This repository was archived by the owner on Dec 17, 2020. It is now read-only.
Conversation
Author
|
Uh, looks like I branched from the wrong branch, so the commits from #44 have been included, but I probably wouldn't want this included without those either. I'm happy to make a new PR with the one commit that differs here if you'd prefer, since they don't actually depend on each other. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
I've separated this from #44 as it's probably more controversial. Adding a
Modifyconstructor allows users to know that a modification will be atomic. If you look at the example in the diff, I've shown how this can be used to allow users to run stateful computations on mutable shared state, concurrently with other threads, or isolated in the traditional manner, without modification to the computation.An alternative implementation would be to implement
GetandPutin terms ofModify, but then the example I've provided would end up callingatomicModifyIORefmany more times than necessary.Keen to hear feedback on this, I'm not confident that this change is actually worth it, and am sure there are problems I haven't considered.