--- trunk/lib/WebPAC/Input.pm	2007/02/04 13:28:30	797
+++ trunk/lib/WebPAC/Input.pm	2007/06/22 00:03:46	873
@@ -16,11 +16,11 @@
 
 =head1 VERSION
 
-Version 0.17
+Version 0.18
 
 =cut
 
-our $VERSION = '0.17';
+our $VERSION = '0.18';
 
 =head1 SYNOPSIS
 
@@ -64,6 +64,9 @@
 	encoding => 'ISO-8859-2',
 	recode => 'char pairs',
 	no_progress_bar => 1,
+	input_config => {
+		mapping => [ 'foo', 'bar', 'baz' ],
+	},
   );
 
 C<module> is low-level file format module. See L<WebPAC::Input::ISIS> and
@@ -233,6 +236,7 @@
 
 	my $ll_db = $class->new(
 		path => $arg->{path},
+		input_config => $arg->{input_config} || $self->{input_config},
 #		filter => sub {
 #			my ($l,$f_nr) = @_;
 #			return unless defined($l);
@@ -282,10 +286,13 @@
 		$log->debug("position: $pos\n");
 
 		my $rec = $ll_db->fetch_rec($pos, sub {
-				my ($l,$f_nr) = @_;
+				my ($l,$f_nr,$debug) = @_;
 #				return unless defined($l);
 #				return $l unless ($rec_regex && $f_nr);
 
+				return unless ( defined($l) && defined($f_nr) );
+
+				warn "-=> $f_nr ## |$l|\n" if ($debug);
 				$log->debug("-=> $f_nr ## $l");
 
 				# codepage conversion and recode_regex
@@ -298,15 +305,22 @@
 					my $c = 0;
 					foreach my $r (@{ $rec_regex->{$f_nr} }) {
 						my $old_l = $l;
-						eval '$l =~ ' . $r;
+						$log->logconfess("expected regex in ", dump( $r )) unless defined($r->{regex});
+						eval '$l =~ ' . $r->{regex};
 						if ($old_l ne $l) {
-							$log->debug("REGEX on $f_nr eval \$l =~ $r\n## old l: [$old_l]\n## new l: [$l]");
+							my $d = "|$old_l| -> |$l| "; # . $r->{regex};
+							$d .= ' +' . $r->{line} . ' ' . $r->{file} if defined($r->{line});
+							$d .= ' ' . $r->{debug} if defined($r->{debug});
+							$log->debug("MODIFY $d");
+							warn "*** $d\n" if ($debug);
+
 						}
 						$log->error("error applying regex: $r") if ($@);
 					}
 				}
 
-				$log->debug("<=- $f_nr ## $l");
+				$log->debug("<=- $f_nr ## |$l|");
+				warn "<=- $f_nr ## $l\n" if ($debug);
 				return $l;
 		});
 
@@ -509,7 +523,8 @@
 
 	my $out = join("\n",
 		map {
-			my $f = $_ || die "no field";
+			my $f = $_;
+			die "no field in ", dump( $s->{fld} ) unless defined( $f );
 			my $v = $s->{fld}->{$f} || die "no s->{fld}->{$f}";
 			$max_fld = $v if ($v > $max_fld);
 
@@ -529,7 +544,7 @@
 			}
 
 			$o;
-		} sort { $a cmp $b } keys %{ $s->{fld} }
+		} sort { $a <=> $b } keys %{ $s->{fld} }
 	);
 
 	$log->debug( sub { dump($s) } );
@@ -548,21 +563,22 @@
 
 	return unless $self->{ll_db};
 
-	if ($self->{ll_db}->can('dump_rec')) {
+	if ($self->{ll_db}->can('dump_ascii')) {
 		return $self->{ll_db}->dump_ascii( $self->{pos} );
 	} else {
 		return dump( $self->{ll_db}->fetch_rec( $self->{pos} ) );
 	}
 }
 
-=head2 modify_record_regexps
+=head2 _get_regex
 
-Generate hash with regexpes to be applied using l<filter>.
+Helper function called which create regexps to be execute on code.
 
-  my $regexpes = $input->modify_record_regexps(
-		900 => { '^a' => { ' : ' => '^b' } },
-		901 => { '*' => { '^b' => ' ; ' } },
-  );
+  _get_regex( 900, 'regex:[0-9]+' ,'numbers' );
+  _get_regex( 900, '^b', ' : ^b' );
+
+It supports perl regexps with C<regex:> prefix to from value and has
+additional logic to skip empty subfields.
 
 =cut
 
@@ -579,14 +595,30 @@
 		$from = '\Q' . $from . '\E';
 	}
 	if ($sf =~ /^\^/) {
+		my $need_subfield_data = '*';	# no
+		# if from is also subfield, require some data in between
+		# to correctly skip empty subfields
+		$need_subfield_data = '+' if ($from =~ m/^\\Q\^/);
 		return
-			's/\Q'. $sf .'\E([^\^]*?)'. $from .'([^\^]*?)/'. $sf .'$1'. $to .'$2/';
+			's/\Q'. $sf .'\E([^\^]' . $need_subfield_data . '?)'. $from .'([^\^]*?)/'. $sf .'$1'. $to .'$2/';
 	} else {
 		return
 			's/'. $from .'/'. $to .'/g';
 	}
 }
 
+
+=head2 modify_record_regexps
+
+Generate hash with regexpes to be applied using L<filter>.
+
+  my $regexpes = $input->modify_record_regexps(
+		900 => { '^a' => { ' : ' => '^b' } },
+		901 => { '*' => { '^b' => ' ; ' } },
+  );
+
+=cut
+
 sub modify_record_regexps {
 	my $self = shift;
 	my $modify_record = {@_};
@@ -604,10 +636,11 @@
 			foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) {
 				my $to = $modify_record->{$f}->{$sf}->{$from};
 				#die "no field?" unless defined($to);
-				$log->debug("transform: |$from| -> |$to|");
+				my $d = "|$from| -> |$to|";
+				$log->debug("transform: $d");
 
 				my $regex = _get_regex($sf,$from,$to);
-				push @{ $regexpes->{$f} }, $regex;
+				push @{ $regexpes->{$f} }, { regex => $regex, debug => $d };
 				$log->debug("regex: $regex");
 			}
 		}
@@ -618,7 +651,7 @@
 
 =head2 modify_file_regexps
 
-Generate hash with regexpes to be applied using l<filter> from
+Generate hash with regexpes to be applied using L<filter> from
 pseudo hash/yaml format for regex mappings.
 
 It should be obvious:
@@ -668,7 +701,11 @@
 			$log->debug("transform: |$from| -> |$to|");
 
 			my $regex = _get_regex($sf,$from,$to);
-			push @{ $regexpes->{$f} }, $regex;
+			push @{ $regexpes->{$f} }, {
+				regex => $regex,
+				file => $modify_path,
+				line => $.,
+			};
 			$log->debug("regex: $regex");
 		}
 	}