Home / ubuntu / Questions / 187477
Accepted
nkint
Asked: 2012-09-14 10:45:44 +0800 CST

How to rename files with count?

  • 772

I have a directory like this:

randomized_quad015.png
randomized_quad030.png
randomized_quad045.png
randomized_quad060.png
randomized_quad075.png
randomized_quad090.png
randomized_quad1005.png
randomized_quad1020.png
randomized_quad1035.png
randomized_quad1050.png
randomized_quad105.png
randomized_quad1065.png
randomized_quad1080.png
randomized_quad1095.png
randomized_quad1110.png
randomized_quad1125.png
randomized_quad1140.png

and I want to rename the first files adding 0 in front of number, like:

randomized_quad0015.png

But I don't know how. Some help?

bash

4 Answers

  1. rename is a command-line tool that lets you rename files, using a Perl command to transform the name. A Perl command that's commonly used in this circumstance is s/REGEXP/REPLACEMENT/.

    If I interpret your question correctly, you want to add a leading 0 in the names that have fewer than 4 digits, so that all the names end up having 4 digits. This slightly complicated expression handles the general case where you have file names with 1, 2 or 3 digits: it aligns all of them to 4 digits. The e modifier at the end causes the replacement to be interpreted not as a string but as a Perl expression. This expression repeats the digit 0 as many times as necessary to pad the number to 4 digits.

    rename 's/(?=([0-9]+)\.)/"0" x (4 - length($1))/e' *.png

    If the files you want to rename all have three digits, there are simpler ways, such as

    rename 's/[0-9]+\./0$&/' *[!0-9][0-9][0-9][0-9].png

    or

    rename 's/(?<=[^0-9])[0-9]{3}\./0$&/' *.png
    • 6
  2. Best Answer

    As an alternative to the various Perl-based methods in other answers, you can do this particular example, and lots of other similar ones with plain bash:

    for i in randomized_quad???.png; do   
  mv $i randomized_quad0${i#randomized_quad}
done

    This loops over the filenames with only three characters in the number region, and the variable expression in braces takes each filename, snips "randomized_quad" off the front (with the # operator), and adds "randomized_quad0" on the front.

    Other excitingly useful operators like this include:

    • % - cut something off the end
    • / - search and replace (/ for the first match, // for all matches)
    • : - for extracting substrings

    A more complete list with examples is here. Also, note that these use POSIX-style regular expressions, so . will match any character, not just a "." If you want to match a ".", you should 'escape' it, with a backslash:

    TestVariable1="filename.png"
TestVariable2="filename0png"

echo "${TestVariable1%.png}"
> filename

echo "${TestVariable2%.png}"
> filename

echo "${TestVariable1%\.png}"
> filename

echo "${TestVariable2%\.png}"
> filename0png
    • 4 
  3. rename 's/quad([0-9]{3})\.png/quad0$1/' randomized_quad*
    • 0

  4. With a recent version of bash you could write:

    for f in randomized_quad*.png; do
    if [[ $f =~ [1-9][0-9]* ]]; then
        new=$(printf "randomized_quad%04d.png" ${BASH_REMATCH[0]})
        [[ $f != $new ]] && mv -v $f $new
    fi
done
    • 0