From deb9aa80747003db75d8efc558c7a3c66a406e36 Mon Sep 17 00:00:00 2001
From: Evan Carroll <ecarroll@x60s.(none)>
Date: Tue, 17 Jun 2008 15:57:58 -0500
Subject: [PATCH] added the ability to supply a first row shorter than the set
 of data

---
 lib/DataExtract/FixedWidth.pm                   | 23 +++++++++++-------
 t/11-Heuristic-woHeader-wCols-wUndefHeaderCol.t | 31 +++++++++++++++++++++++++
 2 files changed, 45 insertions(+), 9 deletions(-)
 create mode 100644 t/11-Heuristic-woHeader-wCols-wUndefHeaderCol.t

diff --git a/lib/DataExtract/FixedWidth.pm b/lib/DataExtract/FixedWidth.pm
index 76a2da5..e06d3a7 100644
--- a/lib/DataExtract/FixedWidth.pm
+++ b/lib/DataExtract/FixedWidth.pm
@@ -88,16 +88,21 @@ has 'skip_header_data' => (
 
 sub _heuristic_trigger {
 	my ( $self, $data ) = @_;
-	
+
 	chomp @$data;
 
+	my $maxLength = 0;
+	for ( @$data ) {
+		$maxLength = length if length > $maxLength
+	}
+
 	$self->header_row( $data->[0] )
 		unless $self->has_header_row
 	;
-	
+
 	{
 		my @unpack;
-		my $mask = ' ' x length $data->[0];
+		my $mask = ' ' x $maxLength;
 		$mask |= $_ for @$data;
 
 		push @unpack, length($1)
@@ -149,7 +154,7 @@ sub _build_colchar_map {
 	my $self = shift;
 	my $ccm = {};
 
-	## If we can generate from heurisitic data and don't have a header_row	
+	## If we can generate from heurisitic data and don't have a header_row
 	if (
 		$self->has_header_row
 		&& !defined $self->header_row
@@ -165,7 +170,7 @@ sub _build_colchar_map {
 	## Generate from header_row
 	else {
 		croak 'Can not render the map of columns to start-chars without the header_row'
-			unless defined $self->has_header_row 
+			unless defined $self->has_header_row
 		;
 
 		foreach my $col ( $self->cols ) {
@@ -329,13 +334,13 @@ DataExtract::FixedWidth - The one stop shop for parsing static column width text
 
 	## We assume the columns have no spaces in the header.
 	my $de = DataExtract::FixedWidth->new({ header_row => $header_row });
-
+	
 	## We explicitly tell what column names to pick out of the header.
 	my $de = DataExtract::FixedWidth->new({
 		header_row => $header_row
 		cols       => [qw/COL1NAME COL2NAME COL3NAME/, 'COL WITH SPACE IN NAME']
 	});
-
+	
 	## We supply data to heuristic and assume
 	## * first row is the header (to avoid this assumption
 	##   set the header_row to undef. )
@@ -368,9 +373,9 @@ DataExtract::FixedWidth - The one stop shop for parsing static column width text
 		unpack_string => $template
 		, header_row  => $header_row
 	});
-
+	
 	$de->parse( $data_row );
-
+	
 	$de->parse_hash( $data_row );
 
 =head1 DESCRIPTION
diff --git a/t/11-Heuristic-woHeader-wCols-wUndefHeaderCol.t b/t/11-Heuristic-woHeader-wCols-wUndefHeaderCol.t
new file mode 100644
index 0000000..8e756ef
--- /dev/null
+++ b/t/11-Heuristic-woHeader-wCols-wUndefHeaderCol.t
@@ -0,0 +1,31 @@
+#!/usr/bin/env perl
+use strict;
+use warnings;
+use feature ':5.10';
+
+##
+## In version .06 this test failed
+## I got the longer a19a9a14a13a5A* instead of a19a9a14a13A*
+## This was because the length for the euristic was set statically to the
+## first row rather than the longest row
+##
+
+use DataExtract::FixedWidth;
+use IO::File;
+
+use Test::More tests => 1;
+
+use File::Spec;
+my $file = File::Spec->catfile( 't', 'data', 'larochenew.TXT' );
+my $fh = IO::File->new( $file );
+my @lines = grep /\w/, $fh->getlines;
+
+my $defw = DataExtract::FixedWidth->new({
+	heuristic => \@lines
+	, cols    => [ qw/vin stock color price miles/ ]
+	, header_row => undef
+});
+
+is ( $defw->unpack_string, 'a19a9a14a13A*', 'Heuristic not affected by being short' );
+
+1;
-- 
2.11.4.GIT