Re: PQnotifies() in 7.3 broken? - Mailing list pgsql-hackers

From Bruce Momjian
Subject Re: PQnotifies() in 7.3 broken?
Date
Msg-id 200212131913.gBDJDLb24382@candle.pha.pa.us
Whole thread Raw
In response to Re: PQnotifies() in 7.3 broken?  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: PQnotifies() in 7.3 broken?
List pgsql-hackers
Tom Lane wrote:
> Bruce Momjian <pgman@candle.pha.pa.us> writes:
> > The point is that we are changing it for 7.3.1, so though 7.3 libpq is
> > almost identical to 7.3.1 libpq, we are going to bump the major and
> > force recompile.  The binary API change was from 7.2 to 7.3, not 7.3 to
> > 7.3.1.  Do people still want a major bump in 7.3.1?
> 
> Yes.  We already discussed this: the consensus was that it's better to
> inconvenience the early adopters of 7.3 than leave a time bomb sitting
> there throughout the 7.3.* cycle.
> 
> BTW, if we do make a major bump, that would suggest we ought to get
> 7.3.1 out as soon as possible, so as to minimize the uptake of 7.3.
> I have been thinking that a mid-Dec release of 7.3.1 would be a good
> plan anyway, since we already know of several moderately serious bugs
> in 7.3.

OK, let me see if I understand the ramifications.

If you install 7.3.1 _on_top_of 7.3, both major versions will exist, and
you your old binaries will continue to work.  However, if you delete the
old libraries, then install, anything compiled against 7.3 will not work
until it is recompiled.

Also, any new linking against a 7.3.1 that has both major version
numbers will use the newer major?  Is that right?

(I honestly never understood all this before.)

--  Bruce Momjian                        |  http://candle.pha.pa.us pgman@candle.pha.pa.us               |  (610)
359-1001+  If your life is a hard drive,     |  13 Roberts Road +  Christ can be your backup.        |  Newtown Square,
Pennsylvania19073
 


pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: Big 7.4 items
Next
From: "Mike Mascari"
Date:
Subject: Re: Big 7.4 items