New upstream version 2019.09.28

[youtubedl] / youtube_dl / extractor / xvideos.py

diff --git a/youtube_dl/extractor/xvideos.py b/youtube_dl/extractor/xvideos.py
index 7e00448246beb9ab9b7c25f33b05e6f4f1bb8283..166bcf44380813e53686724e2ee629a232fcf190 100644 (file)
--- a/youtube_dl/extractor/xvideos.py
+++ b/youtube_dl/extractor/xvideos.py
@@ -3,58 +3,108 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urllib_parse_unquote
 from ..utils import (
-    compat_urllib_parse,
-    ExtractorError,
     clean_html,
+    determine_ext,
+    ExtractorError,
+    int_or_none,
+    parse_duration,
 )
 
 
 class XVideosIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?xvideos\.com/video([0-9]+)(?:.*)'
-    _TEST = {
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:
+                            (?:www\.)?xvideos\.com/video|
+                            flashservice\.xvideos\.com/embedframe/|
+                            static-hw\.xvideos\.com/swf/xv-player\.swf\?.*?\bid_video=
+                        )
+                        (?P<id>[0-9]+)
+                    '''
+    _TESTS = [{
         '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': 108,
             'age_limit': 18,
         }
-    }
+    }, {
+        'url': 'https://flashservice.xvideos.com/embedframe/4588838',
+        'only_matching': True,
+    }, {
+        'url': 'http://static-hw.xvideos.com/swf/xv-player.swf?id_video=4588838',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1)
+        video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
-
-        self.report_extraction(video_id)
+        webpage = self._download_webpage(
+            'https://www.xvideos.com/video%s/' % video_id, video_id)
 
         mobj = re.search(r'<h1 class="inlineError">(.+?)</h1>', webpage)
         if mobj:
             raise ExtractorError('%s said: %s' % (self.IE_NAME, clean_html(mobj.group(1))), expected=True)
 
-        # Extract video URL
-        video_url = compat_urllib_parse.unquote(
-            self._search_regex(r'flv_url=(.+?)&', webpage, 'video URL'))
+        title = self._html_search_regex(
+            (r'<title>(?P<title>.+?)\s+-\s+XVID',
+             r'setVideoTitle\s*\(\s*(["\'])(?P<title>(?:(?!\1).)+)\1'),
+            webpage, 'title', default=None,
+            group='title') or self._og_search_title(webpage)
+
+        thumbnails = []
+        for preference, thumbnail in enumerate(('', '169')):
+            thumbnail_url = self._search_regex(
+                r'setThumbUrl%s\(\s*(["\'])(?P<thumbnail>(?:(?!\1).)+)\1' % thumbnail,
+                webpage, 'thumbnail', default=None, group='thumbnail')
+            if thumbnail_url:
+                thumbnails.append({
+                    'url': thumbnail_url,
+                    'preference': preference,
+                })
+
+        duration = int_or_none(self._og_search_property(
+            'duration', webpage, default=None)) or 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,
+                'format_id': 'flv',
+            })
 
-        # Extract title
-        video_title = self._html_search_regex(
-            r'<title>(.*?)\s+-\s+XVID', webpage, 'title')
+        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,
+                })
 
-        # Extract video thumbnail
-        video_thumbnail = self._search_regex(
-            r'url_bigthumb=(.+?)&amp', webpage, 'thumbnail', fatal=False)
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'url': video_url,
-            'uploader': None,
-            'upload_date': None,
-            'title': video_title,
-            'ext': 'flv',
-            'thumbnail': video_thumbnail,
-            'description': None,
+            'formats': formats,
+            'title': title,
+            'duration': duration,
+            'thumbnails': thumbnails,
             'age_limit': 18,
         }