How to rename files using a mask as simply and quickly as possible?

Question

Python

Hello dear.

How do I change filenames from
1, 2, 3, ... 10, 11, 12, ... 100, 101, 102
to

001, 002, 003, ... 010, 011, 012, ... 100, 101, 102

It is desirable that the program, or whatever it was, could itself understand what the largest number of digits I have (or enter with pens for each folder with files, it’s not at all critical) and how many zeros should be put at the beginning.

Connoisseurs, please tell me how this can be done through a regular expression (preferably) in, for example, Total Commander or Python, or bat, or bash.

PS It is necessary to renumber (I wonder if there is such a word?) both from the mask (1)*, (2)*, (3)*, ... (100)*, (101)*and the mask of file names like *№1, *№2, *№3, ... *№100, *№101, *№102.

Answer 1

Well, on bash manually it's easy:
for i in {1..99}; do a=00$i; mv $i ${a: -3}; done

Answer 2

1. Make a list of files you are interested in using the glob or pathlib module.
2. Using the regular expression "\d+" (module re), you extract sequences of digits from file names. If there are several, you will have to come up with a criterion for which one to take (for example, the last one). You determine the maximum number of digits.
3. Iterate through the list of names again with re.sub() . The second argument to re.sub() can be not only a string to replace, but also a function - you just make a function that pads the input string with zeros to the desired width and returns it.
4. If the new line (with replacement) differs from the original one, then rename the file with the original name to the file with the new name, otherwise you leave it as it is.

Answer 3

Came up with a working solution.
It should work with different filenames in the current directory, as long as there is only one piece with a number in the file.
That is, it will cope with such a set of files: file1.jpg file02.jpg file0035.jpg 36file.jpg, if only the numbers do not intersect
And it will NOT cope with this: file1_1.jpg 01_file_02.jpg

max=$(ls -1 | grep -oP "[0-9]*"|sort -n|tail -n1)
length=$(sed "s/^0*//g"<<<"$max"|xargs expr length)
echo "Max number found: [$max], max length: [$length]"

ls -1|while read oldname; do
  oldnum="$(grep -oP "[0-9]*"<<<"oldname")"
  newnum=$(printf "%0${length}d" $(sed "s/^0*//g"<<<"$oldnum"))
  newname=$(sed "s/$oldnum/$newnum/g"<<<"$oldname")
  if [ -f "$newname" ]; then
    echo "Kinly check the following files manually: $oldname $newname"
  else
    echo "Renaming [$oldname] to [$newname]"
    # mv "$oldname" "$newname"
done

ps after checking the output, uncomment the line with "# mv "

Answer 4

Well, here is the essence, but what more specifically you need for other cases, I did not understand.

Answer 5

The easiest option is to do it just for the find, the only thing is that in your example there are only two options, so there is no need to make any loops, it’s easier to run the condition twice. For tests, replace mv with cp or test daddy.

find . -type f -name '[0-9][0-9]\.txt' -exec bash -c 'mv $0 0${0#./}' {} \;
find . -type f -name '[0-9]\.txt' -exec bash -c 'mv $0 00${0#./}' {} \;

Withdrawal up to

[[email protected] test]$ ls
10.txt  1.txt  2.txt  333.txt

Output after

[[email protected] test]$ ls
001.txt  002.txt  010.txt 333.txt

find-search is recursive
.where
-type f- only files
-name- the file name in this regular routine
-execrun with found, the option of find
-bash - bash itself (since we need to remove ./ from the file name and find itself can do this, but we need to leave the previous option for mv and here already all )
{}substitute the output of find into exec
\;escape and close find
Or better yet, tell me why you are doing this, maybe we are going the wrong way.