[telegraaf] fix extraction

This commit is contained in:
Remita Amine 2019-11-03 22:04:03 +01:00
parent 726e8eef59
commit 564275e26f
1 changed files with 43 additions and 32 deletions

View File

@ -4,21 +4,25 @@ from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
determine_ext, determine_ext,
remove_end, int_or_none,
parse_iso8601,
try_get,
) )
class TelegraafIE(InfoExtractor): class TelegraafIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?telegraaf\.nl/tv/(?:[^/]+/)+(?P<id>\d+)/[^/]+\.html' _VALID_URL = r'https?://(?:www\.)?telegraaf\.nl/video/(?P<id>\d+)'
_TEST = { _TEST = {
'url': 'http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html', 'url': 'https://www.telegraaf.nl/video/734366489/historisch-scheepswrak-slaat-na-100-jaar-los',
'info_dict': { 'info_dict': {
'id': '24353229', 'id': 'gaMItuoSeUg2',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Tikibad ontruimd wegens brand', 'title': 'Historisch scheepswrak slaat na 100 jaar los',
'description': 'md5:05ca046ff47b931f9b04855015e163a4', 'description': 'md5:6f53b7c4f55596722ac24d6c0ec00cfb',
'thumbnail': r're:^https?://.*\.jpg$', 'thumbnail': r're:^https?://.*\.jpg',
'duration': 33, 'duration': 55,
'timestamp': 1572805527,
'upload_date': '20191103',
}, },
'params': { 'params': {
# m3u8 download # m3u8 download
@ -27,23 +31,30 @@ class TelegraafIE(InfoExtractor):
} }
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) article_id = self._match_id(url)
webpage = self._download_webpage(url, video_id) video_id = self._download_json(
'https://www.telegraaf.nl/graphql', article_id, query={
'query': '''{
article(uid: %s) {
videos {
videoId
}
}
}''' % article_id,
})['data']['article']['videos'][0]['videoId']
player_url = self._html_search_regex( item = self._download_json(
r'<iframe[^>]+src="([^"]+")', webpage, 'player URL') 'https://content.tmgvideo.nl/playlist/item=%s/playlist.json' % video_id,
player_page = self._download_webpage( video_id)['items'][0]
player_url, video_id, note='Download player webpage') title = item['title']
playlist_url = self._search_regex(
r'playlist\s*:\s*"([^"]+)"', player_page, 'playlist URL')
playlist_data = self._download_json(playlist_url, video_id)
item = playlist_data['items'][0]
formats = [] formats = []
locations = item['locations'] locations = item.get('locations') or {}
for location in locations.get('adaptive', []): for location in locations.get('adaptive', []):
manifest_url = location['src'] manifest_url = location.get('src')
if not manifest_url:
continue
ext = determine_ext(manifest_url) ext = determine_ext(manifest_url)
if ext == 'm3u8': if ext == 'm3u8':
formats.extend(self._extract_m3u8_formats( formats.extend(self._extract_m3u8_formats(
@ -54,25 +65,25 @@ class TelegraafIE(InfoExtractor):
else: else:
self.report_warning('Unknown adaptive format %s' % ext) self.report_warning('Unknown adaptive format %s' % ext)
for location in locations.get('progressive', []): for location in locations.get('progressive', []):
src = try_get(location, lambda x: x['sources'][0]['src'])
if not src:
continue
label = location.get('label')
formats.append({ formats.append({
'url': location['sources'][0]['src'], 'url': src,
'width': location.get('width'), 'width': int_or_none(location.get('width')),
'height': location.get('height'), 'height': int_or_none(location.get('height')),
'format_id': 'http-%s' % location['label'], 'format_id': 'http' + ('-%s' % label if label else ''),
}) })
self._sort_formats(formats) self._sort_formats(formats)
title = remove_end(self._og_search_title(webpage), ' - VIDEO')
description = self._og_search_description(webpage)
duration = item.get('duration')
thumbnail = item.get('poster')
return { return {
'id': video_id, 'id': video_id,
'title': title, 'title': title,
'description': description, 'description': item.get('description'),
'formats': formats, 'formats': formats,
'duration': duration, 'duration': int_or_none(item.get('duration')),
'thumbnail': thumbnail, 'thumbnail': item.get('poster'),
'timestamp': parse_iso8601(item.get('datecreated'), ' '),
} }