'remixlhk', value)
+
+        login_page = self._download_webpage(
+            'https://login.vk.com/?act=login', None,
+            note='Logging in as %s' % username,
+            data=urlencode_postdata(login_form))
+
+        if re.search(r'onLoginFailed', login_page):
+            raise ExtractorError(
+                'Unable to login, incorrect username and/or password', expected=True)
+
+    def _real_initialize(self):
+        self._login()
+
+
+class VKIE(VKBaseIE):
+    IE_NAME = 'vk'
+    IE_DESC = 'VK'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:
+                            (?:
+                                (?:(?:m|new)\.)?vk\.com/video_|
+                                (?:www\.)?daxab.com/
+                            )
+                            ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)|
+                            (?:
+                                (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video|
+                                (?:www\.)?daxab.com/embed/
+                            )
+                            (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))?
+                        )
+                    '''

     _TESTS = [
         {
             'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
     _TESTS = [
         {
             'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',


@@ -31,9 +101,10 @@ class VKIE(InfoExtractor):
                 'id': '162222515',
                 'ext': 'flv',
                 'title': 'ProtivoGunz - Хуёвая песня',
                 'id': '162222515',
                 'ext': 'flv',
                 'title': 'ProtivoGunz - Хуёвая песня',

-                'uploader': 're:Noize MC.*',
+                'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',

                 'duration': 195,
                 'upload_date': '20120212',
                 'duration': 195,
                 'upload_date': '20120212',

+                'view_count': int,

             },
         },
         {
             },
         },
         {


@@ -45,7 +116,8 @@ class VKIE(InfoExtractor):
                 'uploader': 'Tom Cruise',
                 'title': 'No name',
                 'duration': 9,
                 'uploader': 'Tom Cruise',
                 'title': 'No name',
                 'duration': 9,

-                'upload_date': '20130721'
+                'upload_date': '20130721',
+                'view_count': int,

             }
         },
         {
             }
         },
         {


@@ -59,7 +131,9 @@ class VKIE(InfoExtractor):
                 'title': 'Lin Dan',
                 'duration': 101,
                 'upload_date': '20120730',
                 'title': 'Lin Dan',
                 'duration': 101,
                 'upload_date': '20120730',

-            }
+                'view_count': int,
+            },
+            'skip': 'This video has been removed from public access.',

         },
         {
             # VIDEO NOW REMOVED
         },
         {
             # VIDEO NOW REMOVED


@@ -73,7 +147,8 @@ class VKIE(InfoExtractor):
                 'uploader': 'Триллеры',
                 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
                 'duration': 8352,
                 'uploader': 'Триллеры',
                 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
                 'duration': 8352,

-                'upload_date': '20121218'
+                'upload_date': '20121218',
+                'view_count': int,

             },
             'skip': 'Requires vk account credentials',
         },
             },
             'skip': 'Requires vk account credentials',
         },


@@ -100,49 +175,101 @@ class VKIE(InfoExtractor):
                 'title': 'Книга Илая',
                 'duration': 6771,
                 'upload_date': '20140626',
                 'title': 'Книга Илая',
                 'duration': 6771,
                 'upload_date': '20140626',

+                'view_count': int,

             },
             'skip': 'Only works from Russia',
         },
             },
             'skip': 'Only works from Russia',
         },

+        {
+            # video (removed?) only available with list id
+            'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
+            'md5': '091287af5402239a1051c37ec7b92913',
+            'info_dict': {
+                'id': '171201961',
+                'ext': 'mp4',
+                'title': 'ТюменцевВВ_09.07.2015',
+                'uploader': 'Anton Ivanov',
+                'duration': 109,
+                'upload_date': '20150709',
+                'view_count': int,
+            },
+        },
+        {
+            # youtube embed
+            'url': 'https://vk.com/video276849682_170681728',
+            'info_dict': {
+                'id': 'V3K4mi0SYkc',
+                'ext': 'webm',
+                'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
+                'description': 'md5:d9903938abdc74c738af77f527ca0596',
+                'duration': 178,
+                'upload_date': '20130116',
+                'uploader': "Children's Joy Foundation",
+                'uploader_id': 'thecjf',
+                'view_count': int,
+            },
+        },
+        {
+            # video key is extra_data not url\d+
+            'url': 'http://vk.com/video-110305615_171782105',
+            'md5': 'e13fcda136f99764872e739d13fac1d1',
+            'info_dict': {
+                'id': '171782105',
+                'ext': 'mp4',
+                'title': 'S-Dance, репетиции к The way show',
+                'uploader': 'THE WAY SHOW | 17 апреля',
+                'upload_date': '20160207',
+                'view_count': int,
+            },
+        },

         {
             # removed video, just testing that we match the pattern
             'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
             'only_matching': True,
         },
         {
             # 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):
-        (username, password) = self._get_login_info()
-        if username is None:
-            return
-
-        login_form = {
-            'act': 'login',
-            'role': 'al_frame',
-            'expire': '1',
-            'email': username,
-            'pass': password,
+        {
+            # age restricted video, requires vk account credentials
+            'url': 'https://vk.com/video205387401_164765225',
+            'only_matching': True,
+        },
+        {
+            # pladform embed
+            'url': 'https://vk.com/video-76116461_171554880',
+            'only_matching': True,
+        },
+        {
+            'url': 'http://new.vk.com/video205387401_165548505',
+            'only_matching': True,

         }
         }

-
-        request = compat_urllib_request.Request('https://login.vk.com/?act=login',
-                                                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):
-            raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
-
-    def _real_initialize(self):
-        self._login()
+    ]

 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('videoid')
 
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('videoid')
 

-        if not video_id:
+        if video_id:
+            info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
+            # Some videos (removed?) can only be downloaded with list id specified
+            list_id = mobj.group('list_id')
+            if list_id:
+                info_url += '&list=%s' % list_id
+        else:
+            info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query')

             video_id = '%s_%s' % (mobj.group('oid'), mobj.group('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)
 
         info_page = self._download_webpage(info_url, video_id)
 

+        error_message = self._html_search_regex(
+            [r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
+                r'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'],
+            info_page, 'error message', default=None)
+        if error_message:
+            raise ExtractorError(error_message, expected=True)
+
+        if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
+            raise ExtractorError(
+                'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
+                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.',
         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.',


@@ -152,51 +279,78 @@ class VKIE(InfoExtractor):
             'use --username and --password options to provide account credentials.',
 
             r'<!>Unknown error':
             'use --username and --password options to provide account credentials.',
 
             r'<!>Unknown error':

-            'Video %s does not exist.'
+            'Video %s does not exist.',
+
+            r'<!>Видео временно недоступно':
+            'Video %s is temporarily unavailable.',
+
+            r'<!>Access denied':
+            'Access denied to video %s.',

         }
 
         for error_re, error_msg in ERRORS.items():
             if re.search(error_re, info_page):
                 raise ExtractorError(error_msg % video_id, expected=True)
 
         }
 
         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('Youtube video detected')
-            return self.url_result(m_yt.group(1), 'Youtube')
+        youtube_url = self._search_regex(
+            r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"',
+            info_page, 'youtube iframe', default=None)
+        if youtube_url:
+            return self.url_result(youtube_url, 'Youtube')
+
+        vimeo_url = VimeoIE._extract_vimeo_url(url, info_page)
+        if vimeo_url is not None:
+            return self.url_result(vimeo_url)
+
+        pladform_url = PladformIE._extract_url(info_page)
+        if pladform_url:
+            return self.url_result(pladform_url)

 
         m_rutube = re.search(
 
         m_rutube = re.search(

-            r'\ssrc="((?:https?:)?//rutube\.ru\\?/video\\?/embed(?:.*?))\\?"', info_page)
+            r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page)

         if m_rutube is not None:
         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)
 
             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)
+        m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)

         if m_opts:
         if m_opts:

-            m_opts_url = re.search(r"url\s*:\s*'([^']+)", m_opts.group(1))
+            m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", 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)
 
             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 = self._search_regex(r'var\s+vars\s*=\s*({.+?});', info_page, 'vars')

         data = json.loads(data_json)
 
         # Extract upload date
         upload_date = None
         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)
+        mobj = re.search(r'id="mv_date(?:_views)?_wrap"[^>]*>([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))
 
         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,
-            'width': int(k[len('url'):]),
-        } for k, v in data.items()
-            if k.startswith('url')]
+        view_count = None
+        views = self._html_search_regex(
+            r'"mv_views_count_number"[^>]*>(.+?\bviews?)<',
+            info_page, 'view count', default=None)
+        if views:
+            view_count = str_to_int(self._search_regex(
+                r'([\d,.]+)', views, 'view count', fatal=False))
+
+        formats = []
+        for k, v in data.items():
+            if not k.startswith('url') and not k.startswith('cache') and k != 'extra_data' or not v:
+                continue
+            height = int_or_none(self._search_regex(
+                r'^(?:url|cache)(\d+)', k, 'height', default=None))
+            formats.append({
+                'format_id': k,
+                'url': v,
+                'height': height,
+            })

         self._sort_formats(formats)
 
         return {
         self._sort_formats(formats)
 
         return {


@@ -207,29 +361,166 @@ class VKIE(InfoExtractor):
             'uploader': data.get('md_author'),
             'duration': data.get('duration'),
             'upload_date': upload_date,
             'uploader': data.get('md_author'),
             'duration': data.get('duration'),
             'upload_date': upload_date,

+            'view_count': view_count,

         }
 
 
         }
 
 

-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\?.*)?'
+class VKUserVideosIE(VKBaseIE):
+    IE_NAME = 'vk:uservideos'
+    IE_DESC = "VK - User's Videos"
+    _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'

     _TEMPLATE_URL = 'https://vk.com/videos'
     _TEMPLATE_URL = 'https://vk.com/videos'

-    _TEST = {
+    _TESTS = [{

         'url': 'http://vk.com/videos205387401',
         'info_dict': {
             'id': '205387401',
         'url': 'http://vk.com/videos205387401',
         'info_dict': {
             'id': '205387401',

+            'title': "Tom Cruise's Videos",

         },
         'playlist_mincount': 4,
         },
         'playlist_mincount': 4,

-    }
+    }, {
+        'url': 'http://vk.com/videos-77521',
+        'only_matching': True,
+    }, {
+        'url': 'http://vk.com/videos-97664626?section=all',
+        'only_matching': True,
+    }, {
+        'url': 'http://m.vk.com/videos205387401',
+        'only_matching': True,
+    }, {
+        'url': 'http://new.vk.com/videos205387401',
+        'only_matching': True,
+    }]

 
     def _real_extract(self, url):
         page_id = self._match_id(url)
 
     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 = [
+
+        webpage = self._download_webpage(url, page_id)
+
+        entries = [

             self.url_result(
                 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
             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)
+            for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
+
+        title = unescapeHTML(self._search_regex(
+            r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
+            webpage, 'title', default=page_id))
+
+        return self.playlist_result(entries, page_id, title)
+
+
+class VKWallPostIE(VKBaseIE):
+    IE_NAME = 'vk:wallpost'
+    _VALID_URL = r'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P<id>-?\d+_\d+)))'
+    _TESTS = [{
+        # public page URL, audio playlist
+        'url': 'https://vk.com/bs.official?w=wall-23538238_35',
+        'info_dict': {
+            'id': '23538238_35',
+            'title': 'Black Shadow - Wall post 23538238_35',
+            'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c',
+        },
+        'playlist': [{
+            'md5': '5ba93864ec5b85f7ce19a9af4af080f6',
+            'info_dict': {
+                'id': '135220665_111806521',
+                'ext': 'mp3',
+                'title': 'Black Shadow - Слепое Верование',
+                'duration': 370,
+                'uploader': 'Black Shadow',
+                'artist': 'Black Shadow',
+                'track': 'Слепое Верование',
+            },
+        }, {
+            'md5': '4cc7e804579122b17ea95af7834c9233',
+            'info_dict': {
+                'id': '135220665_111802303',
+                'ext': 'mp3',
+                'title': 'Black Shadow - Война - Негасимое Бездны Пламя!',
+                'duration': 423,
+                'uploader': 'Black Shadow',
+                'artist': 'Black Shadow',
+                'track': 'Война - Негасимое Бездны Пламя!',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        }],
+        'skip': 'Requires vk account credentials',
+    }, {
+        # single YouTube embed, no leading -
+        'url': 'https://vk.com/wall85155021_6319',
+        'info_dict': {
+            'id': '85155021_6319',
+            'title': 'Sergey Gorbunov - Wall post 85155021_6319',
+        },
+        'playlist_count': 1,
+        'skip': 'Requires vk account credentials',
+    }, {
+        # wall page URL
+        'url': 'https://vk.com/wall-23538238_35',
+        'only_matching': True,
+    }, {
+        # mobile wall page URL
+        'url': 'https://m.vk.com/wall-23538238_35',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        post_id = self._match_id(url)
+
+        wall_url = 'https://vk.com/wall%s' % post_id
+
+        post_id = remove_start(post_id, '-')
+
+        webpage = self._download_webpage(wall_url, post_id)
+
+        error = self._html_search_regex(
+            r'>Error</div>\s*<div[^>]+class=["\']body["\'][^>]*>([^<]+)',
+            webpage, 'error', default=None)
+        if error:
+            raise ExtractorError('VK said: %s' % error, expected=True)
+
+        description = clean_html(get_element_by_class('wall_post_text', webpage))
+        uploader = clean_html(get_element_by_class(
+            'fw_post_author', webpage)) or self._og_search_description(webpage)
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        entries = []
+
+        for audio in re.finditer(r'''(?sx)
+                            <input[^>]+
+                                id=(?P<q1>["\'])audio_info(?P<id>\d+_\d+).*?(?P=q1)[^>]+
+                                value=(?P<q2>["\'])(?P<url>http.+?)(?P=q2)
+                                .+?
+                            </table>''', webpage):
+            audio_html = audio.group(0)
+            audio_id = audio.group('id')
+            duration = parse_duration(get_element_by_class('duration', audio_html))
+            track = self._html_search_regex(
+                r'<span[^>]+id=["\']title%s[^>]*>([^<]+)' % audio_id,
+                audio_html, 'title', default=None)
+            artist = self._html_search_regex(
+                r'>([^<]+)</a></b>\s*&ndash', audio_html,
+                'artist', default=None)
+            entries.append({
+                'id': audio_id,
+                'url': audio.group('url'),
+                'title': '%s - %s' % (artist, track) if artist and track else audio_id,
+                'thumbnail': thumbnail,
+                'duration': duration,
+                'uploader': uploader,
+                'artist': artist,
+                'track': track,
+            })
+
+        for video in re.finditer(
+                r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):
+            entries.append(self.url_result(
+                compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key()))
+
+        title = 'Wall post %s' % post_id
+
+        return self.playlist_result(
+            orderedSet(entries), post_id,
+            '%s - %s' % (uploader, title) if uploader else title,
+            description)