Re: Refactoring: Use soft error reporting for *_opt_error functions

On Mon, Sep 01, 2025 at 12:21:18PM +0100, Dean Rasheed wrote:
> On Mon, 1 Sept 2025 at 10:36, Amul Sul <sulamul(at)gmail(dot)com> wrote:
>> I believe we should update all *_opt_error functions to use the new
>> soft error reporting infrastructure instead of boolean flags -- did
>> the same in the attached patch. I am not sure if this patch should be
>> part of that thread[1]. It's a significant improvement in itself, as
>> it would make the code more compact and consistent.

Handling that as a separate patch seems OK here. Thanks for caring.

> Agreed. That does look neater.

Yep. More consistent.

+/* forward declarations to avoid node.h include */
+typedef struct Node Node;

s/declarations/declaration/. Singular required, only one declaration.

Looking at the surroundings, would it make sense to do the same for
pg_lsn_in_internal()? It requires a safe fallback when parsing the
LSN from the recovery_target_lsn GUC.
--
Michael