Re: [HACKERS] [COMMITTERS] pgsql: Add missing support for new nodefields

Hello Robert,

> IMHO, what would be a lot more useful than something that generates
> {read,equal,copy,out}funcs.c automatically would be something that
> just checks them for trivial errors of omission.

Hmmm. Checking for errors is actually more complicated than generating the
function: basically you have to generate the function, at least
implicitely, then parse the actual functions, then compare the two, then 
generate meaningful messages. Thrice the work.

> The idea would be that if you added a field that wasn't supposed to be
> copied, you'd have to add something to copyfuncs.c that said, e.g.
>
> /* NOTCOPIED: mymember */

Yep, I was thinking of maybe use directives added to header files to 
handle some special cases, but the real special cases would maybe more 
readily turned to manual to keep a simpler generation script.

I do not fancy relying on another representation/language because of Tom's 
objection that it would mean another language to learn, and I do not think 
that it is desirable in pg.

-- 
Fabien.