Question

I am trying to use awk to read some input at the field position at 3, $3, field 3 is a string

awk -F'","' '{print $1}'  input.txt

my file input.txt looks like this

field1,field2,field3,field4,field5

the problem is that these fields are separated by commas, some of them are double quoted while others are not. And field 5 is double quoted and contains every type of symbols. Example:

imfield1,imfield2,"imfield3",imfield4,"im"",""fi"",el,""d5"

can awk handle a situation like this??
In more gner, how can I get the whole string by typIng $5 ?

Answer 1

You can use Lorance Stinson’s Awk CSV parser, in which case it’s as simple as:

function parse_csv(..) {
    ..
}

{
    num_fields = parse_csv($0, csv, ",", "\"", "\"", "\\n", 1);
    print csv[2]
}

If you’re not hell-bent on Awk, Python also comes with a nice CSV parser:

import csv, sys

for row in csv.reader(sys.stdin):
    print row[2]

Or from the command line (bit tricky in one line):

python -c 'import csv,sys;[sys.stdout.write(row[2]+"\n") for row in csv.reader(sys.stdin)]' < input.txt