[postgis-users] st_union says geometries have different srids
pramsey at cleverelephant.ca
Mon Mar 17 11:54:25 PDT 2008
NP, the architecture is a red herring, since the reporter saw the
problem in 64-bit land and I have now seen it on two different 32-bit
compilations (OS/X (thanks for the explanation!) and Linux)
On 3/17/08, William Kyngesburye <woklist at kyngchaos.com> wrote:
> On Mar 17, 2008, at 12:10 PM, Paul Ramsey wrote:
> >> Nope not yet; I'm still in the process of setting up various build
> >> environments. Does OS X support "uname -a" to show exactly what
> >> architecture
> >> you are running on?
> > Heron:~ pramsey$ uname -a
> > Darwin Heron.local 9.2.0 Darwin Kernel Version 9.2.0: Tue Feb 5
> > 16:13:22 PST 2008; root:xnu-1228.3.13~1/RELEASE_I386 i386
> > Nothing definitive that I can see. I'm putting together a centos
> > virtual box now to see if the test fails there too.
> OSX 10.5 is both 32bit and 64bit, but it also depends on the processor
> - PPCs are all 64bit capable, but can run in 32bit mode (all previous
> OSX and Classic versions do this), core duos on the first Intel models
> are 32bit, core 2 duos are 64bit. The 10.5 system includes all of
> these in one fat binary.
> BUT, a default compile on 10.5 builds 32bit binaries, so unless you
> explicitly said to build 64bit, your Postgres and PostGIS are 32bit.
> Or, are you using my binaries? Those include 64bit and 32bit
> binaries, and will run 64bit when possible.
> I have a Postgres 8.2 x86_64 for OSX 10.5 on my Postgres page with
> matching PostGIS 1.3.2, if you want to test it. I've upgraded my 10.5
> Mac to PG 8.3 already and don't have a spare to install 8.2.
> William Kyngesburye <kyngchaos*at*kyngchaos*dot*com>
> "Oh, look, I seem to have fallen down a deep, dark hole. Now what
> does that remind me of? Ah, yes - life."
> - Marvin
More information about the postgis-users