debian/control: Add B-D on pandoc and zip. Closes: #828905.

[youtubedl] / youtube_dl / extractor / xvideos.py

diff --git a/youtube_dl/extractor/xvideos.py b/youtube_dl/extractor/xvideos.py
index 90138d7e523a405c20bae8352c6233b5868860f5..1dfe031cab8e9ccf4fa6fb23fbd90d40759ac930 100644 (file)
--- a/youtube_dl/extractor/xvideos.py
+++ b/youtube_dl/extractor/xvideos.py
@@ -1,53 +1,72 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urllib_parse_unquote
 from ..utils import (
-    compat_urllib_parse,
+    clean_html,
+    ExtractorError,
+    determine_ext,
 )
 
 
 class XVideosIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?xvideos\.com/video([0-9]+)(?:.*)'
+    _VALID_URL = r'https?://(?:www\.)?xvideos\.com/video(?P<id>[0-9]+)(?:.*)'
     _TEST = {
-        u'url': u'http://www.xvideos.com/video939581/funny_porns_by_s_-1',
-        u'file': u'939581.flv',
-        u'md5': u'1d0c835822f0a71a7bf011855db929d0',
-        u'info_dict': {
-            u"title": u"Funny Porns By >>>>S<<<<<< -1",
-            u"age_limit": 18,
+        'url': 'http://www.xvideos.com/video4588838/biker_takes_his_girl',
+        'md5': '4b46ae6ea5e6e9086e714d883313c0c9',
+        'info_dict': {
+            'id': '4588838',
+            'ext': 'flv',
+            'title': 'Biker Takes his Girl',
+            'age_limit': 18,
         }
     }
 
     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)
+        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, u'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)
 
-        # Extract title
-        video_title = self._html_search_regex(r'<title>(.*?)\s+-\s+XVID',
-            webpage, u'title')
+        formats = []
 
-        # Extract video thumbnail
-        video_thumbnail = self._search_regex(r'http://(?:img.*?\.)xvideos.com/videos/thumbs/[a-fA-F0-9]+/[a-fA-F0-9]+/[a-fA-F0-9]+/[a-fA-F0-9]+/([a-fA-F0-9.]+jpg)',
-            webpage, u'thumbnail', fatal=False)
+        video_url = compat_urllib_parse_unquote(self._search_regex(
+            r'flv_url=(.+?)&', webpage, 'video URL', default=''))
+        if video_url:
+            formats.append({'url': video_url})
 
-        info = {
+        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))
+
+        self._sort_formats(formats)
+
+        return {
             'id': video_id,
-            'url': video_url,
-            'uploader': None,
-            'upload_date': None,
+            'formats': formats,
             'title': video_title,
-            'ext': 'flv',
             'thumbnail': video_thumbnail,
-            'description': None,
             'age_limit': 18,
         }
-
-        return [info]