Checking if substring is in filename in bash

Question

I'm trying to create a script that identifies the names of files in a directory and then checks to see if a string is a substring of the name. I'm doing this in bash and cannot use the grep command. Any thoughts?

I have the following code to check if a user submission matches a file name or a string in the name.

read -p name 

for file in sample/*; do
    echo $(basename "$file")
    if [[$(basename "$file") ~= $name]];
    then echo "invalid"
    fi
done

Possible dupe of https://stackoverflow.com/questions/10881157/read-file-names-from-directory-in-bash — Anand G, Jan 30 '18 at 07:32
Possible duplicate of [Read file names from directory in Bash](https://stackoverflow.com/questions/10881157/read-file-names-from-directory-in-bash) — M. Becerra, Jan 30 '18 at 07:44
Tangentially see also https://stackoverflow.com/questions/9581064/why-should-there-be-a-space-after-and-before-in-bash — tripleee, Jan 30 '18 at 07:47

score 2 · Answer 1 · answered Jan 30 '18 at 07:44

You can just interpolate the user input into the wildcard.

printf '%s\n' sample/*"$name"*

If you want to loop over the matches, try

for file in sample/*"$name"*; do
    # cope with nullglob
    test -e "$file" || break
    : do things with "$file"
done

If you just need to check that the name isn't a substring of an existing file's name:

valid=true
for file in sample/*"$name"*; do       
    test -e "$file" && valid=false
done
echo "$name is valid? $valid"

The shell by default does not expand a wildcard which doesn't match any files; so in this case, your loop will run once, but the loop variable will not match any existing file. You might also want to look at the nullglob option in Bash to make it loop zero times in this case.

Checking if substring is in filename in bash

1 Answers1