Re: CURRENT OF cursor without OIDs
От | Mikheev, Vadim |
---|---|
Тема | Re: CURRENT OF cursor without OIDs |
Дата | |
Msg-id | 3705826352029646A3E91C53F7189E32016752@sectorbase2.sectorbase.com обсуждение исходный текст |
Ответ на | CURRENT OF cursor without OIDs (Ian Lance Taylor <ian@airs.com>) |
Список | pgsql-hackers |
> Oops I'm referring to client side cursors in our ODBC > driver. We have no cross-transaction cursors yet though > I'd like to see a backend cross-transaction cursor also. Ops, sorry. BTW, what are "visibility" rules for ODBC cross-tx cursor? No Repeatable reads, no Serializability? Do you hold some locks over table while cursor opened (I noticed session locking in lmgr recently)? Could ODBC cross-tx cursors be implemented using server cross-tx cursors? > > I think we'll be able to restore old tid along with other tuple > > data from rollback segments, so I don't see any problem from > > osmgr... > > How do we detect the change of tuples from clients ? What version of tuple client must see? New one? > TIDs are invariant under osmgr. xmin is about to be > unreliable for the purpose. Seems I have to learn more about ODBC cross-tx cursors -:( Anyway, *MSQL*, Oracle, Informix - all have osmgr. Do they have cross-tx cursors in their ODBC drivers? Vadim
В списке pgsql-hackers по дате отправления: