HEADS UP! RCNG in the tree...

[Matthew Dillon]

    I like both the idea of not making mergemaster dependant on a new utility,
    and also incorporating that new utility into the tree.  Mergemaster
    really ought to be runnable on older systems, that's one of the purposes
    of having it in the first place!

    I'll leave it to Robert to coordinate both pieces of work.

					-Matt
					Matthew Dillon 
					<dillon at xxxxxxxxxxxxx>

:Robert Garrett wrote:
:
:> Please try the following patch, basically this uses the same
:> find_mode() routine as the 4.8 mergemaster did.
:> rather than expect the output of the new stat
:> 
:> Rob
:> 
:> ===================================================================
:> RCS file: /usr/home/dcvs/src/usr.sbin/mergemaster/mergemaster.sh,v
:> retrieving revision 1.3
:> diff -u -r1.3 mergemaster.sh
:> --- mergemaster/mergemaster.sh  24 Jul 2003 06:35:39 -0000      1.3
:> +++ mergemaster/mergemaster.sh  25 Jul 2003 11:14:36 -0000
:> @@ -670,10 +670,9 @@
:> 
:>  # 4095 = "obase=10;ibase=8;07777" | bc
:>  find_mode () {
:> -  local OCTAL
:> -  OCTAL=$(( ~$(echo "obase=10; ibase=8; ${CONFIRMED_UMASK}" | bc) & 4095
:> &
:> -    $(echo "obase=10; ibase=8; $(stat -f "%OMp%OLp" ${1})" | bc) ))
:> -  printf "%04o\n" ${OCTAL}
:> + OCTAL=`perl -e 'printf "%04o\n", (((stat("$ARGV[0]"))[2] & 07777) &~ \
:> +    oct("$ARGV[1]"))' "${1}" "${CONFIRMED_UMASK}"`
:> +  echo "${OCTAL}"
:>  }
:lol, I didn't see your patch.. my bad 
:
:Rob
: