--- trunk/t/3-normalize-xml.t	2005/11/19 23:48:24	70
+++ trunk/t/3-normalize-xml.t	2006/01/08 21:16:22	370
@@ -1,6 +1,6 @@
 #!/usr/bin/perl -w
 
-use Test::More tests => 74;
+use Test::More tests => 132;
 use Test::Exception;
 use Cwd qw/abs_path/;
 use blib;
@@ -13,12 +13,21 @@
 
 ok(my $abs_path = abs_path($0), "abs_path");
 $abs_path =~ s#/[^/]*$#/#;
-diag "abs_path: $abs_path";
+#diag "abs_path: $abs_path";
 
 throws_ok { new WebPAC::Normalize::XML( lookup_regex => 'foo' ) } qr/pair/, "lookup_regex without lookup";
 throws_ok { new WebPAC::Normalize::XML( lookup => 'bar' ) } qr/pair/, "lookup without lookup_regex";
 
-ok(my $n = new WebPAC::Normalize::XML( debug => 0 ), "new");
+ok(my $n = new WebPAC::Normalize::XML(
+	debug => 0,
+	filter => {
+		regex => sub {
+			my ($val, $regex) = @_;
+			eval "\$val =~ $regex";
+			return $val;
+		},
+	},
+), "new");
 
 throws_ok { $n->open() } qr/tag/, "open without tag";
 throws_ok { $n->open( tag => 'isis' ) } qr/xml_file/, "open without xml_file";
@@ -112,24 +121,141 @@
 		if (ref($item) eq 'HASH') {
 			foreach my $sf (keys %$item) {
 				my $found = 0;
-				ok($n->get_data(\$rec, $fld, $sf, $r, \$found), "v${fld}^${sf}");
+				ok($n->get_data(\$rec, $fld, $sf, $r, \$found), "v${fld}^${sf} / $r");
 				ok($found, "found");
 			}
 			my $found = 0;
-			ok(! $n->get_data(\$rec, $fld, 'x', $r, \$found), "no v${fld}^x");
+			ok(! $n->get_data(\$rec, $fld, 'x', $r, \$found), "no v${fld}^x / $r");
 			ok(! $found, "not found");
 		} else {
 			my $found = 0;
-			ok($n->get_data(\$rec, $fld, '', $r, \$found), "v${fld}");
+			ok($n->get_data(\$rec, $fld, undef, $r, \$found), "v${fld} / $r");
 			ok($found, "found");
 		}
 	}
 	my $found = 0;
-	ok(! $n->get_data(\$rec, '999', '', $r, \$found), "no v${fld}");
+	ok(! $n->get_data(\$rec, '999', '', $r, \$found), "no v${fld} / $r");
 	ok(! $found, "not found");
 }
 
 ok(my $ds = $n->data_structure( $rec ), "data_structure");
 
-diag Dumper($rec, $ds);
+#diag Dumper($rec, $ds);
+
+# fake load of our test normalisation data
+$n->{tag} = 'isis';
+
+#diag Dumper($n->{import_xml}->{indexer});
+
+
+$rec = {
+	'900' => [ {
+		'a' => '1',
+		'b' => '2',
+		'c' => '3',
+		'x' => 'yap',
+	} ],
+};
+
+my $import = {
+	'Tag' => { 'isis' => [
+			{ content => 'v900^a + v900^b = v900^c [v900^x]' },
+	] },
+};
+
+sub parse_test($$$$$) {
+	my ($import,$tag,$rec,$i,$r) = @_;
+	$n->{import_xml}->{indexer} = $import;
+	# erase internal cache (yak!)
+	delete($n->{tags_by_order});
+	push @{$rec->{'000'}}, 42 unless ($rec->{'000'});
+	#diag "test normalisation of: ",Dumper($n->{import_xml}->{indexer}, $rec);
+	ok(my $ds = $n->data_structure( $rec ), "data_structure");
+	diag Dumper($ds) if ($i == 0);
+	cmp_ok($ds->{$tag}->{display}->[$i], 'eq', $r, "parse $tag/$i - $r");
+}
+
+parse_test($import, 'Tag', $rec, 0, '1 + 2 = 3 [yap]');
+
+delete($rec->{'900'}->[0]->{'b'});
+parse_test($import, 'Tag', $rec, 0, '1 = 3 [yap]');
+
+$rec->{'900'}->[0]->{'b'} = 5;
+$rec->{'900'}->[0]->{'c'} = 6;
+parse_test($import, 'Tag', $rec, 0, '1 + 5 = 6 [yap]');
+
+delete($rec->{'900'}->[0]->{'c'});
+$rec->{'900'}->[0]->{'x'} = 'hmmm';
+parse_test($import, 'Tag', $rec, 0, '1 + 5 [hmmm]');
+
+$rec->{'900'}->[0]->{'x'} = 'nope!';
+delete($rec->{'900'}->[0]->{'a'});
+parse_test($import, 'Tag', $rec, 0, '5 [nope!]');
+
+$rec = {
+	'900' => [ {
+		'b' => 'b1',
+		'x' => 'b1',
+	},{
+		'a' => 'a1',
+		'b' => 'b2',
+		'x' => 'a1_b2',
+	},{
+		'b' => 'b3',
+		'c' => 'c1',
+		'x' => 'b3_c1',
+	},{
+		'a' => 'a2',
+		'b' => 'b4',
+		'c' => 'c2',
+		'x' => 'a2_b4_c2',
+	} ],
+};
+
+$import = {
+	'Tag' => { 'isis' => [
+			{ content => '0 v900^a 1 v900^b 2 v900^c 3 v900^x 4' },
+	] },
+};
+
+parse_test($import, 'Tag', $rec, 0, '0 b1 3 b1 4');
+parse_test($import, 'Tag', $rec, 1, '0 a1 1 b2 3 a1_b2 4');
+parse_test($import, 'Tag', $rec, 2, '0 b3 2 c1 3 b3_c1 4');
+parse_test($import, 'Tag', $rec, 3, '0 a2 1 b4 2 c2 3 a2_b4_c2 4');
+
+sub parse_test_arr($$$$) {
+	my ($import, $tag, $rec, $arr) = @_;
+	my $i = 0;
+	foreach my $res (@{$arr}) {
+		parse_test($import, $tag, $rec, $i, $res);
+		$i++;
+	}
+}
+
+$import = {
+	'Tag_a' => { 'isis' => [
+		{ content => 'v900^a' },
+	] },
+	'Tag_b' => { 'isis' => [
+		{ content => 'v900^b' },
+	] },
+	'Tag_c' => { 'isis' => [
+		{ content => 'v900^c' },
+	] },
+	'Tag_x' => { 'isis' => [
+		{ content => 'v900^x' },
+	] },
+	'Tag_s1' => { 'isis' => [
+		{ content => 's900^b = s900^x' },
+	] },
+	'Tag_s2' => { 'isis' => [
+		{ content => 's900^a s900^b s900^c s900^x' },
+	] },
+};
+
+parse_test_arr($import, 'Tag_a', $rec, [ '','a1','','a2' ] );
+parse_test_arr($import, 'Tag_b', $rec, [ 'b1','b2','b3','b4' ] );
+parse_test_arr($import, 'Tag_c', $rec, [ '','','c1','c2' ] );
+parse_test_arr($import, 'Tag_x', $rec, [ 'b1','a1_b2','b3_c1','a2_b4_c2' ] );
+parse_test_arr($import, 'Tag_s1', $rec, [ 'b1 = b1', 'b1 = b1', 'b1 = b1', 'b1 = b1' ] );