midgard/youtube-dl
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

[vrt] Extract all videos on page

Browse source
This commit is contained in:
Midgard 2021-07-15 18:13:56 +02:00
parent 5a0e54f64f
commit 380b0da4da
Signed by: midgard
GPG key ID: 511C112F1331BBB4
1 changed files with 43 additions and 25 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

44
youtube_dl/extractor/vrt.py
View file

@ -1,6 +1,7 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import json
import re import re
from .common import InfoExtractor from .common import InfoExtractor
@ -54,34 +55,51 @@ class VRTIE(InfoExtractor):
def _real_extract(self, url): def _real_extract(self, url):
site, display_id = re.match(self._VALID_URL, url).groups() site, display_id = re.match(self._VALID_URL, url).groups()
webpage = self._download_webpage(url, display_id) webpage = self._download_webpage(url, display_id)
attrs = extract_attributes(self._search_regex(
r'(<[^>]+class="vrtvideo( [^"]*)?"[^>]*>)', webpage, 'vrt video'))
page_title = strip_or_none(get_element_by_class(
'vrt-title', webpage) or self._html_search_meta(
['og:title', 'twitter:title', 'name'], webpage))
page_description = self._html_search_meta(
['og:description', 'twitter:description', 'description'], webpage)
if page_description == '…':
page_description = None
page_timestamp = unified_timestamp(self._html_search_meta(
'article:published_time', webpage))
def entries():
matches = re.findall(r'(<[^>]+class="vrtvideo(?: [^"]*)?"[^>]*>)', webpage)
if not matches:
self._downloader.report_warning('no vrt videos found on page')
return
for tag_string in matches:
attrs = extract_attributes(tag_string)
asset_id = attrs['data-video-id'] asset_id = attrs['data-video-id']
publication_id = attrs.get('data-publication-id') publication_id = attrs.get('data-publication-id')
if publication_id: if publication_id:
asset_id = publication_id + '$' + asset_id asset_id = publication_id + '$' + asset_id
client = attrs.get('data-client-code') or self._CLIENT_MAP[site] client = attrs.get('data-client-code') or self._CLIENT_MAP[site]
title = strip_or_none(get_element_by_class( analytics_str = attrs.get('data-analytics')
'vrt-title', webpage) or self._html_search_meta( if analytics_str:
['og:title', 'twitter:title', 'name'], webpage)) analytics = json.loads(analytics_str)
description = self._html_search_meta( timestamp = unified_timestamp(analytics.get("onTime")) or page_timestamp
['og:description', 'twitter:description', 'description'], webpage) title = analytics.get("episode") or analytics.get("program") or page_title
if description == '…': else:
description = None timestamp = page_timestamp
timestamp = unified_timestamp(self._html_search_meta( title = page_title
'article:published_time', webpage))
return { yield {
'_type': 'url_transparent', '_type': 'url_transparent',
'id': asset_id, 'id': asset_id,
'display_id': display_id, 'display_id': display_id,
'title': title, 'title': title,
'description': description,
'thumbnail': attrs.get('data-posterimage'), 'thumbnail': attrs.get('data-posterimage'),
'timestamp': timestamp, 'timestamp': timestamp,
'duration': float_or_none(attrs.get('data-duration'), 1000), 'duration': float_or_none(attrs.get('data-duration'), 1000),
'url': 'https://mediazone.vrt.be/api/v1/%s/assets/%s' % (client, asset_id), 'url': 'https://mediazone.vrt.be/api/v1/%s/assets/%s' % (client, asset_id),
'ie_key': 'Canvas', 'ie_key': 'Canvas',
} }
return self.playlist_result(
entries(), display_id, page_title, page_description)