Merge branch 'master' of https://github.com/DarkstaIkers/youtube-dl into DarkstaIkers...

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

diff --git a/youtube_dl/extractor/xvideos.py b/youtube_dl/extractor/xvideos.py
index 2a45dc574263f7e651020e591fcc40bdf987367d..30825daae956e12f450e32bc66aea737a5c293e0 100644 (file)
--- a/youtube_dl/extractor/xvideos.py
+++ b/youtube_dl/extractor/xvideos.py
@@ -3,12 +3,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_parse,
-)
+from ..compat import compat_urllib_parse_unquote
 from ..utils import (
     clean_html,
     ExtractorError,
+    determine_ext,
 )
 
 
@@ -16,10 +15,10 @@ 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',
             'age_limit': 18,
         }
@@ -33,18 +32,41 @@ class XVideosIE(InfoExtractor):
         if mobj:
             raise ExtractorError('%s said: %s' % (self.IE_NAME, clean_html(mobj.group(1))), expected=True)
 
-        video_url = compat_urllib_parse.unquote(
-            self._search_regex(r'flv_url=(.+?)&', webpage, 'video URL'))
         video_title = self._html_search_regex(
             r'<title>(.*?)\s+-\s+XVID', webpage, 'title')
         video_thumbnail = self._search_regex(
             r'url_bigthumb=(.+?)&amp', webpage, 'thumbnail', 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,
+                'format_id': 'flv',
+            })
+
+        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)
+
         return {
             'id': video_id,
-            'url': video_url,
+            'formats': formats,
             'title': video_title,
-            'ext': 'flv',
             'thumbnail': video_thumbnail,
             'age_limit': 18,
         }