script conversion from Perl to shell [closed] - perl

Question

Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
Questions asking for code must demonstrate a minimal understanding of the problem being solved. Include attempted solutions, why they didn't work, and the expected results. See also: Stack Overflow question checklist
Closed 9 years ago.
Improve this question
following is the code in perl.
Can we write the same thing in shell scripts ??
If yes how ?
I have used associative arrays but unable to achieve what this is doing
open MYFILE, "<", "$ARGV[0]" or die "Can't open $ARGV[0] file \n";
############ to retieve the info and put them in associative arrray ##############
$line = <MYFILE>;
#line1 = split(/,/ , $line);
$length = #line1;
$count = 0;
while($count < $length)
{
$line1[$count] =~ s/^\"//;
$line1[$count] =~ s/\"$//;
$count++;
}
$line = <MYFILE>;
#line2 = split(/,/ , $line);
$length = #line2;
$count = 0;
while($count < $length)
{
$line2[$count] =~ s/^\"//;
$line2[$count] =~ s/\"$//;
$count++;
}
$count = 0;
while($count < $length)
{
$array{$line1[$count]}=$line2[$count];
$count++;
}

Answer 1

Of course you can translate that to a shell script: Just wrap the perl script in a here-doc, pass it to perl, and put #!/bin/sh at the top…
#!/bin/sh
perl - <<'END' $1
...
END
But more seriously, you might achieve enlightenment by rewriting the code in a different fashion. What you are doing is reading a line, splitting it at commata, and removing quotation marks at the beginning and end of each field:
sub get_fields {
map { s/^"//; s/"$//; $_ } split /,/, $_[0];
}
my #keys = get_fields scalar <>; # 1st line
my #vals = get_fields scalar <>; # 2nd line
my %hash;
#hash{ #line1 } = #line2;
Except for the slice operation at the end, you can now more easily rewrite the code because it uses data flow instead of structured programming as the predominant paradigm. Not to mention that my code is shorter by an order of magnitude (in base 3).
If you are writing code for production purposes, don't do this. It will break. I assume you are processing CSV. Stick with Perl, and use Text::CSV. Then:
use strict; use warnings; use autodie;
use Text::CSV;
my $csv = Text::CSV->new({ binary => 1 });
open my $fh, "<:utf8", $ARGV[0];
my $keys = $csv->getline($fh);
my $vals = $csv->getline($fh);
my %hash;
#hash{#$keys} = #$vals;
It isn't even much longer, but very unlikely to break (It doesn't split on commas inside quotes).