]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/smotri.py
debian/control: State compliancy with Debian Policy 4.4.0.
[youtubedl] / youtube_dl / extractor / smotri.py
index 13e7e71cb37b4d7b5ec2e5ab2c341551e7e05f28..45995f30f301ff1a442c584505bf3d0d848ca069 100644 (file)
@@ -1,7 +1,6 @@
-# encoding: utf-8
+# coding: utf-8
 from __future__ import unicode_literals
 
-import os.path
 import re
 import json
 import hashlib
@@ -9,25 +8,26 @@ import uuid
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_urllib_parse,
-    compat_urllib_request,
     ExtractorError,
-    url_basename,
     int_or_none,
+    sanitized_Request,
+    unified_strdate,
+    urlencode_postdata,
+    xpath_text,
 )
 
 
 class SmotriIE(InfoExtractor):
     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})'
+    _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',
-            'md5': '2a7b08249e6f5636557579c368040eb9',
+            'md5': '02c0dfab2102984e9c5bb585cc7cc321',
             'info_dict': {
                 'id': 'v261036632ab',
                 'ext': 'mp4',
@@ -35,7 +35,6 @@ class SmotriIE(InfoExtractor):
                 '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',
             },
         },
@@ -50,11 +49,10 @@ class SmotriIE(InfoExtractor):
                 '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
+        # video-password, not approved by moderator
         {
             'url': 'http://smotri.com/video/view/?id=v1390466a13c',
             'md5': 'f6331cef33cad65a0815ee482a54440b',
@@ -66,13 +64,30 @@ class SmotriIE(InfoExtractor):
                 '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',
             },
+            'skip': 'Video is not approved by moderator',
         },
-        # age limit + video-password
+        # video-password
+        {
+            'url': 'http://smotri.com/video/view/?id=v6984858774#',
+            'md5': 'f11e01d13ac676370fc3b95b9bda11b0',
+            'info_dict': {
+                'id': 'v6984858774',
+                'ext': 'mp4',
+                'title': 'Дача Солженицина ПАРОЛЬ 223322',
+                'uploader': 'psavari1',
+                'uploader_id': 'psavari1',
+                'upload_date': '20081103',
+                'thumbnail': r're:^https?://.*\.jpg$',
+            },
+            'params': {
+                'videopassword': '223322',
+            },
+        },
+        # age limit + video-password, not approved by moderator
         {
             'url': 'http://smotri.com/video/view/?id=v15408898bcf',
             'md5': '91e909c9f0521adf5ee86fbe073aad70',
@@ -85,16 +100,34 @@ class SmotriIE(InfoExtractor):
                 'upload_date': '20101001',
                 'thumbnail': 'http://frame3.loadup.ru/75/75/1540889.1.3.jpg',
                 'age_limit': 18,
-                'description': 'этот ролик не покажут по ТВ, видео этот ролик не покажут по ТВ',
             },
             'params': {
                 'videopassword': '333'
-            }
+            },
+            'skip': 'Video is not approved by moderator',
+        },
+        # age limit + video-password
+        {
+            'url': 'http://smotri.com/video/view/?id=v7780025814',
+            'md5': 'b4599b068422559374a59300c5337d72',
+            'info_dict': {
+                'id': 'v7780025814',
+                'ext': 'mp4',
+                'title': 'Sexy Beach (пароль 123)',
+                'uploader': 'вАся',
+                'uploader_id': 'asya_prosto',
+                'upload_date': '20081218',
+                'thumbnail': r're:^https?://.*\.jpg$',
+                'age_limit': 18,
+            },
+            'params': {
+                'videopassword': '123'
+            },
         },
         # swf player
         {
             'url': 'http://pics.smotri.com/scrubber_custom8.swf?file=v9188090500',
-            'md5': '4d47034979d9390d14acdf59c4935bc2',
+            'md5': '31099eeb4bc906712c5f40092045108d',
             'info_dict': {
                 'id': 'v9188090500',
                 'ext': 'mp4',
@@ -102,17 +135,11 @@ class SmotriIE(InfoExtractor):
                 '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(
@@ -131,50 +158,59 @@ class SmotriIE(InfoExtractor):
     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')
-        real_video_id = mobj.group('realvideoid')
-
-        # 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)
-
-        status = video_json['status']
-        if status == self._VIDEO_NOT_FOUND:
-            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_id = self._match_id(url)
+
+        video_form = {
+            'ticket': video_id,
+            'video_url': '1',
+            'frame_url': '1',
+            'devid': 'LoadupFlashPlayer',
+            'getvideoinfo': '1',
+        }
+
+        video_password = self._downloader.params.get('videopassword')
+        if video_password:
+            video_form['pass'] = hashlib.md5(video_password.encode('utf-8')).hexdigest()
+
+        video = self._download_json(
+            'http://smotri.com/video/view/url/bot/',
+            video_id, 'Downloading video JSON',
+            data=urlencode_postdata(video_form),
+            headers={'Content-Type': 'application/x-www-form-urlencoded'})
+
+        video_url = video.get('_vidURL') or video.get('_vidURL_mp4')
+
+        if not video_url:
+            if video.get('_moderate_no'):
+                raise ExtractorError(
+                    'Video %s has not been approved by moderator' % video_id, expected=True)
+
+            if video.get('error'):
+                raise ExtractorError('Video %s does not exist' % video_id, expected=True)
+
+            if video.get('_pass_protected') == 1:
+                msg = ('Invalid video password' if video_password
+                       else 'This video is protected by a password, use the --video-password option')
+                raise ExtractorError(msg, expected=True)
+
+        title = video['title']
+        thumbnail = video.get('_imgURL')
+        upload_date = unified_strdate(video.get('added'))
+        uploader = video.get('userNick')
+        uploader_id = video.get('userLogin')
+        duration = int_or_none(video.get('duration'))
 
         # Video JSON does not provide enough meta data
         # We will extract some from the video web page instead
-        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')
+        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(
@@ -182,128 +218,90 @@ class SmotriIE(InfoExtractor):
                 (video_id, warning))
 
         # Adult content
-        if re.search('EroConfirmText">', video_page) is not None:
+        if 'EroConfirmText">' in webpage:
             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)')
+                r'<a[^>]+href="/video/view/\?id=%s&confirm=([^"]+)"' % video_id,
+                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(
-            'Общее количество просмотров.*?<span class="Number">(\\d+)</span>',
-            video_page, 'view count', fatal=False, flags=re.MULTILINE|re.DOTALL)
+        view_count = self._html_search_regex(
+            r'(?s)Общее количество просмотров.*?<span class="Number">(\d+)</span>',
+            webpage, 'view count', fatal=False)
 
         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': int_or_none(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
         }
 
 
 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_\'-]+)'
-    
+    _VALID_URL = r'https?://(?:www\.)?smotri\.com/community/video/(?P<id>[0-9A-Za-z_\'-]+)'
+    _TEST = {
+        'url': 'http://smotri.com/community/video/kommuna',
+        'info_dict': {
+            'id': 'kommuna',
+        },
+        'playlist_mincount': 4,
+    }
+
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        community_id = mobj.group('communityid')
+        community_id = self._match_id(url)
 
-        url = 'http://smotri.com/export/rss/video/by/community/-/%s/video.xml' % community_id
-        rss = self._download_xml(url, community_id, 'Downloading community RSS')
+        rss = self._download_xml(
+            'http://smotri.com/export/rss/video/by/community/-/%s/video.xml' % community_id,
+            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(
-            '^Видео сообщества "([^"]+)"$', description_text, 'community title')
+        entries = [
+            self.url_result(video_url.text, SmotriIE.ie_key())
+            for video_url in rss.findall('./channel/item/link')]
 
-        return self.playlist_result(entries, community_id, community_title)
+        return self.playlist_result(entries, community_id)
 
 
 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_\'-]+)'
+    _VALID_URL = r'https?://(?:www\.)?smotri\.com/user/(?P<id>[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')
+        user_id = self._match_id(url)
 
-        url = 'http://smotri.com/export/rss/user/video/-/%s/video.xml' % user_id
-        rss = self._download_xml(url, user_id, 'Downloading user RSS')
+        rss = self._download_xml(
+            'http://smotri.com/export/rss/user/video/-/%s/video.xml' % user_id,
+            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(
-            '^Видео режиссера (.*)$', description_text,
-            'user nickname')
+        description_text = xpath_text(rss, './channel/description') or ''
+        user_nickname = self._search_regex(
+            '^Видео режиссера (.+)$', description_text,
+            'user nickname', fatal=False)
 
         return self.playlist_result(entries, user_id, user_nickname)
 
@@ -311,25 +309,27 @@ class SmotriUserIE(InfoExtractor):
 class SmotriBroadcastIE(InfoExtractor):
     IE_DESC = 'Smotri.com broadcasts'
     IE_NAME = 'smotri:broadcast'
-    _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/live/(?P<broadcastid>[^/]+))/?.*'
+    _VALID_URL = r'https?://(?:www\.)?(?P<url>smotri\.com/live/(?P<id>[^/]+))/?.*'
+    _NETRC_MACHINE = 'smotri'
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        broadcast_id = mobj.group('broadcastid')
+        broadcast_id = mobj.group('id')
 
         broadcast_url = 'http://' + mobj.group('url')
         broadcast_page = self._download_webpage(broadcast_url, broadcast_id, 'Downloading broadcast page')
 
         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)
+            raise ExtractorError(
+                'Broadcast %s does not exist' % broadcast_id, expected=True)
 
         # Adult content
         if re.search('EroConfirmText">', broadcast_page) is not None:
 
             (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)
+                self.raise_login_required(
+                    'Erotic broadcasts allowed only for registered users')
 
             login_form = {
                 'login-hint53': '1',
@@ -338,50 +338,59 @@ class SmotriBroadcastIE(InfoExtractor):
                 'password': password,
             }
 
-            request = compat_urllib_request.Request(broadcast_url + '/?no_redirect=1', compat_urllib_parse.urlencode(login_form))
+            request = sanitized_Request(
+                broadcast_url + '/?no_redirect=1', urlencode_postdata(login_form))
             request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-            broadcast_page = self._download_webpage(request, broadcast_id, 'Logging in and confirming age')
+            broadcast_page = self._download_webpage(
+                request, broadcast_id, 'Logging in and confirming age')
 
-            if re.search('>Неверный логин или пароль<', broadcast_page) is not None:
-                raise ExtractorError('Unable to log in: bad username or password', expected=True)
+            if '>Неверный логин или пароль<' in broadcast_page:
+                raise ExtractorError(
+                    'Unable to log in: bad username or password', expected=True)
 
             adult_content = True
         else:
             adult_content = False
 
         ticket = self._html_search_regex(
-            'window\.broadcast_control\.addFlashVar\\(\'file\', \'([^\']+)\'\\);',
-            broadcast_page, 'broadcast ticket')
+            (r'data-user-file=(["\'])(?P<ticket>(?!\1).+)\1',
+             r"window\.broadcast_control\.addFlashVar\('file'\s*,\s*'(?P<ticket>[^']+)'\)"),
+            broadcast_page, 'broadcast ticket', group='ticket')
 
-        url = 'http://smotri.com/broadcast/view/url/?ticket=%s' % ticket
+        broadcast_url = 'http://smotri.com/broadcast/view/url/?ticket=%s' % ticket
 
-        broadcast_password = self._downloader.params.get('videopassword', None)
+        broadcast_password = self._downloader.params.get('videopassword')
         if broadcast_password:
-            url += '&pass=%s' % hashlib.md5(broadcast_password.encode('utf-8')).hexdigest()
+            broadcast_url += '&pass=%s' % hashlib.md5(broadcast_password.encode('utf-8')).hexdigest()
 
-        broadcast_json_page = self._download_webpage(url, broadcast_id, 'Downloading broadcast JSON')
+        broadcast_json_page = self._download_webpage(
+            broadcast_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('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('Broadcast %s is offline' % broadcast_id, expected=True)
 
             rtmp_url = broadcast_json['_server']
-            if not rtmp_url.startswith('rtmp://'):
+            mobj = re.search(r'^rtmp://[^/]+/(?P<app>.+)/?$', rtmp_url)
+            if not mobj:
                 raise ExtractorError('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']
+            broadcast_app = '%s/%s' % (mobj.group('app'), broadcast_json['_vidURL'])
+            broadcast_thumbnail = broadcast_json.get('_imgURL')
+            broadcast_title = self._live_title(broadcast_json['title'])
+            broadcast_description = broadcast_json.get('description')
+            broadcaster_nick = broadcast_json.get('nick')
+            broadcaster_login = broadcast_json.get('login')
             rtmp_conn = 'S:%s' % uuid.uuid4().hex
         except KeyError:
             if protected_broadcast:
@@ -399,6 +408,9 @@ class SmotriBroadcastIE(InfoExtractor):
             'age_limit': 18 if adult_content else 0,
             'ext': 'flv',
             'play_path': broadcast_playpath,
+            'player_url': 'http://pics.smotri.com/broadcast_play.swf',
+            'app': broadcast_app,
             'rtmp_live': True,
-            'rtmp_conn': rtmp_conn
+            'rtmp_conn': rtmp_conn,
+            'is_live': True,
         }