Horton

Thread

Commits

  1. Deduplicate handling of binary and text modes in logicalrep_read_tuple().

  2. Allow logical replication to transfer data in binary format.

  1. Deduplicate logicalrep_read_tuple()

    Bharath Rupireddy <bharath.rupireddyforpostgres@gmail.com> — 2023-01-18T07:26:06Z

    Hi,

logicalrep_read_tuple() duplicates code for LOGICALREP_COLUMN_TEXT and
LOGICALREP_COLUMN_BINARY introduced by commit 9de77b5. While it
doesn't hurt anyone, deduplication makes code a bit leaner by 57 bytes
[1]. I've attached a patch for $SUBJECT.

Thoughts?

[1] size ./src/backend/replication/logical/proto.o
PATCHED:
   text    data     bss     dec     hex filename
  15558       0       0   15558    3cc6
./src/backend/replication/logical/proto.o

HEAD:
   text    data     bss     dec     hex filename
  15615       0       0   15615    3cff
./src/backend/replication/logical/proto.o

--
Bharath Rupireddy
PostgreSQL Contributors Team
RDS Open Source Databases
Amazon Web Services: https://aws.amazon.com

  2. Re: Deduplicate logicalrep_read_tuple()

    Peter Smith <smithpb2250@gmail.com> — 2023-01-19T03:06:26Z

    On Wed, Jan 18, 2023 at 6:26 PM Bharath Rupireddy
<bharath.rupireddyforpostgres@gmail.com> wrote:
>
> Hi,
>
> logicalrep_read_tuple() duplicates code for LOGICALREP_COLUMN_TEXT and
> LOGICALREP_COLUMN_BINARY introduced by commit 9de77b5. While it
> doesn't hurt anyone, deduplication makes code a bit leaner by 57 bytes
> [1]. I've attached a patch for $SUBJECT.
>
> Thoughts?
>

The code looks the same but there is a subtle comment difference where
previously only LOGICALREP_COLUMN_BINARY case said:
 /* not strictly necessary but per StringInfo practice */

So if you de-duplicate the code then should that comment be modified to say
/* not strictly necessary for LOGICALREP_COLUMN_BINARY but per
StringInfo practice */

------
Kind Regards,
Peter Smith.
Fujitsu Australia

  3. Re: Deduplicate logicalrep_read_tuple()

    Bharath Rupireddy <bharath.rupireddyforpostgres@gmail.com> — 2023-03-03T10:43:23Z

    On Thu, Jan 19, 2023 at 8:36 AM Peter Smith <smithpb2250@gmail.com> wrote:
>
> On Wed, Jan 18, 2023 at 6:26 PM Bharath Rupireddy
> <bharath.rupireddyforpostgres@gmail.com> wrote:
> >
> > Hi,
> >
> > logicalrep_read_tuple() duplicates code for LOGICALREP_COLUMN_TEXT and
> > LOGICALREP_COLUMN_BINARY introduced by commit 9de77b5. While it
> > doesn't hurt anyone, deduplication makes code a bit leaner by 57 bytes
> > [1]. I've attached a patch for $SUBJECT.
> >
> > Thoughts?
> >
>
> The code looks the same but there is a subtle comment difference where
> previously only LOGICALREP_COLUMN_BINARY case said:
>  /* not strictly necessary but per StringInfo practice */
>
> So if you de-duplicate the code then should that comment be modified to say
> /* not strictly necessary for LOGICALREP_COLUMN_BINARY but per
> StringInfo practice */

Thanks. Done so in the attached v2.

-- 
Bharath Rupireddy
PostgreSQL Contributors Team
RDS Open Source Databases
Amazon Web Services: https://aws.amazon.com

  4. Re: Deduplicate logicalrep_read_tuple()

    Amit Kapila <amit.kapila16@gmail.com> — 2023-03-03T11:04:30Z

    On Fri, Mar 3, 2023 at 4:13 PM Bharath Rupireddy
<bharath.rupireddyforpostgres@gmail.com> wrote:
>
> On Thu, Jan 19, 2023 at 8:36 AM Peter Smith <smithpb2250@gmail.com> wrote:
> >
> > On Wed, Jan 18, 2023 at 6:26 PM Bharath Rupireddy
> > <bharath.rupireddyforpostgres@gmail.com> wrote:
> > >
> > > Hi,
> > >
> > > logicalrep_read_tuple() duplicates code for LOGICALREP_COLUMN_TEXT and
> > > LOGICALREP_COLUMN_BINARY introduced by commit 9de77b5. While it
> > > doesn't hurt anyone, deduplication makes code a bit leaner by 57 bytes
> > > [1]. I've attached a patch for $SUBJECT.
> > >
> > > Thoughts?
> > >
> >
> > The code looks the same but there is a subtle comment difference where
> > previously only LOGICALREP_COLUMN_BINARY case said:
> >  /* not strictly necessary but per StringInfo practice */
> >
> > So if you de-duplicate the code then should that comment be modified to say
> > /* not strictly necessary for LOGICALREP_COLUMN_BINARY but per
> > StringInfo practice */
>
> Thanks. Done so in the attached v2.
>

LGTM. Unless Peter or someone has any comments on this, I'll push this
early next week.

-- 
With Regards,
Amit Kapila.

  5. Re: Deduplicate logicalrep_read_tuple()

    Peter Smith <smithpb2250@gmail.com> — 2023-03-05T07:52:19Z

    On Fri, Mar 3, 2023 at 10:04 PM Amit Kapila <amit.kapila16@gmail.com> wrote:
>
> On Fri, Mar 3, 2023 at 4:13 PM Bharath Rupireddy
> <bharath.rupireddyforpostgres@gmail.com> wrote:
> >
> > On Thu, Jan 19, 2023 at 8:36 AM Peter Smith <smithpb2250@gmail.com> wrote:
> > >
> > > On Wed, Jan 18, 2023 at 6:26 PM Bharath Rupireddy
> > > <bharath.rupireddyforpostgres@gmail.com> wrote:
> > > >
> > > > Hi,
> > > >
> > > > logicalrep_read_tuple() duplicates code for LOGICALREP_COLUMN_TEXT and
> > > > LOGICALREP_COLUMN_BINARY introduced by commit 9de77b5. While it
> > > > doesn't hurt anyone, deduplication makes code a bit leaner by 57 bytes
> > > > [1]. I've attached a patch for $SUBJECT.
> > > >
> > > > Thoughts?
> > > >
> > >
> > > The code looks the same but there is a subtle comment difference where
> > > previously only LOGICALREP_COLUMN_BINARY case said:
> > >  /* not strictly necessary but per StringInfo practice */
> > >
> > > So if you de-duplicate the code then should that comment be modified to say
> > > /* not strictly necessary for LOGICALREP_COLUMN_BINARY but per
> > > StringInfo practice */
> >
> > Thanks. Done so in the attached v2.
> >
>
> LGTM. Unless Peter or someone has any comments on this, I'll push this
> early next week.
>

No more comments. Patch v2 LGTM.

------
Kind Regards,
Peter Smith.
Fujitsu Australia

  6. Re: Deduplicate logicalrep_read_tuple()

    Amit Kapila <amit.kapila16@gmail.com> — 2023-03-06T10:44:24Z

    On Sun, Mar 5, 2023 at 1:22 PM Peter Smith <smithpb2250@gmail.com> wrote:
>
> On Fri, Mar 3, 2023 at 10:04 PM Amit Kapila <amit.kapila16@gmail.com> wrote:
> >
> > >
> > > Thanks. Done so in the attached v2.
> > >
> >
> > LGTM. Unless Peter or someone has any comments on this, I'll push this
> > early next week.
> >
>
> No more comments. Patch v2 LGTM.
>

Pushed.

-- 
With Regards,
Amit Kapila.