-import operator
+# coding: utf-8
+
import re
from .common import InfoExtractor
from ..utils import (
- parse_xml_doc,
+ int_or_none,
unified_strdate,
)
class ZDFIE(InfoExtractor):
- _VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek(?P<hash>#)?\/(.*beitrag\/video\/)(?P<video_id>[^/\?]+)(?:\?.*)?'
+ _VALID_URL = r'^https?://www\.zdf\.de/ZDFmediathek(?P<hash>#)?/(.*beitrag/(?:video/)?)(?P<video_id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'
+
+ _TEST = {
+ u"url": u"http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt",
+ u"file": u"2037704.webm",
+ u"info_dict": {
+ u"upload_date": u"20131127",
+ u"description": u"Union und SPD haben sich auf einen Koalitionsvertrag geeinigt. Aber was bedeutet das für die Bürger? Sehen Sie hierzu das ZDFspezial \"Ende des Machtpokers - Große Koalition für Deutschland\".",
+ u"uploader": u"spezial",
+ u"title": u"ZDFspezial - Ende des Machtpokers"
+ },
+ u"skip": u"Videos on ZDF.de are depublicised in short order",
+ }
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('video_id')
xml_url = u'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
- info_xml = self._download_webpage(
- xml_url, video_id, note=u'Downloading video info')
- doc = parse_xml_doc(info_xml)
+ doc = self._download_xml(
+ xml_url, video_id,
+ note=u'Downloading video info',
+ errnote=u'Failed to download video info')
title = doc.find('.//information/title').text
description = doc.find('.//information/detail').text
(?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
''', format_id)
- PROTO_ORDER = ['http', 'rtmp', 'rtsp']
- try:
- proto_pref = -PROTO_ORDER.index(format_m.group('proto'))
- except ValueError:
- proto_pref = 999
+ ext = format_m.group('container')
+ proto = format_m.group('proto').lower()
quality = fnode.find('./quality').text
- QUALITY_ORDER = ['veryhigh', '300', 'high', 'med', 'low']
- try:
- quality_pref = -QUALITY_ORDER.index(quality)
- except ValueError:
- quality_pref = 999
-
abr = int(fnode.find('./audioBitrate').text) // 1000
vbr = int(fnode.find('./videoBitrate').text) // 1000
- pref = (is_available, proto_pref, quality_pref, vbr, abr)
+
+ format_note = u''
+ if not format_note:
+ format_note = None
return {
- 'format_id': format_id,
+ 'format_id': format_id + u'-' + quality,
'url': video_url,
- 'ext': format_m.group('container'),
+ 'ext': ext,
'acodec': format_m.group('acodec'),
'vcodec': format_m.group('vcodec'),
'abr': abr,
'vbr': vbr,
- 'width': int(fnode.find('./width').text),
- 'height': int(fnode.find('./height').text),
- 'quality_name': quality,
- 'filesize': int(fnode.find('./filesize').text),
- 'format_note': None if is_available else u'(unavailable)',
- '_pref': pref,
+ 'width': int_or_none(fnode.find('./width').text),
+ 'height': int_or_none(fnode.find('./height').text),
+ 'filesize': int_or_none(fnode.find('./filesize').text),
+ 'format_note': format_note,
+ 'protocol': proto,
+ '_available': is_available,
}
format_nodes = doc.findall('.//formitaeten/formitaet')
- formats = sorted(map(xml_to_format, format_nodes),
- key=operator.itemgetter('_pref'))
+ formats = list(filter(
+ lambda f: f['_available'],
+ map(xml_to_format, format_nodes)))
+
+ self._sort_formats(formats)
return {
'id': video_id,