[sexykarma] Improve and simplify

youtube_dl/extractor/sexykarma.py

@@ -1,77 +1,106 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
 import re
-import datetime
+
+from .common import InfoExtractor
+from ..utils import (
+    unified_strdate,
+    parse_duration,
+    int_or_none,
+)
+
 
 class SexyKarmaIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?sexykarma\.com/gonewild/video/.+\-(?P<id>[a-zA-Z0-9\-]+)(.html)'
+    _VALID_URL = r'https?://(?:www\.)?sexykarma\.com/gonewild/video/(?P<display_id>[^/]+)-(?P<id>[a-zA-Z0-9]+)\.html'
     _TESTS = [{
         'url': 'http://www.sexykarma.com/gonewild/video/taking-a-quick-pee-yHI70cOyIHt.html',
         'md5': 'b9798e7d1ef1765116a8f516c8091dbd',
         'info_dict': {
             'id': 'yHI70cOyIHt',
+            'display_id': 'taking-a-quick-pee',
             'ext': 'mp4',
             'title': 'Taking a quick pee.',
-            'uploader_id': 'wildginger7',
+            'description': '',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'duration': int,
+            'uploader': 'wildginger7',
-            'view_count': int,
             'upload_date': '20141007',
+            'duration': 81,
+            'view_count': int,
+            'comment_count': int,
+            'categories': list,
         }
     }, {
         'url': 'http://www.sexykarma.com/gonewild/video/pot-pixie-tribute-8Id6EZPbuHf.html',
         'md5': 'dd216c68d29b49b12842b9babe762a5d',
         'info_dict': {
             'id': '8Id6EZPbuHf',
+            'display_id': 'pot-pixie-tribute',
             'ext': 'mp4',
             'title': 'pot_pixie tribute',
-            'uploader_id': 'banffite',
+            'description': 'tribute',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'duration': int,
+            'uploader': 'banffite',
-            'view_count': int,
             'upload_date': '20141013',
+            'duration': 16,
+            'view_count': int,
+            'comment_count': int,
+            'categories': list,
         }
     }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
 
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(url, display_id)
-              
-        title = self._html_search_regex(r'<h2 class="he2"><span>(.*?)</span>', webpage, 'title')
-        uploader_id = self._html_search_regex(r'class="aupa">\n*(.*?)</a>', webpage, 'uploader')
-        url = self._html_search_regex(r'<p><a href="(.*?)" ?\n*target="_blank"><font color', webpage, 'url')
-        thumbnail = self._html_search_regex(r'<div id="player" style="z-index:1;"> <span id="edge"></span> <span id="container"><img[\n ]*src="(.+?)"', webpage, 'thumbnail')
-        
-        str_duration = self._html_search_regex(r'<tr>[\n\s]*<td>Time: </td>[\n\s]*<td align="right"><span>(.+)\n*', webpage, 'duration')
-        duration = self._to_seconds(str_duration)
 
-        str_views = self._html_search_regex(r'<tr>[\n\s]*<td>Views: </td>[\n\s]*<td align="right"><span>(.+)</span>', webpage, 'view_count')
+        video_url = self._html_search_regex(
-        view_count = int(str_views)
+            r'<p>Save this video to your computer: </p><p><a href="([^"]+)"',
-        # print view_count
+            webpage, 'url')
 
-        date = self._html_search_regex(r'class="aup">Added: <strong>(.*?)</strong>', webpage, 'date')
+        title = self._html_search_regex(
-        d = datetime.datetime.strptime(date, '%B %d, %Y')
+            r'<h2 class="he2"><span>(.*?)</span>',
-        upload_date = d.strftime('%Y%m%d')
+            webpage, 'title')
+        description = self._html_search_meta(
+            'description', webpage, 'description', fatal=False, default='')
+        thumbnail = self._html_search_regex(
+            r'<span id="container"><img\s+src="([^"]+)"',
+            webpage, 'thumbnail', fatal=False)
 
-        categories = re.findall(r'http://www.sexykarma.com/gonewild/search/video/(?:.+?)"><span>(.*?)</span>', webpage)
+        uploader = self._html_search_regex(
+            r'class="aupa">\s*(.*?)</a>',
+            webpage, 'uploader')
+        upload_date = unified_strdate(self._html_search_regex(
+            r'Added: <strong>(.+?)</strong>', webpage, 'upload date', fatal=False))
+
+        duration = parse_duration(self._search_regex(
+            r'<td>Time:\s*</td>\s*<td align="right"><span>\s*(.+?)\s*</span>',
+            webpage, 'duration', fatal=False))
+
+        view_count = int_or_none(self._search_regex(
+            r'<td>Views:\s*</td>\s*<td align="right"><span>\s*(\d+)\s*</span>',
+            webpage, 'view count', fatal=False))
+        comment_count = int_or_none(self._search_regex(
+            r'<td>Comments:\s*</td>\s*<td align="right"><span>\s*(\d+)\s*</span>',
+            webpage, 'comment count', fatal=False))
+
+        categories = self._html_search_meta(
+            'keywords', webpage, 'categories',
+            fatal=False, default='').split(',')
 
         return {
             'id': video_id,
+            'display_id': display_id,
+            'url': video_url,
             'title': title,
-            'uploader_id': uploader_id,
+            'description': description,
-            'url': url,
             'thumbnail': thumbnail,
+            'uploader': uploader,
+            'upload_date': upload_date,
             'duration': duration,
             'view_count': view_count,
-            'upload_date': upload_date,
+            'comment_count': comment_count,
             'categories': categories,
         }
-
-    def _to_seconds(self, timestr):
-        seconds= 0
-        for part in timestr.split(':'):
-            seconds= seconds*60 + int(part)
-        return seconds