Naming and Renaming Files

Numerically incrementing names example

In this example, we are studying a book and creating files with incremental digits so we have a history of the progress in order. Good for review or for sharing with others since the order and the improvements or advancements made in the implementation were done in a particular order. So far we have these files:

$ ls -1
gear_1.scm
gear_2.scm
...
gear_9.scm

Then we progress in the chapters and end up with a few more incrementally named gear_??.rb files:

$ ls
gear_10.rb  gear_12.rb  gear_1.rb  gear_3.rb  gear_5.rb  gear_7.rb  gear_9.rb
gear_11.rb  gear_13.rb  gear_2.rb  gear_4.rb  gear_6.rb  gear_8.rb

$ ls -1
gear_10.rb
gear_11.rb
gear_12.rb
gear_13.rb
gear_1.rb
gear_2.rb
gear_3.rb
gear_4.rb
gear_5.rb
gear_6.rb
gear_7.rb
gear_8.rb
gear_9.rb

Oh noes! gear_10.rb to gear_13.rb all come before gear_1.rb to gear_9.rb. Even though we can review the files in order, it would be nice if the were really listed from 1 to 13 instead of from 10 to 13 and from 1 to 9.

The thing is that the “gear_” substring all have the same ordering because all the files have that common substring. So the algorithm needs to compare more chars to figure out the order. 1 ends in a tie for both “gear_1.rb” and “gear_10.rb”, because so far it has compared “gear_1” an “gear_1”. The algorithm has to take a look further chars. For “gear_1.rb” the next char is the “.” (dot), and for “gear_10.rb” it is the “0” (zero).

Now type man ascii (man 7 ascii) on your Unix-like operating system and you’ll see “0” has a lower value than “.”.

Oct   Dec   Hex   Char                      │ Oct   Dec   Hex   Char
────────────────────────────────────────────┼───────────────────────────
000   0     00    NUL '\0' (null character) │ 100   64    40    @
...
056   46    2E    .                         │ 156   110   6E    n
....

Let’s pretend we are the algorithm:

g e a r _ 1 . r b
- - - - - -
          ^

g e a r _ 1 0 . r b
- - - - - -
          ^

Thehe substring up to “gear_1” in both filenames is not enough to determine the order. Try the next char:

g e a r _ 1 . r b
- - - - - -
            ^

g e a r _ 1 0 . r b
- - - - - - -
            ^

We finally have different chars to compare, and “0” comes before “.” in ASCII/UTF-8, so “gear_10.rb” is listed before “gear_1.rb”.

UTF-8 is compatible with ASCII so it will work the same on our modern UTF-8 enabled systems and applications.

So now that we know that, we can rename the files so they always have two incrementing digits (assuming there will be no more than 99 files for a chapter or an entire book, as when we get to 100, the problem would show up again).

Using the `rename' command

The parameters for our case are these:

  • gear_ which means the substring we want to keep untouched.

  • gear_0 to mean we wan to add a “0” after the “_”.

  • gear_?.rb which means we only want to rename the files that had the single digit between “_” and “.rb”

$ rename gear_ gear_0 gear_?.scm

$ ls -1
gear_01.scm
gear_02.scm
gear_03.scm
gear_04.scm
gear_05.scm
gear_06.scm
gear_07.scm
gear_08.scm
gear_09.scm
gear_10.scm
gear_11.scm
gear_12.scm
gear_13.scm