[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bootstrap weirdness
From: |
Ralf Wildenhues |
Subject: |
bootstrap weirdness |
Date: |
Thu, 17 Mar 2005 18:34:13 +0100 |
User-agent: |
Mutt/1.4.1i |
HEAD and branch-2-0 ./bootstrap contains
| # Make a dummy libtoolize script for autoreconf:
| test -f clcommit.m4sh && cat > ./config/libtoolize <<'EOF'
| #! /bin/sh
| # This is a dummy file for bootstrapping CVS libtool.
| echo "$0: Bootstrap detected, no files installed." | sed 's,^.*/,,g'
| exit 0
| EOF
| chmod 755 ./config/libtoolize
|
| # Running the installed `libtoolize' will trash the local (newer) libtool.m4
| # among others. Call the dummy script we made earlier.
| LIBTOOLIZE=`pwd`/config/libtoolize
| export LIBTOOLIZE
Now first, I don't understand why certain commands should depend on
whether clcommit.m4sh is available (is this a cheap "we are a CVS
checkout" test?) . Second, if there is a reason, the following
commands ought to depend on that as well (ie., the chmod and setting of
LIBTOOLIZE).
Now, if this is not a CVS checkout: Should we be allowed to call
bootstrap at all? If not, then that should fail right away.
If so, however, I see the possibility that, because we do not build
a dummy libtoolize, things could fail in autoreconf. Right?
Regards,
Ralf
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- bootstrap weirdness,
Ralf Wildenhues <=