automake-patches
[Top][All Lists]
Advanced

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

Re: rewrite of m4/depout.m4


From: Ralf Wildenhues
Subject: Re: rewrite of m4/depout.m4
Date: Fri, 27 Apr 2007 00:58:52 +0200
User-agent: Mutt/1.5.13 (2006-08-11)

Hi Stepan,

* Stepan Kasal wrote on Thu, Apr 26, 2007 at 05:42:58PM CEST:
> On Tue, Apr 17, 2007 at 07:02:13PM +0200, Ralf Wildenhues wrote:
> > 
> > I think the killer argument against this change is the point that a
> > buggy Makefile has worse consequences, starting from spurious errors
> > at config.status time, and not ending in an infinite configure recheck
> > loop.
> 
> well, this is a pity: running make from config.status would be cool.  :-(

Do you have resources to test that none of the known make
implementations invoke makefile rebuilding rules when invoked as
  $feed_makefile | make -f -

?  Can you come up with anything else that can go wrong?  What about:
at the time that
  ./config.status Makefile depfiles

is issued, Makefile exists but it happens to include add.mk which does
not exist yet (because "config.status add.mk" generates it later)?
What else can go wrong?  What if `make' goes berserk, but `gmake' is
sane (but the user did not tell MAKE=gmake to configure?).

> But we could apply a more minimalistic approach: we might try to grep
> only the variables which are needed to evaluate the directory names
> and then run make with that specially crafted makefile.

You mean: while the 8000 line script that is `automake' is not smart
enough to distinguish variables from rules, a grep "script" ...

> We are using awk already, so we can use it for this task, too.

... or an awk "program" will do?

> But since we have got _that_ far, we might also use awk to perform
> the recursive evaluation of the directory name, and output the list
> of the directories to be created.

We might as well use automake.  Or make.  Or none.  I don't think
there's a viable third alternative.  But I haven't given up on Jim's
suggestion quite yet, I just haven't had time do test it yet.

Cheers,
Ralf




reply via email to

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