Example contents of a file:

001FILE
002FILE
003FILE
099FILE
111FILE

Is it possible to search for [000-111]FILE and still keep the number part of the result in tact? The number portion is a changing variable but the FILE part stays the same. I want to search for all [000-111]FILE and space them out to "001 FILE" "002 FILE" etc.

If I use sed 's/[000-111]FILE/&/g' It will basically replace with the same variable as found in search.

I know how to search for the variable but don't know how to keep a portion of the result. The cut function would be nice but I can't seem to be able to call it in through sed.

Searching for FILE and adding a space is not an option as some of the other contents have the same names and cannot be effected.

Any ideas?

Thank you but this file has other numbered variables too so I cannot search for numbers alone or characters alone. I must be able to search for 001FILE as a whole and then split the output into 001 FILE.

On another note, it doesn't have to be sed. I'm working on Sun Solaris 10 so which ever utility available by default will work too.

Any other ideas?

Not sure I fully understand the requirements, but try this variation:

If that isn't what you want, give us some more sample input data (including the lines you don't want to change) and the output you expect from that data.

Hi. I was trying to keep the question simple but I guess I can't anymore so here is the actual data.

The above is just 2 line of a very large file. The problem is, I need a space between the 0.000 and TRUE so I can import it correctly into excel.

The numbers are not constant and is a changing variable. I need to be able to target just the lines containing [0-1].[0-9][0-9][0-9]TRUE and put a space/tab to separate the number and the TRUE.

Thanks.

sed -e '/[0-1][0-9][0-9]TRUE/s/[0-1][0-9][0-9]/& /g' file

That worked!
Thank you all very much.