On Mon, Feb 16, 2026 at 03:24:16PM +0530, Amit Kapila wrote: > On Fri, Feb 13, 2026 at 3:42 AM Bruce Momjian <[email protected]> wrote: > > pg_upgrade had a similar challenge, and I found that pushing as much of > > the changes _out_ of pg_upgrade and to other parts of the system, e.g,, > > pg_dump, was a big help. I am not sure if that is possible for > > replicated DDL, but if it is, I would pursue it. > > > > The other reason to pursue a deparsing approach is to allow "Schema > and Object Redirection" between source and target. For example, users > may want to redirect operations from a source object (e.g., > HR.ACCOUNTS on the source) to a different target object (e.g., > HR_BACKUP.ACCT on the target). Many people in the past supported such > flexibility in DDL replication which is one of the reasons we pursued > a deparsing approach in the past though that project/approach is > paused for the time being. I admit that there were many unresolved > challenges apart from the maintenance part of that approach.
I understand the desire to get two features from one set of changes, but taking one hard feature and trying to get a second one accomplished usually, for me, ends in failure. -- Bruce Momjian <[email protected]> https://momjian.us EDB https://enterprisedb.com Do not let urgent matters crowd out time for investment in the future.
