###########################################################################
#
# ISISPlug.pm -- A plugin for CDS/ISIS databases
#
# A component of the Greenstone digital library software
# from the New Zealand Digital Library Project at the 
# University of Waikato, New Zealand.
#
# Copyright 1999-2003 New Zealand Digital Library Project
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
#
###########################################################################

package ISISPlug;


use multiread;
use SplitPlug;


# ISISPlug is a sub-class of SplitPlug.
sub BEGIN {
    @ISA = ('SplitPlug');
}


my $arguments = 
    [ { 'name' => "process_exp",
	'desc' => "{BasPlug.process_exp}",
	'type' => "string",
	'reqd' => "no",
	'deft' => &get_default_process_exp() },
      { 'name' => "block_exp",
	'desc' => "{BasPlug.block_exp}",
	'type' => "string",
	'deft' => &get_default_block_exp() },
      { 'name' => "subfield_separator",
	'desc' => "{ISISPlug.subfield_separator}",
	'type' => "string",
	'reqd' => "no",
	'deft' => ", " },
      { 'name' => "entry_separator",
	'desc' => "{ISISPlug.entry_separator}",
	'type' => "string",
	'reqd' => "no",
	'deft' => "<br>" } ];

my $options = { 'name'     => "ISISPlug",
		'desc'     => "{ISISPlug.desc}",
		'inherits' => "Yes",
		'args'     => $arguments };


# This plugin processes files with the suffix ".mst"
sub get_default_process_exp {
    return q^(?i)(\.mst)$^;
}


# This plugin blocks files with the suffix ".fdt" and ".xrf"
sub get_default_block_exp {
    return q^(?i)(\.fdt|\.xrf)$^;
}

    
# This plugin splits the input text at the "----------" lines
sub get_default_split_exp {
    return q^\n----------\n^;
}


sub new {
    my $class = shift(@_);

    my $self = new SplitPlug($class, @_);
    if (!parsargv::parse(\@_, 
			 q^subfield_separator/.*/, ^, \$self->{'subfield_separator'},
			 q^entry_separator/.*/<br>^, \$self->{'entry_separator'},
			 "allow_extra_options")) {
	print STDERR "\nIncorrect options passed to ISISPlug, check your collect.cfg configuration file\n";
	die "\n";
    }

    # To allow for proper inheritance of arguments
    my $option_list = $self->{'option_list'};
    push(@{$option_list}, $options);
    $self->{'plugin_type'} = "ISISPlug";

    return bless $self, $class;
}


sub read_file {
    my $self = shift (@_);
    my ($filename, $encoding, $language, $textref) = @_;

    my ($databasename) = ($filename =~ /([^\.]+)\.mst$/i);

    # The text to split is exported from the database by the IsisGdl program
    open(FILE, "IsisGdl $filename |");

    my $reader = new multiread();
    $reader->set_handle ('ISISPlug::FILE');
    $reader->set_encoding ($encoding);
    $reader->read_file ($textref);

    close(FILE);

    # Parse the associated ISIS database Field Definition Table file (.fdt)
    my $fdtfilename = $databasename . ".fdt";
    my %fdtmapping = &parse_field_definition_table($fdtfilename);

    # Map the tag numbers to tag names, using the FDT mapping
    $$textref =~ s/\ntag=(\d+) /\ntag=$fdtmapping{$1}{'title'} /g;

    # Add a newline at the start so it is split properly
    $$textref = "\n" . $$textref;
}


sub process
{
    my $self = shift (@_);
    my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj) = @_;
    my $outhandle = $self->{'outhandle'};

    my $subfield_separator = $self->{'subfield_separator'};
    my $entry_separator = $self->{'entry_separator'};

    # Report that we're processing the file
    print $outhandle "IsisPlug: processing $file\n"
	if ($self->{'verbosity'}) > 1;

    # Process each line of the ISIS record, one at a time
    foreach $line (split(/\n/, $$textref)) {
	$line =~ /^tag=(.+) data=(.+)$/;
	local $rawtagname = $1;
	local $rawtagdata = $2;
	# print "Raw tag: $rawtagname, Raw data: $rawtagdata\n";

	# Metadata field names: title case, then remove spaces
	local $tagname = "";
	foreach $word (split(/\s+/, $rawtagname)) {
	    substr($word, 0, 1) =~ tr/a-z/A-Z/;
	    $tagname .= $word;
	}

	# Make sure there is nothing bad in the tag names
	$tagname =~ s/&//g;

	# Handle each piece of metadata ('%' separated)
	local $completetagvalue = "";
	foreach $rawtagvalue (split(/%/, $rawtagdata)) {
	    $completetagvalue .= $entry_separator unless ($completetagvalue eq "");

	    # Metadata field values: take care with subfields
	    local $completeentryvalue = "";
	    while ($rawtagvalue ne "") {
		# If there is a subfield specifier, parse it off
		local $subfieldname = "";
		if ($rawtagvalue =~ s/^\^([a-z])//) {
		    $subfieldname = "." . $1;
		}

		# Parse the metadata value off
		$rawtagvalue =~ s/^([^\^]*)//;
		local $metadatafieldname = $tagname . $subfieldname;
		local $metadatafieldvalue = $1;
		# print "Metadata: $metadatafieldname -> $metadatafieldvalue\n";

		# Handle Keywords specially
		if ($metadatafieldname eq "Keywords") {
		    local $keywordmetadatavalue = $metadatafieldvalue;
		    local $keywordlist = "";
		    while ($keywordmetadatavalue =~ s/\<([^\>]+)\>//) {
			local $keyword = $1;
			$doc_obj->add_utf8_metadata($cursection, $metadatafieldname, $keyword); 
			$keywordlist .= ", " unless ($keywordlist eq "");
			$keywordlist .= $keyword;
		    }

		    $metadatafieldvalue = $keywordlist;
		}

		else {
		    $doc_obj->add_utf8_metadata($cursection, $metadatafieldname, $metadatafieldvalue); 
		}

		$completeentryvalue .= $subfield_separator unless ($completeentryvalue eq "");
		$completeentryvalue .= $metadatafieldvalue;
	    }

	    $completetagvalue .= $completeentryvalue;
	}

	# print "Metadata: $tagname.all -> $completetagvalue\n";
	$doc_obj->add_utf8_metadata($cursection, $tagname . ".all", $completetagvalue); 
    }
    # print "\n";

    # Add the full record as the document text
    $$textref =~ s/\</&lt;/g;
    $$textref =~ s/\>/&gt;/g;
    $doc_obj->add_utf8_text ($cursection, $$textref);

    # Document was processed successfully
    return 1;
}


sub parse_field_definition_table
{
    local $fdtfilename = shift(@_);

    local %fdtmapping = ();

    open(FDT_FILE, "<$fdtfilename") || die "Error: Could not open file $fdtfilename.\n";

    local $amongstdefinitions = 0;
    foreach $fdtfileline (<FDT_FILE>) {
	$fdtfileline =~ s/(\s*)$//;  # Remove any nasty spaces at the end of the lines

	if ($amongstdefinitions) {
	    local $fieldtitle     = substr($fdtfileline,  0, 30);
	    local $fieldsubfields = substr($fdtfileline, 30, 20);
	    local $fieldspecs     = substr($fdtfileline, 50);

	    # Remove extra spaces
	    $fieldtitle =~ s/(\s*)$//;
	    $fieldsubfields =~ s/(\s*)$//;

	    # Map from tag number to metadata field title and subfields
	    local $fieldtag = (split(/ /, $fieldspecs))[0];
	    $fdtmapping{$fieldtag} = { 'title' => $fieldtitle,
				       'subfields' => $fieldsubfields };
	}
	elsif ($fdtfileline eq "***") {
	    $amongstdefinitions = 1;
	}
    }

    close(FDT_FILE);

    return %fdtmapping;
}


1;