Hi Craig,
On 01/09/16 06:08, Craig Ringer wrote:
> Hi all
>
> Attached is a rebased and updated logical decoding timeline following
> patch for 10.0.
>
> This is a pre-requisite for the pending work on logical decoding on
> standby servers and simplified failover of logical decoding.
>
I went over this and it looks fine to me, I only rebased the patch on
top of current version (we renamed pg_xlog which broke the tests) and
split the test harness to separate patch. Otherwise I would consider
this to be ready for committer.
I think this should go in early so that there is enough time in the
cycle to uncover potential issues if there are any, even though it looks
all correct to me.
>
> The test harness code will become unnecessary when proper support for
> logical failover or logical decoding on standby is added, so I'm not
> really sure it should be committed.
Yeah as I said above I split out the test harness for this reason. The
good thing is that when the followup patches get in the test harness
should be easy to removed as the changes are very localized.
--
Petr Jelinek http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services