setting the output field separator in awk

问题:

I'n trying this statement in my awk script (in a file containing separate code, so not inline), script name: print-table.awk

BEGIN {FS = "\t";OFS = "," ; print "about to open the file"}
{print $0}
END {print "about to close stream" }

and running it this way from the shell

awk -f print-table.awk table

Where table is a tab separated file, My goal is to declare the field separator (FS) and the output field separator (OFS) within the external function, and calling from the shell simply the

awk -f file input

without setting the field separator in the command line with -F"\t" and without stdout it to a sed statement replacing the tab with a comma,

Any advise how can i do that?

回答1:

You need to convince awk that something has changed to get it to reformat $0 using your OFS. The following works though there may be a more idiomatic way to do it.

BEGIN {FS = "\t";OFS = "," ; print "about to open the file"}
{$1=$1}1
END {print "about to close stream" }

回答2:

You need to alter one of the field in awk:

awk 'BEGIN {FS="\t";OFS=","; print "about to open the file"} {$1=$1}1' file