[Xastir] "Fix" for Cygwin compile problem

[russo at bogodyn.org]

I just found an annoying mistake on my part in acinclude.m4.

If db.h exists then configure will go on to look for a library containing
db_create.  When I was trying to clean up the configure test this weekend
I neglected to remove a comment character from an important line that 
reset LIBS to whatever it was before the probe.  Meaning that -ldb was
always in LIBS after the probe, even if there WAS no -ldb or if -ldb failed
to contain db_create.

I just commited that fix.  But if you are running anonymous CVS and want the
fix immediately, apply the following patch:

-----------------------------PATCH FOLLOWS
diff -u -r1.24 -r1.25
--- acinclude.m4        14 Nov 2004 01:10:25 -0000      1.24
+++ acinclude.m4        15 Nov 2004 18:05:57 -0000      1.25
@@ -2,7 +2,7 @@
 #
 # Copyright (C) 2000-2004  The Xastir Group
 #
-# $Id: acinclude.m4,v 1.24 2004/11/14 01:10:25 tvrusso Exp $
+# $Id: acinclude.m4,v 1.25 2004/11/15 18:05:57 tvrusso Exp $
 
 # test for devices.  Avoid the tests on Cygwin as they hang on some
 # WinXP boxes.
@@ -644,7 +644,7 @@
 ])
 
 # From Cyrus imap distribution (KB3EGH)
-dnl $Id: acinclude.m4,v 1.24 2004/11/14 01:10:25 tvrusso Exp $
+dnl $Id: acinclude.m4,v 1.25 2004/11/15 18:05:57 tvrusso Exp $
 
 dnl These are the Cyrus Berkeley DB macros.  In an ideal world these would be
 dnl identical to the above.
@@ -715,7 +715,7 @@
 #          BDB_LIBADD="$BDB_LIBADD -ldb"; dblib="berkeley"; dbname=db,
 #            dblib="no")
 #        fi
-#      LIBS=$saved_LIBS
+       LIBS=$saved_LIBS
 
        LDFLAGS=$BDB_SAVE_LDFLAGS
 ])
---------------------------------------END OF PATCH--------------------

All I did was remove the comment from the "LIBS=$save_LIBS" line.

My apologies for that.
T.