[postgis-devel] CMake Progress?

[Mateusz Loskot]

On 14 June 2012 17:39, Paul Ramsey <pramsey at cleverelephant.ca> wrote:
> Mark, Mats,
> How is progress on a CMake build coming (or not)? Every time I see a
> "mingw64" failure I want to toss my self into a creek. We have at
> least two, maybe three windows build targets (mingw old, mingw new,
> mingw64) and they all have different characteristics! :) It would be
> nice to get that down to two, less mobile, targets, MSVC32 and MSCV64.
> Hoping against hope,

Paul,

I've attached patch which includes development pushed by Mark and myself
during the sprint in Paris, plus some extras I developed in train on
my way back.

Now, I haven't touched it since, due to my sudden crisis in my relationship
with Windows for my FOSS development :)
I wiped out Windows setup from my machine and got back to roots (Unix).
It will take me a while before I have VirtualBox with Windows and
Visual C++ ready.

But, I'm going to push forward the CMake configuration on my Linux box.
This should be some step forward anyway, as it will make it easier to try
building and testing with Visual C++. I think I will be able to do some breaks
for testing at Cadcorp on Windows too.

Long story short, I'm going to work on the CMake configuration to make PostGIS
build, run and test on Linux first. I'll work against my fork of
PostGIS git repo
at GitHub, I think.

Mark, Paul, what you think? Shall we collaborate for this at GitHub?
Perhaps having normal branch in SVN would be better?
I have no strong preferences, but assuming as there are at least two people,
Mark and myself, who aim to work on CMake, it would be good to have place to
push updates

Best regards,
-- 
Mateusz Loskot, http://mateusz.loskot.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: postgis-cmake-mark-mloskot-paris-2012-unfinished.patch
Type: application/octet-stream
Size: 41121 bytes
Desc: not available
URL: <http://lists.osgeo.org/pipermail/postgis-devel/attachments/20120615/f4b627b5/attachment.obj>