--- trunk/lib/WebPAC/Input.pm	2006/10/25 17:10:08	761
+++ trunk/lib/WebPAC/Input.pm	2007/11/28 22:51:43	1076
@@ -7,7 +7,7 @@
 
 use WebPAC::Common;
 use base qw/WebPAC::Common/;
-use Data::Dumper;
+use Data::Dump qw/dump/;
 use Encode qw/from_to/;
 
 =head1 NAME
@@ -16,11 +16,11 @@
 
 =head1 VERSION
 
-Version 0.14
+Version 0.18
 
 =cut
 
-our $VERSION = '0.14';
+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
@@ -102,11 +105,6 @@
 
 	require $module_path;
 
-	# check if required subclasses are implemented
-	foreach my $subclass (qw/open_db fetch_rec init dump_rec/) {
-		# FIXME
-	}
-
 	$self->{'encoding'} ||= 'ISO-8859-2';
 
 	$self ? return $self : return undef;
@@ -229,15 +227,16 @@
 		$log->debug("using modify_file $p");
 		$rec_regex = $self->modify_file_regexps( $p );
 	} elsif (my $h = $arg->{modify_records}) {
-		$log->debug("using modify_records ", Dumper( $h ));
+		$log->debug("using modify_records ", sub { dump( $h ) });
 		$rec_regex = $self->modify_record_regexps(%{ $h });
 	}
-	$log->debug("rec_regex: ", Dumper($rec_regex)) if ($rec_regex);
+	$log->debug("rec_regex: ", sub { dump($rec_regex) }) if ($rec_regex);
 
 	my $class = $self->{module} || $log->logconfess("can't get low-level module name!");
 
 	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);
@@ -287,10 +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
@@ -303,19 +305,26 @@
 					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;
 		});
 
-		$log->debug(sub { Dumper($rec) });
+		$log->debug(sub { dump($rec) });
 
 		if (! $rec) {
 			$log->warn("record $pos empty? skipping...");
@@ -344,8 +353,8 @@
 			foreach my $fld (keys %{ $rec }) {
 				$self->{_stats}->{fld}->{ $fld }++;
 
-				$log->logdie("invalid record fild $fld, not ARRAY")
-					unless (ref($rec->{ $fld }) eq 'ARRAY');
+				#$log->logdie("invalid record fild $fld, not ARRAY")
+				next unless (ref($rec->{ $fld }) eq 'ARRAY');
 	
 				foreach my $row (@{ $rec->{$fld} }) {
 
@@ -474,10 +483,12 @@
 
 sub seek {
 	my $self = shift;
-	my $pos = shift || return;
+	my $pos = shift;
 
 	my $log = $self->_get_logger();
 
+	$log->logconfess("called without pos") unless defined($pos);
+
 	if ($pos < 1) {
 		$log->warn("seek before first record");
 		$pos = 1;
@@ -512,19 +523,26 @@
 
 	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);
 
 			my $o = sprintf("%4s %d ~", $f, $v);
 
 			if (defined($s->{sf}->{$f})) {
+				my @subfields = keys %{ $s->{sf}->{$f} };
 				map {
 					$o .= sprintf(" %s:%d%s", $_, 
 						$s->{sf}->{$f}->{$_}->{count},
 						$s->{sf}->{$f}->{$_}->{repeatable} ? '*' : '',
 					);
-				} sort keys %{ $s->{sf}->{$f} };
+				} (
+					# first indicators and other special subfields
+					sort( grep { length($_)  > 1 } @subfields ),
+					# then subfileds (single char)
+					sort( grep { length($_) == 1 } @subfields ),
+				);
 			}
 
 			if (my $v_r = $s->{repeatable}->{$f}) {
@@ -532,49 +550,87 @@
 			}
 
 			$o;
-		} sort { $a cmp $b } keys %{ $s->{fld} }
+		} sort { 
+			if ( $a =~ m/^\d+$/ && $b =~ m/^\d+$/ ) {
+				$a <=> $b
+			} else {
+				$a cmp $b
+			}
+		} keys %{ $s->{fld} }
 	);
 
-	$log->debug( sub { Dumper($s) } );
+	$log->debug( sub { dump($s) } );
 
 	return $out;
 }
 
-=head2 dump
+=head2 dump_ascii
 
 Display humanly readable dump of record
 
 =cut
 
-sub dump {
+sub dump_ascii {
 	my $self = shift;
 
-	return $self->{ll_db}->dump_rec( $self->{pos} );
+	return unless $self->{ll_db};
 
+	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
 
 sub _get_regex {
 	my ($sf,$from,$to) = @_;
+
+	# protect /
+	$from =~ s!/!\\/!gs;
+	$to =~ s!/!\\/!gs;
+
+	if ($from =~ m/^regex:(.+)$/) {
+		$from = $1;
+	} else {
+		$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([^\^]*?)\Q'. $from .'\E([^\^]*?)/'. $sf .'$1'. $to .'$2/';
+			's/\Q'. $sf .'\E([^\^]' . $need_subfield_data . '?)'. $from .'([^\^]*?)/'. $sf .'$1'. $to .'$2/';
 	} else {
 		return
-			's/\Q'. $from .'\E/'. $to .'/g';
+			'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 = {@_};
@@ -592,10 +648,11 @@
 			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");
 			}
 		}
@@ -606,7 +663,7 @@
 
 =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:
@@ -656,7 +713,11 @@
 			$log->debug("transform: |$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");
 		}
 	}