Re: Lisp files that load cl-lib in problematical ways

[Dmitry Gutov]

On 26/10/2023 12:31, Adam Porter wrote:
> Hi Eli,
>
> I'm curious about this:
>
> > Btw, some of the cl-lib facilities should not be used in new code if
> > they have equivalent implementations ins seq.el (which is nowadays
> > preloaded) and in map.el.
>
> Since seq.el and map.el use generic methods, don't they have some 
> additional overhead compared to type-specific functions, like cl-lib 
> ones that are specific to lists?  Or is it small enough now that it 
> doesn't matter anymore?

For most functions, the dispatch overhead might only show up on the 
radar for small sequences (otherwise it's dwarfed by useful work). And I 
haven't yet seen a case where it was specifically a problem.

cl-lib has its own performance problems, so the real answer is for the 
programmer to measure all alternatives and sometimes contribute 
performance fixes upstream ;)