[postgis-devel] Moving towards a 1.2.2 release?

[Mark Cave-Ayland]

On Tue, 2007-05-15 at 20:02 -0600, Michael Fuhr wrote:
> On Mon, May 14, 2007 at 07:16:02PM +0100, Mark Cave-Ayland wrote:
> > 3. Compiling and regression testing against a PostgreSQL 8.3 tree. Most
> > versions tend to break the index code, and I'd be surprised if 8.3 was
> > any different. There are likely to be some small fixes needed here.
> 
> Compiling against 8.3devel fails:
> 
> lwgeom_geos_c.c: In function `postgis_geos_version':
> lwgeom_geos_c.c:84: warning: implicit declaration of function `VARATT_SIZEP'
> lwgeom_geos_c.c:84: error: invalid lvalue in assignment
> 
> I think this is the relevant commit:
> 
> http://archives.postgresql.org/pgsql-committers/2007-02/msg00517.php
> 
> I haven't followed the discussion surrounding that commit enough
> to know how best to write code that must compile against 8.3 and
> earlier versions.  The attached patch gets PostGIS to compile and
> pass all regression tests against 8.3devel but it will break builds
> against earlier versions.  In a few places I replaced the constant 4
> with VARHDRSZ but somebody more familiar with the code might want to
> evaluate whether VARHDRSZ is indeed what's intended.


Hi Michael,

Thanks for this! I had in the back of my head that there would be two
changes that may affect PostGIS - one was the VARLENA macros above, and
another was a change in the index opclasses. However, if it passes
regression for you then that's great.

I'll have a look at this over the next few days, and get it into a form
that we can commit to SVN.


Many thanks,

Mark.

-- 
ILande - Open Source Consultancy
http://www.ilande.co.uk