pypdf is an open source, pure-python PDF library. In affected versions an attacker may craft a PDF which leads to an infinite loop if `__parse_content_stream` is executed. That is, for example, the case if the user extracted text from such a PDF. This issue was introduced in pull request #969 and resolved in pull request #1828. Users are advised to upgrade. Users unable to upgrade may modify the line `while peek not in (b"\r", b"\n")` in `pypdf/generic/_data_structures.py` to `while peek not in (b"\r", b"\n", b"")`.
References
Link | Resource |
---|---|
https://github.com/py-pdf/pypdf/pull/1828 | Patch |
https://github.com/py-pdf/pypdf/pull/969 | Issue Tracking Patch |
https://github.com/py-pdf/pypdf/security/advisories/GHSA-4vvm-4w3v-6mr8 | Exploit Patch |
Configurations
Configuration 1 (hide)
|
History
No history.
Information
Published : 2023-06-27 22:15
Updated : 2023-07-06 16:13
NVD link : CVE-2023-36464
Mitre link : CVE-2023-36464
CVE.ORG link : CVE-2023-36464
JSON object : View
Products Affected
pypdf2_project
- pypdf2
pypdf_project
- pypdf
CWE
CWE-835
Loop with Unreachable Exit Condition ('Infinite Loop')