On Thu, Nov 4, 2021 at 12:23 PM Greg Nancarrow <gregn4...@gmail.com> wrote: > > On Thu, Nov 4, 2021 at 3:13 PM Amit Kapila <amit.kapil...@gmail.com> wrote: > > > > On further thinking about this, I think we should define the behavior > > of replication among partitioned (on the publisher) and > > non-partitioned (on the subscriber) tables a bit more clearly. > > > > - If the "publish_via_partition_root" is set for a publication then we > > can always replicate to the table with the same name as the root table > > in publisher. > > - If the "publish_via_partition_root" is *not* set for a publication > > then we can always replicate to the tables with the same name as the > > non-root tables in publisher. > > > > Thoughts? > > > > I'd adjust that wording slightly, because "we can always replicate to > ..." sounds a bit vague, and saying that an option is set or not set > could be misinterpreted, as the option could be "set" to false. > > How about: > > - If "publish_via_partition_root" is true for a publication, then data > is replicated to the table with the same name as the root (i.e. > partitioned) table in the publisher. > - If "publish_via_partition_root" is false (the default) for a > publication, then data is replicated to tables with the same name as > the non-root (i.e. partition) tables in the publisher. >
Sounds good to me. If we follow this then I think the patch by Hou-San is good to solve the first problem as described in his last email [1]? [1] - https://www.postgresql.org/message-id/OS0PR01MB5716C756312959F293A822C794869%40OS0PR01MB5716.jpnprd01.prod.outlook.com -- With Regards, Amit Kapila.
