[Bug] Filter NUL byte (\x00) is not stripped — wrong escape "\\u0000" used instead of "\u0000"

[kelvin-qin]

Contact Information

No response

MaxKB Version

MaxKB 2.10.4-lts

Problem Description

In apps/common/utils/common.py:443, s_list = ["\\u0000"] is a 6-character literal
string (backslash + "u0000"), not the actual NUL byte. .replace() therefore never
matches a real NUL byte, so NUL characters pass through the "special character filter".
This function is used to sanitize user/content input, so NUL bytes survive cleansing.

Steps to Reproduce

1. Call filter_special_character("a\x00b") # input contains a real NUL byte
2. Inspect the return value

The expected correct result

Expected Result

Return "ab" (the NUL byte is removed).

Actual Result (error log / stacktrace)

Return value is 'a\x00b' unchanged — the NUL byte is still present.
Reproduced by executing the real source function:
input repr: 'a\x00b' (len 3)
output repr: 'a\x00b' (len 3)
NUL byte still present after filter?: True

Related log output

Additional Information

Suggested Fix (optional)

• Line 443: change s_list = ["\\u0000"] to s_list = ["\u0000"]
  (single backslash so Python decodes it as the NUL byte), or use:
  s_list = ["\x00"].

在Python中：

"\u0000" 是转义后的反斜杠加上字面文本 u0000，即字符串 \u0000（6个字符：, u, 0, 0, 0, 0）。
真正的NUL字节应写作 "\u0000"（不加双反斜杠），是单个字符 \x00。
因此，当调用 filter_special_character("a\x00b") 时，输入中包含的是真正的NUL字节 (\x00)，而代码尝试替换的是字面字符串 "\u0000"，二者完全不匹配，.replace() 不会有任何效果。