On Mon, Feb 10, 2020 at 1:06 PM Masahiko Sawada <masahiko.saw...@2ndquadrant.com> wrote: > On Mon, 10 Feb 2020 at 12:54, Amit Langote <amitlangot...@gmail.com> wrote: > > > > Sawada-san, > > > > On Mon, Feb 10, 2020 at 12:25 PM Masahiko Sawada > > <masahiko.saw...@2ndquadrant.com> wrote: > > > > > > On Thu, Feb 06, 2020 at 04:25:47PM +0900, Amit Langote wrote: > > > > > > > About the implementation, how about defining a static inline > > > > > > > function, > > > > > > > say is_user_object(), next to FirstNormalObjectId's definition and > > > > > > > make pg_is_user_object() call it? There are a few placed in the > > > > > > > backend code that perform the same computation as > > > > > > > pg_is_user_object(), > > > > > > > which could be changed to use is_user_object() instead. > > > > > > Attached the updated version patch that includes regression tests. And > > > I have registered this to the next commit fest. > > > > Thank you. > > > > Any thoughts on the above suggestion? > > Oops, I had overlooked it. I agree with you. > > How about having it as a macro like: > > #define ObjectIdIsUserObject(oid) ((Oid)(oid) >= FirstNormalObjectId)
I'm fine with a macro. Thanks, Amit