Proposal: Add log1p and expm1 to GHC.Float.Floating

[Edward Kmett]

It sounds like we're converging to a solution whereby they do get defaults,
but you also get a MINIMAL pragma warning you you probably want to override
them for manual Floating instances, so users will get a warning not an
error if they omit implementations, and switching over will not hurt
existing code.

That warning should serve much the same role as the deprecation cycle would
and no existing user code will break.



On Thu, Apr 24, 2014 at 4:38 AM, Johan Tibell <johan.tibell at gmail.com>wrote:

> On Thu, Apr 24, 2014 at 10:29 AM, Gregory Collins <greg at gregorycollins.net
> > wrote:
>
>>
>> On Thu, Apr 24, 2014 at 5:38 AM, wren romano <winterkoninkje at gmail.com>wrote:
>>
>>> I believe we've reached a consensus that no defaults is superior to
>>> error defaults. And I agree! So that's one bird down :)
>>>
>>
>> If I understand correctly, that's going to break user code without a
>> deprecation cycle and as such gets an enthusiastic -1 from me.
>>
>
> If that's the case, -1 from me too. Breaking user code for a change which
> is useful to a tiny minority is not worth it.
>
> Otherwise I'm +1 having the functions in general.
>
>
> _______________________________________________
> Libraries mailing list
> Libraries at haskell.org
> http://www.haskell.org/mailman/listinfo/libraries
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.haskell.org/pipermail/libraries/attachments/20140424/cb26189b/attachment.html>