- One possible solution could be to take each source pathname in turn, take the basename from this path, and then use sed to copy the source file to the destination directory using the same base name and performing the required translation as the copy is performed:
for i in $1/*
do
x=`basename $i`
sed s/SP/SU/ $i >$2/$x
done
exit 0
- It is a simple matter to use grep to see if PATH contains the name of the directory /usr/local/bin, and display an appropriate message:
if echo $PATH | grep "/usr/local/bin" &>/dev/null
then
echo PATH contains /usr/local/bin
else
echo PATH does not contain /usr/local/bin
fi
exit 0
Notice that the output and error output from the grep command are both being discarded, as only the exit status from grep is significant.
- The easiest way to sort this out is to use the numeric expression evaluation facilities of the shell to maintain a line count which can be echoed to the standard output at the start of each output line and then incremented ready for the next line:
linecount=1
while read line
do
echo $linecount $line
linecount=$[$linecount+i]
done
exit 0
- This solution to the problem starts by checking the command line parameter for errors. Then the code works from left to right and top to bottom across the character positions of the output square, working out with some simple logic which of the four characters to display at each position:
if test $# -ne "1"
then
echo usage: drawsquare <n>
exit 1
fi
if test $1 -lt "2" -o $1 -gt "15"
then
echo 'usage: drawsquare <n>'
echo ' (where 2>=n<=15)'
exit 1
fi
xcount=$1
ycount=$1
while test $ycount -gt "0"
do
while test $xcount -gt "0"
do
if test $xcount -eq "1" -o $xcount -eq $1
then
if test $ycount -eq "1" -o $ycount -eq $1
then
echo -n "+"
else
echo -n "|"
fi
else
if test $ycount -eq "1" -o $ycount -eq $1
then
echo -n "-"
else
echo -n " "
fi
fi
xcount=$[$xcount-i]
done
xcount=$1
ycount=$[$ycount-1]
echo
done
exit 0
- For each file in the specified directory, tests are made to see if it is readable and a regular file. If this test is true then the file size is checked and the names of files smaller than the second parameter are displayed along with their sizes:
if test $# -ne 2
then
echo 'usage: listfiles <dirpath> <size>'
exit 1
fi
if ! test -d $1
then
echo 'usage: listfiles <dirpath> <size>'
exit 1
fi
for i in $1/*
do
if test -r $i -a -f $i
then
size=`wc -c >$i`
if test $size -lt $2
then
echo 'basename $i' has size $size bytes
fi
fi
done
