Re: [PATCH] Add pg_get_trigger_ddl() to retrieve the CREATE TRIGGER statement
Josef Šimánek <josef.simanek@gmail.com>
From: Josef Šimánek <josef.simanek@gmail.com>
To: Philip Alger <paalger0@gmail.com>
Cc: Josef Šimánek <retro@ballgag.cz>, Chao Li <li.evan.chao@gmail.com>, jian he <jian.universality@gmail.com>, Jim Jones <jim.jones@uni-muenster.de>, Andrew Dunstan <andrew@dunslane.net>, Cary Huang <cary.huang@highgo.ca>, pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2025-11-03T01:31:52Z
Lists: pgsql-hackers
po 3. 11. 2025 v 2:22 odesílatel Philip Alger <paalger0@gmail.com> napsal:
>
> Hello,
>
>
>>
>> > I think it’s better to pfree(res).
>>
>> Would you mind to share why pfree is needed? I tried to trace this
>> with Valgrind, but even pfree(res) was present or not, there was no
>> leak detected and both compiles without additional warnings. Wouldn't
>> be res "trashed" at the end of the function (after next line) anyway?
>
>
> The wrapper function string_to_text, which is a wrapper for cstring_to_text, includes pfree.
>
> see ruleutils.c
>
> static text *
> string_to_text(char *str)
> {
> text *result;
>
> result = cstring_to_text(str);
> pfree(str);
> return result;
> }
Indeed, IMHO pfree is not needed and the current patch is good as is.
>
> --
> Best,
> Phil Alger