sfactor
sfactor

Reputation: 13062

Parsing tab delimited file with double quotes in Perl

I have a data set that is tab delimited with the user-agent strings in double quotes. I need to parse each of these columns and based on the answer of my other post I used the Text::CSV module.

94410634  0   GET  "Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0; GTB6.6; .NET CLR 2.0.50727; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; AskTB5.5)"   1

The code is a simple one.

#!/usr/bin/perl

use strict;
use warnings;
use Text::CSV;

my $csv = Text::CSV->new(sep_char => "\t");

    while (<>) {
        if ($csv->parse($_)) {
            my @columns = $csv->fields();
            print "@columns\n";
        } else {
            my $err = $csv->error_input;
            print "Failed to parse line: $err";
        }
    }

But i get the Failed to parse line: error when I try it on this dataset. what am I doing wrong? I need to extract the 4th column containing the user-agent strings for further processing.

Upvotes: 2

Views: 1694

Answers (1)

DVK
DVK

Reputation: 129549

  1. Your constructor arguments should be in a hashref, not a hash:

    my $csv = Text::CSV->new( { sep_char => "\t" } );

  2. Are you sure the dataset is exactly what you think it is? May be there's a double quote missing somewhere or there were no tabs?

    To verify the file contents, are you on Unix/Linux or Windows? On unix, please run this: cat -vet my_log_file_name | head -3 and check whether the output has spaces or "^I" sequences where you expect tabs. cat -vet prints out all the special characters as special printable sequences (TAB => ^I, newline => $, etc...)

The following test works perfectly on my ActivePerl:

#!/usr/bin/perl
use strict;
use warnings;
use Text::CSV;

my $s = qq[94410634\t0\tGET\t"Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0; GTB6.6; .NET CLR 2.0.50727; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; AskTB5.5)"\t1\n];;
my $csv = Text::CSV->new({sep_char => "\t"});

if ($csv->parse($s)) {
    my @columns = $csv->fields();
    print "c=$columns[3]\n";
} else {
    my $err = $csv->error_input;
    print "Failed to parse line: $err";
}

Output:

C:\> perl d:\scripts\test4.pl
c=Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0; GTB6.6; ...

Upvotes: 6

Related Questions