[CMake] CMake 2.8.1 RC 3 is ready to try

[Alan W. Irwin]

On 2010-02-18 08:52-0800 Alan W. Irwin wrote:

> [The information about CMAKE_MODULE_PATH being a list]
> is going to help a lot in the
> language support transition between 2.6.x and 2.8.x.

To Arjen, Brad, and Bill:

Setting up CMAKE_MODULE_PATH as a list whose contents depended on CMake
version meant quite a few reorganizational changes had to be made to the
PLplot build system, but according to my initial limited tests with
CMake-2.8.1-RC1 on Linux with gfortran compiler (which doesn't really use
any of Arjen's Windows-related Fortran changes) the reorganization appears
to work OK with Fortran, Ada, and D for shared libraries.

If I have interpreted Brad's various discussions with Arjen correctly, all
Arjen's Windows Fortran support files are now redundant (both for CMake-2.6.x
and CMake-2.8.x) except for the Compaq Visual Fortran file for CMake-2.6.x.
Also, he needs a different version of that file (with different name?) for
CMake-2.8.x (which I believe Arjen has already implemented).  But he still
has a number of deletions and one addition to make to his language support
files consistent with the newly reorganized directories for those support
files in the PLplot svn repository, and then he needs to do extensive
testing of that result to make sure all those changes work well for PLplot
fortran both for CMake-2.6.x and CMake-2.8.1-RC3 for various Windows
platforms. I have no idea what Arjen's schedule is like, but if he states
here he does have time to deal with all of this testing right now, it might
be worth Bill's while to wait until Arjen's tests confirm all is well with
PLplot and various forms of Windows Fortran before releasing CMake-2.8.1.

I have a lot more CMake-2.8.1-RC3 testing to do myself because now a
Java-related error has showed up as well as the D issue I mentioned
previously.  So, Bill, forget what I said about finishing my testing by late
this (Thursday) night.  It's all fairly open-ended so I cannot give you a
definite time scale for when I will be finished at this stage, but I do
promise to work steadily on it until I have either solved the issues (if
either/both are mine) or else implemented a simple test case that shows an
issue with CMake-2.8.1-RC3.  Since resolving these issues is open-ended, it
may not be a good idea to delay the release for me, but perhaps you should
get in touch with me off list to discuss where I am in the process once your
other release issues are solved.  (Or if the other issues are just waiting
for Arjen's testing, then you might not wish to wait for either of us.  :-))

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state implementation
for stellar interiors (freeeos.sf.net); PLplot scientific plotting software
package (plplot.org); the libLASi project (unifont.org/lasi); the Loads of
Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________