find and separate dirname using awk

[B0rderLine]

Hi first timer here...Can anyone help with script below. Seems to work okay, but problem arises if user enters full path of file .. mv will not work.
Apparently I have to use awk or something to determime path component of $1

#! /bin/sh

# Prog that moves file specified by user
# File is moved to "trash" folder where it can be
# recovered from if required# To use for adding removed file
# details to tra.db

TRASH="${HOME}/trash"

if [ $# -ne 1 ]
then
echo "name of file: \c"
read fname

if [ -z "$fname" ]
then
echo " $No file ... Quitting.... "
exit 1
else
echo "checking $fname"
fi
else
fname="$1"
fi

fPID="$fname$$"

# Check file

if [ -f $fname -a -w $fname ]
then
echo " Preparing to remove"
else
echo " No file: $fname "
echo " Check permission and file exists"
exit 1
fi


if [ ! -d $TRASH] # Check "trash" exists
then
mkdir $TRASH
fi

if mv -i $fname ${TRASH}/${fPID}
then
echo "moved file: $fname"
echo ${fname}\ $fPID >> ${HOME}/tra.db
exit 0
else
echo "Error: Can not mv $fname .. Exit and try again"
exit 2
fi

 

[PHV]

Try this:

fPID=`basename "$fname"`$$

Are you sure you want to escape the $ character when writing
to tra.db file ?

Hope This Help
PH.

 

[B0rderLine]

Thanks PHV
Unfortunately I'm not able to use 'basename' on the system.
the escape character when writing to tra.db is actually to insert a space between $fname & $fPID... \ $fPID

I've been advised that the problem can be worked using awk, but I'm really lost when I try it.

 

[PHV]

fPID=`echo "$fname" | awk -F/ '{print $(NF)}'`

Hope This Help
PH.

 

[B0rderLine]

Tks PH
I'll give it a go.. and post again to let u know.

 

[B0rderLine]

PH you're my hero !! Thankyou thankyou thankyou !!

 

[MaxKnight]

Just an extra tip... try "basename".

No need for awk.

 

[B0rderLine]

Thanks for the tip MaxKnight, but as I said in an earlier post.. I don't have that command available to me on the sys I'm using.