Re: Berkeley DB 3.2.9

[Karl Fogel <kf...@galois.collab.net>]

Jim Blandy <ji...@zwingli.cygnus.com> writes:
> The Berkeley DB 3.2.9 API isn't compatible with 3.1.17.  From the
> 3.2.9 ``ChangeLog'':
> 
> Interface Additions and Changes:
> 
>     6.Add a reference to a DB object as the first argument to four application callback functions (the callback functions set by
>        DB->set_bt_compare, DB->set_bt_prefix, DB->set_dup_compare, and
> DB->set_h_hash). [#2179] 
> 
> This means that every FS btree that uses a customized sort order needs
> to have its code changed slightly.  I've included a patch below.
> 
> Should we simply require Berkeley DB 3.2.9, and declare 3.1.17
> obsolete?

I think so.  Temporary inconvenience for us now, but it is after all
the wave of the future...

If you want to check in that patch (I think you forgot to include it
in your mail), upgrading shouldn't be too painful for the rest of us.
All tests are known to pass w/ the patch and BDB 3.2.9, right?

Re: Berkeley DB 3.2.9

[Jim Blandy <ji...@zwingli.cygnus.com>]

Okay --- as of the change I just committed to nodes-table.c, the
Subversion filesystem now requires Berkeley DB 3.2.9.  And with some
other changes I'll be committing in a few minutes, it actually works.

Greg Stein <gs...@lyra.org> writes:

> 
> On Thu, Mar 01, 2001 at 12:41:23PM -0600, Karl Fogel wrote:
> > Jim Blandy <ji...@zwingli.cygnus.com> writes:
> >...
> > > Should we simply require Berkeley DB 3.2.9, and declare 3.1.17
> > > obsolete?
> > 
> > I think so.  Temporary inconvenience for us now, but it is after all
> > the wave of the future...
> > 
> > If you want to check in that patch (I think you forgot to include it
> > in your mail), upgrading shouldn't be too painful for the rest of us.
> > All tests are known to pass w/ the patch and BDB 3.2.9, right?
> 
> +1 on upgrading. Let's take the hit before M2.
> 
> Cheers,
> -g
> 
> -- 
> Greg Stein, http://www.lyra.org/
>

Re: Berkeley DB 3.2.9

[Greg Stein <gs...@lyra.org>]

On Thu, Mar 01, 2001 at 12:41:23PM -0600, Karl Fogel wrote:
> Jim Blandy <ji...@zwingli.cygnus.com> writes:
>...
> > Should we simply require Berkeley DB 3.2.9, and declare 3.1.17
> > obsolete?
> 
> I think so.  Temporary inconvenience for us now, but it is after all
> the wave of the future...
> 
> If you want to check in that patch (I think you forgot to include it
> in your mail), upgrading shouldn't be too painful for the rest of us.
> All tests are known to pass w/ the patch and BDB 3.2.9, right?

+1 on upgrading. Let's take the hit before M2.

Cheers,
-g

-- 
Greg Stein, http://www.lyra.org/