]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/spankwire.py
debian/control: Prefer ffmpeg to libav-tools in Recommends.
[youtubedl] / youtube_dl / extractor / spankwire.py
index 94602e89e56549243ed38ecb107ef842cd8ebd46..5fa6faf18b738aa32e384972bf65ad56188ad9b4 100644 (file)
@@ -3,12 +3,14 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
+    compat_urllib_parse_unquote,
     compat_urllib_parse_urlparse,
     compat_urllib_request,
-    compat_urllib_parse,
-    unified_strdate,
+)
+from ..utils import (
     str_to_int,
+    unified_strdate,
 )
 from ..aes import aes_decrypt_text
 
@@ -25,7 +27,7 @@ class SpankwireIE(InfoExtractor):
             'description': 'Crazy Bitch X rated music video.',
             'uploader': 'oreusz',
             'uploader_id': '124697',
-            'upload_date': '20070508',
+            'upload_date': '20070507',
             'age_limit': 18,
         }
     }
@@ -42,7 +44,7 @@ class SpankwireIE(InfoExtractor):
         title = self._html_search_regex(
             r'<h1>([^<]+)', webpage, 'title')
         description = self._html_search_regex(
-            r'<div\s+id="descriptionContent">([^<]+)<',
+            r'(?s)<div\s+id="descriptionContent">(.+?)</div>',
             webpage, 'description', fatal=False)
         thumbnail = self._html_search_regex(
             r'playerData\.screenShot\s*=\s*["\']([^"\']+)["\']',
@@ -62,14 +64,14 @@ class SpankwireIE(InfoExtractor):
             r'<div id="viewsCounter"><span>([\d,\.]+)</span> views</div>',
             webpage, 'view count', fatal=False))
         comment_count = str_to_int(self._html_search_regex(
-            r'Comments<span[^>]+>\s*\(([\d,\.]+)\)</span>',
+            r'<span\s+id="spCommentCount"[^>]*>([\d,\.]+)</span>',
             webpage, 'comment count', fatal=False))
 
         video_urls = list(map(
-            compat_urllib_parse.unquote,
-            re.findall(r'playerData\.cdnPath[0-9]{3,}\s*=\s*["\']([^"\']+)["\']', webpage)))
+            compat_urllib_parse_unquote,
+            re.findall(r'playerData\.cdnPath[0-9]{3,}\s*=\s*(?:encodeURIComponent\()?["\']([^"\']+)["\']', webpage)))
         if webpage.find('flashvars\.encrypted = "true"') != -1:
-            password = self._html_search_regex(
+            password = self._search_regex(
                 r'flashvars\.video_title = "([^"]+)',
                 webpage, 'password').replace('+', ' ')
             video_urls = list(map(