mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2025-07-02 19:38:32 +00:00
Code Issues Projects Releases Wiki Activity

[ie] Add _search_nuxt_json helper

Browse Source 
Authored by: bashonly
This commit is contained in:
bashonly 2025-06-03 17:43:29 -05:00
parent e1b6062f8c
commit 291301e939
No known key found for this signature in database
GPG Key ID: 783F096F253D15B0
2 changed files with 122 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

81
test/test_InfoExtractor.py
View File

@ -1947,6 +1947,87 @@ def test_search_nextjs_data(self):
with self.assertWarns(DeprecationWarning):
self.assertEqual(self.ie._search_nextjs_data('', None, default='{}'), {})
def test_search_nuxt_json(self):
HTML = '''
<script data-ssr="true" id="__NUXT_DATA__" type="application/json">
[
["ShallowReactive",1],
{"data":2,"state":22,"once":26},
["ShallowReactive",3],
{"$abcdef123456":4},
{"podcast":5,"activeEpisodeData":7},
{"podcast":6,"seasons":14},
{"title":10,"id":11},
["Reactive",8],
{"episode":9,"creators":18,"trick_data":19,"empty_list":21},
{"title":12,"id":13},
"Series Title",
"podcast-id-01",
"Episode Title",
"episode-id-99",
[15,16,17],
1,
2,
3,
[20],
[99,"gotcha"],
"Podcast Creator",
[],
{"$ssite-config":23},
{"env":24,"name":25},
"production",
"podcast-website",
["Set"]
]
</script>'''
DATA = {
'podcast': {
'podcast': {
'title': 'Series Title',
'id': 'podcast-id-01',
},
'seasons': [1, 2, 3],
},
'activeEpisodeData': {
'episode': {
'title': 'Episode Title',
'id': 'episode-id-99',
},
'creators': ['Podcast Creator'],
'trick_data': [99, 'gotcha'],
'empty_list': [],
},
}
FULL = {
'data': {
'$abcdef123456': DATA,
},
'state': {
'$ssite-config': {
'env': 'production',
'name': 'podcast-website',
},
},
'once': ['Set'],
}
BAD_HTML = '''
<script data-ssr="true" id="__NUXT_DATA__" type="application/json">
[
["ShallowReactive",1],
{"data":2},
{"improper_raw_list":3},
[15,16,17]
]
</script>'''
self.assertEqual(self.ie._search_nuxt_json(HTML, None), DATA)
self.assertEqual(self.ie._search_nuxt_json(HTML, None, traverse=None), FULL)
self.assertEqual(self.ie._search_nuxt_json('', None, fatal=False), {})
self.assertEqual(self.ie._search_nuxt_json(BAD_HTML, None, fatal=False), {})
self.assertEqual(self.ie._search_nuxt_json(HTML, None, fatal=False, allow_recursion=1), {})
with self.assertRaisesRegex(ExtractorError, r'recursion limit reached'):
self.ie._search_nuxt_json(HTML, None, allow_recursion=1)
if __name__ == '__main__':
unittest.main()

41
yt_dlp/extractor/common.py
View File

@ -1795,6 +1795,47 @@ def _search_nuxt_data(self, webpage, video_id, context_name='__NUXT__', *, fatal
ret = self._parse_json(js, video_id, transform_source=functools.partial(js_to_json, vars=args), fatal=fatal)
return traverse_obj(ret, traverse) or {}
def _search_nuxt_json(self, webpage, video_id, script_id='__NUXT_DATA__', *, fatal=True,
traverse=('data', ..., {dict}, any), allow_recursion=100):
"""Parses Nuxt.js metadata when it has already been rendered into a JSON array"""
ERROR_MSG = 'Unable to extract NUXT JSON data'
array = self._search_json(
fr'<script\b[^>]+\bid="{re.escape(script_id)}"[^>]*>', webpage, script_id,
video_id, contains_pattern=r'\[(?s:.+)\]', default=NO_DEFAULT if fatal else [{}])
def extract_element(element, allow_recursion):
if allow_recursion < 0:
msg = f'{ERROR_MSG}: recursion limit reached'
if fatal:
raise ExtractorError(msg)
self.report_warning(msg, video_id=video_id, only_once=True)
return None
allow_recursion -= 1
try:
if isinstance(element, list) and element:
if element[0] in ('ShallowReactive', 'Reactive') and isinstance(element[1], int):
return extract_element(array[element[1]], allow_recursion)
if all(isinstance(ele, int) for ele in element):
return [extract_element(array[ele], allow_recursion) for ele in element]
if isinstance(element, dict):
ret = {}
for k, v in element.items():
if isinstance(v, int):
ret[k] = extract_element(array[v], allow_recursion)
else:
ret[k] = v
return ret
except IndexError as e:
if not fatal:
return None
raise ExtractorError(f'{ERROR_MSG}: {e}')
return element
return traverse_obj(extract_element(array[0], allow_recursion), traverse) or {}
@staticmethod
def _hidden_inputs(html):
html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html)