Splitting CSV file and excluding column in output using bash, sed or awk

awk, bash, csv, linux, sed

Solution

Here's a one-liner for you in `awk`:

`awk -F "," '{ split ($8,array," "); sub ("\"","",array[1]); sub (NR,"",$0); sub (",","",$0); print $0 > array[1] }' file.txt`

Desired output achieved, although perhaps some of this code could be made more succinct. HTH.

EDIT:

Read code from left to right:

`-F ","` Yes this sets the delimiter.

`split ($8,array," ")` This splits the eighth column on the space and puts this info in an array called `array`.

`sub ("\"","",array[1])` We take the first array element (this is a slice that's going to become our output file name) and substitute out the leading `"` symbol (We need to escape the `"` symbol so we put the `\` character in front).

`sub (NR,"",$0)` This conveniently removes the line number from the beginning of your file (`NR` is row number and `$0` is of course the whole line of input before delimitation).

`sub (",","",$0)` This removes the comma after the row number.

Now that we have a clean filename and a clean row of data we can write `$0` to `array[1]`: `print $0 > array[1]`.

FIX:

So if you'd prefer a underscore instead of a hypon, all we need to fix is `array[1]`. I've just added in a global substitution: `gsub ("-","_",array[1])`.

The updated code is:

`awk -F "," '{ split ($8,array," "); sub ("\"","",array[1]); gsub ("-","_",array[1]); sub (NR,"",$0); sub (",","",$0); print $0 > array[1] }' file.txt`

HTH.

Problem

Original source