[cspan] Make ‘www’ optional and improve the regex for extracting the id (fixes #2194)

This commit is contained in:
Jaime Marquínez Ferrándiz 2014-01-22 11:06:03 +01:00
parent c39f7013e1
commit 407ae733ab

View file

@ -10,7 +10,7 @@ from ..utils import (
class CSpanIE(InfoExtractor): class CSpanIE(InfoExtractor):
_VALID_URL = r'http://www\.c-spanvideo\.org/program/(.*)' _VALID_URL = r'http://(?:www\.)?c-spanvideo\.org/program/(?P<name>.*)'
IE_DESC = 'C-SPAN' IE_DESC = 'C-SPAN'
_TEST = { _TEST = {
'url': 'http://www.c-spanvideo.org/program/HolderonV', 'url': 'http://www.c-spanvideo.org/program/HolderonV',
@ -24,9 +24,9 @@ class CSpanIE(InfoExtractor):
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(1) prog_name = mobj.group('name')
webpage = self._download_webpage(url, prog_name) webpage = self._download_webpage(url, prog_name)
video_id = self._search_regex(r'programid=(.*?)&', webpage, 'video id') video_id = self._search_regex(r'prog(?:ram)?id=(.*?)&', webpage, 'video id')
title = self._html_search_regex( title = self._html_search_regex(
r'<!-- title -->\n\s*<h1[^>]*>(.*?)</h1>', webpage, 'title') r'<!-- title -->\n\s*<h1[^>]*>(.*?)</h1>', webpage, 'title')