On 8/29/18 2:16 PM, Richard via Boost wrote:
In article <0499fe61-4113-95f1-c56c-d5778b0b0fa2@rrsd.com>, Robert Ramey via Boost
writes: On 8/29/18 9:14 AM, Cem Bassoy via Boost wrote:
I like Edwards proposal. However, not all Boost users are C++ experts, especially those who want to use 'higher-level' boost libraries such as GIL. Comparing Boost's installation processes with Qt's one, many users do not want to be concerned with cloning/compiling/linking issues.
Any boost library should be straight forward to build if it's not header only.
Agreed, but bjam/jam/whatever is an impediment here. Boost is the only thing I've ever encountered that used this. CMake is the most widely used build environment now for C/C++ and the sooner we move to replacing bjam/jam/whatever with CMake, the sooner all this build suffering ends.
If we're distributing something that is problematic to build & test, we have a problem. Let's fix that!
It's been that way for... what, a decade? I have only seen signs of hope more recently that the boost organizing committee decided to bless CMake as the direction to go.
Right - there is a consensus that Boost should support CMake. The BOD has passed some sort of resolution meant to make this happen. The only real requirement is that the proposed solution be available for review. But ... Nothing has happened. Turns out, it's a big job that requires a lot of effort which no one has been prepared to take this on. We need ideas, persons with initiative and insight, and who knows what else. Robert Ramey