[vporn] Fix issues, extract all formats and metadata

youtube_dl/extractor/vporn.py

@@ -1,52 +1,99 @@
-# coding: utf-8
 from __future__ import unicode_literals
 
 import re
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    parse_duration,
+    str_to_int,
+)
+
 
 class VpornIE(InfoExtractor):
-    _VALID_URL = r'http?://(?:www\.)?vporn\.com/[a-z]+/(?P<title_dash>[a-z-]+)/(?P<id>\d+)/?'
+    _VALID_URL = r'https?://(?:www\.)?vporn\.com/[^/]+/(?P<display_id>[^/]+)/(?P<id>\d+)'
     _TEST = {
         'url': 'http://www.vporn.com/masturbation/violet-on-her-th-birthday/497944/',
         'md5': 'facf37c1b86546fa0208058546842c55',
         'info_dict': {
             'id': '497944',
+            'display_id': 'violet-on-her-th-birthday',
             'ext': 'mp4',
-            'title': 'Violet On Her 19th Birthday',
+            'title': 'Violet on her 19th birthday',
             'description': 'Violet dances in front of the camera which is sure to get you horny.',
-            'duration': 393,
             'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'kileyGrope',
+            'categories': ['Masturbation', 'Teen'],
+            'duration': 393,
+            'age_limit': 18,
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        title = self._html_search_regex(
+            r'videoname\s*=\s*\'([^\']+)\'', webpage, 'title').strip()
+        description = self._html_search_regex(
+            r'<div class="description_txt">(.*?)</div>', webpage, 'description', fatal=False)
+        thumbnail = self._html_search_regex(
+            r'flashvars\.imageUrl\s*=\s*"([^"]+)"', webpage, 'description', fatal=False, default=None)
+        if thumbnail:
+            thumbnail = 'http://www.vporn.com' + thumbnail
+
+        uploader = self._html_search_regex(
+            r'(?s)UPLOADED BY.*?<a href="/user/[^"]+">([^<]+)</a>',
+            webpage, 'uploader', fatal=False)
 
-        webpage = self._download_webpage(url, video_id)
+        categories = re.findall(r'<a href="/cat/[^"]+">([^<]+)</a>', webpage)
-        title = self._html_search_regex(r'<title>(.*?) - Vporn Video</title>', webpage, 'title')
-        video_url = self._html_search_regex(r'flashvars.videoUrlMedium  = "(.*?)"', webpage, 'video_url')
-        description = self._html_search_regex(r'<div class="description_txt">(.*?)</div>', webpage, 'description')
-        thumbnail = 'http://www.vporn.com' + self._html_search_regex(r'flashvars.imageUrl = "(.*?)"', webpage, 'description')
 
-        mobj = re.search(r'<span class="f_right">duration (?P<minutes>\d+) min (?P<seconds>\d+) sec </span>', webpage)
+        duration = parse_duration(self._search_regex(
-        duration = int(mobj.group('minutes')) * 60 + int(mobj.group('seconds')) if mobj else None
+            r'duration (\d+ min \d+ sec)', webpage, 'duration', fatal=False))
 
-        mobj = re.search(r'<span>((?P<thousands>\d+),)?(?P<units>\d+) VIEWS</span>', webpage)
+        view_count = str_to_int(self._html_search_regex(
-        try:
+            r'<span>([\d,\.]+) VIEWS</span>', webpage, 'view count', fatal=False))
-            view_count = int(mobj.group('units'))
+        like_count = str_to_int(self._html_search_regex(
-            view_count += int(mobj.group('thousands')) * 1000
+            r'<span id="like" class="n">([\d,\.]+)</span>', webpage, 'like count', fatal=False))
-        except:
+        dislike_count = str_to_int(self._html_search_regex(
-            pass
+            r'<span id="dislike" class="n">([\d,\.]+)</span>', webpage, 'dislike count', fatal=False))
+        comment_count = str_to_int(self._html_search_regex(
+            r'<h4>Comments \(<b>([\d,\.]+)</b>\)</h4>', webpage, 'comment count', fatal=False))
+
+        formats = []
+
+        for video in re.findall(r'flashvars\.videoUrl([^=]+?)\s*=\s*"([^"]+)"', webpage):
+            video_url = video[1]
+            fmt = {
+                'url': video_url,
+                'format_id': video[0],
+            }
+            m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)_(?P<vbr>\d+)k\.mp4$', video_url)
+            if m:
+                fmt.update({
+                    'width': int(m.group('width')),
+                    'height': int(m.group('height')),
+                    'vbr': int(m.group('vbr')),
+                })
+            formats.append(fmt)
+
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'url': video_url,
+            'display_id': display_id,
-            'thumbnail': thumbnail,
             'title': title,
             'description': description,
-            'duration': int_or_none(duration),
+            'thumbnail': thumbnail,
-            'view_count': int_or_none(view_count),
+            'uploader': uploader,
+            'categories': categories,
+            'duration': duration,
+            'view_count': view_count,
+            'like_count': like_count,
+            'dislike_count': dislike_count,
+            'comment_count': comment_count,
+            'age_limit': 18,
+            'formats': formats,
         }