]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/redtube.py
d/p/remove-autoupdate-mechanism.patch: Update patch with metadata at the top.
[youtubedl] / youtube_dl / extractor / redtube.py
index d6054d7175fd49a22117dd357bea7905f6e739be..879bcf81d8136ff4bb1f90a9312ff5c7812fdfb8 100644 (file)
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
-from ..utils import ExtractorError
+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]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:(?:www\.)?redtube\.com/|embed\.redtube\.com/\?.*?\bid=)(?P<id>[0-9]+)'
+    _TESTS = [{
         'url': 'http://www.redtube.com/66418',
-        'md5': '7b8c22b5e7098a3e1c09709df1126d2d',
+        'md5': 'fc08071233725f26b8f014dba9590005',
         'info_dict': {
             'id': '66418',
             'ext': 'mp4',
             'title': 'Sucked on a toilet',
+            'upload_date': '20110811',
+            '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(url, video_id)
+        webpage = self._download_webpage(
+            'http://www.redtube.com/%s' % video_id, video_id)
 
         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_url = self._html_search_regex(
-            r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL')
-        video_title = self._html_search_regex(
-            r'<h1 class="videoTitle[^"]*">(.+?)</h1>',
-            webpage, 'title')
-        video_thumbnail = self._og_search_thumbnail(webpage)
+        title = self._html_search_regex(
+            (r'<h(\d)[^>]+class="(?:video_title_text|videoTitle)[^"]*">(?P<title>(?:(?!\1).)+)</h\1>',
+             r'(?:videoTitle|title)\s*:\s*(["\'])(?P<title>(?:(?!\1).)+)\1',),
+            webpage, 'title', group='title',
+            default=None) or self._og_search_title(webpage)
+
+        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)
+
+        thumbnail = self._og_search_thumbnail(webpage)
+        upload_date = unified_strdate(self._search_regex(
+            r'<span[^>]+>ADDED ([^<]+)<',
+            webpage, 'upload date', fatal=False))
+        duration = int_or_none(self._og_search_property(
+            'video:duration', webpage, default=None) or self._search_regex(
+                r'videoDuration\s*:\s*(\d+)', webpage, 'duration', default=None))
+        view_count = str_to_int(self._search_regex(
+            (r'<div[^>]*>Views</div>\s*<div[^>]*>\s*([\d,.]+)',
+             r'<span[^>]*>VIEWS</span>\s*</td>\s*<td>\s*([\d,.]+)'),
+            webpage, 'view count', fatal=False))
 
         # No self-labeling, but they describe themselves as
         # "Home of Videos Porno"
@@ -37,9 +104,12 @@ class RedTubeIE(InfoExtractor):
 
         return {
             'id': video_id,
-            'url': video_url,
             'ext': 'mp4',
-            'title': video_title,
-            'thumbnail': video_thumbnail,
+            'title': title,
+            'thumbnail': thumbnail,
+            'upload_date': upload_date,
+            'duration': duration,
+            'view_count': view_count,
             'age_limit': age_limit,
+            'formats': formats,
         }