From 009a3408f5195442c4bd0471a91a0b38d283382d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
 <jaime.marquinez.ferrandiz@gmail.com>
Date: Sun, 2 Feb 2014 18:24:20 +0100
Subject: [PATCH] [cspan] Fix extraction (fixes #2291)

The webpage urls have changed.
The title and thumbnail are now extracted from an xml.
---
 youtube_dl/extractor/cspan.py | 43 ++++++++++++++++++++++-------------
 1 file changed, 27 insertions(+), 16 deletions(-)

diff --git a/youtube_dl/extractor/cspan.py b/youtube_dl/extractor/cspan.py
index e54009622a..d65046f588 100644
--- a/youtube_dl/extractor/cspan.py
+++ b/youtube_dl/extractor/cspan.py
@@ -1,49 +1,60 @@
 from __future__ import unicode_literals
 
-import json
 import re
 
 from .common import InfoExtractor
 from ..utils import (
     unescapeHTML,
+    find_xpath_attr,
 )
 
 
 class CSpanIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?c-spanvideo\.org/program/(?P<name>.*)'
+    _VALID_URL = r'http://(?:www\.)?c-span\.org/video/\?(?P<id>\d+)'
     IE_DESC = 'C-SPAN'
     _TEST = {
-        'url': 'http://www.c-spanvideo.org/program/HolderonV',
-        'file': '315139.mp4',
+        'url': 'http://www.c-span.org/video/?313572-1/HolderonV',
         'md5': '8e44ce11f0f725527daccc453f553eb0',
         'info_dict': {
+            'id': '315139',
+            'ext': 'mp4',
             'title': 'Attorney General Eric Holder on Voting Rights Act Decision',
-            'description': 'Attorney General Eric Holder spoke to reporters following the Supreme Court decision in [Shelby County v. Holder] in which the court ruled that the preclearance provisions of the Voting Rights Act could not be enforced until Congress established new guidelines for review.',
+            'description': 'Attorney General Eric Holder spoke to reporters following the Supreme Court decision in Shelby County v. Holder in which the court ruled that the preclearance provisions of the Voting Rights Act could not be enforced until Congress established new guidelines for review.',
         },
         'skip': 'Regularly fails on travis, for unknown reasons',
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        prog_name = mobj.group('name')
-        webpage = self._download_webpage(url, prog_name)
-        video_id = self._search_regex(r'prog(?:ram)?id=(.*?)&', webpage, 'video id')
+        page_id = mobj.group('id')
+        webpage = self._download_webpage(url, page_id)
+        video_id = self._search_regex(r'data-progid=\'(\d+)\'>', webpage, 'video id')
 
-        title = self._html_search_regex(
-            r'<!-- title -->\n\s*<h1[^>]*>(.*?)</h1>', webpage, 'title')
-        description = self._og_search_description(webpage)
+        description = self._html_search_regex(
+            [
+                # The full description
+                r'<div class=\'expandable\'>(.*?)<a href=\'#\'',
+                # If the description is small enough the other div is not
+                # present, otherwise this is a stripped version
+                r'<p class=\'initial\'>(.*?)</p>'
+            ],
+            webpage, 'description', flags=re.DOTALL)
 
         info_url = 'http://c-spanvideo.org/videoLibrary/assets/player/ajax-player.php?os=android&html5=program&id=' + video_id
-        data_json = self._download_webpage(
-            info_url, video_id, 'Downloading video info')
-        data = json.loads(data_json)
+        data = self._download_json(info_url, video_id)
 
         url = unescapeHTML(data['video']['files'][0]['path']['#text'])
 
+        doc = self._download_xml('http://www.c-span.org/common/services/flashXml.php?programid=' + video_id,
+            video_id)
+
+        def find_string(s):
+            return find_xpath_attr(doc, './/string', 'name', s).text
+
         return {
             'id': video_id,
-            'title': title,
+            'title': find_string('title'),
             'url': url,
             'description': description,
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'thumbnail': find_string('poster'),
         }