|
Hylafax Mailing List Archives
|
[Date Prev][Date Next][Thread Prev][Thread Next]
[Date Index]
[Thread Index]
Re: [hylafax-users] sed-3.02.80
On 2002.04.23 17:41 Seemant Kulleen wrote:
> Hi all,
>
> I am a developer for Gentoo linux, and I am trying to create a build
> script for hylafax-4.1.2 for Gentoo linux users. We use GNU sed-3.02.80.
>
> Without Daniel Robbins' (:P) patch, configure complains with the c++
> error
> that has been discussed on this list before. After the patch, configure
> succeeds, but the Makefile that it generates is horrendous. I have
> attached that to this e-mail.
Daniel Robbins' patch just wasn't complete.
> If anyone can supply any clues for a successful compilation of
> hylafax-4.1.2 with GNU sed-3.02.80 I would be appreciative.
See the attached. The problem is with configure's usage of 'sed -f' (or
rather, the file syntax being used). Beats me, though, if I know what it
is.
Lee.
diff -Nru hylafax.orig/configure hylafax/configure
--- hylafax.orig/configure Fri Apr 19 06:50:49 2002
+++ hylafax/configure Wed Apr 24 02:08:20 2002
@@ -1353,7 +1353,9 @@
WARNING="Warning, this file was automatically created by the HylaFAX configure script"
$RM confsed1; dumpvars "$VAR1" | sort>confsed1
$RM confsed2; dumpvars "$VAR2" | sort>confsed2
+ unset POSIXLY_CORRECT
$RM xdefs; sed -f confsed1 $SRCDIR/defs.in | sed -f confsed2 >$1
+ export POSIXLY_CORRECT=1
}
# NB: save original values for use in checks below
@@ -4524,7 +4526,9 @@
# the comparison (this assumes dates are used in lines
# of the form DATE: @DATE@).
#
+ unset POSIXLY_CORRECT
$RM $F.new; $SED -f confsed1 $SRCDIR/$F.in | $SED -f confsed2 > $F.new
+ export POSIXLY_CORRECT=1
$RM confx; $SED '/DATE:/d' $F.new >confx
$RM confy; $SED '/DATE:/d' $F >confy 2>/dev/null
if cmp -s confx confy >/dev/null 2>&1; then
@@ -4535,9 +4539,12 @@
fi
else
Note "Creating $F from $SRCDIR/$F.in"
+ unset POSIXLY_CORRECT
if $SED -f confsed1 $SRCDIR/$F.in | $SED -f confsed2 >$F.new; then
+ export POSIXLY_CORRECT=1
$RM $F; $MV $F.new $F; $CHMOD 444 $F
else
+ export POSIXLY_CORRECT=1
cat 1>&2 <<EOF
Help, there was a problem crafting $F from $F.in.
The command:
