[cspan] Fix extraction (fixes #2291)

The webpage urls have changed.
The title and thumbnail are now extracted from an xml.
This commit is contained in:
Jaime Marquínez Ferrándiz 2014-02-02 18:24:20 +01:00
parent b58e3c8918
commit 009a3408f5

View File

@ -1,49 +1,60 @@
from __future__ import unicode_literals from __future__ import unicode_literals
import json
import re import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
unescapeHTML, unescapeHTML,
find_xpath_attr,
) )
class CSpanIE(InfoExtractor): 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' IE_DESC = 'C-SPAN'
_TEST = { _TEST = {
'url': 'http://www.c-spanvideo.org/program/HolderonV', 'url': 'http://www.c-span.org/video/?313572-1/HolderonV',
'file': '315139.mp4',
'md5': '8e44ce11f0f725527daccc453f553eb0', 'md5': '8e44ce11f0f725527daccc453f553eb0',
'info_dict': { 'info_dict': {
'id': '315139',
'ext': 'mp4',
'title': 'Attorney General Eric Holder on Voting Rights Act Decision', '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', 'skip': 'Regularly fails on travis, for unknown reasons',
} }
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
prog_name = mobj.group('name') page_id = mobj.group('id')
webpage = self._download_webpage(url, prog_name) webpage = self._download_webpage(url, page_id)
video_id = self._search_regex(r'prog(?:ram)?id=(.*?)&', webpage, 'video id') video_id = self._search_regex(r'data-progid=\'(\d+)\'>', webpage, 'video id')
title = self._html_search_regex( description = self._html_search_regex(
r'<!-- title -->\n\s*<h1[^>]*>(.*?)</h1>', webpage, 'title') [
description = self._og_search_description(webpage) # 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 info_url = 'http://c-spanvideo.org/videoLibrary/assets/player/ajax-player.php?os=android&html5=program&id=' + video_id
data_json = self._download_webpage( data = self._download_json(info_url, video_id)
info_url, video_id, 'Downloading video info')
data = json.loads(data_json)
url = unescapeHTML(data['video']['files'][0]['path']['#text']) 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 { return {
'id': video_id, 'id': video_id,
'title': title, 'title': find_string('title'),
'url': url, 'url': url,
'description': description, 'description': description,
'thumbnail': self._og_search_thumbnail(webpage), 'thumbnail': find_string('poster'),
} }