On 04.12.24 05:55, jian he wrote:
> On Fri, Nov 29, 2024 at 6:13 PM Peter Eisentraut <peter@eisentraut.org> wrote:
>>
>> - Added support for virtual columns in trigger column lists. (For that,
>> I renamed ExecInitStoredGenerated() to ExecInitGenerated(), which
>> handles the computation of ri_extraUpdatedCols.)
>>
>
> why not duplicate some code from ExecInitStoredGenerated to
> ExecGetExtraUpdatedCols?
This answers itself: I'd rather not duplicate code. I don't see that as
an improvement.
> * now the expression is that something initiated for the virtual
> generated column. which may not be necessary for virtual columns.
> let's make ResultRelInfo->ri_GeneratedExprsI,
> ResultRelInfo->ri_GeneratedExprsU be NULL for virtual columns.
>
> currently it may look like this:
> (gdb) p resultRelInfo->ri_GeneratedExprsU
> $20 = (ExprState **) 0x34f9638
> (gdb) p resultRelInfo->ri_GeneratedExprsU[0]
> $21 = (ExprState *) 0x0
> (gdb) p resultRelInfo->ri_GeneratedExprsU[1]
> $22 = (ExprState *) 0x0
> (gdb) p resultRelInfo->ri_GeneratedExprsU[2]
> $23 = (ExprState *) 0x40
I have fixed that in v11.
> * ExecInitStoredGenerated main used in ExecComputeStoredGenerated.
> * we also need to slightly change ExecInitGenerated's comments.
also fixed
> * in InitResultRelInfo, do we need explicit set ri_Generated_valid to false?
Doesn't seem necessary. The struct is initialized to zero at the beginning.
