dircmp.sh revision 7c478bd95313f5f23a4c958a745db2134aa03244
#!/usr/bin/ksh
#
# CDDL HEADER START
#
# The contents of this file are subject to the terms of the
# Common Development and Distribution License, Version 1.0 only
# (the "License"). You may not use this file except in compliance
# with the License.
#
# You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
# or http://www.opensolaris.org/os/licensing.
# See the License for the specific language governing permissions
# and limitations under the License.
#
# When distributing Covered Code, include this CDDL HEADER in each
# file and include the License file at usr/src/OPENSOLARIS.LICENSE.
# If applicable, add the following below this CDDL HEADER, with the
# fields enclosed by brackets "[]" replaced with your own identifying
# information: Portions Copyright [yyyy] [name of copyright owner]
#
# CDDL HEADER END
#
# Copyright (c) 1984, 1986, 1987, 1988, 1989 AT&T
# All Rights Reserved
#
# Copyright (c) 1996-2000 by Sun Microsystems, Inc.
# All rights reserved.
#ident "%Z%%M% %I% %E% SMI"
PATH=/usr/bin
USAGE="usage: dircmp [-d] [-s] [-wn] dir1 dir2"
trap "rm -f /usr/tmp/dc$$*;exit" 1 2 3 15
typeset -i exitstat=0
typeset -i sizediff
typeset -i cmpdiff
typeset -i Sflag=0
typeset -i Dflag=0
typeset -i fsize1
typeset -i fsize2
typeset -l LFBOUND=2147483648
width=72
#
# function to generate consistent "diff" output whether or not files are intact
#
function dodiffs {
type=`LC_MESSAGES=C file $D1/"$a"`
case "$type" in
*text) ;;
*script) ;;
*empty*) echo $D1/`basename "$a"` is an empty file |
pr -h "diff of $a in $D1 and $D2" >> /usr/tmp/dc$$g
continue
;;
*cannot*) echo $D1/`basename "$a"` does not exist |
pr -h "diff of $a in $D1 and $D2" >> /usr/tmp/dc$$g
continue
;;
*) echo $D1/`basename "$a"` is an object file |
pr -h "diff of $a in $D1 and $D2" >> /usr/tmp/dc$$g
continue
;;
esac
type=`LC_MESSAGES=C file $D2/"$a"`
case "$type" in
*text) ;;
*script) ;;
*empty*) echo $D2/`basename "$a"` is an empty file |
pr -h "diff of $a in $D1 and $D2" >> /usr/tmp/dc$$g
continue
;;
*cannot*) echo $D2/`basename "$a"` does not exist |
pr -h "diff of $a in $D1 and $D2" >> /usr/tmp/dc$$g
continue
;;
*) echo $D2/`basename "$a"` is an object file |
pr -h "diff of $a in $D1 and $D2" >> /usr/tmp/dc$$g
continue
;;
esac
#
# If either is a "large file" use bdiff (LF aware),
# else use diff.
#
if (( fsize1 < LFBOUND && fsize2 < LFBOUND ))
then cmd="diff"
else cmd="bdiff"
fi
($cmd "$D1"/"$a" "$D2"/"$a"; echo $? > /usr/tmp/dc$$status) | \
pr -h "diff of $a in $D1 and $D2" >> /usr/tmp/dc$$g
if [[ `cat /usr/tmp/dc$$status` != 0 ]]
then exitstat=$diffstat
fi
}
#
# dircmp entry point
#
while getopts dsw: i
do
case $i in
d) Dflag=1;;
s) Sflag=1;;
w) width=`expr $OPTARG + 0 2>/dev/null`
if [ $? = 2 ]
then echo "dircmp: numeric argument required"
exit 2
fi
;;
\?) echo $USAGE
exit 2;;
esac
done
shift `expr $OPTIND - 1`
#
D0=`pwd`
D1=$1
D2=$2
if [ $# -lt 2 ]
then echo $USAGE
exit 1
elif [ ! -d "$D1" ]
then echo $D1 not a directory !
exit 2
elif [ ! -d "$D2" ]
then echo $D2 not a directory !
exit 2
fi
#
# find all dirs/files in both directory hierarchies. Use "comm" to identify
# which are common to both hierarchies as well as unique to each.
# At this point, print those that are unique.
#
cd "$D1"
find . -print | sort > /usr/tmp/dc$$a
cd "$D0"
cd "$D2"
find . -print | sort > /usr/tmp/dc$$b
comm /usr/tmp/dc$$a /usr/tmp/dc$$b | sed -n \
-e "/^ /w /usr/tmp/dc$$c" \
-e "/^ [^ ]/w /usr/tmp/dc$$d" \
-e "/^[^ ]/w /usr/tmp/dc$$e"
rm -f /usr/tmp/dc$$a /usr/tmp/dc$$b
pr -w${width} -h "$D1 only and $D2 only" -m /usr/tmp/dc$$e /usr/tmp/dc$$d
rm -f /usr/tmp/dc$$e /usr/tmp/dc$$d
#
# Generate long ls listings for those dirs/files common to both hierarchies.
# Use -lgn to avoid problem when user or group names are too long, causing
# expected field separator to be missing
# Avoid other potential problems by piping through sed:
# - Remove: Spaces in size field for block & character special files
# '71, 0' becomes '710'
# - For file name, do not print '-> some_file'
# '/tmp/foo -> FOO' becomes '/tmp/foo'
# The following sed is to read filenames with special characters
sed -e 's/..//' -e 's/\([^-a-zA-Z0-9/_.]\)/\\\1/g' < /usr/tmp/dc$$c > /usr/tmp/dc$$f
cat /usr/tmp/dc$$f | xargs ls -lLgnd | \
sed -e '/^[bc]/ s/, *//' -e '/^l/ s/ -> .*//' > /usr/tmp/dc$$i 2>/dev/null
cd "$D0"
cd "$D1"
cat /usr/tmp/dc$$f | xargs ls -lLgnd | \
sed -e '/^[bc]/ s/, *//' -e '/^l/ s/ -> .*//' > /usr/tmp/dc$$h 2>/dev/null
cd "$D0"
> /usr/tmp/dc$$g
#
# Process the results of the 'ls -lLgnd' to obtain file size info
# and identify a large file's existence.
#
while read -u3 tmp tmp tmp fsize1 tmp tmp tmp a &&
read -u4 tmp tmp tmp fsize2 tmp tmp tmp b; do
#
# A window of opportunity exists where the ls -lLgnd above
# could produce different
# results if any of the files were removed since the find command.
# If the pair of reads above results in different values (file names) for 'a'
# and 'b', then get the file pointers in sync before continuing, and display
# "different" message as customary.
#
if [[ "$a" != "$b" ]]; then
while [[ "$a" < "$b" ]]; do
if (( Sflag != 1 ))
then echo "different $a"
dodiffs
fi
read -u3 tmp tmp tmp fsize1 tmp tmp tmp a
done
while [[ "$a" > "$b" ]]; do
if (( Sflag != 1 ))
then echo "different $b"
dodiffs
fi
read -u4 tmp tmp tmp fsize2 tmp tmp tmp b
done
fi
cmpdiff=0
sizediff=0
if [ -d "$D1"/"$a" ]
then if (( Sflag != 1 ))
then echo "directory $a"
fi
elif [ -f "$D1"/"$a" ]
then
#
# If the file sizes are different, then we can skip the run
# of "cmp" which is only used to determine 'same' or 'different'.
# If the file sizes are the same, we still need to run "cmp"
#
if (( fsize1 != fsize2 ))
then
sizediff=1
else
cmp -s "$D1"/"$a" "$D2"/"$a"
cmpdiff=$?
fi
if (( sizediff == 0 && cmpdiff == 0 ))
then if (( Sflag != 1 ))
then echo "same $a"
fi
else echo "different $a"
if (( Dflag == 1 ))
then
dodiffs
fi
fi
elif (( Sflag != 1 ))
then echo "special $a"
fi
done 3</usr/tmp/dc$$h 4</usr/tmp/dc$$i | pr -r -h "Comparison of $D1 $D2"
if (( Dflag == 1 ))
then cat /usr/tmp/dc$$g
fi
rm -f /usr/tmp/dc$$*
exit $exitstat