#!/bin/bash
if [ -e *.txt ] then
rm *.txt
echo "removing previous files"
else
echo "files already deleted"
fi
I'm trying to remove all txt files in my current directory if I have more than 1 txt file I get an error. Not quite sure whats happening.
If you have, say, two files 1.txt and 2.txt, your code basically leads to:
if [ -e 1.txt 2.txt ]
…
which just raises a syntax error, because -e expects a single argument.
You could rely on some CLI tool such as find, but maybe you don't need the if in the first place?
Namely something like:
rm -f *.txt
to avoid failing in case there is no *.txt file,
otherwise:
rm -f -v *.txt
if you also want to get some log of the removed files.
Keeping the if ... fi logic, there's no need for an -e test here. Unix utils are themselves designed to return exit codes, and in this case it's simpler to test the result of rm itself:
if rm *.txt 2> /dev/null
then
echo "removing previous files"
else
echo "files already deleted"
fi
But even though it's not needed, do note that a separate test could be made to work like so:
if [ "$(ls *.txt)" ]
...or in pure bash, (no external util like ls needed), if we temporarilly set the nullglob option:
if (shopt -s nullglob; [ "$(echo *.txt)" ])
echo *.txt prints *.txt litteraly if there is no .txt file (but maybe there's some Bash option to change that default behavior)
s/echo/ls/, which is lame, but it seems to work...shopt -s nullglob; if [ "$(echo *.txt)" ]; then echo "file(s) found"; fi