]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/smotri.py
Prepare for upload.
[youtubedl] / youtube_dl / extractor / smotri.py
index 99f5b19d2dd68e78aaf2f45f79fad60b8bb459dc..9bd5defa7ac5e171904eb681015e9bcf1661acb9 100644 (file)
@@ -1,4 +1,5 @@
 # encoding: utf-8
+from __future__ import unicode_literals
 
 import os.path
 import re
@@ -12,85 +13,121 @@ from ..utils import (
     compat_urllib_request,
     ExtractorError,
     url_basename,
+    int_or_none,
 )
 
 
 class SmotriIE(InfoExtractor):
-    IE_DESC = u'Smotri.com'
-    IE_NAME = u'smotri'
-    _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/video/view/\?id=(?P<videoid>v(?P<realvideoid>[0-9]+)[a-z0-9]{4}))'
+    IE_DESC = 'Smotri.com'
+    IE_NAME = 'smotri'
+    _VALID_URL = r'^https?://(?:www\.)?(?:smotri\.com/video/view/\?id=|pics\.smotri\.com/(?:player|scrubber_custom8)\.swf\?file=)(?P<videoid>v(?P<realvideoid>[0-9]+)[a-z0-9]{4})'
+    _NETRC_MACHINE = 'smotri'
 
     _TESTS = [
         # real video id 2610366
         {
-            u'url': u'http://smotri.com/video/view/?id=v261036632ab',
-            u'file': u'v261036632ab.mp4',
-            u'md5': u'2a7b08249e6f5636557579c368040eb9',
-            u'info_dict': {
-                u'title': u'катастрофа с камер видеонаблюдения',
-                u'uploader': u'rbc2008',
-                u'uploader_id': u'rbc08',
-                u'upload_date': u'20131118',
-                u'description': u'катастрофа с камер видеонаблюдения, видео катастрофа с камер видеонаблюдения',
-                u'thumbnail': u'http://frame6.loadup.ru/8b/a9/2610366.3.3.jpg',
+            'url': 'http://smotri.com/video/view/?id=v261036632ab',
+            'md5': '2a7b08249e6f5636557579c368040eb9',
+            'info_dict': {
+                'id': 'v261036632ab',
+                'ext': 'mp4',
+                'title': 'катастрофа с камер видеонаблюдения',
+                'uploader': 'rbc2008',
+                'uploader_id': 'rbc08',
+                'upload_date': '20131118',
+                'description': 'катастрофа с камер видеонаблюдения, видео катастрофа с камер видеонаблюдения',
+                'thumbnail': 'http://frame6.loadup.ru/8b/a9/2610366.3.3.jpg',
             },
         },
         # real video id 57591
         {
-            u'url': u'http://smotri.com/video/view/?id=v57591cb20',
-            u'file': u'v57591cb20.flv',
-            u'md5': u'830266dfc21f077eac5afd1883091bcd',
-            u'info_dict': {
-                u'title': u'test',
-                u'uploader': u'Support Photofile@photofile',
-                u'uploader_id': u'support-photofile',
-                u'upload_date': u'20070704',
-                u'description': u'test, видео test',
-                u'thumbnail': u'http://frame4.loadup.ru/03/ed/57591.2.3.jpg',
+            'url': 'http://smotri.com/video/view/?id=v57591cb20',
+            'md5': '830266dfc21f077eac5afd1883091bcd',
+            'info_dict': {
+                'id': 'v57591cb20',
+                'ext': 'flv',
+                'title': 'test',
+                'uploader': 'Support Photofile@photofile',
+                'uploader_id': 'support-photofile',
+                'upload_date': '20070704',
+                'description': 'test, видео test',
+                'thumbnail': 'http://frame4.loadup.ru/03/ed/57591.2.3.jpg',
             },
         },
         # video-password
         {
-            u'url': u'http://smotri.com/video/view/?id=v1390466a13c',
-            u'file': u'v1390466a13c.mp4',
-            u'md5': u'f6331cef33cad65a0815ee482a54440b',
-            u'info_dict': {
-                u'title': u'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
-                u'uploader': u'timoxa40',
-                u'uploader_id': u'timoxa40',
-                u'upload_date': u'20100404',
-                u'thumbnail': u'http://frame7.loadup.ru/af/3f/1390466.3.3.jpg',
-                u'description': u'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1, видео TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
+            'url': 'http://smotri.com/video/view/?id=v1390466a13c',
+            'md5': 'f6331cef33cad65a0815ee482a54440b',
+            'info_dict': {
+                'id': 'v1390466a13c',
+                'ext': 'mp4',
+                'title': 'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
+                'uploader': 'timoxa40',
+                'uploader_id': 'timoxa40',
+                'upload_date': '20100404',
+                'thumbnail': 'http://frame7.loadup.ru/af/3f/1390466.3.3.jpg',
+                'description': 'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1, видео TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
             },
-            u'params': {
-                u'videopassword': u'qwerty',
+            'params': {
+                'videopassword': 'qwerty',
             },
         },
         # age limit + video-password
         {
-            u'url': u'http://smotri.com/video/view/?id=v15408898bcf',
-            u'file': u'v15408898bcf.flv',
-            u'md5': u'91e909c9f0521adf5ee86fbe073aad70',
-            u'info_dict': {
-                u'title': u'этот ролик не покажут по ТВ',
-                u'uploader': u'zzxxx',
-                u'uploader_id': u'ueggb',
-                u'upload_date': u'20101001',
-                u'thumbnail': u'http://frame3.loadup.ru/75/75/1540889.1.3.jpg',
-                u'age_limit': 18,
-                u'description': u'этот ролик не покажут по ТВ, видео этот ролик не покажут по ТВ',
+            'url': 'http://smotri.com/video/view/?id=v15408898bcf',
+            'md5': '91e909c9f0521adf5ee86fbe073aad70',
+            'info_dict': {
+                'id': 'v15408898bcf',
+                'ext': 'flv',
+                'title': 'этот ролик не покажут по ТВ',
+                'uploader': 'zzxxx',
+                'uploader_id': 'ueggb',
+                'upload_date': '20101001',
+                'thumbnail': 'http://frame3.loadup.ru/75/75/1540889.1.3.jpg',
+                'age_limit': 18,
+                'description': 'этот ролик не покажут по ТВ, видео этот ролик не покажут по ТВ',
             },
-            u'params': {
-                u'videopassword': u'333'
+            'params': {
+                'videopassword': '333'
             }
-        }
+        },
+        # swf player
+        {
+            'url': 'http://pics.smotri.com/scrubber_custom8.swf?file=v9188090500',
+            'md5': '4d47034979d9390d14acdf59c4935bc2',
+            'info_dict': {
+                'id': 'v9188090500',
+                'ext': 'mp4',
+                'title': 'Shakira - Don\'t Bother',
+                'uploader': 'HannahL',
+                'uploader_id': 'lisaha95',
+                'upload_date': '20090331',
+                'description': 'Shakira - Don\'t Bother, видео Shakira - Don\'t Bother',
+                'thumbnail': 'http://frame8.loadup.ru/44/0b/918809.7.3.jpg',
+            },
+        },
     ]
-    
+
     _SUCCESS = 0
     _PASSWORD_NOT_VERIFIED = 1
     _PASSWORD_DETECTED = 2
     _VIDEO_NOT_FOUND = 3
 
+    @classmethod
+    def _extract_url(cls, webpage):
+        mobj = re.search(
+            r'<embed[^>]src=(["\'])(?P<url>http://pics\.smotri\.com/(?:player|scrubber_custom8)\.swf\?file=v.+?\1)',
+            webpage)
+        if mobj is not None:
+            return mobj.group('url')
+
+        mobj = re.search(
+            r'''(?x)<div\s+class="video_file">http://smotri\.com/video/download/file/[^<]+</div>\s*
+                    <div\s+class="video_image">[^<]+</div>\s*
+                    <div\s+class="video_id">(?P<id>[^<]+)</div>''', webpage)
+        if mobj is not None:
+            return 'http://smotri.com/video/view/?id=%s' % mobj.group('id')
+
     def _search_meta(self, name, html, display_name=None):
         if display_name is None:
             display_name = name
@@ -106,71 +143,71 @@ class SmotriIE(InfoExtractor):
 
         # Download video JSON data
         video_json_url = 'http://smotri.com/vt.php?id=%s' % real_video_id
-        video_json_page = self._download_webpage(video_json_url, video_id, u'Downloading video JSON')
+        video_json_page = self._download_webpage(video_json_url, video_id, 'Downloading video JSON')
         video_json = json.loads(video_json_page)
-        
+
         status = video_json['status']
         if status == self._VIDEO_NOT_FOUND:
-            raise ExtractorError(u'Video %s does not exist' % video_id, expected=True)
-        elif status == self._PASSWORD_DETECTED:  # The video is protected by a password, retry with
+            raise ExtractorError('Video %s does not exist' % video_id, expected=True)
+        elif status == self._PASSWORD_DETECTED: # The video is protected by a password, retry with
                                                 # video-password set
             video_password = self._downloader.params.get('videopassword', None)
             if not video_password:
-                raise ExtractorError(u'This video is protected by a password, use the --video-password option', expected=True)
+                raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
             video_json_url += '&md5pass=%s' % hashlib.md5(video_password.encode('utf-8')).hexdigest()
-            video_json_page = self._download_webpage(video_json_url, video_id, u'Downloading video JSON (video-password set)')
+            video_json_page = self._download_webpage(video_json_url, video_id, 'Downloading video JSON (video-password set)')
             video_json = json.loads(video_json_page)
             status = video_json['status']
             if status == self._PASSWORD_NOT_VERIFIED:
-                raise ExtractorError(u'Video password is invalid', expected=True)
-        
+                raise ExtractorError('Video password is invalid', expected=True)
+
         if status != self._SUCCESS:
-            raise ExtractorError(u'Unexpected status value %s' % status)
-        
+            raise ExtractorError('Unexpected status value %s' % status)
+
         # Extract the URL of the video
         video_url = video_json['file_data']
-        
+
         # Video JSON does not provide enough meta data
         # We will extract some from the video web page instead
-        video_page_url = 'http://' + mobj.group('url')
-        video_page = self._download_webpage(video_page_url, video_id, u'Downloading video page')
+        video_page_url = 'http://smotri.com/video/view/?id=%s' % video_id
+        video_page = self._download_webpage(video_page_url, video_id, 'Downloading video page')
 
         # Warning if video is unavailable
         warning = self._html_search_regex(
             r'<div class="videoUnModer">(.*?)</div>', video_page,
-            u'warning message', default=None)
+            'warning message', default=None)
         if warning is not None:
             self._downloader.report_warning(
-                u'Video %s may not be available; smotri said: %s ' %
+                'Video %s may not be available; smotri said: %s ' %
                 (video_id, warning))
 
         # Adult content
-        if re.search(u'EroConfirmText">', video_page) is not None:
+        if re.search('EroConfirmText">', video_page) is not None:
             self.report_age_confirmation()
             confirm_string = self._html_search_regex(
                 r'<a href="/video/view/\?id=%s&confirm=([^"]+)" title="[^"]+">' % video_id,
-                video_page, u'confirm string')
+                video_page, 'confirm string')
             confirm_url = video_page_url + '&confirm=%s' % confirm_string
-            video_page = self._download_webpage(confirm_url, video_id, u'Downloading video page (age confirmed)')
+            video_page = self._download_webpage(confirm_url, video_id, 'Downloading video page (age confirmed)')
             adult_content = True
         else:
             adult_content = False
-        
+
         # Extract the rest of meta data
-        video_title = self._search_meta(u'name', video_page, u'title')
+        video_title = self._search_meta('name', video_page, 'title')
         if not video_title:
             video_title = os.path.splitext(url_basename(video_url))[0]
 
-        video_description = self._search_meta(u'description', video_page)
-        END_TEXT = u' на сайте Smotri.com'
+        video_description = self._search_meta('description', video_page)
+        END_TEXT = ' на сайте Smotri.com'
         if video_description and video_description.endswith(END_TEXT):
             video_description = video_description[:-len(END_TEXT)]
-        START_TEXT = u'Смотреть онлайн ролик '
+        START_TEXT = 'Смотреть онлайн ролик '
         if video_description and video_description.startswith(START_TEXT):
             video_description = video_description[len(START_TEXT):]
-        video_thumbnail = self._search_meta(u'thumbnail', video_page)
+        video_thumbnail = self._search_meta('thumbnail', video_page)
 
-        upload_date_str = self._search_meta(u'uploadDate', video_page, u'upload date')
+        upload_date_str = self._search_meta('uploadDate', video_page, 'upload date')
         if upload_date_str:
             upload_date_m = re.search(r'(?P<year>\d{4})\.(?P<month>\d{2})\.(?P<day>\d{2})T', upload_date_str)
             video_upload_date = (
@@ -183,8 +220,8 @@ class SmotriIE(InfoExtractor):
             )
         else:
             video_upload_date = None
-        
-        duration_str = self._search_meta(u'duration', video_page)
+
+        duration_str = self._search_meta('duration', video_page)
         if duration_str:
             duration_m = re.search(r'T(?P<hours>[0-9]{2})H(?P<minutes>[0-9]{2})M(?P<seconds>[0-9]{2})S', duration_str)
             video_duration = (
@@ -197,19 +234,19 @@ class SmotriIE(InfoExtractor):
             )
         else:
             video_duration = None
-        
+
         video_uploader = self._html_search_regex(
-            u'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info[^"]+">(.*?)</a>',
-            video_page, u'uploader', fatal=False, flags=re.MULTILINE|re.DOTALL)
-        
+            '<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info[^"]+">(.*?)</a>',
+            video_page, 'uploader', fatal=False, flags=re.MULTILINE|re.DOTALL)
+
         video_uploader_id = self._html_search_regex(
-            u'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info\\(.*?\'([^\']+)\'\\);">',
-            video_page, u'uploader id', fatal=False, flags=re.MULTILINE|re.DOTALL)
-        
+            '<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info\\(.*?\'([^\']+)\'\\);">',
+            video_page, 'uploader id', fatal=False, flags=re.MULTILINE|re.DOTALL)
+
         video_view_count = self._html_search_regex(
-            u'Общее количество просмотров.*?<span class="Number">(\\d+)</span>',
-            video_page, u'view count', fatal=False, flags=re.MULTILINE|re.DOTALL)
-                
+            'Общее количество просмотров.*?<span class="Number">(\\d+)</span>',
+            video_page, 'view count', fatal=False, flags=re.MULTILINE|re.DOTALL)
+
         return {
             'id': video_id,
             'url': video_url,
@@ -220,60 +257,76 @@ class SmotriIE(InfoExtractor):
             'upload_date': video_upload_date,
             'uploader_id': video_uploader_id,
             'duration': video_duration,
-            'view_count': video_view_count,
+            'view_count': int_or_none(video_view_count),
             'age_limit': 18 if adult_content else 0,
             'video_page_url': video_page_url
         }
 
 
 class SmotriCommunityIE(InfoExtractor):
-    IE_DESC = u'Smotri.com community videos'
-    IE_NAME = u'smotri:community'
+    IE_DESC = 'Smotri.com community videos'
+    IE_NAME = 'smotri:community'
     _VALID_URL = r'^https?://(?:www\.)?smotri\.com/community/video/(?P<communityid>[0-9A-Za-z_\'-]+)'
+    _TEST = {
+        'url': 'http://smotri.com/community/video/kommuna',
+        'info_dict': {
+            'id': 'kommuna',
+            'title': 'КПРФ',
+        },
+        'playlist_mincount': 4,
+    }
     
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         community_id = mobj.group('communityid')
 
         url = 'http://smotri.com/export/rss/video/by/community/-/%s/video.xml' % community_id
-        rss = self._download_xml(url, community_id, u'Downloading community RSS')
+        rss = self._download_xml(url, community_id, 'Downloading community RSS')
 
         entries = [self.url_result(video_url.text, 'Smotri')
                    for video_url in rss.findall('./channel/item/link')]
 
         description_text = rss.find('./channel/description').text
         community_title = self._html_search_regex(
-            u'^Видео сообщества "([^"]+)"$', description_text, u'community title')
+            '^Видео сообщества "([^"]+)"$', description_text, 'community title')
 
         return self.playlist_result(entries, community_id, community_title)
 
 
 class SmotriUserIE(InfoExtractor):
-    IE_DESC = u'Smotri.com user videos'
-    IE_NAME = u'smotri:user'
+    IE_DESC = 'Smotri.com user videos'
+    IE_NAME = 'smotri:user'
     _VALID_URL = r'^https?://(?:www\.)?smotri\.com/user/(?P<userid>[0-9A-Za-z_\'-]+)'
+    _TESTS = [{
+        'url': 'http://smotri.com/user/inspector',
+        'info_dict': {
+            'id': 'inspector',
+            'title': 'Inspector',
+        },
+        'playlist_mincount': 9,
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         user_id = mobj.group('userid')
 
         url = 'http://smotri.com/export/rss/user/video/-/%s/video.xml' % user_id
-        rss = self._download_xml(url, user_id, u'Downloading user RSS')
+        rss = self._download_xml(url, user_id, 'Downloading user RSS')
 
         entries = [self.url_result(video_url.text, 'Smotri')
                    for video_url in rss.findall('./channel/item/link')]
 
         description_text = rss.find('./channel/description').text
         user_nickname = self._html_search_regex(
-            u'^Видео режиссера (.*)$', description_text,
-            u'user nickname')
+            '^Видео режиссера (.*)$', description_text,
+            'user nickname')
 
         return self.playlist_result(entries, user_id, user_nickname)
 
 
 class SmotriBroadcastIE(InfoExtractor):
-    IE_DESC = u'Smotri.com broadcasts'
-    IE_NAME = u'smotri:broadcast'
+    IE_DESC = 'Smotri.com broadcasts'
+    IE_NAME = 'smotri:broadcast'
     _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/live/(?P<broadcastid>[^/]+))/?.*'
 
     def _real_extract(self, url):
@@ -281,46 +334,40 @@ class SmotriBroadcastIE(InfoExtractor):
         broadcast_id = mobj.group('broadcastid')
 
         broadcast_url = 'http://' + mobj.group('url')
-        broadcast_page = self._download_webpage(broadcast_url, broadcast_id, u'Downloading broadcast page')
+        broadcast_page = self._download_webpage(broadcast_url, broadcast_id, 'Downloading broadcast page')
 
-        if re.search(u'>Режиссер с логином <br/>"%s"<br/> <span>не существует<' % broadcast_id, broadcast_page) is not None:
-            raise ExtractorError(u'Broadcast %s does not exist' % broadcast_id, expected=True)
+        if re.search('>Режиссер с логином <br/>"%s"<br/> <span>не существует<' % broadcast_id, broadcast_page) is not None:
+            raise ExtractorError('Broadcast %s does not exist' % broadcast_id, expected=True)
 
         # Adult content
-        if re.search(u'EroConfirmText">', broadcast_page) is not None:
+        if re.search('EroConfirmText">', broadcast_page) is not None:
 
             (username, password) = self._get_login_info()
             if username is None:
-                raise ExtractorError(u'Erotic broadcasts allowed only for registered users, '
-                    u'use --username and --password options to provide account credentials.', expected=True)
-
-            # Log in
-            login_form_strs = {
-                u'login-hint53': '1',
-                u'confirm_erotic': '1',
-                u'login': username,
-                u'password': password,
+                raise ExtractorError('Erotic broadcasts allowed only for registered users, '
+                    'use --username and --password options to provide account credentials.', expected=True)
+
+            login_form = {
+                'login-hint53': '1',
+                'confirm_erotic': '1',
+                'login': username,
+                'password': password,
             }
-            # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
-            # chokes on unicode
-            login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
-            login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
-            login_url = broadcast_url + '/?no_redirect=1'
-            request = compat_urllib_request.Request(login_url, login_data)
+
+            request = compat_urllib_request.Request(broadcast_url + '/?no_redirect=1', compat_urllib_parse.urlencode(login_form))
             request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-            broadcast_page = self._download_webpage(
-                request, broadcast_id, note=u'Logging in and confirming age')
+            broadcast_page = self._download_webpage(request, broadcast_id, 'Logging in and confirming age')
 
-            if re.search(u'>Неверный логин или пароль<', broadcast_page) is not None:
-                raise ExtractorError(u'Unable to log in: bad username or password', expected=True)
+            if re.search('>Неверный логин или пароль<', broadcast_page) is not None:
+                raise ExtractorError('Unable to log in: bad username or password', expected=True)
 
             adult_content = True
         else:
             adult_content = False
 
         ticket = self._html_search_regex(
-            u'window\.broadcast_control\.addFlashVar\\(\'file\', \'([^\']+)\'\\);',
-            broadcast_page, u'broadcast ticket')
+            'window\.broadcast_control\.addFlashVar\\(\'file\', \'([^\']+)\'\\);',
+            broadcast_page, 'broadcast ticket')
 
         url = 'http://smotri.com/broadcast/view/url/?ticket=%s' % ticket
 
@@ -328,22 +375,22 @@ class SmotriBroadcastIE(InfoExtractor):
         if broadcast_password:
             url += '&pass=%s' % hashlib.md5(broadcast_password.encode('utf-8')).hexdigest()
 
-        broadcast_json_page = self._download_webpage(url, broadcast_id, u'Downloading broadcast JSON')
+        broadcast_json_page = self._download_webpage(url, broadcast_id, 'Downloading broadcast JSON')
 
         try:
             broadcast_json = json.loads(broadcast_json_page)
 
             protected_broadcast = broadcast_json['_pass_protected'] == 1
             if protected_broadcast and not broadcast_password:
-                raise ExtractorError(u'This broadcast is protected by a password, use the --video-password option', expected=True)
+                raise ExtractorError('This broadcast is protected by a password, use the --video-password option', expected=True)
 
             broadcast_offline = broadcast_json['is_play'] == 0
             if broadcast_offline:
-                raise ExtractorError(u'Broadcast %s is offline' % broadcast_id, expected=True)
+                raise ExtractorError('Broadcast %s is offline' % broadcast_id, expected=True)
 
             rtmp_url = broadcast_json['_server']
             if not rtmp_url.startswith('rtmp://'):
-                raise ExtractorError(u'Unexpected broadcast rtmp URL')
+                raise ExtractorError('Unexpected broadcast rtmp URL')
 
             broadcast_playpath = broadcast_json['_streamName']
             broadcast_thumbnail = broadcast_json['_imgURL']
@@ -354,8 +401,8 @@ class SmotriBroadcastIE(InfoExtractor):
             rtmp_conn = 'S:%s' % uuid.uuid4().hex
         except KeyError:
             if protected_broadcast:
-                raise ExtractorError(u'Bad broadcast password', expected=True)
-            raise ExtractorError(u'Unexpected broadcast JSON')
+                raise ExtractorError('Bad broadcast password', expected=True)
+            raise ExtractorError('Unexpected broadcast JSON')
 
         return {
             'id': broadcast_id,