gpsd-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [gpsd-dev] cmake:


From: Matt
Subject: Re: [gpsd-dev] cmake:
Date: Fri, 23 Jan 2015 19:36:17 +1100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Icedove/31.3.0

At risk of adding fuel to the flames, use of scons is also a problem when trying to compile for mingw, as porting scons and all of its dependencies to mingw appears to be a mammoth task, from which I'd never return to work on gpsd.

In the past I just wrote a Makefile.mingw . It worked fine, it ended up being clean and simple, and it was much, much easier than any other way I can see to create a mingw executable.

Bernd, why couldn't you just create a Makefile and use it instead of scons if you wanted to?

If it's OK to modify upstream source before packaging and distributing the resultant binaries, surely it's even more OK to change the build system. Does your choice of build system have to be the same as upstream's for some reason?

Please forgive me if I am ignorant of Debian policy on this matter.

On 15/01/15 01:34, Eric S. Raymond wrote:
Bernd Zeimetz <address@hidden>:
My problem with scons is that I wasted hours of my life in trying to figure
out things in scons that would just work out of the box in a correct way.
Or that would have been easy to debug in other build systems by reading
a stupid Makefile...
Pretty much anything other than bare makefiles would have presented
you with the same "out of the box" problem, though.  Can you honestly
say the complexity barrier to entry of cmake is any lower than scons
presents to a new user?  I doubt it...

Separately, I wish I had only wasted hours of my life on autotools
since 1991.  It has to have run into *weeks* of my life by now.  That
felt like months.

I'm sorry you had a bad initial experience with scons. Can you identify
specific design decisions or missing features that give you problems?




reply via email to

[Prev in Thread] Current Thread [Next in Thread]