Re: handling of heap rewrites in logical decoding

On 16 March 2018 at 10:54, Craig Ringer <craig(at)2ndquadrant(dot)com> wrote:

> On 16 March 2018 at 08:51, Peter Eisentraut <peter(dot)eisentraut(at)2ndquadrant(dot)
> com> wrote:
>
>> On 2/28/18 13:52, Peter Eisentraut wrote:> Second version, which uses an
>> OID. I added another field to the output> plugin options (next to the
>> output_type), to indicate whether the plugin> wants to receive these
>> changes. I added some test cases to> test_decoding to show how it works
>> either way. It's a bit messy to pass> this setting through to the
>> ReorderBuffer; maybe there is a better idea.> But the result seems
>> pretty useful.
>> Here is a rebased update of this patch. No functionality changes
>> compared to v2.
>>
>
>
> Picking this up for review.
>
>
Why was relrewrite / Anum_pg_class_relrewrite inserted at position 28, not
appended to pg_class?

I don't see how it'd cause any harm, I'm just curious about the rationale.

Otherwise, I'm totally happy with what I see here. It's always nice to get
to the end of a patch and think "wait, that's all?"

Ready to go.

--
Craig Ringer http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services