UNIX for Dummies Questions & Answers

I have a file which is fixed width columns. This is an offset buffer - rather than space or tab delimited. There are upto about 8 columns and I need to get all of the column 5's values into another file.

The problem is that because the delimiter is a space - and some fields are blank - the 5th field is sometimes taken as the 6th - so I lose quite a bit of data.

I want to find a way to output just the values from the 50th - 60th positions on the line into another file - for all lines in a large text file.

Ideally I'd like to get two columns for each line.....but I imagine that should be easy enough if I can figure out how to get the column I want.

I tried " awk -F" " '{print $5 "," $1 }' " - but this gives me the problem...any ideas?

Try the cut command which will allow a cut from specific positions
cut -c50-60 filename

Yep - that's exactly what i need - cheers.

For two columns, it would be:

cut -c20-30,50-60 myfile > newfile

but reversing these:

cut -c50-60,20-30 myfile > newfile

will produce same results! man page indicates to use multiple cuts, then paste back together to get the columns rearranged. Now, how easy would it have been for the cut command to just output the data in the desired order?

So if you want column 2 followed by column 1, I would use awk:

awk '{print substr($0,50,11), substr($0,20,11)'} myfile > newfile

That comma will give you a space separator. Omit it to get the two columns concatenated.