On Thu, Mar 30, 2023 at 11:12 AM Peter Smith wrote:
>
> > >
> > > 5.
> > > AFAIK, the "if (change->data.tp.oldtuple)" can only be true for UPDATE
> > > or DELETE, so the code would be better to include a sanity Assert.
> > >
> > > SUGGESTION
> > > if (change->data.tp.oldtuple)
> > > {
> > >
Hi Hou-san,
I looked again at v4-0001.
On Thu, Mar 30, 2023 at 2:01 PM houzj.f...@fujitsu.com
wrote:
>
> On Thursday, March 30, 2023 9:15 AM Peter Smith wrote:
> >
...
> >
> > 2.
> > /* Convert tuple if needed. */
> > if (relentry-> attrmap)
> > {
> > ...
> > }
> >
> > The "Convert tuple
On Thursday, March 30, 2023 9:15 AM Peter Smith wrote:
>
> Hi Hou-san,
>
> I tried to compare the logic of patch v3-0001 versus the original HEAD code.
>
> IMO this patch logic is not exactly the same as before -- there are
> some subtle differences. I am not sure if these differences
Hi Hou-san,
I tried to compare the logic of patch v3-0001 versus the original HEAD code.
IMO this patch logic is not exactly the same as before -- there are
some subtle differences. I am not sure if these differences represent
real problems or not.
Below are all my review comments:
==
1.
On Monday, March 20, 2023 5:20 pmhouzj.f...@fujitsu.com wrote:
>
> On Thursday, March 16, 2023 12:30 PM Amit Kapila
> wrote:
>
> >
> > On Wed, Mar 15, 2023 at 2:00 PM houzj.f...@fujitsu.com
> > wrote:
> > >
> > > I noticed that there are some duplicated codes in pgoutput_change()
> > function
On Friday, March 17, 2023 11:49 AM Peter Smith wrote:
>
> On Wed, Mar 15, 2023 at 7:30 PM houzj.f...@fujitsu.com
> wrote:
> >
> > Hi,
> >
> > I noticed that there are some duplicated codes in pgoutput_change()
> > function which can be simplified, and here is an attempt to do that.
>
> Hi
On Thursday, March 16, 2023 12:30 PM Amit Kapila
wrote:
>
> On Wed, Mar 15, 2023 at 2:00 PM houzj.f...@fujitsu.com
> wrote:
> >
> > I noticed that there are some duplicated codes in pgoutput_change()
> function
> > which can be simplified, and here is an attempt to do that.
> >
>
> For
On Wed, Mar 15, 2023 at 7:30 PM houzj.f...@fujitsu.com
wrote:
>
> Hi,
>
> I noticed that there are some duplicated codes in pgoutput_change() function
> which can be simplified, and here is an attempt to do that.
>
> Best Regards,
> Hou Zhijie
Hi Hou-san.
I had a quick look at the 0001 patch.
On Wed, Mar 15, 2023 at 2:00 PM houzj.f...@fujitsu.com
wrote:
>
> I noticed that there are some duplicated codes in pgoutput_change() function
> which can be simplified, and here is an attempt to do that.
>
For REORDER_BUFFER_CHANGE_DELETE, when the old tuple is missing, after
this patch, we
