[Top][All Lists]

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

Re: AC_SUBST usage to define include or linker flags

From: Reuben Thomas
Subject: Re: AC_SUBST usage to define include or linker flags
Date: Fri, 23 Oct 2009 00:37:12 +0100

2009/10/22 Dustin J. Mitchell <address@hidden>:
> On Thu, Oct 22, 2009 at 8:44 AM, Pierre-Henri Trivier
> <address@hidden> wrote:
>> However, not all m4 files do it, and not all of them use the same
>> conventions. So my questions are :
> The archive has really only just started to get into the business of
> defining conventions, and in fact this was one of the topics we have
> been discussing.  The method you describe (XXX_INCLUDES, etc.) is,
> IMHO, most flexible.  Macros which simply add their arguments to the
> global CPPFLAGS, LIBS, etc. tend to over-link projects.  For example,
> if one executable in a project requires libxslt, then only that
> executable should be linked with XSLT_LIBS; but if the xslt macro puts
> -lxslt in the global LIBS, then every executable ends up with an
> unused dependency.
>> * is it considered good pratice to use AC_SUBST to define such symbols ? If
>> not, how are you supposed to transfer the information of "I found the
>> include files, and they are in this folder" or "I found the lib, and it is
>> in that folder".
> Yes, AC_SUBST and AC_DEFINE are the main ways that autoconf scripts
> communicate with the build process.

I think the real question here was whether the macros are expected to
AC_SUBST variables of this sort or whether it's up to the user's usually.

Belief marks the line at which our thinking stops (Carse)

reply via email to

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