bash command rename file in a new filename numbers natural

Here are the most common and practical bash ways to rename files to sequential natural numbers (1, 2, 3, …), usually with leading zeros so they sort correctly (natural / version sort friendly).

1. Simplest & most reliable (bash loop + printf padding)

i=1
for file in *; do
    # Skip directories if you only want files
    [[ -f "$file" ]] || continue

    ext="${file##*.}"
    new=$(printf "%04d.%s" "$i" "$ext")    # %04d = 4 digits with leading zeros

    mv -i -- "$file" "$new"
    ((i++))
done

• Change %04d to %03d, %05d, etc. depending how many digits you need.
• Use -n instead of -i if you don’t want interactive overwrite confirmation.
• To sort by name naturally first (recommended): replace * with $(ls -v) or $(ls | sort -V)

2. One-liner version (very popular)

n=1; for f in *; do [[ -f "$f" ]] && mv -i -- "$f" "$(printf "%04d.${f##*.}" $((n++)))"; done

3. Sort by modification time (newest gets highest number)

n=1
for f in $(ls -t); do          # -t = newest first
    [[ -f "$f" ]] || continue
    mv -i -- "$f" "$(printf "%04d.${f##*.}" $n)"
    ((n++))
done

Reverse order (oldest first = 0001):

n=1
for f in $(ls -tr); do ...     # -tr = oldest first

4. Using Perl rename command (very powerful, often pre-installed on Ubuntu/Debian)

# Number all .jpg files, padded to 4 digits
rename -n 'our $i; s/.*/sprintf("%04d.jpg", ++$i)/e' *.jpg

Remove -n (dry-run) when you’re ready to really rename.

Alternative syntax (keeps original extension):

rename -n 'our $i; s/.+/sprintf("%04d", ++$i) . "." . $&/e' *

5. Using ls -v (natural sort) + cat -n style

ls -v | cat -n | while read -r num name; do
    [[ -f "$name" ]] || continue
    mv -i -- "$name" "$(printf "%04d.${name##*.}" "$num")"
done

• ls -v → natural/version sort (1, 2, …, 10, 11 instead of 1, 10, 11, 2)

Quick recommendations