pypdf 是一个免费且开源的纯 Python PDF 库。在 6.21.3 版本之前,pypdf 在处理包含大量范围的扁平页标签数字树的特制 PDF 文件时,检索页标签所需的时间会异常过长。该问题源于 属性在解析每一页时,都会从开头重复扫描同一个 数组,导致计算量随页面数量和标签范围数量共同增长。此问题已在 6.21.3 版本中修复。
Although we use advanced large model technology, its output may still contain inaccurate or outdated information.Shenlong tries to ensure data accuracy, but please verify and judge based on the actual situation.
| # | POC Description | Source Link | Shenlong Link |
|---|
No public POC found.
Login to generate AI POC| CVE-2026-119203 | 7.1 HIGH | pypdf: Possible infinite loop when hitting EOF in standard xref table during trailer searc |
| CVE-2026-119204 | 7.1 HIGH | pypdf: Possible large memory usage when parsing page tree |
| CVE-2026-119207 | 7.1 HIGH | pypdf: Possible large memory usage for exceptions in strict mode |
| CVE-2026-119205 | 7.1 HIGH | pypdf: Possible infinite loop when cloning dictionaries |
| CVE-2026-119209 | 7.1 HIGH | pypdf: Possible large memory usage when parsing cross-reference streams |
| CVE-2026-119208 | 7.1 HIGH | pypdf: Possible long runtimes/large memory usage for encrypted PDF files |