[heise] Fix extraction

Now they use an XML format instead of JSON.
pull/4046/head
Philipp Hagemeister 2014-10-27 01:33:49 +01:00
parent 4a6c94288a
commit ecfe623422
1 changed files with 21 additions and 27 deletions

View File

@ -4,6 +4,7 @@ from __future__ import unicode_literals
from .common import InfoExtractor
from ..utils import (
get_meta_content,
int_or_none,
parse_iso8601,
)
@ -28,20 +29,26 @@ class HeiseIE(InfoExtractor):
'timestamp': 1411812600,
'upload_date': '20140927',
'description': 'In uplink-Episode 3.3 geht es darum, wie man sich von Cloud-Anbietern emanzipieren kann, worauf man beim Kauf einer Tastatur achten sollte und was Smartphones über uns verraten.',
'thumbnail': 're:https?://.*\.jpg$',
}
}
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
json_url = self._search_regex(
r'json_url:\s*"([^"]+)"', webpage, 'json URL')
config = self._download_json(json_url, video_id)
container_id = self._search_regex(
r'<div class="videoplayerjw".*?data-container="([0-9]+)"',
webpage, 'container ID')
sequenz_id = self._search_regex(
r'<div class="videoplayerjw".*?data-sequenz="([0-9]+)"',
webpage, 'sequenz ID')
data_url = 'http://www.heise.de/videout/feed?container=%s&sequenz=%s' % (container_id, sequenz_id)
doc = self._download_xml(data_url, video_id)
info = {
'id': video_id,
'thumbnail': config.get('poster'),
'thumbnail': self._og_search_thumbnail(webpage),
'timestamp': parse_iso8601(get_meta_content('date', webpage)),
'description': self._og_search_description(webpage),
}
@ -49,32 +56,19 @@ class HeiseIE(InfoExtractor):
title = get_meta_content('fulltitle', webpage)
if title:
info['title'] = title
elif config.get('title'):
info['title'] = config['title']
else:
info['title'] = self._og_search_title(webpage)
formats = []
for t, rs in config['formats'].items():
if not rs or not hasattr(rs, 'items'):
self._downloader.report_warning(
'formats: {0}: no resolutions'.format(t))
continue
for height_str, obj in rs.items():
format_id = '{0}_{1}'.format(t, height_str)
if not obj or not obj.get('url'):
self._downloader.report_warning(
'formats: {0}: no url'.format(format_id))
continue
formats.append({
'url': obj['url'],
'format_id': format_id,
'height': self._int(height_str, 'height'),
})
for source_node in doc.findall('.//{http://rss.jwpcdn.com/}source'):
label = source_node.attrib['label']
height = int_or_none(self._search_regex(
r'^(.*?_)?([0-9]+)p$', label, 'height', default=None))
formats.append({
'url': source_node.attrib['file'],
'format_note': label,
'height': height,
})
self._sort_formats(formats)
info['formats'] = formats