BASH check if the name of a file exists in a text document

# filter lines that second column is jane # print third column awk '$2 == "jane"{print $3}' list.txt| # for each third column execute bash # inside bash check if first argument exists # if it does, print some text xargs -n1 -d'\n' bash -c '[[ -e "$1" ]] && echo "$1" exists' _ 

With some help from a friend we figured it out and this code works for what I'm trying to do.

#!/bin/bash > oldFiles.txt files=$(grep " jane " list.txt | cut -d' ' -f3) for f in $files; do if [ -e $f ]; then echo $f >> oldFiles.txt; fi done 

#!/bin/sh next() { [[ $(wc -l stack | cut -d' ' -f1) -gt 0 ]] && main end } main() { line=$(sed -n '1p' stack) [[ -s "${line}" ]] && echo "File exists." sed -i '1d' stack next } end() { rm -v ./stack exit 0 } sed -n '/jane/p' list.txt | cut -d' ' -f3 > stack next 

There are three functions here, defined first. The first two commands which actually run first are at the bottom; first the stack is generated with sed and cut, and then the next function is called.

The next function uses wc to see if the line number in the stack file is greater than zero. If it is, main is called. If not, end is called.

If main is called, the first line in the file is assigned to the line variable, and is then tested to see if that filename exists, and echoes if it does. The top line in the file is then deleted by sed, (which reduces the line count of the file by 1) and the script then runs the next function again.

When the line length of the stack file reaches zero, end is called, which deletes the empty stack file, and exits with an error code of zero, indicating success.