PERL Fixed Width to CSV based on Input Files - perl

Question

EDITED: I'm attempting to create a brief script that calls for an input fixed width file and a file with the start position and length of each attribute and then outputs the file as CSV instead of fixed width. I haven't messed with removing whitespace yet and am currently focusing on building the file reader portion.
Fixed:
My current issue is that this code returns data from the third row for $StartPosition and from the fourth row for $Length when they should both be first found on the first row of COMMA. I have no idea what is prompting this behavior.
Next issue: It only reads the first record in practice_data.txt I'm guessing it's something where I need to tell COMMA to go back to the beginning?
while (my $sourceLine = <SOURCE>) {
$StartPosition = 0;
$Length = 0;
$Output = "";
$NextRecord ="";
while (my $commaLine = <COMMA>) {
my $Comma = index($commaLine, ',');
print "Comma location found at $Comma \n";
$StartPosition = substr($commaLine, 0, $Comma);
print "Start position is $StartPosition \n";
$Comma = $Comma + 1
$Length = substr($commaLine, $Comma);
print "Length is $Length \n";
$NextRecord = substr($sourceLine, $StartPosition, $Length);
$Output = "$Output . ',' . $NextRecord";
}
print OUTPUT "$Output \n";
}
practice_data.txt
1234512345John Doe 123 Mulberry Lane Columbus Ohio 43215Johnny Jane
5432154321Jason McKinny 423 Thursday Lane Columbus Ohio 43212Jase Jamie
4321543212Mike Jameson 289 Front Street Cleveland Ohio 43623James Sarah
Each record is 100 characters long.
Definitions.txt:
0,10
10,10
20,10
30,20
50,10
60,10
70,5
75,15
90,10

Answer 1

It always helps to provide enough information so that we can at least do some testing without having to read your code and imagine what the data must look like.
I suggest you use unpack, after first building a template from the file that holds the field specifications. Note that the A field specifier trims trailing spaces from the data.
It is all but essential to use the Text::CSV module to parse or generate well-formed CSV data. And I have used the autodie pragma to avoid having to explicitly check and report on the status of every I/O operation.
I have used this data
my_source_data.txt
12345678 ABCDE1234FGHIJK
my_field_spec.txt
0,8
10,5
15,4
19,6
And this program
use strict;
use warnings;
use 5.010;
use autodie;
use Text::CSV;
my #template;
open my $field_fh, '<', 'my_field_spec.txt';
while ( <$field_fh> ) {
my (#info) = /\d+/g;
die unless #info == 2;
push #template, sprintf '#%dA%d', #info;
}
my $template = "#template";
open my $source_fh, '<', 'my_source_data.txt';
my $csv = Text::CSV->new( { binary => 1, eol => $/ } );
while ( <$source_fh> ) {
my #fields = unpack $template;
$csv->print(\*STDOUT, \#fields);
}
output
12345678,ABCDE,1234,FGHIJK

Answer 2

It looks like you're slightly confused on how to read the contents of the COMMA filehandle.. Each time you read <COMMA>, you're reading another line from that file. Instead, read a line into a scalar like my $line = <FH> and use that instead:
while (my $source_line = <SOURCE>) {
$StartPosition = 0;
$Length = 0;
$Output = "";
$Input = $_;
$NextRecord ="";
while (my $comma_line = <COMMA>) {
my $Comma = index($comma_line, ',');
print "Comma location found at $Comma \n";
$StartPosition = substr($comma_line, 0, $Comma);
print "Start position is $StartPosition \n";
$Length = substr($comma_line, $Comma);
print "Length is $Length \n";
$NextRecord = substr($Input, $StartPosition, $Length) + ',';
$Output = "$Output$NextRecord";
}
print OUTPUT "$Output \n";
}