Re: What to name the current heap after pluggable storage / what torename?
От | Andres Freund |
---|---|
Тема | Re: What to name the current heap after pluggable storage / what torename? |
Дата | |
Msg-id | 20190319050821.kc4iwyos64jqw7rl@alap3.anarazel.de обсуждение исходный текст |
Ответ на | Re: What to name the current heap after pluggable storage / what to rename? (Haribabu Kommi <kommi.haribabu@gmail.com>) |
Список | pgsql-hackers |
Hi, On 2019-03-19 15:25:44 +1100, Haribabu Kommi wrote: > On Tue, Mar 19, 2019 at 2:32 PM Andres Freund <andres@anarazel.de> wrote: > > I'm kinda wondering about replacing the TableTuple prefix with TableMod, > > seems less confusing to me. > One more way, how about just TupleUpdated and etc. Removing of Table? > The structure name also suggests as TM (IMO, it is TupleModication?) Enum members are in the global namespace. I could go for a TM_ prefix however. > I'm also wondering about replacing > > *MayBeModified with *OK. > How about TupleModified? I am fine with *OK also. Hm, OK just seems nicer for lock_tuple, that's why I was wondering about it, but I think my concern there is a bit pedantic. > > Right now I have > > > > typedef enum TM_Result HTSU_Result; > > > > #define HeapTupleMayBeUpdated TableTupleMayBeModified > > #define HeapTupleInvisible TableTupleInvisible > > #define HeapTupleSelfUpdated TableTupleSelfModified > > #define HeapTupleUpdated TableTupleUpdated > > #define HeapTupleDeleted TableTupleDeleted > > #define HeapTupleBeingUpdated TableTupleBeingModified > > #define HeapTupleWouldBlock TableTupleWouldBlock > > > > in heapam.h (whereas the above is in tableam.h), for backward > > compat. But I'm not sure it's worth it. > > > > These old macros are pretty much used in the internal code, and I doubt > that any one depends directly on those macros. I vote for removal of > these backward compatibility macros. Cool. Greetings, Andres Freund
В списке pgsql-hackers по дате отправления: