debian/watch: Add back pgpsigurlmangle.

[youtubedl] / youtube_dl / extractor / firstpost.py

diff --git a/youtube_dl/extractor/firstpost.py b/youtube_dl/extractor/firstpost.py
index 7e3d1afd215bfc8ef39c6751e116def05bf858ed..e8936cb2468f78bd2c1b59008a13e9411204380e 100644 (file)
--- a/youtube_dl/extractor/firstpost.py
+++ b/youtube_dl/extractor/firstpost.py
@@ -1,13 +1,10 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 
 
 class FirstpostIE(InfoExtractor):
-    IE_NAME = 'Firstpost.com'
-    _VALID_URL = r'http://(?:www\.)?firstpost\.com/[^/]+/.*-(?P<id>[0-9]+)\.html'
+    _VALID_URL = r'https?://(?:www\.)?firstpost\.com/[^/]+/.*-(?P<id>[0-9]+)\.html'
 
     _TEST = {
         'url': 'http://www.firstpost.com/india/india-to-launch-indigenous-aircraft-carrier-monday-1025403.html',
@@ -16,23 +13,38 @@ class FirstpostIE(InfoExtractor):
             'id': '1025403',
             'ext': 'mp4',
             'title': 'India to launch indigenous aircraft carrier INS Vikrant today',
-            'description': 'Its flight deck is over twice the size of a football field, its power unit can light up the entire Kochi city and the cabling is enough to cover the distance between here to Delhi.',
+            'description': 'md5:feef3041cb09724e0bdc02843348f5f4',
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
+        page = self._download_webpage(url, video_id)
+
+        title = self._html_search_meta('twitter:title', page, 'title', fatal=True)
+        description = self._html_search_meta('twitter:description', page, 'title')
+
+        data = self._download_xml(
+            'http://www.firstpost.com/getvideoxml-%s.xml' % video_id, video_id,
+            'Downloading video XML')
+
+        item = data.find('./playlist/item')
+        thumbnail = item.find('./image').text
 
-        webpage = self._download_webpage(url, video_id)
-        video_url = self._html_search_regex(
-            r'<div.*?name="div_video".*?flashvars="([^"]+)">',
-            webpage, 'video URL')
+        formats = [
+            {
+                'url': details.find('./file').text,
+                'format_id': details.find('./label').text.strip(),
+                'width': int(details.find('./width').text.strip()),
+                'height': int(details.find('./height').text.strip()),
+            } for details in item.findall('./source/file_details') if details.find('./file').text
+        ]
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'url': video_url,
-            'title': self._og_search_title(webpage),
-            'description': self._og_search_description(webpage),
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'formats': formats,
         }