[orf] Use new extraction method (Fixes #2057)
Philipp Hagemeister
11 years ago
parent
c11a0611d9
commit
5d73273f6f
@ -1,54 +1,98 @@
|
|||||||
# coding: utf-8
|
# coding: utf-8
|
||||||
|
from __future__ import unicode_literals
|
||||||
|
|
||||||
import re
|
|
||||||
import xml.etree.ElementTree
|
|
||||||
import json
|
import json
|
||||||
|
import re
|
||||||
|
|
||||||
from .common import InfoExtractor
|
from .common import InfoExtractor
|
||||||
from ..utils import (
|
from ..utils import (
|
||||||
compat_urlparse,
|
HEADRequest,
|
||||||
ExtractorError,
|
unified_strdate,
|
||||||
find_xpath_attr,
|
|
||||||
)
|
)
|
||||||
|
|
||||||
|
|
||||||
class ORFIE(InfoExtractor):
|
class ORFIE(InfoExtractor):
|
||||||
_VALID_URL = r'https?://tvthek\.orf\.at/(programs/.+?/episodes|topics/.+?)/(?P<id>\d+)'
|
_VALID_URL = r'https?://tvthek\.orf\.at/(?:programs/.+?/episodes|topics/.+?|program/[^/]+)/(?P<id>\d+)'
|
||||||
|
|
||||||
|
_TEST = {
|
||||||
|
'url': 'http://tvthek.orf.at/program/matinee-Was-Sie-schon-immer-ueber-Klassik-wissen-wollten/7317210/Was-Sie-schon-immer-ueber-Klassik-wissen-wollten/7319746/Was-Sie-schon-immer-ueber-Klassik-wissen-wollten/7319747',
|
||||||
|
'file': '7319747.mp4',
|
||||||
|
'md5': 'bd803c5d8c32d3c64a0ea4b4eeddf375',
|
||||||
|
'info_dict': {
|
||||||
|
'title': 'Was Sie schon immer über Klassik wissen wollten',
|
||||||
|
'description': 'md5:0ddf0d5f0060bd53f744edaa5c2e04a4',
|
||||||
|
'duration': 3508,
|
||||||
|
'upload_date': '20140105',
|
||||||
|
},
|
||||||
|
'skip': 'Blocked outside of Austria',
|
||||||
|
}
|
||||||
|
|
||||||
def _real_extract(self, url):
|
def _real_extract(self, url):
|
||||||
mobj = re.match(self._VALID_URL, url)
|
mobj = re.match(self._VALID_URL, url)
|
||||||
playlist_id = mobj.group('id')
|
playlist_id = mobj.group('id')
|
||||||
webpage = self._download_webpage(url, playlist_id)
|
webpage = self._download_webpage(url, playlist_id)
|
||||||
|
|
||||||
flash_xml = self._search_regex('ORF.flashXML = \'(.+?)\'', webpage, u'flash xml')
|
data_json = self._search_regex(
|
||||||
flash_xml = compat_urlparse.parse_qs('xml='+flash_xml)['xml'][0]
|
r'initializeAdworx\((.+?)\);\n', webpage, 'video info')
|
||||||
flash_config = xml.etree.ElementTree.fromstring(flash_xml.encode('utf-8'))
|
all_data = json.loads(data_json)
|
||||||
playlist_json = self._search_regex(r'playlist\': \'(\[.*?\])\'', webpage, u'playlist').replace(r'\"','"')
|
sdata = all_data[0]['values']['segments']
|
||||||
playlist = json.loads(playlist_json)
|
|
||||||
|
def quality_to_int(s):
|
||||||
videos = []
|
m = re.search('([0-9]+)', s)
|
||||||
ns = '{http://tempuri.org/XMLSchema.xsd}'
|
if m is None:
|
||||||
xpath = '%(ns)sPlaylist/%(ns)sItems/%(ns)sItem' % {'ns': ns}
|
return -1
|
||||||
webpage_description = self._og_search_description(webpage)
|
return int(m.group(1))
|
||||||
for (i, (item, info)) in enumerate(zip(flash_config.findall(xpath), playlist), 1):
|
|
||||||
# Get best quality url
|
entries = []
|
||||||
rtmp_url = None
|
for sd in sdata:
|
||||||
for q in ['Q6A', 'Q4A', 'Q1A']:
|
video_id = sd['id']
|
||||||
video_url = find_xpath_attr(item, '%sVideoUrl' % ns, 'quality', q)
|
formats = [{
|
||||||
if video_url is not None:
|
'preference': -10 if fd['delivery'] == 'hls' else None,
|
||||||
rtmp_url = video_url.text
|
'format_id': '%s-%s-%s' % (
|
||||||
break
|
fd['delivery'], fd['quality'], fd['quality_string']),
|
||||||
if rtmp_url is None:
|
'url': fd['src'],
|
||||||
raise ExtractorError(u'Couldn\'t get video url: %s' % info['id'])
|
'protocol': fd['protocol'],
|
||||||
description = self._html_search_regex(
|
'quality': quality_to_int(fd['quality']),
|
||||||
r'id="playlist_entry_%s".*?<p>(.*?)</p>' % i, webpage,
|
} for fd in sd['playlist_item_array']['sources']]
|
||||||
u'description', default=webpage_description, flags=re.DOTALL)
|
|
||||||
videos.append({
|
# Check for geoblocking.
|
||||||
|
# There is a property is_geoprotection, but that's always false
|
||||||
|
geo_str = sd.get('geoprotection_string')
|
||||||
|
if geo_str:
|
||||||
|
try:
|
||||||
|
http_url = next(
|
||||||
|
f['url']
|
||||||
|
for f in formats
|
||||||
|
if re.match(r'^https?://.*\.mp4$', f['url']))
|
||||||
|
except StopIteration:
|
||||||
|
pass
|
||||||
|
else:
|
||||||
|
req = HEADRequest(http_url)
|
||||||
|
response = self._request_webpage(
|
||||||
|
req, video_id,
|
||||||
|
note='Testing for geoblocking',
|
||||||
|
errnote=((
|
||||||
|
'This video seems to be blocked outside of %s. '
|
||||||
|
'You may want to try the streaming-* formats.')
|
||||||
|
% geo_str),
|
||||||
|
fatal=False)
|
||||||
|
|
||||||
|
self._sort_formats(formats)
|
||||||
|
|
||||||
|
upload_date = unified_strdate(sd['created_date'])
|
||||||
|
entries.append({
|
||||||
'_type': 'video',
|
'_type': 'video',
|
||||||
'id': info['id'],
|
'id': video_id,
|
||||||
'title': info['title'],
|
'title': sd['header'],
|
||||||
'url': rtmp_url,
|
'formats': formats,
|
||||||
'ext': 'flv',
|
'description': sd.get('description'),
|
||||||
'description': description,
|
'duration': int(sd['duration_in_seconds']),
|
||||||
})
|
'upload_date': upload_date,
|
||||||
|
'thumbnail': sd.get('image_full_url'),
|
||||||
return videos
|
})
|
||||||
|
|
||||||
|
return {
|
||||||
|
'_type': 'playlist',
|
||||||
|
'entries': entries,
|
||||||
|
'id': playlist_id,
|
||||||
|
}
|
||||||
|
|||||||
Loading…
Reference in New Issue