[xvideos] Add video duration

[youtube-dl] / youtube_dl / extractor / xvideos.py

diff --git a/youtube_dl/extractor/xvideos.py b/youtube_dl/extractor/xvideos.py
index 1dfe031cab8e9ccf4fa6fb23fbd90d40759ac930..ce3a05e419dfa773b7bae2cc3035a407f0e48318 100644 (file)
--- a/youtube_dl/extractor/xvideos.py
+++ b/youtube_dl/extractor/xvideos.py
@@ -8,6 +8,7 @@ from ..utils import (
     clean_html,
     ExtractorError,
     determine_ext,
+    parse_duration,
 )
 
 
@@ -15,11 +16,12 @@ class XVideosIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?xvideos\.com/video(?P<id>[0-9]+)(?:.*)'
     _TEST = {
         'url': 'http://www.xvideos.com/video4588838/biker_takes_his_girl',
-        'md5': '4b46ae6ea5e6e9086e714d883313c0c9',
+        'md5': '14cea69fcb84db54293b1e971466c2e1',
         'info_dict': {
             'id': '4588838',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Biker Takes his Girl',
+            'duration': 120,
             'age_limit': 18,
         }
     }
@@ -36,30 +38,32 @@ class XVideosIE(InfoExtractor):
             r'<title>(.*?)\s+-\s+XVID', webpage, 'title')
         video_thumbnail = self._search_regex(
             r'url_bigthumb=(.+?)&amp', webpage, 'thumbnail', fatal=False)
+        video_duration = parse_duration(self._search_regex(
+            r'<span class="duration">.*?(\d[^<]+)', webpage, 'duration', fatal=False))
 
         formats = []
 
         video_url = compat_urllib_parse_unquote(self._search_regex(
             r'flv_url=(.+?)&', webpage, 'video URL', default=''))
         if video_url:
-            formats.append({'url': video_url})
+            formats.append({
+                'url': video_url,
+                'format_id': 'flv',
+            })
 
-        player_args = self._search_regex(
-            r'(?s)new\s+HTML5Player\((.+?)\)', webpage, ' html5 player', default=None)
-        if player_args:
-            for arg in player_args.split(','):
-                format_url = self._search_regex(
-                    r'(["\'])(?P<url>https?://.+?)\1', arg, 'url',
-                    default=None, group='url')
-                if not format_url:
-                    continue
-                ext = determine_ext(format_url)
-                if ext == 'mp4':
-                    formats.append({'url': format_url})
-                elif ext == 'm3u8':
-                    formats.extend(self._extract_m3u8_formats(
-                        format_url, video_id, 'mp4',
-                        entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
+        for kind, _, format_url in re.findall(
+                r'setVideo([^(]+)\((["\'])(http.+?)\2\)', webpage):
+            format_id = kind.lower()
+            if format_id == 'hls':
+                formats.extend(self._extract_m3u8_formats(
+                    format_url, video_id, 'mp4',
+                    entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
+            elif format_id in ('urllow', 'urlhigh'):
+                formats.append({
+                    'url': format_url,
+                    'format_id': '%s-%s' % (determine_ext(format_url, 'mp4'), format_id[3:]),
+                    'quality': -2 if format_id.endswith('low') else None,
+                })
 
         self._sort_formats(formats)
 
@@ -67,6 +71,7 @@ class XVideosIE(InfoExtractor):
             'id': video_id,
             'formats': formats,
             'title': video_title,
+            'duration': video_duration,
             'thumbnail': video_thumbnail,
             'age_limit': 18,
         }