I have written a shell script to get any column. The script is:
#!/bin/sh
awk '{print $c}' c=${1:-1}
So that i can call it as
ls -l | column 2
But how do i implement it for multiple columns? Say, it i want something like :
ls -l | column 2 3
Add a comment
|
3 Answers
In this case, I wouldn't use awk at all:
columns() { tr -s '[:blank:]' ' ' | cut -d ' ' -f "$@"; }
This uses tr to squeeze all sequences of whitespace to a single space, then cut to extract the fields you're interested in.
ls -l | columns 1,5,9-
Note, you shouldn't parse the output of ls.
1 Comment
Ed Morton
+1 for the right solution. I'd probably use
tr -s '[:blank:]' '\t' just so you don't need to specify -d ' ' in the cut args but nbd.awk -v c="3,5" '
BEGIN{ split(c,a,/[^[:digit:]]+/) }
{ for(i=1;i in a;i++) printf "%s%s", (i==1?"":OFS), $(a[i]); print "" }
'
Use any non-digit(s) you like as the column number separator (e.g. comma or space).
Comments
ls -l | nawk -v r="3,5" 'BEGIN{split(r,a,",")}{for(i in a)printf $a[i]" ";print "\n"}'
Now you can simply change your r variable in shell script and pass it on.
or you can configure it in your shell script and use r=$your_list_var
What ever fields numbers are present in $your_list_var will be printed by awk command.
The example above print 3rd and 5th fields of ls -l output.
2 Comments
glenn jackman
+1, but put the
split in the BEGIN block so you don't have to do it for every line.Ed Morton
Will fail cryptically when the input file contains printf formatting characters and won't necessarily output fields in the input order, nor in the order of the argument. It will also add a space to the end of each line and a spurious newline between each output line.