Annotate changelog with bug being closed.

[youtubedl] / youtube_dl / extractor / redtube.py

diff --git a/youtube_dl/extractor/redtube.py b/youtube_dl/extractor/redtube.py
index ebc4e23261bfe029e6e30150c2ca3b264d24ab81..f70a75256c638f4a3ce9cda3b9577176e49f3cca 100644 (file)
--- a/youtube_dl/extractor/redtube.py
+++ b/youtube_dl/extractor/redtube.py
@@ -1,29 +1,112 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    str_to_int,
+    unified_strdate,
+)
 
 
 class RedTubeIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?redtube\.com/(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:(?:www\.)?redtube\.com/|embed\.redtube\.com/\?.*?\bid=)(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'http://www.redtube.com/66418',
+        'md5': '7b8c22b5e7098a3e1c09709df1126d2d',
+        'info_dict': {
+            'id': '66418',
+            'ext': 'mp4',
+            'title': 'Sucked on a toilet',
+            'upload_date': '20120831',
+            'duration': 596,
+            'view_count': int,
+            'age_limit': 18,
+        }
+    }, {
+        'url': 'http://embed.redtube.com/?bgcolor=000000&id=1443286',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//embed\.redtube\.com/\?.*?\bid=\d+)',
+            webpage)
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(
+            'http://www.redtube.com/%s' % video_id, video_id)
 
-    def _real_extract(self,url):
-        mobj = re.match(self._VALID_URL, url)
+        if any(s in webpage for s in ['video-deleted-info', '>This video has been removed']):
+            raise ExtractorError('Video %s has been removed' % video_id, expected=True)
 
-        video_id = mobj.group('id')
-        video_extension = 'mp4'        
-        webpage = self._download_webpage(url, video_id)
+        title = self._html_search_regex(
+            (r'<h1 class="videoTitle[^"]*">(?P<title>.+?)</h1>',
+             r'videoTitle\s*:\s*(["\'])(?P<title>)\1'),
+            webpage, 'title', group='title')
 
-        self.report_extraction(video_id)
+        formats = []
+        sources = self._parse_json(
+            self._search_regex(
+                r'sources\s*:\s*({.+?})', webpage, 'source', default='{}'),
+            video_id, fatal=False)
+        if sources and isinstance(sources, dict):
+            for format_id, format_url in sources.items():
+                if format_url:
+                    formats.append({
+                        'url': format_url,
+                        'format_id': format_id,
+                        'height': int_or_none(format_id),
+                    })
+        medias = self._parse_json(
+            self._search_regex(
+                r'mediaDefinition\s*:\s*(\[.+?\])', webpage,
+                'media definitions', default='{}'),
+            video_id, fatal=False)
+        if medias and isinstance(medias, list):
+            for media in medias:
+                format_url = media.get('videoUrl')
+                if not format_url or not isinstance(format_url, compat_str):
+                    continue
+                format_id = media.get('quality')
+                formats.append({
+                    'url': format_url,
+                    'format_id': format_id,
+                    'height': int_or_none(format_id),
+                })
+        if not formats:
+            video_url = self._html_search_regex(
+                r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL')
+            formats.append({'url': video_url})
+        self._sort_formats(formats)
 
-        video_url = self._html_search_regex(r'<source src="(.+?)" type="video/mp4">',
-            webpage, u'video URL')
+        thumbnail = self._og_search_thumbnail(webpage)
+        upload_date = unified_strdate(self._search_regex(
+            r'<span[^>]+class="added-time"[^>]*>ADDED ([^<]+)<',
+            webpage, 'upload date', fatal=False))
+        duration = int_or_none(self._search_regex(
+            r'videoDuration\s*:\s*(\d+)', webpage, 'duration', default=None))
+        view_count = str_to_int(self._search_regex(
+            r'<span[^>]*>VIEWS</span></td>\s*<td>([\d,.]+)',
+            webpage, 'view count', fatal=False))
 
-        video_title = self._html_search_regex('<h1 class="videoTitle slidePanelMovable">(.+?)</h1>',
-            webpage, u'title')
+        # No self-labeling, but they describe themselves as
+        # "Home of Videos Porno"
+        age_limit = 18
 
-        return [{
-            'id':       video_id,
-            'url':      video_url,
-            'ext':      video_extension,
-            'title':    video_title,
-        }]
+        return {
+            'id': video_id,
+            'ext': 'mp4',
+            'title': title,
+            'thumbnail': thumbnail,
+            'upload_date': upload_date,
+            'duration': duration,
+            'view_count': view_count,
+            'age_limit': age_limit,
+            'formats': formats,
+        }