i have ~100 data files which contain ~10000 rows - the files are slightly different size. each row contains various data values including a unique number (id) in column x which identifies the source which the data refers to. I want to extract a whole row if id= a specific number. i then want to do that for all 100 files to get a file which gives all the data for that id. any idea how i do this using a combination of, say, cshell, awk and sed commands? many thanks gav

When I have to do that I try to write it down in plain words such as you have above and then try to break it down in parts. I then use those parts for the commands. Important parts of what you said. Be sure you know your shell be it bash or what not. -from files (or create file that contains all the file names if you wish.) -ascii -search each line using delimiter. Save delimiter in holding area of like delimiter being id= if changes. -take line to some working area or even the final file. I could do it maybe but it sounds like too much work. I read it wrong and answer it wrong too. So get off my case you peanut.