I am not too deep in the material and rarely use Bash scripts. Even with some research, I couldn't quickly learn everything in Bash so that I could search an entire directory with its sub-directories for files and then output their type. I've now gotten a bit into the direction of functions, but again don't quite know how to do this recursively. Also, I want to consider only files and not folders. Here is something I have already done on my own:

for item in "$1"/*
do
    if ! [ -d $item ]; then
        echo $(file $item)
    fi
done;

So when the script is called, the path is passed as an argument. The path is then searched for non-directories and their type is output with the command file. But how is this recursive and also implementable for sub-directories? I have also tried it by iterating over ls -R, but then names of folders are still appended and I can no longer check via my way if it is a folder or a file. Edit: I can't use find!

I am glad about any help!

CodePudding user response：

You can use bash extended globbing capabilities:

$ shopt -s dotglob globstar
$ for i in **/*; do [ -d "$i" ] && continue; file "$i"; done

CodePudding user response：

This may help: How to recursively list subdirectories in Bash without using "find" or "ls" commands?

That said, I modified it to accept user input as follows:

#!/bin/bash

recurse() {
 for i in "$1"/*;do
    if [ -d "$i" ];then
        echo "dir: $i"
        recurse "$i"
    elif [ -f "$i" ]; then
        echo "file: $i"
    fi
 done
}

recurse $1

If you didn't want the files portion (which it appears you don't) then just remove the elif and line below it. I left it in as the original post had it also. Hope this helps.

CodePudding user response：

Suggesting

   file $(find -type f 2>/dev/null)

Explanation:

1. find command that search recursively into current folder.

2. Filter only files from find command: -type f .

3. Ignore any find permission/access errors with 2>/dev/null

4. Run file command on each found file.