# CVE-2026-102995: pypdf memory exhaustion in ToUnicode font mapping parsing

Source: https://junglewise.ai/threats/cve-2026-102995-pypdf-is-a-free-and-open-source-pure-python-pdf-library-prior-to  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-09-30. Updated: 2026-09-30  
License: CC BY 4.0 with a link back

CVE-2026-102995 · Severity: high · CVSS 4 · Published 2026-09-30

pypdf is a Python library for reading and manipulating PDF files. A malicious PDF can exploit improper token length validation in font ToUnicode mapping parsing to consume excessive memory during operations like text extraction, potentially causing denial of service.

Technologies: [Py-Pdf Pypdf](https://junglewise.ai/threats/technologies/py-pdf-pypdf), [pypdf (PyPI)](https://junglewise.ai/threats/technologies/pypdf). Vendors: [Py-Pdf](https://junglewise.ai/threats/vendors/py-pdf), [PyPI](https://junglewise.ai/threats/vendors/pypi).

## Quick answers

**Is CVE-2026-102995 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-102995 being exploited in the wild.

**How severe is CVE-2026-102995?** High severity, with a CVSS score of 4, published 30 September 2026.

**Which products and versions are affected?** Py-Pdf Pypdf prior to 6.18.1; pypdf (PyPI) < 6.18.1, ECOSYSTEM:introduced=0,fixed=6.18.1.

**Is there a fix for CVE-2026-102995?** Yes. pypdf (PyPI): upgrade to 6.18.1 or later. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

**How is CVE-2026-102995 attacked?** It is attacked over the network, in a software library.

**Am I affected by CVE-2026-102995?** Check whether you run Py-Pdf Pypdf or pypdf (PyPI) at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## How to fix

- pypdf (PyPI): upgrade to 6.18.1 or later

  ```sh
  pip install "pypdf>=6.18.1"
  ```


## Technical details

The vulnerability exists in pypdf/_cmap.py's parse_bfchar function, which fails to properly limit the length of tokens decoded from PDF font /ToUnicode mappings. An attacker can craft a PDF with oversized source-code or destination-string tokens that cause the parser to retain excessive memory. The flaw is triggered during PDF parsing without requiring user interaction beyond opening the file.

## Affected products

- py-pdf pypdf prior to 6.18.1

## Identifiers

- CVE-2026-102995
- GHSA-fp3h-c4fm-7vvf
- PYSEC-2026-4154

## Timeline

- 2026-09-30: disclosed
- 2026-09-11: patched

## References

- [https://github.com/py-pdf/pypdf/commit/319d0b823ce2c311a2435e9fd93c13994d32bb51](https://github.com/py-pdf/pypdf/commit/319d0b823ce2c311a2435e9fd93c13994d32bb51)
- [https://github.com/py-pdf/pypdf/pull/4071](https://github.com/py-pdf/pypdf/pull/4071)

## Related threats

- [pypdf memory exhaustion in cross-reference stream parsing](https://junglewise.ai/threats/cve-2026-119209-pypdf-is-a-free-and-open-source-pure-python-pdf-library-prior-to)
- [pypdf denial of service via forged encrypted PDF page count](https://junglewise.ai/threats/cve-2026-119208-pypdf-is-a-free-and-open-source-pure-python-pdf-library-prior-to)
- [pypdf excessive memory consumption in strict mode PDF parsing](https://junglewise.ai/threats/cve-2026-119207-pypdf-is-a-free-and-open-source-pure-python-pdf-library-prior-to)
- [pypdf denial of service via crafted page labels](https://junglewise.ai/threats/cve-2026-119206-pypdf-is-a-free-and-open-source-pure-python-pdf-library-prior-to)
- [pypdf infinite loop in DictionaryObject cloning](https://junglewise.ai/threats/cve-2026-119205-pypdf-is-a-free-and-open-source-pure-python-pdf-library-prior-to)