[empflix] Revert to XML parser

Don't rely on the XML being broken (if they fix it, our code wouldn't work anymore).
Instead, use the transform function we already have :)

This partially reverts commit c7bee2a725.
pull/8/head
Philipp Hagemeister 10 years ago
parent a204c85408
commit 8f1ea7cbb6

@ -3,6 +3,7 @@ from __future__ import unicode_literals
import re import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import fix_xml_ampersands
class EmpflixIE(InfoExtractor): class EmpflixIE(InfoExtractor):
@ -35,20 +36,17 @@ class EmpflixIE(InfoExtractor):
r'flashvars\.config = escape\("([^"]+)"', r'flashvars\.config = escape\("([^"]+)"',
webpage, 'flashvars.config') webpage, 'flashvars.config')
# XML is malformed cfg_xml = self._download_xml(
cfg_xml = self._download_webpage( cfg_url, video_id, note='Downloading metadata',
cfg_url, video_id, note='Downloading metadata') transform_source=fix_xml_ampersands)
formats = [ formats = [
{ {
'url': item[1], 'url': item.find('videoLink').text,
'format_id': item[0], 'format_id': item.find('res').text,
} for item in re.findall( } for item in cfg_xml.findall('./quality/item')
r'<item>\s*<res>([^>]+)</res>\s*<videoLink>([^<]+)</videoLink>\s*</item>', cfg_xml)
] ]
thumbnail = cfg_xml.find('./startThumb').text
thumbnail = self._html_search_regex(
r'<startThumb>([^<]+)</startThumb>', cfg_xml, 'thumbnail', fatal=False)
return { return {
'id': video_id, 'id': video_id,

Loading…
Cancel
Save