Closed etiennebacher closed 9 months ago
I can't replicate #219 because there has to be an exact timing of read conflict, and that only happens rarely. However, your code makes sense, so I think we can merge and close the issue.
I'm running into this with polars from time to time so I'm currently testing on it
I can't reproduce the error with polars
so let's merge this
Close #219
I think the error when we have
freeze
+parallel
is that we write tofreeze.rds
for each file individually, right after they're rendered. This is an issue when several processes want to write to the same file at the same time.This PR should fix this by updating all hashes at once after all man pages (or vignettes) have been rendered.
@vincentarelbundock can you try with this branch to see if it solves #219?