etc/yt-dlp
0
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-26 02:55:17 +00:00
Code Issues Projects Releases Wiki Activity

[extractor/npr] Implement e50c3500b4 differently

Browse source 
Closes #4141
This commit is contained in:
pukkandan 2022-06-23 01:42:39 +05:30
parent 28163422a6
commit 0c36dc00d7
No known key found for this signature in database
GPG key ID: 7EEE9E1E817D0A39
2 changed files with 25 additions and 33 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

42
yt_dlp/extractor/common.py
View file

@ -1392,27 +1392,25 @@ def _twitter_search_player(self, html):
return self._html_search_meta('twitter:player', html, return self._html_search_meta('twitter:player', html,
'twitter card player') 'twitter card player')
def _search_json_ld(self, html, video_id, expected_type=None, **kwargs): def _yield_json_ld(self, html, video_id, *, fatal=True, default=NO_DEFAULT):
json_ld_list = list(re.finditer(JSON_LD_RE, html)) """Yield all json ld objects in the html"""
default = kwargs.get('default', NO_DEFAULT) if default is not NO_DEFAULT:
# JSON-LD may be malformed and thus `fatal` should be respected. fatal = False
# At the same time `default` may be passed that assumes `fatal=False` for mobj in re.finditer(JSON_LD_RE, html):
# for _search_regex. Let's simulate the same behavior here as well. json_ld_item = self._parse_json(mobj.group('json_ld'), video_id, fatal=fatal)
fatal = kwargs.get('fatal', True) if default is NO_DEFAULT else False for json_ld in variadic(json_ld_item):
json_ld = [] if isinstance(json_ld, dict):
for mobj in json_ld_list: yield json_ld
json_ld_item = self._parse_json(
mobj.group('json_ld'), video_id, fatal=fatal) def _search_json_ld(self, html, video_id, expected_type=None, *, fatal=True, default=NO_DEFAULT):
if not json_ld_item: """Search for a video in any json ld in the html"""
continue if default is not NO_DEFAULT:
if isinstance(json_ld_item, dict): fatal = False
json_ld.append(json_ld_item) info = self._json_ld(
elif isinstance(json_ld_item, (list, tuple)): list(self._yield_json_ld(html, video_id, fatal=fatal, default=default)),
json_ld.extend(json_ld_item) video_id, fatal=fatal, expected_type=expected_type)
if json_ld: if info:
json_ld = self._json_ld(json_ld, video_id, fatal=fatal, expected_type=expected_type) return info
if json_ld:
return json_ld
if default is not NO_DEFAULT: if default is not NO_DEFAULT:
return default return default
elif fatal: elif fatal:
@ -1500,7 +1498,7 @@ def extract_video_object(e):
assert is_type(e, 'VideoObject') assert is_type(e, 'VideoObject')
author = e.get('author') author = e.get('author')
info.update({ info.update({
'url': traverse_obj(e, 'contentUrl', 'embedUrl', expected_type=url_or_none), 'url': url_or_none(e.get('contentUrl')),
'title': unescapeHTML(e.get('name')), 'title': unescapeHTML(e.get('name')),
'description': unescapeHTML(e.get('description')), 'description': unescapeHTML(e.get('description')),
'thumbnails': [{'url': url} 'thumbnails': [{'url': url}

16
yt_dlp/extractor/npr.py
View file

@ -1,9 +1,5 @@
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import int_or_none, qualities, traverse_obj, url_or_none
int_or_none,
qualities,
url_or_none,
)
class NprIE(InfoExtractor): class NprIE(InfoExtractor):
@ -74,10 +70,6 @@ def _real_extract(self, url):
})['list']['story'][0] })['list']['story'][0]
playlist_title = story.get('title', {}).get('$text') playlist_title = story.get('title', {}).get('$text')
# Fetch the JSON-LD from the npr page.
json_ld = self._search_json_ld(
self._download_webpage(url, playlist_id), playlist_id, 'NewsArticle', fatal=False)
KNOWN_FORMATS = ('threegp', 'm3u8', 'smil', 'mp4', 'mp3') KNOWN_FORMATS = ('threegp', 'm3u8', 'smil', 'mp4', 'mp3')
quality = qualities(KNOWN_FORMATS) quality = qualities(KNOWN_FORMATS)
@ -124,8 +116,10 @@ def _real_extract(self, url):
stream_url, stream_id, 'mp4', 'm3u8_native', stream_url, stream_id, 'mp4', 'm3u8_native',
m3u8_id='hls', fatal=False)) m3u8_id='hls', fatal=False))
if not formats and json_ld.get('url'): if not formats:
formats.extend(self._extract_m3u8_formats(json_ld['url'], media_id, 'mp4', m3u8_id='hls', fatal=False)) raw_json_ld = self._yield_json_ld(self._download_webpage(url, playlist_id), playlist_id, fatal=False)
m3u8_url = traverse_obj(list(raw_json_ld), (..., 'subjectOf', ..., 'embedUrl'), get_all=False)
formats = self._extract_m3u8_formats(m3u8_url, media_id, 'mp4', m3u8_id='hls', fatal=False)
self._sort_formats(formats) self._sort_formats(formats)