Re: A single general modification function for Map and IntMap proposal

Hi all,

-----Original message----- From: Nikita Volkov Sent: 30 Apr 2013, 17:18

Because of the above I have very often found myself in requirement for the following function:

withItem :: (Ord k) => k -> (Maybe i -> (r, Maybe i)) -> Map k i -> (r, Map k i) withItem k f m = let item = Map.lookup k m (r, item') = f item m' = Map.update (const item') k m in (r, m')

last time we talked about adding lens :: k -> Map k a -> (Maybe a -> Map k a, Maybe a) the performance of the direct implementation was actually worse than using lookup+insert or such, see the thread http://www.haskell.org/pipermail/libraries/2012-January/017423.html and the benchmark results at http://www.haskell.org/pipermail/libraries/2012-January/017435.html I am also a bit worried whether withItem is the "right" general function. For example Shachaf Ben-Kiki mentions the at :: (Ord k, Functor f) => k -> (Maybe i -> f (Maybe i)) -> Map k i -> f (Map k i) from the lens package in other mail, but maybe there are others.

1. Implement an efficient version of "withItem" for lazy and strict versions of "Map" and "IntMap".

Maybe we will need a benchmark to see whether withItem is really faster than combination of lookup+update.

3. Begin the deprecation process of the following functions: insertWith, insertWithKey, insertLookupWithKey, adjust, adjustWithKey, update, updateWithKey, updateLookupWithKey, alter.

I am against deprecating insertWith, insertWithKey, updateWithKey, updateWithKey, because they are useful. I am against deprecating adjust, adjustWithKey and alter, mostly because of compatibility reasons, although I agree that their name is not descriptive. I am indifferent about insertLookupWithKey and updateLookupWithKey. It would be nice to show that their implementation is really faster than lookup+insert / lookup+update combo. Cheers, Milan