Potential DoS via quadratic complexity in unicodedata.normalize()

Description

unicodedata.normalize() can take excessive CPU time when processing specially crafted Unicode input containing long runs of combining characters with alternating Canonical Combining Class values. This affects all normalization forms.

Affected Products

Also Affects

Downstream vendors/products affected by this vulnerability

References

• https://mail.python.org/archives/list/security-announce@python.org/thread/PP5HB4K7727OBBM76KA2ILID76K3OZGZ/(vendor-advisory)
• https://github.com/python/cpython/pull/149080(patch)
• https://github.com/python/cpython/issues/149079(issue-tracking)
• https://github.com/python/cpython/commit/6b505d1f41f8f3ea0fe5a4786d3a8fff1875cfc0(patch)
• https://github.com/python/cpython/commit/991224b1e8311c85f198f6dd8208bf8cff7fc26f(patch)
• https://github.com/python/cpython/commit/ba785b88add96acbf403d65cb157fb2743a33a32(patch)
• https://github.com/python/cpython/commit/c5512bd7c1dc28055660565275012766941d3066(patch)
• https://github.com/python/cpython/commit/90748760d38ca3ac5fc6788a69becab905c95598(patch)

Related News (4 articles)