ujson.dumps() (or ujson.dump() or ujson.encode()) have a reject_bytes=False option. When set, they may accept malformed or truncated UTF-8 byte sequences, silently rewriting them into different Unicode characters instead of rejecting them. This leads to input validation bypass and data integrity issues.
Details
The expected behavior is that for x being any bytes string, x == ujson.loads(ujson.dumps(x, reject_bytes=False)).encode(errors="surrogatepass") should always either be true or ujson.dumps() will throw an exception. In reality, some strings which should've been errors are silently rewritten as other strings:
Invalid continuation bytes are replaced with valid ones: b'\xcf\x13' -> b'\xcf\x93'
Unterminated sequence completes the sequence: b'\xc3' -> b'\xc3\x80'
... or leads to reading past the end of string: b'\xf0\x90\x94' -> b"\xf0\x90\x94\x80inxcontrib'"
Impact
An application relying on reject_bytes=False for UTF-8 handling may experience:
Data integrity issues
Experience validation bypass if said validation occurs before serialisation