Space leak in Data.Map.partitionWithKey? about containers HOT 9 CLOSED

Which version of containers are you using? If it's 0.5, are you using the Strict or the Lazy API?

from containers.

No, I'm using 0.4.2.1 -- I couldn't use 0.5 (dependency problems -- upper limits, heh :). But I did check, and I think I concluded that the Strict and Lazy API in 0.5 use the same implementation of partitionWithKey?

from containers.

You're right, they're using the same implementation. Would it be possible for you to move to 0.5 or should I try to fix both the 0.4 and 0.5 releases?

from containers.

I use data-accessor a lot, which needs a new release to allow for containers-0.5. But that's minor; the main problem is that template-haskell is compiled against 0.4.2.1, and cabal tells me that rebuilding template-haskell can lead to all kinds of problems, so I haven't been brave enough to try it.

from containers.

In that case I think you're out of luck. Even if containers is to blame here and I make a new release in the 0.4.2 series, you'd still have to recompile template-haskell. That being said, I'm still interested in fixing this issue.

from containers.

Fair point. Mind you, mostly I'd just like to understand what's going on -- I don't like "fixing" problems without understanding the fix.

from containers.

GitHub mistakenly closed this when I made a yet-to-be merged commit in another branch. Reopening until the patch is merged into the master branch.

from containers.

Aaah. The fix is obvious in hindsight, I should have noticed that. Thanks for looking into this!

from containers.

So GitHub will continue to close this issue as I update my pull request. I will leave it closed. You'll have to trust that I will eventually merge the patch into master.

from containers.