X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fsmotri.py;h=beea58d6317727133f85b74c14097445cf785dc5;hb=1538eff6d8ee73f2ae4ac1dd832a9ef978c63074;hp=ea42d5320293e5f26856b485732d82240296ad00;hpb=25939ffe56905784c9260e2fa33231f03ccf08a5;p=youtube-dl

diff --git a/youtube_dl/extractor/smotri.py b/youtube_dl/extractor/smotri.py
index ea42d5320..beea58d63 100644
--- a/youtube_dl/extractor/smotri.py
+++ b/youtube_dl/extractor/smotri.py
@@ -1,59 +1,66 @@
 # encoding: utf-8
 
+import os.path
 import re
 import json
 import hashlib
+import uuid
 
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
-    ExtractorError
+    compat_urllib_parse,
+    compat_urllib_request,
+    ExtractorError,
+    url_basename,
 )
 
 
 class SmotriIE(InfoExtractor):
     IE_DESC = u'Smotri.com'
     IE_NAME = u'smotri'
-    _VALID_URL = r'^(?:http://)?(?:www\.)?(?P<url>smotri\.com/video/view/\?id=(?P<videoid>v(?P<realvideoid>[0-9]+)[a-z0-9]{4}))'
-    
+    _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/video/view/\?id=(?P<videoid>v(?P<realvideoid>[0-9]+)[a-z0-9]{4}))'
+
     _TESTS = [
         # real video id 2610366
         {
             u'url': u'http://smotri.com/video/view/?id=v261036632ab',
             u'file': u'v261036632ab.mp4',
-            u'md5': u'46a72e83a6ad8862b64fa6953fa93f8a',
+            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'thumbnail': u'http://frame6.loadup.ru/8b/a9/2610366.3.3.jpg'
+                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',
             },
         },
         # real video id 57591
         {
             u'url': u'http://smotri.com/video/view/?id=v57591cb20',
             u'file': u'v57591cb20.flv',
-            u'md5': u'9eae59f6dda7087bf39a140e2fff5757',
+            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'thumbnail': u'http://frame4.loadup.ru/03/ed/57591.2.3.jpg'
-            },            
+                u'description': u'test, видео test',
+                u'thumbnail': u'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'fe4dd9357558d5ee3c8fc0ef0d39de66',
+            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'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',
             },
             u'params': {
                 u'videopassword': u'qwerty',
@@ -63,15 +70,16 @@ class SmotriIE(InfoExtractor):
         {
             u'url': u'http://smotri.com/video/view/?id=v15408898bcf',
             u'file': u'v15408898bcf.flv',
-            u'md5': u'c66a5d61379ac6fde06f07eebe436316',
+            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'age_limit': 18,
+                u'description': u'этот ролик не покажут по ТВ, видео этот ролик не покажут по ТВ',
+            },
             u'params': {
                 u'videopassword': u'333'
             }
@@ -82,14 +90,15 @@ class SmotriIE(InfoExtractor):
     _PASSWORD_NOT_VERIFIED = 1
     _PASSWORD_DETECTED = 2
     _VIDEO_NOT_FOUND = 3
-    
+
     def _search_meta(self, name, html, display_name=None):
         if display_name is None:
             display_name = name
         return self._html_search_regex(
             r'<meta itemprop="%s" content="([^"]+)" />' % re.escape(name),
             html, display_name, fatal=False)
-        
+        return self._html_search_meta(name, html, display_name)
+
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('videoid')
@@ -103,12 +112,12 @@ class SmotriIE(InfoExtractor):
         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
+        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)
-            video_json_url += '&md5pass=%s' % hashlib.md5(video_password).hexdigest()
+            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 = json.loads(video_json_page)
             status = video_json['status']
@@ -120,18 +129,26 @@ class SmotriIE(InfoExtractor):
         
         # Extract the URL of the video
         video_url = video_json['file_data']
-        video_ext = determine_ext(video_url)
         
         # 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')
-        
+
+        # Warning if video is unavailable
+        warning = self._html_search_regex(
+            r'<div class="videoUnModer">(.*?)</div>', video_page,
+            u'warning messagef', default=None)
+        if warning is not None:
+            self._downloader.report_warning(
+                u'Video %s may not be available; smotri said: %s ' %
+                (video_id, warning))
+
         # Adult content
         if re.search(u'EroConfirmText">', video_page) is not None:
             self.report_age_confirmation()
             confirm_string = self._html_search_regex(
-                ur'<a href="/video/view/\?id=%s&confirm=([^"]+)" title="[^"]+">' % video_id,
+                r'<a href="/video/view/\?id=%s&confirm=([^"]+)" title="[^"]+">' % video_id,
                 video_page, u'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)')
@@ -142,98 +159,215 @@ class SmotriIE(InfoExtractor):
         # Extract the rest of meta data
         video_title = self._search_meta(u'name', video_page, u'title')
         if not video_title:
-            video_title = video_url.rsplit('/', 1)[-1]
-        
+            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'
+        if video_description and video_description.endswith(END_TEXT):
+            video_description = video_description[:-len(END_TEXT)]
+        START_TEXT = u'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)
-        
-        upload_date_str = self._search_meta(u'uploadDate', video_page, u'upload date')        
-        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 = (
-            (
-                upload_date_m.group('year') +
-                upload_date_m.group('month') +
-                upload_date_m.group('day')
+
+        upload_date_str = self._search_meta(u'uploadDate', video_page, u'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 = (
+                (
+                    upload_date_m.group('year') +
+                    upload_date_m.group('month') +
+                    upload_date_m.group('day')
+                )
+                if upload_date_m else None
             )
-            if upload_date_m else None
-        )
+        else:
+            video_upload_date = None
         
         duration_str = self._search_meta(u'duration', video_page)
-        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 = (
-            (
-                (int(duration_m.group('hours')) * 60 * 60) +
-                (int(duration_m.group('minutes')) * 60) +
-                int(duration_m.group('seconds'))
+        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 = (
+                (
+                    (int(duration_m.group('hours')) * 60 * 60) +
+                    (int(duration_m.group('minutes')) * 60) +
+                    int(duration_m.group('seconds'))
+                )
+                if duration_m else None
             )
-            if duration_m else None
-        )
+        else:
+            video_duration = None
         
         video_uploader = self._html_search_regex(
-            ur'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info[^"]+">(.*?)</a>',
+            u'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info[^"]+">(.*?)</a>',
             video_page, u'uploader', fatal=False, flags=re.MULTILINE|re.DOTALL)
         
         video_uploader_id = self._html_search_regex(
-            ur'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info\(.*?\'([^\']+)\'\);">',
+            u'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info\\(.*?\'([^\']+)\'\\);">',
             video_page, u'uploader id', fatal=False, flags=re.MULTILINE|re.DOTALL)
         
         video_view_count = self._html_search_regex(
-            ur'Общее количество просмотров.*?<span class="Number">(\d+)</span>',
+            u'Общее количество просмотров.*?<span class="Number">(\\d+)</span>',
             video_page, u'view count', fatal=False, flags=re.MULTILINE|re.DOTALL)
                 
         return {
             'id': video_id,
             'url': video_url,
             'title': video_title,
-            'ext': video_ext,
             'thumbnail': video_thumbnail,
             'description': video_description,
             'uploader': video_uploader,
             'upload_date': video_upload_date,
             'uploader_id': video_uploader_id,
-            'video_duration': video_duration,
+            'duration': video_duration,
             'view_count': 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'
-    _VALID_URL = r'^(?:http://)?(?:www\.)?smotri\.com/community/video/(?P<communityid>[0-9A-Za-z_\'-]+)'
+    _VALID_URL = r'^https?://(?:www\.)?smotri\.com/community/video/(?P<communityid>[0-9A-Za-z_\'-]+)'
     
     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')
-    
+
         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(
-            ur'^Видео сообщества "([^"]+)"$', rss.find('./channel/description').text, u'community title')
+            u'^Видео сообщества "([^"]+)"$', description_text, u'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'
-    _VALID_URL = r'^(?:http://)?(?:www\.)?smotri\.com/user/(?P<userid>[0-9A-Za-z_\'-]+)'
-    
+    _VALID_URL = r'^https?://(?:www\.)?smotri\.com/user/(?P<userid>[0-9A-Za-z_\'-]+)'
+
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, 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')
-        
+
         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(
-            ur'^Видео режиссера (.*)$', rss.find('./channel/description').text, u'user nickname')
+            u'^Видео режиссера (.*)$', description_text,
+            u'user nickname')
 
         return self.playlist_result(entries, user_id, user_nickname)
-        
\ No newline at end of file
+
+
+class SmotriBroadcastIE(InfoExtractor):
+    IE_DESC = u'Smotri.com broadcasts'
+    IE_NAME = u'smotri:broadcast'
+    _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/live/(?P<broadcastid>[^/]+))/?.*'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        broadcast_id = mobj.group('broadcastid')
+
+        broadcast_url = 'http://' + mobj.group('url')
+        broadcast_page = self._download_webpage(broadcast_url, broadcast_id, u'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)
+
+        # Adult content
+        if re.search(u'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,
+            }
+            # 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.add_header('Content-Type', 'application/x-www-form-urlencoded')
+            broadcast_page = self._download_webpage(
+                request, broadcast_id, note=u'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)
+
+            adult_content = True
+        else:
+            adult_content = False
+
+        ticket = self._html_search_regex(
+            u'window\.broadcast_control\.addFlashVar\\(\'file\', \'([^\']+)\'\\);',
+            broadcast_page, u'broadcast ticket')
+
+        url = 'http://smotri.com/broadcast/view/url/?ticket=%s' % ticket
+
+        broadcast_password = self._downloader.params.get('videopassword', None)
+        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')
+
+        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)
+
+            broadcast_offline = broadcast_json['is_play'] == 0
+            if broadcast_offline:
+                raise ExtractorError(u'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')
+
+            broadcast_playpath = broadcast_json['_streamName']
+            broadcast_thumbnail = broadcast_json['_imgURL']
+            broadcast_title = broadcast_json['title']
+            broadcast_description = broadcast_json['description']
+            broadcaster_nick = broadcast_json['nick']
+            broadcaster_login = broadcast_json['login']
+            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')
+
+        return {
+            'id': broadcast_id,
+            'url': rtmp_url,
+            'title': broadcast_title,
+            'thumbnail': broadcast_thumbnail,
+            'description': broadcast_description,
+            'uploader': broadcaster_nick,
+            'uploader_id': broadcaster_login,
+            'age_limit': 18 if adult_content else 0,
+            'ext': 'flv',
+            'play_path': broadcast_playpath,
+            'rtmp_live': True,
+            'rtmp_conn': rtmp_conn
+        }