This is the mail archive of the automake@gnu.org mailing list for the automake project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]

Re: making pre-build headers

To: "John W. Eaton" <jwe at bevo dot che dot wisc dot edu>
Subject: Re: making pre-build headers
From: Bruce Korb <bkorb at sco dot COM>
Date: Tue, 31 Oct 2000 13:29:52 -0800
CC: Paul Martinolich <martinol at datasync dot com>, autoconf at gnu dot org, automake at gnu dot org
List-Id: Discussion list for automake <automake.gnu.org>
Organization: The Santa Cruz Operation
References: <200010312030.OAA20555@shell.datasync.com> <39FF31C6.46ADB689@sco.com> <14847.14076.634725.63532@foobar.bogus.domain>

"John W. Eaton" wrote:
> 
> On 31-Oct-2000, Bruce Korb <bkorb@sco.COM> wrote:
> 
> | It is ugly.  Especially...
> 
> Isn't that what the move-if-change script is for?

Its a nice try at the problem, but it doesn't work, for
the reason you noted.

The move-if-change saves the timestamp update so the compile
won't happen again, but it also leaves the header out-of-date
with respect to the sources it depends on.  Hence, the
mumble-stamp mumbo-jumbo.  The solution requires either
using check sum stamps for dependency analysis, or the
``dont-worry-about-this-dependency'' hint.  I really do not
want make rebuilding my generated sources, even if it 
does throw it away.

References:
- making pre-build headers
  - From: Paul Martinolich
- Re: making pre-build headers
  - From: Bruce Korb
- Re: making pre-build headers
  - From: John W. Eaton

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]