source: main/trunk/greenstone2/perllib/docprint.pm@ 34220

###########################################################################
#
# docprint.pm --
# A component of the Greenstone digital library software
# from the New Zealand Digital Library Project at the 
# University of Waikato, New Zealand.
#
# Copyright (C) 2006 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.
#
###########################################################################

# This is used to output an XML representation of a doc_obj - this will be 
# Greenstone XML format.
# This is used by GreenstoneXMLPlugout and doc.pm
# It's also used by GreenstoneSQLPlugout to output a doc_obj's meta and/or
# fulltxt to an SQL db, while subroutine  unescape_text() is used by
# GreenstoneSQLPlugin for unescaping txt read back in.

package docprint;

use constant OUTPUT_NONE => 0;
use constant OUTPUT_META_ONLY => 1;
use constant OUTPUT_TEXT_ONLY => 2;
use constant OUTPUT_ALL => 3;

use strict;

sub get_section_xml {
    return &get_section_xml_from_root(@_);
}

sub get_section_xml_from_root {
    my ($doc_obj, $options) = @_;
    return &recursive_get_section_xml($doc_obj, $doc_obj->get_top_section(), $options);
}

sub recursive_get_section_xml {    
    my ($doc_obj, $section, $options) = @_;
    
    # 'output' can be OUTPUT_ALL|OUTPUT_META_ONLY|OUTPUT_TEXT_ONLY|OUTPUT_NONE
    # If not provided, it defaults to OUTPUT_ALL.
    # If OUTPUT_ALL, the metadata and full text both go into doc.xml
    # If OUTPUT_META_ONLY, the metadata goes into doc.xml and full text goes elsewhere (mysql db).
    # If OUTPUT_TEXT_ONLY, the full text goes into doc.xml and metadata goes elsewhere (mysql db).
    # If OUTPUT_NONE, the full text and metadata goes elsewhere (mysql db)
    # In the last 3 cases, an XML comment is left behind as a 'breadcrumb' to indicate
    # that the "missing" doc information is stored elsewhere.
    if(!defined $options) { 
    $options = { 'output' => OUTPUT_ALL };
    }
    
    my $section_ptr = $doc_obj->_lookup_section ($section);
    return "" unless defined $section_ptr;

    my $all_text = "<Section>\n";
    $all_text .= "  <Description>\n";

    # scalar comparisons on a constant is allowed (but not string evaluation of scalars)
    # https://www.perlmonks.org/?node_id=559456
    if($options->{'output'} == OUTPUT_ALL || $options->{'output'} == OUTPUT_META_ONLY) {
    # output metadata
    foreach my $data (@{$section_ptr->{'metadata'}}) {
        my $escaped_value = &escape_text($data->[1]);
        $all_text .= '    <Metadata name="' . $data->[0] . '">' . $escaped_value . "</Metadata>\n";
    }
    } else {
    $all_text .= "<!-- metadata is stored elsewhere (MySQL database) -->\n";
    }

    $all_text .= "  </Description>\n";

    # output the text
    $all_text .= "  <Content>";
    if($options->{'output'} == OUTPUT_ALL || $options->{'output'} == OUTPUT_TEXT_ONLY) {
    $all_text .= &escape_text($section_ptr->{'text'});
    } else {
    $all_text .= "<!-- full text is stored elsewhere (MySQL database) -->\n";
    }
    $all_text .= "</Content>\n";
    
    # output all the subsections
    foreach my $subsection (@{$section_ptr->{'subsection_order'}}) {
    $all_text .= &recursive_get_section_xml($doc_obj, "$section.$subsection", $options);
    }
    
    $all_text .=  "</Section>\n";

    # make sure no nasty control characters have snuck through
    # (XML::Parser will barf on anything it doesn't consider to be
    # valid UTF-8 text, including things like \c@, \cC etc.)
    # Will treat tab chars, \x09, as a special case right after this
    $all_text =~ s/[\x00-\x08\x0B\x0C\x0E-\x1F]//g;
    
    # $all_text gets written out into an xml context and represents the html version of a doc,
    # allowing the use of html entities for the tab character (&#09;)
    # Tabs (ASCII \x09) may be meaningful spacing in such cases whether the html emanated from a
    # text file, original html or other doc. Particularly when tabs are nested in <pre> tags.
    # Instead of removing tabs, replacing tabs with their entity reference will allow <pre> tags
    # to continue preserving any tabs in the final html display.
    # Hopefully with this, XML::Parser will not choke on tabs, and we get tab stop spaces preserved
    # in the html output.
    # This may be the best location to do this replacement and not in TextPlugin, because an html
    # source doc may contain <pre> elements with tab stops, so then HTMLPlugin would have to do the
    # replacement too.
    $all_text =~ s/\x09/&#09;/g;
    
    return $all_text;
}

# pass by ref version: gets a ref to a string and returns ref to the modified string
# so use this for large strings (fulltext)
sub escape_textref {
    my ($textref) = @_;
    # special characters in the xml encoding
    $$textref =~ s/&&/& &/g;
    $$textref =~ s/&/&amp;/g; # this has to be first...
    $$textref =~ s/</&lt;/g;
    $$textref =~ s/>/&gt;/g;
    $$textref =~ s/\"/&quot;/g;

    return $textref;
}

# pass by value - the behaviour of the escape_text method used so far is to pass copies of strings
sub escape_text {
    my ($text) = @_;
    my $textref = &escape_textref(\$text);
    return $$textref;
}

# Unescape variants are used by GreenstoneSQLPlugin when reading text back from SQL db
# Pass by ref version
sub unescape_textref {
    my ($textref) = @_;
    # special characters in the xml encoding
    $$textref =~ s/& &/&&/g;
    $$textref =~ s/&amp;/&/g; # this has to be first...
    $$textref =~ s/&lt;/</g;
    $$textref =~ s/&gt;/>/g;
    $$textref =~ s/&quot;/"/g;

    return $textref;
}

# Pass by value version
sub unescape_text {
    my ($text) = @_;
    my $textref = &unescape_textref(\$text);
    return $$textref;
}

1;