Could you elaborate on which configurations CMake current can NOT support and the failure mechanism under CMake? What are the motivations for moving to CMake 3.10? (I like new versions of CMake but does moving to CMake 3.10 actually solve any problems? It definitely helps on the Fortran side of things)
(Beating the dead horse...) and having a public place where developers can track the issue and provide fixes would help the community help you. At the moment, I have no way of knowing if someone else has tried to fix those issues or not. I would hate to put in a bunch of time on a fix and _then_ learn that someone else was already working on it.
Michael Jackson | Owner, President
Michael A. Jackson 400 S. Pioneer Blvd
Owner, President Springboro, Ohio 45066
BlueQuartz Software, LLC EMail: email@example.com
Voice: 937-790-1601 Web: http://www.bluequartz.net
From: Hdf-forum <firstname.lastname@example.org> on behalf of Elena Pourmal <email@example.com>
Reply-To: HDF Users Discussion List <firstname.lastname@example.org>
Date: Monday, January 8, 2018 at 2:29 PM
To: HDF Users Discussion List <email@example.com>
Subject: [Hdf-forum] Moving to CMake 3.10.* : Re: Issues with INSTALL using NMake Makefiles & CMake (1.8.19, 1.8.20)
This year The HDF Group will be moving HDF software from CMake 3.2.* to CMake 3.10.1.
The reason is we are having too many issues reported with the current CMake version required to build HDF (HDF5, HDF4, HDFView, misc. tools, etc.)
Originally CMake was supported only for Windows, but now we have more and more users using CMake for building parallel, and Fortran configurations on non-Windows platforms. I.e., we need to come up with the solution that covers not only Windows and basic configurations of HDF. Said this we are welcoming any patches that will help us to finish the task.
We will keep you updated on our progress, and as usual, we will have updated software available for testing before the releases.