How a Missing $ Made My Live‑Stream Shell Script Delete All .sh Files

In a live‑stream the presenter attempted to write a Bash script that (1) finds every file ending with .sh in the current directory and sub‑directories, (2) renames each to have a .shell extension, and (3) removes the second line of each file.

Correct script

#!/bin/bash
ALL_SH_FILE=$(find . -type f -name "*.sh")
for file in ${ALL_SH_FILE[*]}
do
    filename=$(echo $file | awk -F'.sh' '{print $1}')
    new_filename="${filename}.shell"
    mv "$file" "$new_filename"
    sed -i '2d' "$new_filename"
 done

Step‑by‑step explanation

Find .sh files : the find . -type f -name "*.sh" command lists all regular files whose names end with .sh.

Rename the extension : for each file, the script extracts the name without the .sh suffix (using either Bash parameter expansion ${file%.sh*} or awk -F'.sh') and appends .shell, then moves the file with mv "$file" "$new_filename".

Delete the second line : sed -i '2d' "$new_filename" edits the file in place, removing its second line.

The bug that caused the disaster

During the live demo the presenter mistakenly wrote the rename command as:

mv "$file" "new_filename"

Because the leading $ was omitted, the literal string new_filename was used as the destination for every .sh file, overwriting the same file repeatedly and effectively deleting all original scripts.

Outcome and lesson

The missing $ turned the intended batch rename into a destructive operation, demonstrating how a single character error in shell commands can have catastrophic results. The author stresses the need for careful review, especially when using powerful commands like rm, mv, and cp on Linux.