On Tue, Jul 19, 2022 at 3:38 PM Andres Freund <andres@anarazel.de> wrote:
> Or alternatively, perhaps we can just make pg_clean_ascii() return NULL
> if allocation failed and then guc_strdup() the result in guc.c?
The guc_strdup() approach really reduces the amount of code, so that's
what I did in v3. I'm not following why we need to return NULL on
failure, though -- both palloc() and guc_malloc() ERROR on failure, so
is it okay to keep those semantics the same?
> If we end up needing a two phase approach, why use the same function for
> both phases? That seems quite awkward.
Mostly so the byte counting always agrees between the two phases, no
matter how the implementation evolves. But it's hopefully moot now.
--Jacob