pepepper/youtube-dl
1
0
Fork 0
mirror of https://github.com/ytdl-org/youtube-dl synced 2024-12-23 12:40:11 +09:00
Code Issues Projects Releases Wiki Activity

Use program-detail endpoint and remove a test

Browse Source
This commit is contained in:
Bart Broere 2024-03-06 11:52:08 +01:00
parent d426a92a60
commit 3b3d73cbe6
1 changed files with 24 additions and 37 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

59
youtube_dl/extractor/npo.py
View File

@ -25,18 +25,6 @@ class NPOIE(InfoExtractor):
_TESTS = [{ _TESTS = [{
'url': 'https://npo.nl/start/serie/zembla/seizoen-2015/wie-is-de-mol-2/', 'url': 'https://npo.nl/start/serie/zembla/seizoen-2015/wie-is-de-mol-2/',
# TODO fill in other test attributes # TODO fill in other test attributes
}, {
'url': 'http://www.npo.nl/de-mega-mike-mega-thomas-show/27-02-2009/VARA_101191800',
'md5': 'da50a5787dbfc1603c4ad80f31c5120b',
'info_dict': {
'id': 'VARA_101191800',
'ext': 'm4v',
'title': 'De Mega Mike & Mega Thomas show: The best of.',
'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
'upload_date': '20090227',
'duration': 2400,
},
'skip': 'Video gone',
}, { }, {
'url': 'https://npo.nl/start/serie/vpro-tegenlicht/seizoen-11/zwart-geld-de-toekomst-komt-uit-afrika', 'url': 'https://npo.nl/start/serie/vpro-tegenlicht/seizoen-11/zwart-geld-de-toekomst-komt-uit-afrika',
'md5': 'f8065e4e5a7824068ed3c7e783178f2c', 'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
@ -66,27 +54,21 @@ class NPOIE(InfoExtractor):
url = url[:-10] url = url[:-10]
url = url.rstrip('/') url = url.rstrip('/')
slug = url.split('/')[-1] slug = url.split('/')[-1]
page = self._download_webpage(url, slug, 'Finding productId using slug: %s' % slug)
# TODO find out what proper HTML parsing utilities are available in youtube-dl program_metadata = self._download_json('https://npo.nl/start/api/domain/program-detail',
next_data = page.split('<script id="__NEXT_DATA__" type="application/json">')[1].split('</script>')[0] slug,
# TODO The data in this script tag feels like GraphQL, so there might be an easier way query={'slug': slug})
# to get the product id, maybe using a GraphQL endpoint product_id = program_metadata.get('productId')
next_data = self._parse_json(next_data, slug) images = program_metadata.get('images')
product_id, title, description, thumbnail = None, None, None, None thumbnail = None
for query in next_data['props']['pageProps']['dehydratedState']['queries']: for image in images:
if isinstance(query['state']['data'], list): thumbnail = image.get('url')
for entry in query['state']['data']: break
if entry['slug'] == slug: title = program_metadata.get('title')
product_id = entry.get('productId') descriptions = program_metadata.get('description', {})
title = entry.get('title') description = descriptions.get('long') or descriptions.get('short') or descriptions.get('brief')
synopsis = entry.get('synopsis', {}) duration = program_metadata.get('durationInSeconds')
description = (synopsis.get('long')
or synopsis.get('short')
or synopsis.get('brief'))
thumbnails = entry.get('images')
for thumbnail_entry in thumbnails:
if 'url' in thumbnail_entry:
thumbnail = thumbnail_entry.get('url')
if not product_id: if not product_id:
raise ExtractorError('No productId found for slug: %s' % slug) raise ExtractorError('No productId found for slug: %s' % slug)
@ -96,9 +78,9 @@ class NPOIE(InfoExtractor):
'id': slug, 'id': slug,
'formats': formats, 'formats': formats,
'title': title or slug, 'title': title or slug,
'description': description, 'description': description or title or slug,
'thumbnail': thumbnail, 'thumbnail': thumbnail,
# TODO fill in other metadata that's available 'duration': duration,
} }
def _download_by_product_id(self, product_id, slug, url=None): def _download_by_product_id(self, product_id, slug, url=None):
@ -106,7 +88,8 @@ class NPOIE(InfoExtractor):
formats = [] formats = []
for profile in ( for profile in (
'dash', 'dash',
# 'hls', # TODO test what needs to change for 'hls' support # 'hls' is available too, but implementing it doesn't add much
# As far as I know 'dash' is always available
): ):
stream_link = self._download_json( stream_link = self._download_json(
'https://prod.npoplayer.nl/stream-link', video_id=slug, 'https://prod.npoplayer.nl/stream-link', video_id=slug,
@ -131,6 +114,7 @@ class BNNVaraIE(NPOIE):
_VALID_URL = r'https?://(?:www\.)?bnnvara\.nl/videos/[0-9]*' _VALID_URL = r'https?://(?:www\.)?bnnvara\.nl/videos/[0-9]*'
_TESTS = [{ _TESTS = [{
'url': 'https://www.bnnvara.nl/videos/27455', 'url': 'https://www.bnnvara.nl/videos/27455',
# TODO fill in other test attributes
}] }]
def _real_extract(self, url): def _real_extract(self, url):
@ -170,6 +154,7 @@ class ONIE(NPOIE):
_VALID_URL = r'https?://(?:www\.)?ongehoordnederland.tv/.*' _VALID_URL = r'https?://(?:www\.)?ongehoordnederland.tv/.*'
_TESTS = [{ _TESTS = [{
'url': 'https://ongehoordnederland.tv/2024/03/01/korte-clips/heeft-preppen-zin-betwijfel-dat-je-daar-echt-iets-aan-zult-hebben-bij-oorlog-lydia-daniel/', 'url': 'https://ongehoordnederland.tv/2024/03/01/korte-clips/heeft-preppen-zin-betwijfel-dat-je-daar-echt-iets-aan-zult-hebben-bij-oorlog-lydia-daniel/',
# TODO fill in other test attributes
}] }]
def _real_extract(self, url): def _real_extract(self, url):
@ -196,6 +181,7 @@ class VPROIE(NPOIE):
_VALID_URL = r'https?://(?:www\.)?vpro.nl/.*' _VALID_URL = r'https?://(?:www\.)?vpro.nl/.*'
_TESTS = [{ _TESTS = [{
'url': 'https://www.vpro.nl/programmas/tegenlicht/kijk/afleveringen/2015-2016/offline-als-luxe.html', 'url': 'https://www.vpro.nl/programmas/tegenlicht/kijk/afleveringen/2015-2016/offline-als-luxe.html',
# TODO fill in other test attributes
}] }]
def _real_extract(self, url): def _real_extract(self, url):
@ -224,6 +210,7 @@ class ZAPPIE(NPOIE):
_TESTS = [{ _TESTS = [{
'url': 'https://www.zapp.nl/programmas/zappsport/gemist/AT_300003973', 'url': 'https://www.zapp.nl/programmas/zappsport/gemist/AT_300003973',
# TODO fill in other test attributes
}] }]
def _real_extract(self, url): def _real_extract(self, url):