Re: Walker/mutator prototype.
| От | Tom Lane |
|---|---|
| Тема | Re: Walker/mutator prototype. |
| Дата | |
| Msg-id | 3603.1071445501@sss.pgh.pa.us обсуждение |
| Ответ на | Re: Walker/mutator prototype. (Greg Stark <gsstark@mit.edu>) |
| Список | pgsql-hackers |
Greg Stark <gsstark@mit.edu> writes:
> Kurt Roeckx <Q@ping.be> writes:
>> I did start by changing all the context's to void *, but you'll
>> loose the real type that it gets called with, so the other calls
>> will not generate warnings anymore because of wrong type.
> But at least you'll get a warning if someone passes a non-pointer or an
> incorrect number of arguments altogether.
Note that in practice, the walker/mutator routines are not called from
random places, but by a *very* small number of macros used in clauses.c.
Thus, the probability that someone will introduce a bug into the call
sites is small, and the probability that they'd not discover it
instantly is even smaller.
Given that consideration, I don't see what the point is of trying to
tighten these prototypes. ISTM it adds notational clutter for
essentially zero gain.
regards, tom lane
В списке pgsql-hackers по дате отправления: