Mailing-List:	contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Subscribe:	<mailto:cygwin-subscribe AT sourceware DOT cygnus DOT com>
List-Archive:	<http://sourceware.cygnus.com/ml/cygwin/>
List-Post:	<mailto:cygwin AT sourceware DOT cygnus DOT com>
List-Help:	<mailto:cygwin-help AT sourceware DOT cygnus DOT com>, <http://sourceware.cygnus.com/ml/#faqs>
Sender:	cygwin-owner AT sourceware DOT cygnus DOT com
Delivered-To:	mailing list cygwin AT sourceware DOT cygnus DOT com
MIME-Version:	1.0
Message-ID:	<14589.58703.938479.132274@baynetworks.com>
Date:	Wed, 19 Apr 2000 12:56:47 -0400 (EDT)
To:	"Christopher Jones" <cbjones AT nortelnetworks DOT com>
Cc:	Bernard Dautrevaux <Dautrevaux AT microprocess DOT com>,
	cygwin AT sourceware DOT cygnus DOT com, bug-make AT gnu DOT org
Subject:	RE: New Cygwin Net Release (make error)
In-Reply-To:	<C9A8E1D07093D111B76A0000F8C9918A03333053@zrtpd003.us.nortel.com>
References:	<C9A8E1D07093D111B76A0000F8C9918A03333053 AT zrtpd003 DOT us DOT nortel DOT com>
X-Mailer:	VM 6.75 under Emacs 20.6.1
Reply-To:	"Paul D. Smith" <psmith AT gnu DOT org>
From:	"Paul D. Smith" <psmith AT gnu DOT org>
Organization:	GNU's Not Unix!

%% "Christopher Jones" <cbjones AT nortelnetworks DOT com> writes:

HTML isn't necessary on the mailing list; plain text is just fine, thanks.

  >> From Bernard
  >> From Brian

  >>> #--begin Makefile--- 
  >>> SOMEVAR=$(ANOTHERVAR)/etc 
  >>> include $(DEVDIR)/Makefile.mstr 
  >>> #--end Makefile--- 

  >>> DEVDIR is defined in the environment.  ANOTHERVAR however is
  >>> defined in the included makefile.  I thought make was supposed to
  >>> try to do full expansion on a second read of the makefile?  Seems
  >>> to have worked before.

  >> In fact make is suppose to expand SOMEVAR only when it sees
  >> $(SOMEVAR), so yo u only have to be sure that ANOTHERVAR is set
  >> before trying to use SOMEVAR.

Right.

  >> However be careful: if SOMEVAR is used in a dependency (either
  >> target or dependant part), it will be expanded as soon as it is
  >> seen.

This is true, but not relevant to this situation.  A recursive variable
like SOMEVAR is expanded again _every_ time it's seen, not just the first
time.  So even if it's been expanded before, that won't be a problem
here.

  cj> So I went off and tried to duplicate my problem with the smallest
  cj> Makefile possible and have succeeded.  Turns out the problem is
  cj> with a target macro's value indicating a dos style filename,
  cj> e.g. e:/dir/file (yes, with UNIX slashes) even with MAKE_MODE=UNIX
  cj> set in the environment.  Using something like //e/dir/file works
  cj> of course.  I think that on win32 e:/dir/file is supposed to be
  cj> allowed without confusing the make parser but I cannot find much
  cj> in the info documentation to help make this clearer (it did work
  cj> in the past with make 3.75, the cygwin 1.1 net release uses 3.77).

It would be nice if you'd included the "smallest Makefile possible", as
well as the error you got and what you expected it to do :).

From the description above I'm pretty lost as to what the problem you're
seeing actually is.

-- 
-------------------------------------------------------------------------------
 Paul D. Smith <psmith AT gnu DOT org>          Find some GNU make tips at:
 http://www.gnu.org                      http://www.ultranet.com/~pauld/gmake/
 "Please remain calm...I may be mad, but I am a professional." --Mad Scientist

--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com

- Raw text -