On Tue, Aug 7, 2012 at 1:52 PM, José Armando García Sancio <[email protected]> wrote: > On Tue, Aug 7, 2012 at 12:27 PM, Timon Gehr <[email protected]> wrote: >> On 08/07/2012 08:37 PM, José Armando García Sancio wrote: >>> >>> On Tue, Aug 7, 2012 at 10:39 AM, Timon Gehr<[email protected]> wrote: >>>> >>>> On 08/07/2012 07:59 AM, José Armando García Sancio wrote: >>>> >>>>> >>>>> On Sun, Aug 5, 2012 at 2:32 PM, Kapps<[email protected]> wrote: >>>>>> >>>>>> >>>>>> On Sunday, 5 August 2012 at 14:32:50 UTC, Adam D. Ruppe wrote: >>>>>>> >>>>>>> >>>>>>> >>>>>>> On Sunday, 5 August 2012 at 04:12:23 UTC, Jonathan M Davis wrote: > Interesting. I can't speak for the GC and allocation you mention since > I never tried to implement this myself but I should point out that C++ > STL's algorithm "module" (transform, for_each, etc) works this way. > Having said that, if D can't infer the type of lambda parameters > (which I thought the current version of D could) and create a > function/delegate based on the body then this is a no go. > > Thanks for you insight. > -Jose
Just incase it is not clear. I have zero interest in changing or promoting that we change the signature of the std.algorithm module. This is more of a selfish module design exercise on my part. Thanks, -Jose
