std.rational -- update and progress towards review

Brad Roberts braddr at puremagic.com
Fri Oct 4 12:00:18 PDT 2013


On 10/4/13 7:16 AM, Joseph Rushton Wakeling wrote:
> On 03/10/13 16:38, Dicebot wrote:
>> On Thursday, 3 October 2013 at 12:16:51 UTC, Joseph Rushton Wakeling wrote:
>>> It's not about your opinions of the code per se, so much as about your
>>> experience of what is likely to provide a smooth review process ...
>>
>> Experience? :D I have not yet finished a single complete review process.
>
> You've managed a few, no? :-P
>
> Anyway, here's the state of play: I have two distinct branches that both implement std.rational as a
> new module in Phobos.
>
> https://github.com/WebDrake/phobos/tree/rational implements things as I think they should be, with
> several generic functions/templates parcelled out to std.traits and std.numeric.
>
> https://github.com/WebDrake/phobos/tree/rational-standalone implements things as a standalone module
> with all non-essential functions (or local duplications) marked as private.
>
> I would be happy for either or preferably both side-by-side to be subject to review now.  I think
> both are at the point where my asking on the forums is not going to get this code the scrutiny it
> needs.  That said, my concern is that there will be some significant changes requested and it may be
> knocked back this time -- which is why I've tried asking questions on the forums in the first place.
>
> So really, as review manager, it's your call.  If you'd like me to keep following up on my concerns
> and delay submission, I'll do that, but if you're happy to move forward, let's do it. :-)
>
> Thanks & best wishes,
>
>      -- Joe

Ideally, the unrelated but required non-rational code would be delt with before the review, then the 
issue is moot.  If you've got important or useful changes to other parts of phobos, separate them 
and get them delt with.


More information about the Digitalmars-d mailing list