On Sun, Oct 11, 2026 at 11:12 PM Amit Kapila <[email protected]> wrote: > > On Sun, Oct 11, 2026 at 2:37 AM Dilip Kumar <[email protected]> wrote: > > > > Okay that makes sense. Here is the updated version of the patch > > changes compared to previous version > > > > Thanks for the new version. We can combine 0001 and 0002 now but most > importantly you seem to forget including the code optimizations > suggested by Hou-San and me in emails [1][2]. > > [1] - > https://www.postgresql.org/message-id/CAFvd2n92EXzt7VNRGxQCMkK1t2%2B6RA9SwQ5j9DHMwJcWCNoK1Q%40mail.gmail.com > [2] - > https://www.postgresql.org/message-id/CAA4eK1L-yBio2zbSVsKQ9LCy8orszXviqORHp%3Dsg2jSq%3D76D2A%40mail.gmail.com >
Okay, I missed them, I think both of these optimizations make sense, especially hinding the complexity of decision on conflict log destination inside get_conflictlog_dest_and_table() simplify the code to a great extent. So I will include them in the next version and will also merge 0001 and 0002 -- Regards, Dilip Kumar Google
