mariadb/cmd-line-utils/libedit

An approximate method to merge from upstream is:

	# Fetch latest from upstream (we also include some compat stuff)
	$ CVS_RSH=ssh; export CVS_RSH
	$ CVSROOT="anoncvs@anoncvs.netbsd.org:/cvsroot"
	$ cvs co -d libedit -P src/lib/libedit
	$ mkdir libedit/np
	$ for f in src/common/lib/libc/string/strlcat.c \
	>          src/common/lib/libc/string/strlcpy.c \
	>          src/include/vis.h \
	>          src/lib/libc/gen/unvis.c \
	>          src/lib/libc/gen/vis.c \
	>          src/tools/compat/fgetln.c
	> do
	>   cvs co -P ${f}
        >   mv ${f} libedit/np
	> done
	$ rm -rf src
	$ cd libedit

	# Remove files we don't need/use
	$ rm -rf CVS TEST Makefile shlib_version *.[0-9]
	$ (cd readline; rm -rf CVS Makefile)

	# Rename files to match our naming
	$ mv makelist makelist.sh
	$ mv terminal.h el_terminal.h

	# Remove NetBSD-specific bits
	$ for file in $(find . -type f)
	> do
	>   cp ${file} ${file}.orig
	>   sed -e 's/#include "terminal.h"/#include "el_terminal.h"/g' \
	>       -e 's/sig_handler/el_sig_handler/g' \
	>       -e 's/isprint/el_isprint/g' \
	>       -e '/^__RCSID/d' \
	>       ${file}.orig >${file}
	>   rm ${file}.orig
	> done

then merge remaining bits by hand.  All MySQL-specific changes should be
marked with XXXMYSQL to make them easier to identify and merge.  To generate
a 'clean' diff against upstream you can use the above commands but use

	cvs co -D "2011/10/04 15:27:04" [..]

to fetch the baseline of most recent merge.

Please feed any fixes to Jonathan Perkin <jonathan.perkin@oracle.com> who will
endeavour to merge them upstream and keep diffs minimal.