bash find and remove text

Shell Programming and Scripting
1

Here's the story:
I have two txt files of filenames one is like this

W00CHZ0103340-I1CZ31
W00CHZ0103340-I1CZ32
W00CHZ0103340-I1CZ33
W00CHZ0103341-I1CZ35
W00CHZ0103342-I1CZ46
W00CHZ0103343-I1CZ37
W00CHZ0103344-I1CZ39
W00CHZ0103345-I1CZ43
W00CHZ0103345-I1CZ44
...
the other like this

W00CHZ0103340/images/W00CHZ0103340-I1CZ31/
W00CHZ0103340/images/W00CHZ0103340-I1CZ32/
W00CHZ0103340/images/W00CHZ0103340-I1CZ33/
W00CHZ0103341/images/W00CHZ0103341-I1CZ35/
W00CHZ0103342/images/W00CHZ0103342-I1CZ46/
W00CHZ0103343/images/W00CHZ0103343-I1CZ37/
W00CHZ0103344/images/W00CHZ0103344-I1CZ39/
W00CHZ0103345/images/W00CHZ0103345-I1CZ43/
W00CHZ0103345/images/W00CHZ0103345-I1CZ44/

I want to remove all the text up to and including images/ and then the final / then i would be left with identical lists. On list comes from an ls command and the other from some type of manifest that i am checking against. I need to do this so that i can then compare the two for unique entries and thereby find out what's missing. Please Help me

2

on the second file, run this:

#!/bin/bash
cat secondfile.txt | while read line
do
FN=`echo $line | awk -F\/ '{print $(NF-1)}' `
echo $FN
done

3

With sed:

sed 's!.*\(.*\)/!\1!' file2

or with awk:

awk -F/ '{print $3}' file2

Regards

4

now i have just one more question. I left work already where i need to try and use this script but it will drive me nuts all weekend if i don't get this thing licked. Will the above responses work even if the parameters following the W above change? for example sometimes the filenames change to W23704/images...etc It always starts with W but the proceeding characters can be a hadge podge of numbers and letters and not always the same number. Again, any help is really appreciated. I just want to know that i've got it come monday.

5

No problem. The sed command prints the piece between the last 2 slashes and the awk command changes the field seperator to a slash and prints the 3th field.

Regards

6

There's no need for cat, and calling awk for each and every line is ridiculously inefficient.

Adding to the inefficiency is storing the reault of each awk call in a variable with command substitution then printing the variable. Why not let awk's output be printed directly?

All it takes is a single call to awk, with the filename as an argument:

awk -F/ '{ print $(NF-1) }' secondfile.txt