Add cache-dir to command-line and pyproject.toml #1351
Conversation
src/main.rs
Outdated
| let mut cache_enabled: bool = !cli.no_cache; | ||
| if cache_enabled && cache::init().is_err() { | ||
| if cache_enabled && cache::init(cache_dir).is_err() { |
There was a problem hiding this comment.
I think we may need to call cache::init for every Settings object now, since different files could end up using different cache directories.
There was a problem hiding this comment.
Not sure how to go about this. Was thinking at first to do that in commands::run before running lint_path. But not only would that then happen per file, it can also happen at the same time due to par_iter.
My only other idea is to initialize the cache at the beginning for all resolved settings, keeping track of that and checking that per file.
There was a problem hiding this comment.
I think the latter option probably makes sense, to initialize for all resolved settings once we’ve gathered them up.
There was a problem hiding this comment.
Conceptually it’d be nice to just do it in the cache setter (like, right before we write), but maybe that’s hard to get right in a multi-threaded context?
Refs #1313