UNIX for Dummies Questions & Answers

I have a comma-separated record and I'd like to use sed to pull the Nth record from it.

It seems like it'd need to be something like this:

Code:

sed -n 's/'"\,$1\,"'/&/p'

Am I close?

Try cut...

Code:

cut -d ',' -f $N

...where $N is a number.

or awk (or nawk or gawk)...

Code:

awk '{print $N}'

Just to fill in on the sed question: I'm afraid you're not terribly close. If your sed understands the \{m,n\} repetition operator, you could use that:

Code:

sed 's/^\([^,]*,\)\{'"$N"'\}\([^,]*\),.*/\2/' file

This matches start of line, followed by a parenthesized group of any character which is not comma, any number of times (one field) followed by comma; the whole group, $N times; followed by a group of one field (again, arbitrary number of non-comma characters) followed by a comma and anything; if matched, replace with just the contents of the second group. (This assumes all lines will match this pattern; if not, you'll need some minor tweaks, but that goes for the other solutions, as well.)