Re: Refactor StringInfo usage in subscriptioncmds.c - Mailing list pgsql-hackers

On 11/6/25 13:09, Amit Kapila wrote:
> On Thu, Nov 6, 2025 at 5:24 PM David Rowley <dgrowleyml@gmail.com> wrote:
>> On Fri, 7 Nov 2025 at 00:38, Mats Kindahl <mats.kindahl@gmail.com> wrote:
>>> As discussed in [1] the functions check_publications_origin_tables() and
>>> check_publications_origin_sequences() are building error messages using
>>> dynamically allocated StringInfo instances only to avoid duplicating a
>>> call of ereport().
>> Looks better and more traditional to me, plus git diff --stat reports:
>>
>> 1 file changed, 28 insertions(+), 33 deletions(-)
>>
>> It has my vote. I've included Vignesh and Amit to see if they're also
>> ok with it.
>>
> LGTM as well. We can avoid doing pfree(pubnames.data) as it is not
> allocated in any long term context but I am fine if we want to do it
> as there are places where we do such retail pfree as well.

I have no strong opinion either way, but since it is not needed and it 
is not a frequent call it seemed prudent to free it after.

Best wishes,
Mats Kindahl