This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: other/2126: configure script fails with --silent option
- From: Wolfgang Bangerth <bangerth at ticam dot utexas dot edu>
- To: neroden at gcc dot gnu dot org, <bangerth at ticam dot utexas dot edu>, <gcc-bugs at gcc dot gnu dot org>, <pirie at u dot washington dot edu>, <gcc-gnats at gcc dot gnu dot org>
- Date: Fri, 3 Jan 2003 13:52:06 -0600 (CST)
- Subject: Re: other/2126: configure script fails with --silent option
On 30 Dec 2002 neroden@gcc.gnu.org wrote:
> Synopsis: configure script fails with --silent option
>
> State-Changed-From-To: analyzed->closed
> State-Changed-By: neroden
> State-Changed-When: Mon Dec 30 04:17:28 2002
> State-Changed-Why:
> Fixed in 3.4 (mainline) as a side effect of autoconfiscation of the
> top level. (If someone submits a patch for 3.3 or 3.2.2 I'll try to
> make sure it gets reviewed.)
I even sent a mail directly to you with a patch in it :-) (This date: Thu,
5 Dec 2002 14:16:37 -0600 (CST)).
I append the mail & patch from the mail, but beware that it is not tested.
Regards
Wolfgang
-------------------------------------------------------------------------
Wolfgang Bangerth email: bangerth@ticam.utexas.edu
www: http://www.ticam.utexas.edu/~bangerth
>From bangerth@ticam.utexas.edu Thu Dec 5 14:16:37 2002 -0600
Date: Thu, 5 Dec 2002 14:16:37 -0600 (CST)
From: Wolfgang Bangerth <bangerth@ticam.utexas.edu>
To: neroden@gcc.gnu.org
Subject: PR 2126
Message-ID:
<Pine.LNX.4.44.0212051413060.18107-100000@gandalf.ticam.utexas.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Status: O
X-Status:
X-Keywords:
X-UID: 98
Nathanael,
you are presently cleaning up the configury machinery anyway, so maybe you
can do this on your way as well: PR 2126 states this:
Two command line options, --silent and --site=SITE are supported.
Code on line 193 attempts to identify options that require, but are
missing, values. --site requires a value; --silent does not. --si*
matches both, so --silent is rejected as not having a required value.
Changing the code on line 193 from '--si*' to '--sit*' will trap --site
with no value but allow --silent to pass.
Current workaround is to either patch the configure script before
running it, or use the --quiet option.
I checked that the problem still is in the sources. The report also gives
a patch, which is here in updated version:
Index: configure
===================================================================
RCS file: /cvsroot/gcc/gcc/configure,v
retrieving revision 1.49
diff -c -r1.49 configure
*** configure 29 Sep 2002 16:11:24 -0000 1.49
--- configure 5 Dec 2002 20:15:16 -0000
***************
*** 197,203 ****
;;
# These options have mandatory values. Since we didn't find an = sign,
# the value must be in the next argument
! --bu* | --cache* | --ex* | --ho* | --pre* | --program-p* | --program-s* | --program-t* | --si* | --sr* | --ta* | --tm* | --x-* | --bi* | --sb* | --li* | --da* | --sy* | --sh* | --lo* | --in* | --ol* | --ma*)
optarg=$1
shift
arguments="$arguments $option=$optarg"
--- 197,203 ----
;;
# These options have mandatory values. Since we didn't find an = sign,
# the value must be in the next argument
! --bu* | --cache* | --ex* | --ho* | --pre* | --program-p* | --program-s* | --program-t* | --sit* | --sr* | --ta* | --tm* | --x-* | --bi* | --sb* | --li* | --da* | --sy* | --sh* | --lo* | --in* | --ol* | --ma*)
optarg=$1
shift
arguments="$arguments $option=$optarg"
I checked that all other places have indeed --sit* instead of --si*.
If you should apply such a patch, I can close the report for you if you
want.
Thanks
Wolfgang