Initial extractor for phoenix.de

This commit is contained in:
FliegendeWurst 2019-09-16 22:53:49 +02:00
parent c15c8e25af
commit 813e0fcdfb
No known key found for this signature in database
GPG Key ID: CA38E82B54B32A88
1 changed files with 31 additions and 150 deletions

View File

@ -1,171 +1,52 @@
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..utils import (
int_or_none,
unified_strdate,
xpath_text,
determine_ext,
ExtractorError,
)
from ..utils import ExtractorError
class PhoenixIE(InfoExtractor):
IE_NAME = 'phoenix.de'
_VALID_URL = r'''(?x)https?://(?:www\.)?phoenix\.de/content/
(?:
phoenix/die_sendungen/(?:[^/]+/)?
)?
(?P<id>[0-9]+)'''
_VALID_URL = r'''https?://(?:www\.)?phoenix.de/\D+(?P<id>\d+)\.html'''
_TESTS = [
{
'url': 'http://www.phoenix.de/content/884301',
'md5': 'ed249f045256150c92e72dbb70eadec6',
'url': 'https://www.phoenix.de/sendungen/dokumentationen/unsere-welt-in-zukunft---stadt-a-1283620.html',
'md5': '5e765e838aa3531c745a4f5b249ee3e3',
'info_dict': {
'id': '884301',
'id': '0OB4HFc43Ns',
'ext': 'mp4',
'title': 'Michael Krons mit Hans-Werner Sinn',
'description': 'Im Dialog - Sa. 25.10.14, 00.00 - 00.35 Uhr',
'upload_date': '20141025',
'uploader': 'Im Dialog',
'title': 'Unsere Welt in Zukunft - Stadt',
'description': 'md5:9bfb6fd498814538f953b2dcad7ce044',
'upload_date': '20190912',
'uploader': 'phoenix',
'uploader_id': 'phoenix',
}
},
{
'url': 'http://www.phoenix.de/content/phoenix/die_sendungen/869815',
'only_matching': True,
},
{
'url': 'http://www.phoenix.de/content/phoenix/die_sendungen/diskussionen/928234',
'url': 'https://www.phoenix.de/drohnenangriffe-in-saudi-arabien-a-1286995.html?ref=aktuelles',
'only_matching': True,
},
# an older page: https://www.phoenix.de/sendungen/gespraeche/phoenix-persoenlich/im-dialog-a-177727.html
# seems to not have an embedded video, even though it's uploaded on youtube: https://www.youtube.com/watch?v=4GxnoUHvOkM
]
def extract_from_xml_url(self, video_id, xml_url):
doc = self._download_xml(
xml_url, video_id,
note='Downloading video info',
errnote='Failed to download video info')
status_code = xpath_text(doc, './status/statuscode')
if status_code and status_code != 'ok':
if status_code == 'notVisibleAnymore':
message = 'Video %s is not available' % video_id
else:
message = '%s returned error: %s' % (self.IE_NAME, status_code)
raise ExtractorError(message, expected=True)
title = xpath_text(doc, './/information/title', 'title', True)
urls = []
formats = []
for fnode in doc.findall('.//formitaeten/formitaet'):
video_url = xpath_text(fnode, 'url')
if not video_url or video_url in urls:
continue
urls.append(video_url)
is_available = 'http://www.metafilegenerator' not in video_url
geoloced = 'static_geoloced_online' in video_url
if not is_available or geoloced:
continue
format_id = fnode.attrib['basetype']
format_m = re.match(r'''(?x)
(?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
(?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
''', format_id)
ext = determine_ext(video_url, None) or format_m.group('container')
if ext == 'meta':
continue
elif ext == 'smil':
formats.extend(self._extract_smil_formats(
video_url, video_id, fatal=False))
elif ext == 'm3u8':
# the certificates are misconfigured (see
# https://github.com/ytdl-org/youtube-dl/issues/8665)
if video_url.startswith('https://'):
continue
formats.extend(self._extract_m3u8_formats(
video_url, video_id, 'mp4', 'm3u8_native',
m3u8_id=format_id, fatal=False))
elif ext == 'f4m':
formats.extend(self._extract_f4m_formats(
video_url, video_id, f4m_id=format_id, fatal=False))
else:
quality = xpath_text(fnode, './quality')
if quality:
format_id += '-' + quality
abr = int_or_none(xpath_text(fnode, './audioBitrate'), 1000)
vbr = int_or_none(xpath_text(fnode, './videoBitrate'), 1000)
tbr = int_or_none(self._search_regex(
r'_(\d+)k', video_url, 'bitrate', None))
if tbr and vbr and not abr:
abr = tbr - vbr
formats.append({
'format_id': format_id,
'url': video_url,
'ext': ext,
'acodec': format_m.group('acodec'),
'vcodec': format_m.group('vcodec'),
'abr': abr,
'vbr': vbr,
'tbr': tbr,
'width': int_or_none(xpath_text(fnode, './width')),
'height': int_or_none(xpath_text(fnode, './height')),
'filesize': int_or_none(xpath_text(fnode, './filesize')),
'protocol': format_m.group('proto').lower(),
})
geolocation = xpath_text(doc, './/details/geolocation')
if not formats and geolocation and geolocation != 'none':
self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
self._sort_formats(formats)
thumbnails = []
for node in doc.findall('.//teaserimages/teaserimage'):
thumbnail_url = node.text
if not thumbnail_url:
continue
thumbnail = {
'url': thumbnail_url,
}
thumbnail_key = node.get('key')
if thumbnail_key:
m = re.match('^([0-9]+)x([0-9]+)$', thumbnail_key)
if m:
thumbnail['width'] = int(m.group(1))
thumbnail['height'] = int(m.group(2))
thumbnails.append(thumbnail)
upload_date = unified_strdate(xpath_text(doc, './/details/airtime'))
def extract_from_json_api(self, video_id, api_url):
doc = self._download_json(
api_url, video_id,
note="Downloading webpage metadata",
errnote="Failed to load webpage metadata")
for a in doc["absaetze"]:
if a["typ"] == "video-youtube":
return {
'id': video_id,
'title': title,
'description': xpath_text(doc, './/information/detail'),
'duration': int_or_none(xpath_text(doc, './/details/lengthSec')),
'thumbnails': thumbnails,
'uploader': xpath_text(doc, './/details/originChannelTitle'),
'uploader_id': xpath_text(doc, './/details/originChannelId'),
'upload_date': upload_date,
'formats': formats,
'_type': 'url_transparent',
'id': a["id"],
'title': doc["titel"],
'url': "https://www.youtube.com/watch?v=%s" % a["id"],
'ie_key': 'Youtube',
}
raise ExtractorError("No downloadable video found", expected=True)
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
internal_id = self._search_regex(
r'<div class="phx_vod" id="phx_vod_([0-9]+)"',
webpage, 'internal video ID')
api_url = 'http://www.phoenix.de/php/mediaplayer/data/beitrags_details.php?ak=web&id=%s' % internal_id
return self.extract_from_xml_url(video_id, api_url)
page_id = self._match_id(url)
api_url = 'https://www.phoenix.de/response/id/%s' % page_id
return self.extract_from_json_api(page_id, api_url)