Re: Converting tab-complete.c's else-if chain to a switch

On Sat, Jul 27, 2024 at 8:03 AM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> I wrote:
> > I modified the preprocessor to work like that, and I like the results
> > better than what I had.
>
> I thought this version of the patch would be less subject to merge
> conflicts than v1, but it didn't take long for somebody to break it.
> Rebased v3 attached -- no nontrivial changes from v2.
>
> I'd like to get this merged soon ...

+1

This works nicely for me and I couldn't find anything that doesn't
work as expected, the mechanism seems reasonable, etc. I like how
it's moving slowly towards a grammar DSL.

Should we try to make sure that tab-complete.in.c remains compilable
directly with a build mode that does that with 'cp' instead of the
perl script?

One very trivial thing, where you say:

+# where N is the replacement case label, "HeadMatch" is the original
+# function name shorn of "es", and the rest are the function arguments.

... maybe simpler language: 'with "es" removed'? (Thinking of
non-native speakers; I understood that perfectly but I'm anglophone in
a country overrun with sheep.)