Caching

Caching can significantly speed up repeated checks by reducing requests to the same URL during consecutive runs. For instance, caching responses from https://github.com can decrease the load when checking multiple links. Here’s how to cache the results of a Lychee run.

Caching on the command line

To cache the results of a lychee run, you can use the --cache flag. This will save the results to a .lycheecache file in the current directory. The next time you run lychee with the --cache flag, it will use the cached results instead of making a new request.

lychee --cache --verbose --no-progress './**/*.md' './**/*.html'

Tip

Check out the .lycheecache file to see the cached results. It’s just a plaintext file with the URLs and their status codes as well as a UNIX timestamp per entry, which is used to determine the cache’s age.

The great thing is, that you can remove single lines from the cache file to re-check the URLs on the next run. It’s human-readable and editable.

Caching in Docker

If you’re running lychee inside a Docker container, caching is still possible, but a little trickier.

You need to create a volume to cache the results. This way, the results will persist between runs.

You need to create the .lycheecache file in the current directory before running the Docker container.

touch .lycheecache
docker run -it -v $(pwd)/.lycheecache:/.lycheecache lycheeverse/lychee --cache --verbose https://lychee.cli.rs

Caching in GitHub Actions

To see how you can cache the results of a lychee run in GitHub Actions, check out this page.