Hi.

I have a directory full of data files, each of which is named something like data_12345_xxz, the number string running from 00000 to 15000 say.

I need to be able to rename each of these data files as data_12345_bhz, the 'xx' being replaced by a 'bh' - without having to do each individually!

Would appreciate any help!
John Paul O'Donnell

You can do this fairly easily using the rename program which comes with perl. The syntax is a little mysterious if you don't understand perl regular expressions, so it might be a good idea to read about them before using it.

This command just shows what will be renamed to what without actually doing it - good to check that it's the right thing...
...and this command actually does the rename:

The rename command is part of util-linux, not perl, at least on Slackware.

Also, you don't need to match the whole thing, only the part you want to change, e.g.

should suffice.

Excellent, didn't know that one. Thanks. It's author of rename is Larry Wall. I had assumed it came with perl. In Ubuntu Edgy, dpkg claims not to know which package owns this file, which is a little annoying. Ho hum.

Thanks very much. That's worked.

Ah, I didn't know LW wrote rename. The Slack man page is pretty minimal. It's possible it was supplied with perl originally. util-linux has sucked up loads of handy utilities into one package, just as core-utils unified sh-utils and a few other packages I can't remember the name of. I did a quick google for a history of rename but couldn't find anything of much interest.

And, yeah, rename only changes the pattern specified, the rest of the filename is unchanged. It took me a while to get my head around this, since it's completely different to every other "rename" command on various OSes and to sed etc. Dead handy command, though, and very easy to use once you grok it.