Proposal: Remove Num superclass of Bits

Bas van Dijk v.dijk.bas at gmail.com
Sat Oct 29 19:30:51 CEST 2011


On 29 October 2011 19:22, Johan Tibell <johan.tibell at gmail.com> wrote:
> On Sat, Oct 29, 2011 at 9:41 AM, Bas van Dijk <v.dijk.bas at gmail.com> wrote:
>>
>> On 29 October 2011 18:25, Johan Tibell <johan.tibell at gmail.com> wrote:
>> > No. Do we know which packages would break? I imagine that downloading
>> > the
>> > hackage tarball and grepping for 'instance Bits' should tell you.
>>
>> Where can this tarball be downloaded?
>
> http://hackage.haskell.org/packages/hackage.html
>
> "archive of just the latest versions of all the packages (230MB tar file)"

Nice, I will take a look.

>> > Also, does this have any performance impact?
>>
>> Performance should not be affected. After inlining all instances
>> should use the same definitions.
>
> Good. So none of the current instances use the default* functions?

No, all of them do. But since they previously used the default
implementation, which is the same as the default function, the
performance should be the same.

Bas



More information about the Libraries mailing list