From: | Michael Meskes <meskes(at)postgresql(dot)org> |
---|---|
To: | PostgreSQL Hacker <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Library versioning |
Date: | 2000-09-19 21:04:53 |
Message-ID: | 20000919140453.B10602@feivel.credativ.de |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Mon, Sep 18, 2000 at 11:17:19AM -0700, Alfred Perlstein wrote:
> If you re-order an exported emum you are going to complete break binary
> compatibility, either bump the major or add the entries at the end.
The later is what I did so far.
> However there's another problem even if you add at the end, namely
> that if previously the 'default' doesn't describe the new values
> then you also break compatibility, for instance:
> ...
> If you happened to add another error state or something that indicated
> some other action was required you'd also be breaking compatibility.
If the library returns the value, yes. But in my case the library just
accepts two more values. So th eonly problem I can think of is that someone
would try to run a new binary against the old library.
Michael
--
Michael Meskes
Michael(at)Fam-Meskes(dot)De
Go SF 49ers! Go Rhein Fire!
Use Debian GNU/Linux! Use PostgreSQL!
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2000-09-19 21:13:16 | Another hole detected in pg_upgrade |
Previous Message | Matthew | 2000-09-19 20:23:28 | RE: Re: pg_dump tries to do too much per query |