mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2025-07-18 19:28:31 +00:00
Code Issues Projects Releases Wiki Activity

ruff

Browse Source
This commit is contained in:
c-basalt 2025-07-02 14:22:01 -04:00
parent 56f8c764fe
commit bca83e3229
1 changed files with 16 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
yt_dlp/extractor/khanacademy.py
View File

@ -33,7 +33,7 @@ def _search_query_js(self, query_name):
r'return\s*""\+e\+"\."\+\(', runtime_js, 'js resources', None, end_pattern=r'\)\[e\]\+"\.js"', r'return\s*""\+e\+"\."\+\(', runtime_js, 'js resources', None, end_pattern=r'\)\[e\]\+"\.js"',
transform_source=lambda s: re.sub(r'([\da-f]+):', r'"\1":', s)) transform_source=lambda s: re.sub(r'([\da-f]+):', r'"\1":', s))
# traverse all lazy-loaded js to find query-containing js file # iterate all lazy-loaded js to find query-containing js file
main_js = self._download_webpage(self._MAIN_JS_URL, None, 'Downloading khanacademy.js') main_js = self._download_webpage(self._MAIN_JS_URL, None, 'Downloading khanacademy.js')
for lazy_load in re.finditer(r'lazy\(function\(\)\{return Promise\.all\(\[(.+?)\]\)\.then', main_js): for lazy_load in re.finditer(r'lazy\(function\(\)\{return Promise\.all\(\[(.+?)\]\)\.then', main_js):
for js_name in re.finditer(r'X.e\("([0-9a-f]+)"\)', lazy_load[1]): for js_name in re.finditer(r'X.e\("([0-9a-f]+)"\)', lazy_load[1]):
@ -68,6 +68,8 @@ def _sanitize_query(self, query: str):
if not line or line.startswith('#'): if not line or line.startswith('#'):
continue continue
if line == '}': if line == '}':
# unlike fragment, query has no __typename at its ends
# only object inside query has tailing __typename
if indent > 2 or outlines[0].startswith('fragment'): if indent > 2 or outlines[0].startswith('fragment'):
outlines.append(f'{" " * indent}__typename') outlines.append(f'{" " * indent}__typename')
indent -= 2 indent -= 2
@ -79,29 +81,31 @@ def _sanitize_query(self, query: str):
def _compose_query(self, queries, key): def _compose_query(self, queries, key):
def _get_fragments(key): def _get_fragments(key):
fragments = [self._sanitize_query(queries[key]['query'])] fragments = [self._sanitize_query(queries[key]['query'])]
for key in queries[key]['embeds']: for sub in queries[key]['embeds']:
fragments.extend(_get_fragments(key)) fragments.extend(_get_fragments(sub))
return fragments return fragments
# recursively find all fragments then sort them # recursively find and combine all fragments then sort them
queries = _get_fragments(key) queries = _get_fragments(key)
if not (query := next((q for q in queries if q.startswith('query ')), None)): if not (query := next((q for q in queries if q.startswith('query ')), None)):
raise ExtractorError(f'Failed to get query for {key}') raise ExtractorError(f'Failed to get query for {key}')
fragments = sorted(set(q for q in queries if q.startswith('fragment '))) fragments = sorted(set(q for q in queries if q.startswith('fragment ')))
return '\n\n'.join([query, *fragments]) return '\n\n'.join([query, *fragments])
def _string_hash(self, input): def _string_hash(self, input_str):
hash = 5381 str_hash = 5381
for char in input[::-1]: for char in input_str[::-1]:
hash = ((hash * 33) ^ ord(char)) & 0xFFFFFFFF str_hash = ((str_hash * 33) ^ ord(char)) & 0xFFFFFFFF
return hash return str_hash
def _get_query_hash(self, query_name): def _get_query_hash(self, query_name):
if cache := self.cache.load('khanacademy', f'{query_name}-hash'): if cache := self.cache.load('khanacademy', f'{query_name}-hash'):
# change in hash of runtime.js may indicate change of website version # change in hash of runtime.js may indicate change of graph query schema
# consider cached hash as invalidated upon such change
if cache['runtime_js'] == self._RUNTIME_JS_URL: if cache['runtime_js'] == self._RUNTIME_JS_URL:
return cache['hash'] return cache['hash']
# iterate all query objects to find matching query
queries = self._parse_query_src(self._search_query_js(query_name)) queries = self._parse_query_src(self._search_query_js(query_name))
for key, query_obj in queries.items(): for key, query_obj in queries.items():
if f'query {query_name}' in query_obj['query']: if f'query {query_name}' in query_obj['query']:
@ -200,7 +204,7 @@ class KhanAcademyIE(KhanAcademyBaseIE):
'ext': 'mp4', 'ext': 'mp4',
'title': 'Doodling in math: Spirals, Fibonacci, and being a plant [1 of 3]', 'title': 'Doodling in math: Spirals, Fibonacci, and being a plant [1 of 3]',
'description': 'md5:4098102420babcf909097ec1633a52e7', 'description': 'md5:4098102420babcf909097ec1633a52e7',
"upload_date": "20120131", 'upload_date': '20120131',
'timestamp': 1327972656, 'timestamp': 1327972656,
'thumbnail': r're:https://cdn.kastatic.org/.*', 'thumbnail': r're:https://cdn.kastatic.org/.*',
'duration': 355, 'duration': 355,
@ -232,7 +236,7 @@ def _parse_component_props(self, component_props, display_id, listed=True):
'creators': ('authorNames', ..., {str}), 'creators': ('authorNames', ..., {str}),
'timestamp': ('dateAdded', {parse_iso8601}), 'timestamp': ('dateAdded', {parse_iso8601}),
'license': ('kaUserLicense', {str}), 'license': ('kaUserLicense', {str}),
}) }),
} }