X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Ffourtube.py;h=fc4a5a0fbf01801d598e20a9addd29ebef4a298e;hb=06e4874c997fb523bd9d8e675d1ec0e69ab501ed;hp=8e7a7e1569c863c8f171de5603c11fe8ccf3bc34;hpb=03635e2a711483ad3cf0bf5cdbde173fa37593c9;p=youtube-dl.git

diff --git a/youtube_dl/extractor/fourtube.py b/youtube_dl/extractor/fourtube.py
index 8e7a7e156..fc4a5a0fb 100644
--- a/youtube_dl/extractor/fourtube.py
+++ b/youtube_dl/extractor/fourtube.py
@@ -1,57 +1,117 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
-from ..utils import compat_urllib_request
+from ..utils import (
+    parse_duration,
+    parse_iso8601,
+    sanitized_Request,
+    str_to_int,
+)
+
 
 class FourTubeIE(InfoExtractor):
     IE_NAME = '4tube'
-    _VALID_URL = r'(?:https?://)?www\.4tube\.com/videos/(?P<id>\d+)/.*'
+    _VALID_URL = r'https?://(?:www\.)?4tube\.com/videos/(?P<id>\d+)'
 
     _TEST = {
-            'url': 'http://www.4tube.com/videos/209733/hot-babe-holly-michaels-gets-her-ass-stuffed-by-black',
-            'md5': '6516c8ac63b03de06bc8eac14362db4f',
-            'info_dict': {
-                'id': '209733',
-                'ext': 'mp4',
-                'title': 'Hot Babe Holly Michaels gets her ass stuffed by black'
-                }
-            }
+        'url': 'http://www.4tube.com/videos/209733/hot-babe-holly-michaels-gets-her-ass-stuffed-by-black',
+        'md5': '6516c8ac63b03de06bc8eac14362db4f',
+        'info_dict': {
+            'id': '209733',
+            'ext': 'mp4',
+            'title': 'Hot Babe Holly Michaels gets her ass stuffed by black',
+            'uploader': 'WCP Club',
+            'uploader_id': 'wcp-club',
+            'upload_date': '20131031',
+            'timestamp': 1383263892,
+            'duration': 583,
+            'view_count': int,
+            'like_count': int,
+            'categories': list,
+            'age_limit': 18,
+        }
+    }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
 
-        video_id = mobj.group('id')
-        webpage_url = 'http://www.4tube.com/videos/' + video_id
-        webpage = self._download_webpage(webpage_url, video_id)
+        title = self._html_search_meta('name', webpage)
+        timestamp = parse_iso8601(self._html_search_meta(
+            'uploadDate', webpage))
+        thumbnail = self._html_search_meta('thumbnailUrl', webpage)
+        uploader_id = self._html_search_regex(
+            r'<a class="img-avatar" href="[^"]+/channels/([^/"]+)" title="Go to [^"]+ page">',
+            webpage, 'uploader id', fatal=False)
+        uploader = self._html_search_regex(
+            r'<a class="img-avatar" href="[^"]+/channels/[^/"]+" title="Go to ([^"]+) page">',
+            webpage, 'uploader', fatal=False)
 
-        self.report_extraction(video_id)
+        categories_html = self._search_regex(
+            r'(?s)><i class="icon icon-tag"></i>\s*Categories / Tags\s*.*?<ul class="list">(.*?)</ul>',
+            webpage, 'categories', fatal=False)
+        categories = None
+        if categories_html:
+            categories = [
+                c.strip() for c in re.findall(
+                    r'(?s)<li><a.*?>(.*?)</a>', categories_html)]
 
-        playlist_json = self._html_search_regex(r'var playerConfigPlaylist\s+=\s+([^;]+)', webpage, u'Playlist')
-        media_id = self._search_regex(r'idMedia:\s*(\d+)', playlist_json, u"Media Id")
-        thumbnail_url = self._search_regex(r'image:\s*"([^"]*)', playlist_json, u'Thumbnail')
-        sources = self._search_regex(r'sources:\s*\[([^\]]*)\]', playlist_json, u'Sources').split(',')
-        title = self._search_regex(r'title:\s*"([^"]*)', playlist_json, u'Title')
+        view_count = str_to_int(self._search_regex(
+            r'<meta itemprop="interactionCount" content="UserPlays:([0-9,]+)">',
+            webpage, 'view count', fatal=False))
+        like_count = str_to_int(self._search_regex(
+            r'<meta itemprop="interactionCount" content="UserLikes:([0-9,]+)">',
+            webpage, 'like count', fatal=False))
+        duration = parse_duration(self._html_search_meta('duration', webpage))
 
-        token_url = "http://tkn.4tube.com/{0}/desktop/{1}".format(media_id, "+".join(sources))
+        media_id = self._search_regex(
+            r'<button[^>]+data-id=(["\'])(?P<id>\d+)\1[^>]+data-quality=', webpage,
+            'media id', default=None, group='id')
+        sources = [
+            quality
+            for _, quality in re.findall(r'<button[^>]+data-quality=(["\'])(.+?)\1', webpage)]
+        if not (media_id and sources):
+            player_js = self._download_webpage(
+                self._search_regex(
+                    r'<script[^>]id=(["\'])playerembed\1[^>]+src=(["\'])(?P<url>.+?)\2',
+                    webpage, 'player JS', group='url'),
+                video_id, 'Downloading player JS')
+            params_js = self._search_regex(
+                r'\$\.ajax\(url,\ opts\);\s*\}\s*\}\)\(([0-9,\[\] ]+)\)',
+                player_js, 'initialization parameters')
+            params = self._parse_json('[%s]' % params_js, video_id)
+            media_id = params[0]
+            sources = ['%s' % p for p in params[2]]
+
+        token_url = 'http://tkn.4tube.com/{0}/desktop/{1}'.format(
+            media_id, '+'.join(sources))
         headers = {
-                b'Content-Type': b'application/x-www-form-urlencoded',
-                b'Origin': b'http://www.4tube.com',
-                }
-        token_req = compat_urllib_request.Request(token_url, b'{}', headers)
+            b'Content-Type': b'application/x-www-form-urlencoded',
+            b'Origin': b'http://www.4tube.com',
+        }
+        token_req = sanitized_Request(token_url, b'{}', headers)
         tokens = self._download_json(token_req, video_id)
-
         formats = [{
             'url': tokens[format]['token'],
             'format_id': format + 'p',
             'resolution': format + 'p',
             'quality': int(format),
-            } for format in sources]
+        } for format in sources]
+        self._sort_formats(formats)
 
-        return [{
+        return {
             'id': video_id,
             'title': title,
             'formats': formats,
-            'thumbnail': thumbnail_url,
+            'categories': categories,
+            'thumbnail': thumbnail,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'timestamp': timestamp,
+            'like_count': like_count,
+            'view_count': view_count,
+            'duration': duration,
             'age_limit': 18,
-            'webpage_url': webpage_url,
-            }]
+        }