[Haskell-cafe] Specializing argument to forM_, mapM_ & foldM_ in base-4.8

[Adam Bergmark]

If you care about performance you may - I haven't benchmarked - want to use
Vector instead of lists here since that's what aeson uses internally. Then
it's pretty handy that you can still use forM_.

It's possible that the list pattern deconstruction and list construction
gets optimized away, my gut says you need -O2 for that to happen. Here's a
good explanation on how to dump and read core so you can check for yourself
what happens in this case:
http://stackoverflow.com/questions/6121146/reading-ghc-core . Either way
it's definitiely not less efficient to annotate the type instead. You don't
need ScopedTypeVariables here, you can write the type inside an expression
instead: `forM (objs :: Type) [...]`

HTH,
Adam


On Thu, Oct 15, 2015 at 7:16 PM, Amit Aryeh Levy <amit at amitlevy.com> wrote:

> I've been running into a relatively small but frequent annoyance with
> base >= 4.8 (GHC 7.10). `Control.Monad.foldM_`, `Control.Monad.mapM_`
> and `Control.Monad.forM_` are generalized traverse over any `Foldable a`
> rather than just arrays (`[a]`).
>
>  This is great, except I'm finding that, for a lot of my code that works
> well in previous versions, I need to specialize the argument to `[a]`
> now. If other people are encoutering a similar patter, I wonder what are
> your best practices for doing this: ScopedTypeVariables? Deconstruct the
> reconstruct the array? ...
>
>  The most common example is when I deserialize a JSON array with aeson
> and want to traverse over that array (say, to store the objects to a DB):
>
>  ```
> let objArray = eitherDecode myjson
> case objArray of
>     Left err -> ...
>     Right (objs :: [MyObjType]) ->
>         forM_ objs $ \obj -> saveToDb obj
>  ```
>
> ​The above fix requires `ScopedTypeVariables` (which is probably OK).
> Another option is to deconstruct and reconstruct the list:
>
> ```
> Right (o:objs) ->
>     forM_ (o:objs) $ \obj -> saveToDb obj
> ```
>
> Does this get optimized away?
>
> Penny for your thoughts?
>
> Cheers!
> Amit
>
>
> _______________________________________________
> Haskell-Cafe mailing list
> Haskell-Cafe at haskell.org
> http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/haskell-cafe/attachments/20151015/205b5030/attachment.html>