UNIX for Dummies Questions & Answers

Hello,

I have a bunch of files whose names start with 'xx'
The first line of each file looks something like:

a|...|...|...|... , ...


In order to rename all of these files to whatever's between the 4th | and the comma (in the first line of that particular file) , I have been using:

Code:

for file in xx*; do mv "$file" `head -1 "$file" | cut -d '|' -f 5 | cut -d ',' -f 1`.txt; done

But my major problem is that several files have the same string between the 4th | and the comma. When this 'duplication' happens, I want to be able to rename the file to the same thing, but with something appended to the end, like a number or something.

I want to have the same number of files in the end as the number of files I started with, but right now I'm losing many many files since unix is writing overtop of files when the desired name has already been used.



Thank you in advance

Thank you Anbu23,

Though your method worked perfectly for a simple test case that I made, it encountered a serious problem for the set of files I am working with.

For one of the files, the string destined to become the filename of that file was "too long" for some reason... and so the entire procedure stopped, and only 84 out of my 1883 files actually got renamed.

This was very strange though because when I was using my original code:

Code:

for file in xx*; do mv "$file" `head -1 "$file" | cut -d '|' -f 5 | cut -d ',' -f 1`.txt; done

none of the file names were "too long" ... the only problem was that some files were being overwritten due to duplicate naming.

So does anyone know how I could modify my code to avoid the duplicate naming ??

Or a better way of implementing Anbu23's code so as to not have any restriction on the size of a filename ??

I thought that there wasn't any restriction to the size a filename can be in unix.