Re: [HACKERS] getting rid of SnapshotNow
Alvaro Herrera <alvherre@2ndquadrant.com>
From: Alvaro Herrera <alvherre@2ndquadrant.com>
To: Andres Freund <andres@2ndquadrant.com>
Cc: Robert Haas <robertmhaas@gmail.com>, Tom Lane <tgl@sss.pgh.pa.us>,
pgsql-hackers@postgresql.org, pgsql-odbc@postgresql.org
Date: 2013-07-18T16:25:08Z
Lists: pgsql-hackers
Andres Freund escribió: > On 2013-07-18 12:01:39 -0400, Robert Haas wrote: > > On Thu, Jul 18, 2013 at 11:54 AM, Alvaro Herrera > > <alvherre@2ndquadrant.com> wrote: > > > They don't show up in a quick grep of psqlodbc's source code, FWIW. > > > > Hmm. Maybe we should just remove them and see if anyone complains. > > We could always put them back (or make them available via contrib) if > > it's functionality someone actually needs. The last discussion of > > those functions was in 2007 and nobody seemed too sure back then > > either, so maybe the rumor that anyone is actually using this is no > > more than rumor. > > I am pretty sure they are still used. A quick grep on a not too old > checkout prooves that... Note that the sql accessible functions are > named currtid and currtid2 (yes, really)... Ah, yeah, that does show up. I had grepped for 'currtid_'. Sorry. They're all in positioned_load() in results.c. -- Álvaro Herrera http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services