[extractor] Add _search_nuxt_data (#1921)

Authored by: nao20010128nao
This commit is contained in:
The Hatsune Daishi 2021-12-08 01:38:50 +09:00 committed by GitHub
parent 93864403ea
commit 66f4c04e50
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 19 additions and 12 deletions

View file

@ -1513,6 +1513,24 @@ def _search_nextjs_data(self, webpage, video_id, **kw):
webpage, 'next.js data', **kw), webpage, 'next.js data', **kw),
video_id, **kw) video_id, **kw)
def _search_nuxt_data(self, webpage, video_id, context_name='__NUXT__'):
''' Parses Nuxt.js metadata. This works as long as the function __NUXT__ invokes is a pure function. '''
# not all website do this, but it can be changed
# https://stackoverflow.com/questions/67463109/how-to-change-or-hide-nuxt-and-nuxt-keyword-in-page-source
rectx = re.escape(context_name)
js, arg_keys, arg_vals = self._search_regex(
(r'<script>window\.%s=\(function\((?P<arg_keys>.*?)\)\{return\s(?P<js>\{.*?\})\}\((?P<arg_vals>.+?)\)\);?</script>' % rectx,
r'%s\(.*?\(function\((?P<arg_keys>.*?)\)\{return\s(?P<js>\{.*?\})\}\((?P<arg_vals>.*?)\)' % rectx),
webpage, context_name, group=['js', 'arg_keys', 'arg_vals'])
args = dict(zip(arg_keys.split(','), arg_vals.split(',')))
for key, val in args.items():
if val in ('undefined', 'void 0'):
args[key] = 'null'
return self._parse_json(js_to_json(js, args), video_id)['data'][0]
@staticmethod @staticmethod
def _hidden_inputs(html): def _hidden_inputs(html):
html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html) html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html)

View file

@ -3,7 +3,6 @@
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
js_to_json,
try_get, try_get,
unified_timestamp unified_timestamp
) )
@ -14,17 +13,7 @@ class SovietsClosetBaseIE(InfoExtractor):
def parse_nuxt_jsonp(self, nuxt_jsonp_url, video_id, name): def parse_nuxt_jsonp(self, nuxt_jsonp_url, video_id, name):
nuxt_jsonp = self._download_webpage(nuxt_jsonp_url, video_id, note=f'Downloading {name} __NUXT_JSONP__') nuxt_jsonp = self._download_webpage(nuxt_jsonp_url, video_id, note=f'Downloading {name} __NUXT_JSONP__')
js, arg_keys, arg_vals = self._search_regex( return self._search_nuxt_data(nuxt_jsonp, video_id, '__NUXT_JSONP__')
r'__NUXT_JSONP__\(.*?\(function\((?P<arg_keys>.*?)\)\{return\s(?P<js>\{.*?\})\}\((?P<arg_vals>.*?)\)',
nuxt_jsonp, '__NUXT_JSONP__', group=['js', 'arg_keys', 'arg_vals'])
args = dict(zip(arg_keys.split(','), arg_vals.split(',')))
for key, val in args.items():
if val in ('undefined', 'void 0'):
args[key] = 'null'
return self._parse_json(js_to_json(js, args), video_id)['data'][0]
def video_meta(self, video_id, game_name, category_name, episode_number, stream_date): def video_meta(self, video_id, game_name, category_name, episode_number, stream_date):
title = game_name title = game_name