On 10/11/07, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> "Marko Kreen" <markokr@gmail.com> writes:
> > Could you describe bit more? The is_visible_txid() works
> > on data returned by txid_current_snapshot()? How can there
> > be any subtrans id's if txid_current_snapshot() wont return
> > them?
>
> Ah, I see: txid_current() never reports a subxact ID so there's no need to
> consider them elsewhere in txids either. OK, but this desperately needs
> to be documented.
Will do.
> BTW, I notice that use of txid_current will force assignment of an XID
> in a transaction that might not otherwise have one. Does this matter,
> or is the expectation that it's only going to be used in transactions
> that are making DB modifications anyway?
Yes, the behaviour is fine - it is meant to be used in transactions
that do modifications. Even if not, the lazy xid assignment should
stay internal optimization detail of backend and should not be
exposed to users that clearly.
--
marko