[tnaflix] Improve and make generic

youtube_dl/extractor/tnaflix.py

@@ -5,11 +5,17 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     parse_duration,
-    str_to_int,
+    fix_xml_ampersands,
 )
 
+
 class TNAFlixIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?tnaflix\.com/(?P<cat_id>[\w-]+)/(?P<display_id>[\w-]+)/video(?P<id>\d+)'
+
+    _TITLE_REGEX = None
+    _DESCRIPTION_REGEX = r'<h3 itemprop="description">([^<]+)</h3>'
+    _CONFIG_REGEX = r'flashvars\.config\s*=\s*escape\("([^"]+)"'
+
     _TEST = {
         'url': 'http://www.tnaflix.com/porn-stars/Carmella-Decesare-striptease/video553878',
         'md5': 'ecf3498417d09216374fc5907f9c6ec0',
@@ -18,8 +24,9 @@ class TNAFlixIE(InfoExtractor):
             'display_id': 'Carmella-Decesare-striptease',
             'ext': 'mp4',
             'title': 'Carmella Decesare - striptease',
+            'description': '',
             'thumbnail': 're:https?://.*\.jpg$',
-            #'duration': 84,
+            'duration': 91,
             'age_limit': 18,
         }
     }
@@ -31,14 +38,30 @@ class TNAFlixIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
-        redir_url = self._html_search_regex(
+        title = self._html_search_regex(
-            r'flashvars.config = escape\("([^"]+)"', webpage, 'redirection URL')
+            self._TITLE_REGEX, webpage, 'title') if self._TITLE_REGEX else self._og_search_title(webpage)
-        redirection_webpage = self._download_webpage(redir_url, display_id)
+        description = self._html_search_regex(
-        sources = self._search_regex(
+            self._DESCRIPTION_REGEX, webpage, 'description', fatal=False, default='')
-            r'<quality>(.+)</quality>', redirection_webpage, 'sources', flags=re.MULTILINE|re.DOTALL)
+
-        
+        age_limit = self._rta_search(webpage)
+
+        duration = self._html_search_meta('duration', webpage, 'duration', default=None)
+        if duration:
+            duration = parse_duration(duration[1:])
+
+        cfg_url = self._html_search_regex(
+            self._CONFIG_REGEX, webpage, 'flashvars.config')
+
+        cfg_xml = self._download_xml(
+            cfg_url, display_id, note='Downloading metadata',
+            transform_source=fix_xml_ampersands)
+
+        thumbnail = cfg_xml.find('./startThumb').text
+
         formats = []
-        for format_id, video_url in re.findall(r'<res>([^<]+)</res>\s*<videoLink>([^<]+)</videoLink>', sources, flags=re.MULTILINE|re.DOTALL):
+        for item in cfg_xml.findall('./quality/item'):
+            video_url = re.sub('speed=\d+', 'speed=', item.find('videoLink').text)
+            format_id = item.find('res').text
             fmt = {
                 'url': video_url,
                 'format_id': format_id,
@@ -49,20 +72,13 @@ class TNAFlixIE(InfoExtractor):
             formats.append(fmt)
         self._sort_formats(formats)
         
-        title = self._og_search_title(webpage)
-        
-        #duration = self._html_search_regex(r'<meta itemprop="duration" content="T(\d+)M(\d+)S"', webpage, 'duration')
-
-        thumbnail = self._html_search_regex(
-            r'<meta\s+itemprop="thumbnailUrl"\s+content="([^"]+)"',
-            webpage, 'thumbnail', fatal=False)
-
         return {
             'id': video_id,
             'display_id': display_id,
-            'url': video_url,
             'title': title,
+            'description': description,
             'thumbnail': thumbnail,
-            #'duration': duration,
+            'duration': duration,
-            'age_limit': self._rta_search(webpage),
+            'age_limit': age_limit,
+            'formats': formats,
         }