Re: Proposed syntax for coincidental rules (was: Re: GNU make source code

[Paul D. Smith]

%% Alessandro Vesely <address@hidden> writes:

  av> Paul D. Smith wrote:
  >> [...]
  >> Anyway, a critical feature of GNU make is that it's conformant to the
  >> POSIX standard for make.  Any new feature which broke that compatibility
  >> would need to be considered very critically before being accepted.

  av> Oh well, I'm not going to pay $900+ to look at POSIX specs.
  av> However, the SUS is readable for free at the opengroup.org's.

That's what I use.

  av> From their prototypical man page for make it is fairly clear
  av> that Henning Makholm's proposed syntax was not coherent. See
  av> "Proposed syntax for static make-also rule"
  av> http://lists.gnu.org/archive/html/make-alpha/2002-12/msg00000.html

I'm not sure what you mean by "coherent".

Let's be clear about what conformance to the POSIX standard means (at
least to me): it doesn't mean that every makefile written for GNU make
should also be parseable in some way by a POSIX implementation of make.
Clearly, GNU make already fails that requirement just by having "include".


What it means is much weaker: that any makefile that is written to
conform to the POSIX spec, should be correctly parsed and processed by
GNU make.

In that light, Henning's proposal is OK because it won't change the
behavior of any conforming makefile.

-- 
-------------------------------------------------------------------------------
 Paul D. Smith <address@hidden>          Find some GNU make tips at:
 http://www.gnu.org                      http://make.paulandlesley.org
 "Please remain calm...I may be mad, but I am a professional." --Mad Scientist