In `pollen` I create what are essentially compile caches and store them in 
subdirectories called `pollen-cache`. When it's time to invalidate caches 
for the project, I delete all the `pollen-cache` subdirectories and all the 
Racket-created `compiled` subdirectories.

But it has occurred to me that I wouldn't lose anything by moving every 
`pollen-cache` subdirectory into the nearby `compiled` directory.

*What is the thinking (if any) on sharing the `compiled` subdirectories?*

One point of view is that these folders should be private to the Racket 
language, forever.

Another view is that if one is making a Racket-implemented DSL that keeps a 
compile cache, it should live in the same `compiled` subdir (on the idea 
that users are already, say, setting up `.gitignore` to ignore those 
subdirs.) This would be consistent with the general orientation toward 
letting Racket-implemented DSLs reuse the Racket language ecosystem. Having 
multiple conventions for where to store compile caches is unnecessary.

A third view is that the `compiled` subdir should be open to all who want 
to use it, but that there should be a convention of each package making a 
subdirectory therein, so that the `racket` caches don't mess with `pollen` 
caches and vice versa and so on (which would be more of a social convention 
than technical convention)


-- 
You received this message because you are subscribed to the Google Groups 
"Racket Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/racket-dev/404d9bb0-296c-48ba-8c42-58e7b0d41d8f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to