--- trunk/IsisDB.pm	2004/12/29 16:01:41	9
+++ trunk/IsisDB.pm	2004/12/29 20:10:11	12
@@ -7,7 +7,7 @@
 BEGIN {
 	use Exporter ();
 	use vars qw ($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS);
-	$VERSION     = 0.03;
+	$VERSION     = 0.04;
 	@ISA         = qw (Exporter);
 	#Give a hoot don't pollute, do not export more than needed by default
 	@EXPORT      = qw ();
@@ -22,16 +22,29 @@
 
 =head1 SYNOPSIS
 
-  use IsisDB
+  use IsisDB;
+
   my $isis = new IsisDB(
   	isisdb => './cds/cds',
   );
 
+  for(my $mfn = 1; $mfn <= $isis->{'maxmfn'}; $mfn++) {
+	print $isis->to_ascii($mfn),"\n";
+  }
+
 =head1 DESCRIPTION
 
 This module will read CDS/ISIS databases and create hash values out of it.
 It can be used as perl-only alternative to OpenIsis module.
 
+This will module will always be slower that OpenIsis module which use C
+library. However, since it's written in perl, it's platform independent (so
+you don't need C compiler), and can be easily modified.
+
+Unique feature of this module is ability to C<include_deleted> records.
+It will also skip zero sized fields (OpenIsis has a bug in XS bindings, so
+fields which are zero sized will be filled with random junk from memory).
+
 =head1 METHODS
 
 =cut
@@ -50,14 +63,6 @@
 # some binary reads
 #
 
-sub Read32 {
-	my $self = shift;
-
-	my $f = shift || die "Read32 needs file handle";
-	read($$f,$b,4) || die "can't read 4 bytes from $$f from position ".tell($f);
-	return unpack("l",$b);
-}
-
 =head2 new
 
 Open CDS/ISIS database
@@ -65,8 +70,12 @@
  my $isis = new IsisDB(
  	isisdb => './cds/cds',
 	read_fdt => 1,
-	debug => 1,
 	include_deleted => 1,
+	hash_filter => sub {
+		my $v = shift;
+		$v =~ s#foo#bar#g;
+	},
+	debug => 1,
  );
 
 Options are described below:
@@ -83,13 +92,17 @@
 Boolean flag to specify if field definition table should be read. It's off
 by default.
 
-=item debug
+=item include_deleted
 
-Dump a C<lot> of debugging output.
+Don't skip logically deleted records in ISIS.
 
-=item include_deleted
+=item hash_filter
+
+Filter code ref which will be used before data is converted to hash.
 
-Don't skip logically deleted records.
+=item debug
+
+Dump a B<lot> of debugging output.
 
 =back
 
@@ -104,7 +117,7 @@
 
 	croak "new needs database name (isisdb) as argument!" unless ({@_}->{isisdb});
 
-	foreach my $v (qw{isisdb debug include_deleted}) {
+	foreach my $v (qw{isisdb debug include_deleted hash_filter}) {
 		$self->{$v} = {@_}->{$v};
 	}
 
@@ -147,7 +160,11 @@
 	# NXTMFP	offset to next available position in last block
 	# MFTYPE	always 0 for user db file (1 for system)
 	seek(fileMST,4,0);
-	$self->{'NXTMFN'}=$self->Read32(\*fileMST) || carp "NXTNFN is zero";
+
+	my $buff;
+
+	read(fileMST, $buff, 4);
+	$self->{'NXTMFN'}=unpack("l",$buff) || carp "NXTNFN is zero";
 
 	# save maximum MFN
 	$self->{'maxmfn'} = $self->{'NXTMFN'} - 1;
@@ -188,7 +205,6 @@
 		}
 	}
 
-	my $buff;
 	read(fileCNT, $buff, 26);
 	$self->unpack_cnt($buff);
 
@@ -215,7 +231,7 @@
   my $rec = $isis->fetch(55);
 
 Returns hash with keys which are field names and values are unpacked values
-for that field.
+for that field (like C<^asometing^bsomething else>)
 
 =cut
 
@@ -232,8 +248,11 @@
 	print "seeking to $mfnpos in file '$self->{isisdb}.XRF'\n" if ($self->{debug});
 	seek($self->{'fileXRF'},$mfnpos,0);
 
+	my $buff;
+
 	# read XRFMFB abd XRFMFP
-	my $pointer=$self->Read32(\*{$self->{'fileXRF'}});
+	read($self->{'fileXRF'}, $buff, 4);
+	my $pointer=unpack("l",$buff) || carp "pointer is null";
 
 	my $XRFMFB = int($pointer/2048);
 	my $XRFMFP = $pointer - ($XRFMFB*2048);
@@ -258,7 +277,8 @@
 
 	seek($self->{'fileMST'},$offset4,0);
 
-	my $value=$self->Read32(\*{$self->{'fileMST'}});
+	read($self->{'fileMST'}, $buff, 4);
+	my $value=unpack("l",$buff);
 
 	if ($value!=$mfn) {
 print ("Error: The MFN:".$mfn." is not found in MST(".$value.")");    
@@ -272,7 +292,6 @@
 #	$NVF=$self->Read16($fileMST);
 #	$STATUS=$self->Read16($fileMST);
 
-	my $buff;
 	read($self->{'fileMST'}, $buff, 14);
 
 	my ($MFRL,$MFBWB,$MFBWP,$BASE,$NVF,$STATUS) = unpack("slssss", $buff);
@@ -329,6 +348,9 @@
 	read($self->{'fileMST'},$buff,$fld_len);
 
 	for (my $i = 0 ; $i < $NVF ; $i++) {
+		# skip zero-sized fields
+		next if ($FieldLEN[$i] == 0);
+
 		push @{$self->{record}->{$FieldTAG[$i]}}, substr($buff,$FieldPOS[$i],$FieldLEN[$i]);
 	}
 	close(fileMST);
@@ -364,6 +386,69 @@
 	return $out;
 }
 
+=head2 to_hash
+
+Read mfn and convert it to hash
+
+  my $hash = $isis->to_hash($mfn);
+
+It has ability to convert characters (using C<hash_filter> from ISIS
+database before creating structures enabling character remapping or quick
+fixup of data.
+
+This function returns hash which is like this:
+
+  $hash = {
+    '210' => [
+               {
+                 'c' => 'New York University press',
+                 'a' => 'New York',
+                 'd' => 'cop. 1988'
+               }
+             ],
+    '990' => [
+               '2140',
+               '88',
+               'HAY'
+             ],
+  };
+
+You can later use that has to produce any output from ISIS data.
+
+=cut
+
+sub to_hash {
+	my $self = shift;
+
+	my $mfn = shift || confess "need mfn!";
+
+	my $rec;
+	my $row = $self->fetch($mfn);
+
+	foreach my $k (keys %{$row}) {
+		foreach my $l (@{$row->{$k}}) {
+
+			# filter output
+			$l = $self->{'hash_filter'}->($l) if ($self->{'hash_filter'});
+
+			# has subfields?
+			my $val;
+			if ($l =~ m/\^/) {
+				foreach my $t (split(/\^/,$l)) {
+					next if (! $t);
+					$val->{substr($t,0,1)} = substr($t,1);
+				}
+			} else {
+				$val = $l;
+			}
+
+			push @{$rec->{$k}}, $val;
+		}
+	}
+
+	return $rec;
+}
+
 #
 # XXX porting from php left-over:
 #