How-To: Replace characters in filenames in Linux

Occassionally you’re likely to end up with a bunch of files with filenames that are underscore separated or space separated when you want them the other way around, and then you either have to manually rename them by hand, or live with it.

I had this issue the other day so I wrote a script to convert all filenames in the current directory to the specified format.

The script switches are:

  • -t – turns on [T]esting mode, which displays what would happen if you ran the script, but makes absolutely no changes to filenames.
  • -u – converts all [U]nderscores to spaces.
  • -s – converts all [S]paces to underscores.
  • -l – converts all filenames to [L]owercase.
  • -p – converts all filenames to u[P]percase.

It’d be nice if I could get all working recusively and offer an option to Capitalise Each Word in the filenames, but this will do for now.

Like most scripting, this entire thing could probably be done in a single line, but I found it useful to learn how to use getopts and such.

Hope someone else finds it useful! Oh, and if you need a do-everything file renaming script, I read that FixNames script does what it says on the tin.

How to: Get absolute/relative file paths, filenames and extensions from a Bash script

I’ve been learning some Qt stuff today, and the resulting code requires pre-processing of header files to work correctly, which I thought I’d try to automate. As part of this, I needed to find out how to get at all the different elements of a file from a bash script, so I did some googling and got all the info I needed :)

The script itself isn’t particularly useful, but the component parts of how to get at paths, extensions, and plain filenames without extensions definitely is – check it out:

fileParts Shell Script

I created two dummy “.h” files in my home folder and ran the script – this is what it outputs:

That’s gotta come in useful, right? =D