]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/smotri.py
Prepare for upload.
[youtubedl] / youtube_dl / extractor / smotri.py
index 540c557039fd5876ad32af816152fdc912a87e69..0751efc6111c96ca1c089c66183429f9bde6147c 100644 (file)
@@ -1,7 +1,6 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import os.path
 import re
 import json
 import hashlib
@@ -12,57 +11,58 @@ from ..utils import (
     compat_urllib_parse,
     compat_urllib_request,
     ExtractorError,
-    url_basename,
+    int_or_none,
+    unified_strdate,
 )
 
 
 class SmotriIE(InfoExtractor):
     IE_DESC = 'Smotri.com'
     IE_NAME = 'smotri'
-    _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/video/view/\?id=(?P<videoid>v(?P<realvideoid>[0-9]+)[a-z0-9]{4}))'
+    _VALID_URL = r'^https?://(?:www\.)?(?:smotri\.com/video/view/\?id=|pics\.smotri\.com/(?:player|scrubber_custom8)\.swf\?file=)(?P<id>v(?P<realvideoid>[0-9]+)[a-z0-9]{4})'
     _NETRC_MACHINE = 'smotri'
 
     _TESTS = [
         # real video id 2610366
         {
             'url': 'http://smotri.com/video/view/?id=v261036632ab',
-            'file': 'v261036632ab.mp4',
             '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
         {
             'url': 'http://smotri.com/video/view/?id=v57591cb20',
-            'file': 'v57591cb20.flv',
             '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
         {
             'url': 'http://smotri.com/video/view/?id=v1390466a13c',
-            'file': 'v1390466a13c.mp4',
             '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',
             },
             'params': {
                 'videopassword': 'qwerty',
@@ -71,27 +71,51 @@ class SmotriIE(InfoExtractor):
         # age limit + video-password
         {
             'url': 'http://smotri.com/video/view/?id=v15408898bcf',
-            'file': 'v15408898bcf.flv',
             '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',
             },
             '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',
+                '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:
@@ -102,44 +126,44 @@ class SmotriIE(InfoExtractor):
         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')
-        real_video_id = mobj.group('realvideoid')
+        video_id = self._match_id(url)
+
+        video_form = {
+            'ticket': video_id,
+            'video_url': '1',
+            'frame_url': '1',
+            'devid': 'LoadupFlashPlayer',
+            'getvideoinfo': '1',
+        }
+
+        request = compat_urllib_request.Request(
+            'http://smotri.com/video/view/url/bot/', compat_urllib_parse.urlencode(video_form))
+        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
+
+        video = self._download_json(request, video_id, 'Downloading video JSON')
 
-        # 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, 'Downloading video JSON')
-        video_json = json.loads(video_json_page)
+        if video.get('_moderate_no') or not video.get('moderated'):
+            raise ExtractorError('Video %s has not been approved by moderator' % video_id, expected=True)
 
-        status = video_json['status']
-        if status == self._VIDEO_NOT_FOUND:
+        if video.get('error'):
             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('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, '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('Video password is invalid', expected=True)
-
-        if status != self._SUCCESS:
-            raise ExtractorError('Unexpected status value %s' % status)
-
-        # Extract the URL of the video
-        video_url = video_json['file_data']
+
+        video_url = video.get('_vidURL') or video.get('_vidURL_mp4')
+        title = video['title']
+        thumbnail = video['_imgURL']
+        upload_date = unified_strdate(video['added'])
+        uploader = video['userNick']
+        uploader_id = video['userLogin']
+        duration = int_or_none(video['duration'])
 
         # 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, 'Downloading video page')
+        webpage_url = 'http://smotri.com/video/view/?id=%s' % video_id
+        webpage = self._download_webpage(webpage_url, video_id, 'Downloading video page')
 
         # Warning if video is unavailable
         warning = self._html_search_regex(
-            r'<div class="videoUnModer">(.*?)</div>', video_page,
+            r'<div class="videoUnModer">(.*?)</div>', webpage,
             'warning message', default=None)
         if warning is not None:
             self._downloader.report_warning(
@@ -147,84 +171,32 @@ class SmotriIE(InfoExtractor):
                 (video_id, warning))
 
         # Adult content
-        if re.search('EroConfirmText">', video_page) is not None:
+        if re.search('EroConfirmText">', webpage) 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, 'confirm string')
-            confirm_url = video_page_url + '&confirm=%s' % confirm_string
-            video_page = self._download_webpage(confirm_url, video_id, 'Downloading video page (age confirmed)')
+                webpage, 'confirm string')
+            confirm_url = webpage_url + '&confirm=%s' % confirm_string
+            webpage = 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('name', video_page, 'title')
-        if not video_title:
-            video_title = os.path.splitext(url_basename(video_url))[0]
-
-        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 = 'Смотреть онлайн ролик '
-        if video_description and video_description.startswith(START_TEXT):
-            video_description = video_description[len(START_TEXT):]
-        video_thumbnail = self._search_meta('thumbnail', video_page)
-
-        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 = (
-                (
-                    upload_date_m.group('year') +
-                    upload_date_m.group('month') +
-                    upload_date_m.group('day')
-                )
-                if upload_date_m else None
-            )
-        else:
-            video_upload_date = None
-
-        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 = (
-                (
-                    (int(duration_m.group('hours')) * 60 * 60) +
-                    (int(duration_m.group('minutes')) * 60) +
-                    int(duration_m.group('seconds'))
-                )
-                if duration_m else None
-            )
-        else:
-            video_duration = None
-
-        video_uploader = self._html_search_regex(
-            '<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(
-            '<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(
+        view_count = self._html_search_regex(
             'Общее количество просмотров.*?<span class="Number">(\\d+)</span>',
-            video_page, 'view count', fatal=False, flags=re.MULTILINE|re.DOTALL)
+            webpage, 'view count', fatal=False, flags=re.MULTILINE | re.DOTALL)
 
         return {
             'id': video_id,
             'url': video_url,
-            'title': video_title,
-            'thumbnail': video_thumbnail,
-            'description': video_description,
-            'uploader': video_uploader,
-            'upload_date': video_upload_date,
-            'uploader_id': video_uploader_id,
-            'duration': video_duration,
-            'view_count': video_view_count,
+            'title': title,
+            'thumbnail': thumbnail,
+            'uploader': uploader,
+            'upload_date': upload_date,
+            'uploader_id': uploader_id,
+            'duration': duration,
+            'view_count': int_or_none(view_count),
             'age_limit': 18 if adult_content else 0,
-            'video_page_url': video_page_url
         }
 
 
@@ -232,7 +204,15 @@ class SmotriCommunityIE(InfoExtractor):
     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')
@@ -254,6 +234,14 @@ class SmotriUserIE(InfoExtractor):
     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)
@@ -294,7 +282,7 @@ class SmotriBroadcastIE(InfoExtractor):
             (username, password) = self._get_login_info()
             if username is None:
                 raise ExtractorError('Erotic broadcasts allowed only for registered users, '
-                    'use --username and --password options to provide account credentials.', expected=True)
+                                     'use --username and --password options to provide account credentials.', expected=True)
 
             login_form = {
                 'login-hint53': '1',