pdf-link-checker is a simple tool that parses a PDF document and checks for broken hyperlinks. This done by sending a simple HTTP request to each link found in a given document.
pip install git+https://github.com/bootlin/pdf-link-checker.git export PATH=$HOME/.local/bin:$PATH pdf-link-checker my-awesome-slides.pdf
--max-threads
Specifies the maximum number of allowed threads (default: 100).
To speedup the run, pdf-link-checker will launch several threads in order to check several links in parallel. This option allows to set a limit to the number of threads.
--max-requests-per-host
Specifies the maximum number of allowed requests per host.
Some URLs may belong to the same host, and since pdf-link-checker can check many URLs at the same time, we may want to set a limit to the number of requests per host. Otherwise, some hosts may confuse the check with a DoS attack.
You can get support by reporting your issue on this project on GitHub: https://github.com/bootlin/pdf-link-checker/issues
(...because there's no active project without a TODO list!)
- Fix: some documents are failing on doc.initialize().
- Fix: if the URL is a huge document, we should just check and not download it entirely.
- Replace the thread array into a nice thread pool. Each thread from the pool should take an URL from a (protected) queue. We could also have one queue per host and thus handle the max-requests-per-host constraint without a separate parameter.
- 1.2.0
- Repair breakage against newer versions of pdfminer
- 1.1.1
- Remove extra print, just a leftover
- 1.1.0
- Only allow https and ftp URIs. This prevents from failing on mailto: and file:// URIs.
- Add better exception handling to avoid crashing
- Add better timeout and request exception handling
- Fix broken thread management
- Remove stupid double-requests
- Several small fixes
- 1.0.2
- Updated repo location
- Moved from distutils to setuptools
- 1.0.1
- Version bump
- 1.0
- Initial release