source: main/trunk/greenstone2/perllib/doc.pm@ 34394

###########################################################################
#
# doc.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) 1999 New Zealand Digital Library Project
#
# This program is free software; you can redistr   te 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.
#
###########################################################################

# base class to hold documents

package doc;
eval {require bytes};

BEGIN {
    die "GSDLHOME not set\n" unless defined $ENV{'GSDLHOME'};
    unshift (@INC, "$ENV{'GSDLHOME'}/perllib/dynamic/lib/site_perl/5.005/i686-linux");
}

use strict;
use unicode;
use util;
use FileUtils;
use ghtml;
use File::stat;
##use hashdoc;
use docprint;
use JSON;

# We just need pi from the Trig lib
# Import constants pi2, pip2, pip4 (2*pi, pi/2, pi/4).
# But doing
# use Math::Trig ':pi';
# resulted in errors on 32 bit linux VM with perl v8 of "bareword pi not allowed while strict subs in use"
# So, now doing just
use Math::Trig;

# the document type may be indexed_doc, nonindexed_doc, or
# classification

our $OIDcount = 0;


# Declare the Package (level) Variable that will tell us whether we're in the import or buildcol phase.
# This var will be set by import.pl, buildcol.pl and export.pl to the then appropriate value.
# https://perlmaven.com/package-variables-and-lexical-variables-in-perl
# But: https://stackoverflow.com/questions/7542120/perl-how-to-make-variables-from-requiring-script-available-in-required-script
# https://stackoverflow.com/questions/33230015/how-do-i-use-a-variable-in-another-perl-script
#$doc::cmd_line_mode = undef;
# Decided to follow the usage of $OIDcount above to declare the necessary package-level variables 'globally' accessible
# actually accessible only for those packages that import this one (with 'use doc')
my $cmd_line_mode = undef;

# processor_mode keeps track of which buildcol pass we're at: dummy, text (sidx/didx passes) or infodb
my $processor_mode = undef;

# rename_method can be 'url', 'none', 'base64'
sub new {
    my $class = shift (@_);
    my ($source_filename, $doc_type, $rename_method) = @_;


    my $self = bless {'associated_files'=>[],
              'subsection_order'=>[],
              'next_subsection'=>1,
              'subsections'=>{},
              'metadata'=>[],
              'text'=>"",
              'OIDtype'=>"hash"}, $class;

    # used to set lastmodified here, but this can screw up the HASH ids, so
    # the docsave processor now calls set_lastmodified

    $self->set_source_path($source_filename);
    
    if (defined $source_filename) {
    $source_filename = &util::filename_within_collection($source_filename);
    print STDERR "****** doc.pm::new(): no file rename method provided\n" unless $rename_method;
    $self->set_source_filename ($source_filename, $rename_method);
    }

    $self->set_doc_type ($doc_type) if defined $doc_type;

    return $self;
}


sub set_source_path
{
    my $self = shift @_;
    my ($source_filename) = @_;

    if (defined $source_filename) {
    # On Windows the source_filename can be in terse DOS format
    # e.g. test~1.txt

    $self->{'terse_source_path'} = $source_filename;

        # Use the FileUtil library methods as they are aware of more special
        # cases such as HDFS [jmt12]
    if (&FileUtils::fileExists($source_filename))
        {
        # See if we can do better for Windows with a filename
        if ($ENV{'GSDLOS'} =~ /^windows$/i) {
        require Win32;
        $self->{'source_path'} = Win32::GetLongPathName($source_filename);
        }
        else {
        # For Unix-based systems, there is no difference between the two
        $self->{'source_path'} = $source_filename;
        }       

        # On Windows, the code above has ensured source_path is the Win long (full) path name.
        # To help GLI associate metadata with an easily calculated and accurate representation of
        # filenames, we now store the Win long path name, hex encoded.
        # We're not using this field on Linux, as I can't get the hex encodings generated to match
        # what GLI Java code generates. But for symmetry we store this field on Unix too, but we need
        # to hex-encode source_path on Unix too, or it may not be UTF-8 and doc.xml will be invalid
        my $hexencodedlongsourcepath = &unicode::debug_unicode_string($self->{'source_path'});      
        $self->set_utf8_metadata_element ($self->get_top_section(), "gsdlfullsourcepath", $hexencodedlongsourcepath);
       
    }
    else {
        print STDERR "Warning: In doc::set_source_path(), file\n";
        print STDERR "           $source_filename\n";
        print STDERR "         does not exist\n";
        
        # (default) Set it to whatever we were given
        $self->{'source_path'} = $source_filename;
    }   
    }
    else {
    # Previous code for setting source_path allowed for
    # it to be undefined, so continue this practice
    $self->{'terse_source_path'} = undef;
    $self->{'source_path'} = undef;
    }
}


sub get_source_path
{
    my $self = shift @_;

    return $self->{'terse_source_path'};
}

# set lastmodified for OAI purposes, added by GRB, moved by kjdon
sub set_oailastmodified {
    my $self = shift (@_);

    my $source_path = $self->{'terse_source_path'};
   
    if (defined $source_path && (-e $source_path)) {
    my $current_time = time;

    my ($seconds, $minutes, $hours, $day_of_month, $month, $year,
        $wday, $yday, $isdst) = localtime($current_time);

    my $date_modified = sprintf("%d%02d%02d",1900+$year,$month+1,$day_of_month);

    $self->add_utf8_metadata($self->get_top_section(), "oailastmodified", $current_time);
    $self->add_utf8_metadata($self->get_top_section(), "oailastmodifieddate", $date_modified);
    } 
}

# no longer used for OAI purposes, since lastmodified is not what we want as the
# Datestamp of a document. This doc metadata may be useful for general purposes.
sub set_lastmodified {
    my $self = shift (@_);

    my $source_path = $self->{'terse_source_path'};
   
    if (defined $source_path && (-e $source_path)) {

        my $file_stat = stat($source_path);
    my $mtime = $file_stat->mtime;
    my ($seconds, $minutes, $hours, $day_of_month, $month, $year,
        $wday, $yday, $isdst) = localtime($mtime);

    my $date_modified = sprintf("%d%02d%02d",1900+$year,$month+1,$day_of_month);

    $self->add_utf8_metadata($self->get_top_section(), "lastmodified", $mtime);
    $self->add_utf8_metadata($self->get_top_section(), "lastmodifieddate", $date_modified);
    } 
}

# clone the $self object
sub duplicate {
    my $self = shift (@_);

    my $newobj = {};
    
    foreach my $k (keys %$self) {
    $newobj->{$k} = &clone ($self->{$k});
    }

    bless $newobj, ref($self);
    return $newobj;
}

sub clone {
    my ($from) = @_;
    my $type = ref ($from);

    if ($type eq "HASH") {
    my $to = {};
    foreach my $key (keys %$from) {
        $to->{$key} = &clone ($from->{$key});
    }
    return $to;
    } elsif ($type eq "ARRAY") {
    my $to = [];
    foreach my $v (@$from) {
        push (@$to, &clone ($v));
    }
    return $to;
    } else {
    return $from;
    }
}

sub set_OIDtype {
    my $self = shift (@_);
    my ($type, $metadata) = @_;

    if (defined $type && $type =~ /^(hash|hash_on_file|hash_on_ga_xml|hash_on_full_filename|incremental|filename|dirname|full_filename|assigned)$/) {
    $self->{'OIDtype'} = $type;
    } else {
    $self->{'OIDtype'} = "hash";
    }

    if ($type =~ /^assigned$/) {
    if (defined $metadata) {
        $self->{'OIDmetadata'} = $metadata;
    } else {
        $self->{'OIDmetadata'} = "dc.Identifier";
    }
    }
}

# rename_method can be 'url', 'none', 'base64'
sub set_source_filename {
    my $self = shift (@_);
    my ($source_filename, $rename_method) = @_;

    # Since the gsdlsourcefilename element goes into the doc.xml it has 
    # to be utf8. However, it should also *represent* the source filename 
    # (in the import directory) which may not be utf8 at all. 
    # For instance, if this meta element (gsdlsourcefilename) will be used 
    # by other applications that parse doc.xml in order to locate 
    # gsdlsourcefilename. Therefore, the solution is to URLencode or base64 
    # encode the real filename as this is a binary-to-text encoding meaning
    # that the resulting string is ASCII (utf8). Decoding will give the original.
    
#    print STDERR "******URL/base64 encoding the gsdl_source_filename $source_filename ";
##    print STDERR "******URL/base64 encoding the gsdl_source_filename $source_filename\n";

    # URLencode just the gsdl_source_filename, not the directory. Then prepend dir 
    $source_filename = $self->encode_filename($source_filename, $rename_method);
#    my ($srcfilename,$dirname,$suffix) 
#   = &File::Basename::fileparse($source_filename, "\\.[^\\.]+\$");
#    print STDERR "-> $srcfilename -> ";
#    $srcfilename = &util::rename_file($srcfilename.$suffix, $rename_method);
#    $source_filename = &FileUtils::filenameConcatenate($dirname, $srcfilename);
#    print STDERR "$source_filename\n";

##    print STDERR "AFTER URL/base64 encoding gsdlsourcefilename: $source_filename\n"; 
    
    $self->set_utf8_metadata_element ($self->get_top_section(), 
                 "gsdlsourcefilename", 
                 $source_filename);

    # If we set the file renaming method, at least GLI can decode gsdlsourcefilename to get the
    # original filename (relative path to file) back for GLI to notice what ex Meta is associated with it
    $self->set_utf8_metadata_element($self->get_top_section(),
                     "gsdlsourcefilerenamemethod",
                     $rename_method);
}

sub encode_filename {
    my $self = shift (@_);
    my ($source_filename, $rename_method) = @_;

     my ($srcfilename,$dirname,$suffix) 
    = &File::Basename::fileparse($source_filename, "\\.[^\\.]+\$");
#    print STDERR "-> $srcfilename -> ";
    $srcfilename = &util::rename_file($srcfilename.$suffix, $rename_method);

    # encode any subdirs of "import" also, but not import itself
    my $sep = ($ENV{'GSDLOS'} =~ /^windows$/i) ? "\\" : "/";   
    my @dirs = ($sep eq "\\") ? split(/[\\\/]+/, $dirname) : split(/\//, $dirname);
    
    my $dirpath = "";
    
    # Don't encode the first folder ("import" or "tmp"): GLI's DocXMLFile.java looks for literal "import"
    # or "tmp" before it knows what the file rename method is that is to be used decode the rest of gsdlsourcefilename.
    if(scalar (@dirs) >= 1) {
        $dirpath = shift(@dirs);
    }
    foreach my $subdir (@dirs) {
    #print STDERR "@@@@ Found subdir: $subdir\n";
    
    $subdir = &util::rename_file($subdir, $rename_method);
    #print STDERR "@@@@ encoded subdir: $subdir\n";
    
    $dirpath = &FileUtils::filenameConcatenate($dirpath, $subdir);
    }
    if($dirpath ne "") {
    $dirname = $dirpath;
    }
    
    $source_filename = &FileUtils::filenameConcatenate($dirname, $srcfilename);

    return $source_filename;
}

sub set_converted_filename {
    my $self = shift (@_);
    my ($converted_filename) = @_;

    # we know the converted filename is utf8
    $self->set_utf8_metadata_element ($self->get_top_section(), 
                 "gsdlconvertedfilename", 
                 $converted_filename);
}

# returns the source_filename as it was provided
sub get_unmodified_source_filename {
    my $self = shift (@_);

    return $self->{'terse_source_path'};
}

# returns the source_filename with whatever rename_method was given
sub get_source_filename {
    my $self = shift (@_);

    return $self->get_metadata_element ($self->get_top_section(), "gsdlsourcefilename");
}



# returns converted filename if available else returns source filename
sub get_filename_for_hashing {
    my $self = shift (@_);

    my $filename = $self->get_metadata_element ($self->get_top_section(), "gsdlconvertedfilename");

    if (!defined $filename) {
    my $plugin_name = $self->get_metadata_element ($self->get_top_section(), "Plugin");
    # if NULPlug processed file, then don't give a filename
    if (defined $plugin_name && $plugin_name eq "NULPlug") {
        $filename = undef;
    } else { # returns the URL encoded source filename!
        $filename = $self->get_metadata_element ($self->get_top_section(), "gsdlsourcefilename");
    }
    }

    if (!&FileUtils::isFilenameAbsolute($filename)) {
    $filename = &FileUtils::filenameConcatenate($ENV{'GSDLCOLLECTDIR'},$filename);
    }

    return $filename;
}

sub set_doc_type {
    my $self = shift (@_);
    my ($doc_type) = @_;

    $self->set_metadata_element ($self->get_top_section(), 
                 "gsdldoctype", 
                 $doc_type);
}

# returns the gsdldoctype as it was provided
# the default of "indexed_doc" is used if no document
# type was provided
sub get_doc_type {
    my $self = shift (@_);

    my $doc_type = $self->get_metadata_element ($self->get_top_section(), "gsdldoctype");
    return $doc_type if (defined $doc_type);
    return "indexed_doc";
}


# look up the reference to the a particular section
sub _lookup_section {
    my $self = shift (@_);
    my ($section) = @_;

    my ($num);
    my $sectionref = $self;

    while (defined $section && $section ne "") {
    
    ($num, $section) = $section =~ /^\.?(\d+)(.*)$/;
    
    $num =~ s/^0+(\d)/$1/ if defined $num ; # remove leading 0s
    
    $section = "" unless defined $section;
    

    if (defined $num && defined $sectionref->{'subsections'}->{$num}) {
        $sectionref = $sectionref->{'subsections'}->{$num};
    } else {
        return undef;
    }
    }
    
    return $sectionref;
}

# calculate OID by hashing the contents of the document
sub _calc_OID {
    my $self = shift (@_);
    my ($filename) = @_;


    my $osexe = &util::get_os_exe();

    my $hashfile_exe = &FileUtils::filenameConcatenate($ENV{'GSDLHOME'},"bin",
                       $ENV{'GSDLOS'},"hashfile$osexe");

    &util::set_gnomelib_env(); # gnomelib_env (particularly lib/libiconv2.dylib) required to run the hashfile executable on Mac Lions
    # The subroutine will set the gnomelib env once for each subshell launched, by first testing if GEXTGNOME is not already set

    # A different way to set the gnomelib env would be to do it more locally: exporting the necessary vars 
    # (specifically DYLD/LD_LIB_PATH) for gnome_lib as part of the command executed. 
    # E.g. $result=`export LD_LIBRARY_PATH=../ext/gnome-lib/darwin/lib; hashfile...`

    my $result = "NULL";

    
    if (-e "$hashfile_exe") {
#   $result = `\"$hashfile_exe\" \"$filename\"`;
#   $result = `hashfile$osexe \"$filename\" 2>&1`;
    $result = `hashfile$osexe \"$filename\"`;

    ($result) = $result =~ /:\s*([0-9a-f]+)/i;
    } else {
    print STDERR "doc::_calc_OID $hashfile_exe could not be found\n";
    }
    return "HASH$result";
}

# methods dealing with OID, not groups of them.

# if $OID is not provided one is calculated
sub set_OID {
    my $self = shift (@_);
    my ($OID) = @_;
    
    my $use_hash_oid = 0;
    # if an OID wasn't provided calculate one
    if (!defined $OID) {
    $OID = "NULL";
    if ($self->{'OIDtype'} =~ /^hash/) {
        $use_hash_oid = 1;
    } elsif ($self->{'OIDtype'} eq "incremental") {
        $OID = "D" . $OIDcount;
        $OIDcount ++;
    } elsif ($self->{'OIDtype'} eq "filename") {
        my $filename = $self->get_source_filename();
        $OID = &File::Basename::fileparse($filename, qr/\.[^.]*/);
        $OID = &util::tidy_up_oid($OID);
    } elsif ($self->{'OIDtype'} eq "full_filename") {
        my $source_filename = $self->get_source_filename();
        my $dirsep = &util::get_os_dirsep();

        $source_filename =~ s/^import$dirsep//;
        $source_filename =~ s/$dirsep/-/g;
        $source_filename =~ s/\./_/g;

        $OID = $source_filename;
        $OID = &util::tidy_up_oid($OID);
    } elsif ($self->{'OIDtype'} eq "dirname") {
        my $filename = $self->get_source_filename();
        if (defined($filename)) { # && -e $filename) {
        # get the immediate parent directory
        $OID = &File::Basename::dirname($filename);
        if (defined $OID) {
            $OID = &File::Basename::basename($OID);
            $OID = &util::tidy_up_oid($OID);
        } else {
            print STDERR "Failed to find base for filename ($filename)...generating hash id\n";
            $use_hash_oid = 1;
        }
        } else {
        print STDERR "Failed to find a filename, generating hash id\n";
        $use_hash_oid = 1;
        }    
        
    } elsif ($self->{'OIDtype'} eq "assigned") {
        my $identifier = $self->get_metadata_element ($self->get_top_section(), $self->{'OIDmetadata'});
        if (defined $identifier && $identifier ne "") {
        $OID = $identifier;
        $OID = &util::tidy_up_oid($OID);
        } else {
        # need a hash id
        print STDERR "no $self->{'OIDmetadata'} metadata found, generating hash id\n";
        $use_hash_oid = 1;
        }
                
    } else {
        $use_hash_oid = 1;
    }

    if ($use_hash_oid) {
        my $hash_on_file = 1; 
        my $hash_on_ga_xml = 0;

        if ($self->{'OIDtype'} eq "hash_on_ga_xml") {
        $hash_on_file = 0;
        $hash_on_ga_xml = 1;
        }

        if ($self->{'OIDtype'} eq "hash_on_full_filename") {
        $hash_on_file = 0;
        $hash_on_ga_xml = 0;

        my $source_filename = $self->get_source_filename();
        my $dirsep = &util::get_os_dirsep();

        $source_filename =~ s/^import$dirsep//;
        $source_filename =~ s/$dirsep/-/g;
        $source_filename =~ s/\./_/g;
        
        # If the filename is very short then (handled naively)
        # this can cause conjestion in the hash-values
        # computed, leading documents sharing the same leading
        # Hex values in the computed has.
        #
        # The solution taken here is to replace the name of
        # the file name a sufficient number of times (up to
        # the character limit defined in 'rep_limit' and
        # make that the content that is hashed on

        # *** Think twice before changing the following value
        # as it will break backward compatability of computed
        # document HASH values

        my $rep_limit  = 256; 
        my $hash_content = undef;

        if (length($source_filename)<$rep_limit) {
            my $rep_string = "$source_filename|";
            my $rs_len = length($rep_string);

            my $clone_times = int(($rep_limit-1)/$rs_len) +1;
            
            $hash_content = substr($rep_string x $clone_times, 0, $rep_limit);
        }
        else {
            $hash_content = $source_filename;
        }

        my $filename = &util::get_tmp_filename();
        if (!open (OUTFILE, ">:utf8", $filename)) {
            print STDERR "doc::set_OID could not write to $filename\n";
        } else {
            print OUTFILE $hash_content;
            close (OUTFILE);
        }
        $OID = $self->_calc_OID ($filename);

        &FileUtils::removeFiles ($filename);
        }

        if ($hash_on_file) {
        # "hash" OID - feed file to hashfile.exe
        my $filename = $self->get_filename_for_hashing();
        
        # -z: don't want to hash on the file if it is zero size
        if (defined($filename) && -e $filename && !-z $filename) {
            $OID = $self->_calc_OID ($filename);
        } else {
            $hash_on_ga_xml = 1; # switch to back-up plan, and hash on GA file instead
        }
        }

        if ($hash_on_ga_xml) {
        # In addition being asked to explicity calculate the has based on the GA file,
        # can also end up coming into this block is doing 'hash_on_file' but the file
        # itself is of zero bytes (as could be the case with 'doc.nul' file

        my $filename = &util::get_tmp_filename();
        if (!open (OUTFILE, ">:utf8", $filename)) {
            print STDERR "doc::set_OID could not write to $filename\n";
        } else {
            my $doc_text = &docprint::get_section_xml($self);
            print OUTFILE $doc_text;
            close (OUTFILE);
        }
        $OID = $self->_calc_OID ($filename);
        &FileUtils::removeFiles($filename);
        }
    }
    }
    $self->set_metadata_element ($self->get_top_section(), "Identifier", $OID);
}

# this uses hashdoc (embedded c thingy) which is faster but still 
# needs a little work to be sufficiently stable
sub ___set_OID {
    my $self = shift (@_);
    my ($OID) = @_;

    # if an OID wasn't provided then calculate hash value based on document
    if (!defined $OID) 
    {
    my $hash_text = &docprint::get_section_xml($self);
    my $hash_len = length($hash_text);

        $OID = &hashdoc::buffer($hash_text,$hash_len);
    }

    $self->set_metadata_element ($self->get_top_section(), "Identifier", $OID);
}

# returns the OID for this document
sub get_OID {
    my $self = shift (@_);
    my $OID = $self->get_metadata_element ($self->get_top_section(), "Identifier");
    return $OID if (defined $OID);
    return "NULL";
}

sub delete_OID {
    my $self = shift (@_);
    
    $self->set_metadata_element ($self->get_top_section(), "Identifier", "NULL");
}


# methods for manipulating section names

# returns the name of the top-most section (the top
# level of the document
sub get_top_section {
    my $self = shift (@_);
    
    return "";
}

# returns a section
sub get_parent_section {
    my $self = shift (@_);
    my ($section) = @_;

    $section =~ s/(^|\.)\d+$//;

    return $section;
}

# returns the first child section (or the end child 
# if there isn't any)
sub get_begin_child {
    my $self = shift (@_);
    my ($section) = @_;

    my $section_ptr = $self->_lookup_section($section);
    return "" unless defined $section_ptr;

    if (defined $section_ptr->{'subsection_order'}->[0]) {
    return "$section.$section_ptr->{'subsection_order'}->[0]";
    }

    return $self->get_end_child ($section);
}

# returns the next child of a parent section
sub get_next_child {
    my $self = shift (@_);
    my ($section) = @_;
    
    my $parent_section = $self->get_parent_section($section);
    my $parent_section_ptr = $self->_lookup_section($parent_section);
    return undef unless defined $parent_section_ptr;

    my ($section_num) = $section =~ /(\d+)$/;
    return undef unless defined $section_num;

    my $i = 0;
    my $section_order = $parent_section_ptr->{'subsection_order'};
    while ($i < scalar(@$section_order)) {
    last if $section_order->[$i] eq $section_num;
    $i++;
    }

    $i++; # the next child
    if ($i < scalar(@$section_order)) {
    return $section_order->[$i] if $parent_section eq "";
    return "$parent_section.$section_order->[$i]";
    }

    # no more sections in this level
    return undef;
}

# returns a reference to a list of children
sub get_children {
    my $self = shift (@_);
    my ($section) = @_;

    my $section_ptr = $self->_lookup_section($section);
    return [] unless defined $section_ptr;

    my @children = @{$section_ptr->{'subsection_order'}};

    map {$_ = "$section.$_"; $_ =~ s/^\.+//;} @children;
    return \@children;
}

# returns the child section one past the last one (which
# is coded as "0")
sub get_end_child {
    my $self = shift (@_);
    my ($section) = @_;

    return $section . ".0" unless $section eq "";
    return "0";
}

# returns the next section in book order
sub get_next_section {
    my $self = shift (@_);
    my ($section) = @_;

    return undef unless defined $section;

    my $section_ptr = $self->_lookup_section($section);
    return undef unless defined $section_ptr;

    # first try to find first child
    if (defined $section_ptr->{'subsection_order'}->[0]) {
    return $section_ptr->{'subsection_order'}->[0] if ($section eq "");
    return "$section.$section_ptr->{'subsection_order'}->[0]";
    }

    do {
    # try to find sibling
    my $next_child = $self->get_next_child ($section);
    return $next_child if (defined $next_child);

    # move up one level
    $section = $self->get_parent_section ($section);
    } while $section =~ /\d/;

    return undef;
}

sub is_leaf_section {
    my $self = shift (@_);
    my ($section) = @_;

    my $section_ptr = $self->_lookup_section($section);
    return 1 unless defined $section_ptr;

    return (scalar (@{$section_ptr->{'subsection_order'}}) == 0);
}

# methods for dealing with sections

# returns the name of the inserted section
sub insert_section {
    my $self = shift (@_);
    my ($before_section) = @_;

    # get the child to insert before and its parent section
    my $parent_section = "";
    my $before_child = "0";
    my @before_section = split (/\./, $before_section);
    if (scalar(@before_section) > 0) {
    $before_child = pop (@before_section);
    $parent_section = join (".", @before_section);
    }

    my $parent_section_ptr = $self->_lookup_section($parent_section);
    if (!defined $parent_section_ptr) {
    print STDERR "doc::insert_section couldn't find parent section " .
        "$parent_section\n";
    return;
    }

    # get the next section number
    my $section_num = $parent_section_ptr->{'next_subsection'}++;

    my $i = 0;
    while ($i < scalar(@{$parent_section_ptr->{'subsection_order'}}) &&
       $parent_section_ptr->{'subsection_order'}->[$i] ne $before_child) {
    $i++;
    }
    
    # insert the section number into the order list
    splice (@{$parent_section_ptr->{'subsection_order'}}, $i, 0, $section_num);

    # add this section to the parent section
    my $section_ptr = {'subsection_order'=>[],
               'next_subsection'=>1,
               'subsections'=>{},
               'metadata'=>[],
               'text'=>""};
    $parent_section_ptr->{'subsections'}->{$section_num} = $section_ptr;

    # work out the full section number
    my $section = $parent_section;
    $section .= "." unless $section eq "";
    $section .= $section_num;
    
    return $section;
}

# creates a pre-named section
sub create_named_section {
    my $self = shift (@_);
    my ($mastersection) = @_;

    my ($num);
    my $section = $mastersection;
    my $sectionref = $self;

    while ($section ne "") {
    ($num, $section) = $section =~ /^\.?(\d+)(.*)$/;
    $num =~ s/^0+(\d)/$1/; # remove leading 0s
    $section = "" unless defined $section;
    
    if (defined $num) {
        if (!defined $sectionref->{'subsections'}->{$num}) {
        push (@{$sectionref->{'subsection_order'}}, $num);
        $sectionref->{'subsections'}->{$num} = {'subsection_order'=>[],
                            'next_subsection'=>1,
                            'subsections'=>{},
                            'metadata'=>[],
                            'text'=>""};
        if ($num >= $sectionref->{'next_subsection'}) {
            $sectionref->{'next_subsection'} = $num + 1;
        }
        }
        $sectionref = $sectionref->{'subsections'}->{$num};

    } else {
        print STDERR "doc::create_named_section couldn't create section ";
        print STDERR "$mastersection\n";
        last;
    }
    }
}

# returns a reference to a list of subsections
sub list_subsections {
    my $self = shift (@_);
    my ($section) = @_;

    my $section_ptr = $self->_lookup_section ($section);
    if (!defined $section_ptr) {
    print STDERR "doc::list_subsections couldn't find section $section\n";
    return [];
    }

    return [@{$section_ptr->{'subsection_order'}}];
}

sub delete_section {
    my $self = shift (@_);
    my ($section) = @_;

#    my $section_ptr = {'subsection_order'=>[],
#              'next_subsection'=>1,
#              'subsections'=>{},
#              'metadata'=>[],
#              'text'=>""};

    # if this is the top section reset everything
    if ($section eq "") {
    $self->{'subsection_order'} = [];
    $self->{'subsections'} = {};
    $self->{'metadata'} = [];
    $self->{'text'} = "";
    return;
    }

    # find the parent of the section to delete
    my $parent_section = "";
    my $child = "0";
    my @section = split (/\./, $section);
    if (scalar(@section) > 0) {
    $child = pop (@section);
    $parent_section = join (".", @section);
    }

    my $parent_section_ptr = $self->_lookup_section($parent_section);
    if (!defined $parent_section_ptr) {
    print STDERR "doc::delete_section couldn't find parent section " .
        "$parent_section\n";
    return;
    }

    # remove this section from the subsection_order list
    my $i = 0;
    while ($i < scalar (@{$parent_section_ptr->{'subsection_order'}})) {
    if ($parent_section_ptr->{'subsection_order'}->[$i] eq $child) {
        splice (@{$parent_section_ptr->{'subsection_order'}}, $i, 1);
        last;
    }
    $i++;
    }

    # remove this section from the subsection hash
    if (defined ($parent_section_ptr->{'subsections'}->{$child})) {
    undef $parent_section_ptr->{'subsections'}->{$child};
    }
}

#--
# methods for dealing with metadata

# set_metadata_element and get_metadata_element are for metadata
# which should only have one value. add_meta_data and get_metadata
# are for metadata which can have more than one value.

# returns the first metadata value which matches field

# This version of get metadata element works much like the one above,
# except it allows for the namespace portion of a metadata element to
# be ignored, thus if you are searching for dc.Title, the first piece
# of matching metadata ending with the name Title (once any namespace
# is removed) would be returned.
# 28-11-2003 John Thompson
sub get_metadata_element {
    my $self = shift (@_);
    my ($section, $field, $ignore_namespace) = @_;
    my ($data);

    $ignore_namespace = 0 unless defined $ignore_namespace;

    my $section_ptr = $self->_lookup_section($section);
    if (!defined $section_ptr) {
    print STDERR "doc::get_metadata_element couldn't find section ", $section, "\n";
    return;
    }

    # Remove any namespace if we are being told to ignore them
    if($ignore_namespace) {
    $field =~ s/^.*\.//; #$field =~ s/^\w*\.//;
    }

    foreach $data (@{$section_ptr->{'metadata'}}) {

    my $data_name = $data->[0];

    # Remove any namespace if we are being told to ignore them
    if($ignore_namespace) {
        $data_name =~ s/^.*\.//; #$data_name =~ s/^\w*\.//;
    }
    # we always remove ex. (but not any subsequent namespace) - ex. maybe there in doc_obj, but we will never ask for it.
    $data_name =~ s/^ex\.([^.]+)$/$1/; #$data_name =~ s/^ex\.//; 
    
    return $data->[1] if (scalar(@$data) >= 2 && $data_name eq $field);
    }
    
    return undef; # was not found
}

# returns a list of the form [value1, value2, ...]
sub get_metadata {
    my $self = shift (@_);
    my ($section, $field, $ignore_namespace) = @_;
    my ($data);

    $ignore_namespace = 0 unless defined $ignore_namespace;

    my $section_ptr = $self->_lookup_section($section);
    if (!defined $section_ptr) {
        print STDERR "doc::get_metadata couldn't find section ",
        $section, "\n";
        return;
    }

    # Remove any namespace if we are being told to ignore them
    if($ignore_namespace) {
    $field =~ s/^.*\.//;
    }

    my @metadata = ();
    foreach $data (@{$section_ptr->{'metadata'}}) {

    my $data_name = $data->[0];

    # Remove any namespace if we are being told to ignore them
    if($ignore_namespace) {
        $data_name =~ s/^.*\.//;
    }   
    # we always remove ex. (but not any subsequent namespace) - ex. maybe there in doc_obj, but we will never ask for it.
    $data_name =~ s/^ex\.([^.]+)$/$1/;

        push (@metadata, $data->[1]) if ($data_name eq $field);
    }

    return \@metadata;
}

sub get_metadata_hashmap {
    my $self = shift (@_);
    my ($section, $opt_namespace) = @_;
    
    my $section_ptr = $self->_lookup_section($section);
    if (!defined $section_ptr) {
            print STDERR "doc::get_metadata couldn't find section ",
            $section, "\n";
            return;
        }

    my $metadata_hashmap = {};
    foreach my $data (@{$section_ptr->{'metadata'}}) {
            my $metaname = $data->[0];
          
            if ((!defined $opt_namespace) || ($metaname =~ m/^$opt_namespace\./)) {
                if (!defined $metadata_hashmap->{$metaname}) {
                    $metadata_hashmap->{$metaname} = [];
                  }
                my $metaval_list = $metadata_hashmap->{$metaname};
                push(@$metaval_list, $data->[1]); 
              }
          }
    
    return $metadata_hashmap;
}

# returns a list of the form [[field,value],[field,value],...]
sub get_all_metadata {
    my $self = shift (@_);
    my ($section) = @_;

    my $section_ptr = $self->_lookup_section($section);
    if (!defined $section_ptr) {
    print STDERR "doc::get_all_metadata couldn't find section ", $section, "\n";
    return;
    }
    
    return $section_ptr->{'metadata'};
}

# $value is optional
sub delete_metadata {
    my $self = shift (@_);
    my ($section, $field, $value) = @_;

    my $section_ptr = $self->_lookup_section($section);
    if (!defined $section_ptr) {
    print STDERR "doc::delete_metadata couldn't find section ", $section, "$field\n";
    return;
    }

    my $i = 0;
    while ($i < scalar (@{$section_ptr->{'metadata'}})) {
    if (($section_ptr->{'metadata'}->[$i]->[0] eq $field) &&
        (!defined $value || $section_ptr->{'metadata'}->[$i]->[1] eq $value)) {
        splice (@{$section_ptr->{'metadata'}}, $i, 1);
    } else {
        $i++;
    }
    }
}

sub delete_all_metadata {
    my $self = shift (@_);
    my ($section) = @_;

    my $section_ptr = $self->_lookup_section($section);
    if (!defined $section_ptr) {
    print STDERR "doc::delete_all_metadata couldn't find section ", $section, "\n";
    return;
    }
    
    $section_ptr->{'metadata'} = [];
}

sub set_metadata_element {
    my $self = shift (@_);
    my ($section, $field, $value) = @_;

    $self->set_utf8_metadata_element ($section, $field, 
                      &unicode::ascii2utf8(\$value));
}

# set_utf8_metadata_element assumes the text has already been
# converted to the UTF-8 encoding.
sub set_utf8_metadata_element {
    my $self = shift (@_);
    my ($section, $field, $value) = @_;
    
    $self->delete_metadata ($section, $field);
    $self->add_utf8_metadata ($section, $field, $value);
}


# add_metadata assumes the text is in (extended) ascii form. For
# text which has already been converted to the UTF-8 format use
# add_utf8_metadata.
sub add_metadata {
    my $self = shift (@_);
    my ($section, $field, $value) = @_;
    
    $self->add_utf8_metadata ($section, $field,
                  &unicode::ascii2utf8(\$value));
}

sub add_utf8_metadata {
    my $self = shift (@_);
    my ($section, $field, $value) = @_;
    
    #    my ($cpackage,$cfilename,$cline,$csubr,$chas_args,$cwantarray) = caller(1);
    #    my ($lcfilename) = ($cfilename =~ m/([^\\\/]*)$/);
    #    print STDERR "** Calling method: $lcfilename:$cline $cpackage->$csubr\n";

    my $section_ptr = $self->_lookup_section($section);
    if (!defined $section_ptr) {
    print STDERR "doc::add_utf8_metadata couldn't find section ", $section, "\n";
    return;
    }
    if (!defined $value) {
    print STDERR "doc::add_utf8_metadata undefined value for $field\n";
    return;
    }
    if (!defined $field) {
    print STDERR "doc::add_utf8_metadata undefined metadata type \n";
    return;
    }
    
    #print STDERR "###$field=$value\n";

    # For now, supress this check.  Given that text data read in is now 
    # Unicode aware, then the following block of code can (ironically enough) 
    # cause our unicode compliant string to be re-encoded (leading to
    # a double-encoded UTF-8 string, which we definitely don't want!).
    

    # double check that the value is utf-8
    #    if (!&unicode::check_is_utf8($value)) {
    #   print STDERR "doc::add_utf8_metadata - warning: '$field''s value $value wasn't utf8.";
    #   &unicode::ensure_utf8(\$value);
    #   print STDERR " Tried converting to utf8: $value\n";
    #    }

    #If the metadata value is either a latitude or a longitude value then we want to save a shortened version for spacial searching purposes
    if ($field =~ m/^(.+\.)?Latitude$/ || $field =~ m/^(.+\.)?Longitude$/)
    {
            my ($mdprefix,$metaname) = ($field =~ m/(.+)\.(.+)$/);
        if (defined $mdprefix) {
            # Add in a version of Latitude/Longitude without the metadata namespace prefix to keep Runtime happy
            push (@{$section_ptr->{'metadata'}}, [$metaname, $value]);
        }

        my $direction;
        # TODO does the following work, with eq latitude? the above code implies can be xxxxx.Latitude. should use metaname??
        if($value =~ m/^-/)
        {
            $direction = ($field eq "Latitude") ? "S" : "W"; 
        }
        else
        {
            $direction = ($field eq "Latitude") ? "N" : "E"; 
        }
        
        my ($beforeDec, $afterDec) = ($value =~ m/^-?([0-9]+)\.([0-9]+)$/);
        if(defined $beforeDec && defined $afterDec)
        {
            my $name = ($field eq "Latitude") ? "LatShort" : "LngShort";
            push (@{$section_ptr->{'metadata'}}, [$name, $beforeDec . $direction]);
            
            for(my $i = 2; $i <= 4; $i++)
            {
                if(length($afterDec) >= $i)
                {
                    push (@{$section_ptr->{'metadata'}}, [$name, substr($afterDec, 0, $i)]);
                }
            }

            # what is this???? completely does nothing...
            #Only add the metadata if it has not already been added
            my $metaMap = $self->get_metadata_hashmap($section); # metaMap not used and called function has no apparent side-effects,
                                                                 # but this line appears important for ensuring uniqueness of (Latitude, value) meta
                                                                 # in the section. Also for LatShort, Longitude, LngShort.
        }

        # if we have both Latitude and Longitude, then add coordinate metadata (not sure why, but someone else added it)
        my $latitude;
        my $longitude;
        if($field =~ m/^(.+\.)?Longitude$/) {
            # if we are dealing with Longitude meta, and we have already seen Latitude, then add Coordinate metadata 
            $latitude = $self->get_metadata_element ($section, "Latitude");
            if (defined $latitude) {
            $longitude = $value;
            }
        } else {
            # we are dealing with Latitude - if we have already seen Longitude, then add both as Coordinate metadata
            $longitude = $self->get_metadata_element ($section, "Longitude");
            if (defined $longitude) {
            $latitude = $value;
            }
        }
        if (defined $latitude && defined $longitude) {
            ##rint STDERR "adding coords for $section, $latitude, $longitude\n";
            $self->processCoordinate($section, $latitude, $longitude);
        } else {
            ##rint STDERR "lat or long not defined, not adding coords for $section\n";
        }
        
    }
    
    elsif($field eq "GPS.mapOverlay") { # then the $value is a JSON string

        # In order to allow searching map-data-enriched documents by map shape descriptions,
        # and to run rawquery searches for other docs by proximity based on their map data,
        # need to store the shape descriptions and Coordinate info for shapes into the text index.
        # We add the description for each shape in the mapoverlay into the text index as GPSMapOverlayLabel
        # And we add Coordinate (CD) and CoordShort (CS) info for each shape in the mapoverlay in the format (Lat, Lng) as ("37S339 175E342")
        # where the digits before the N/S/E/W direction represents the whole number. And the digits after the direction are the
        # decimal places which can vary between 0 and 2 to 4 digits.
        
        # However, we only want to process GPS.mapOverlay during buildcol and only in the text indexing passes (e.g. sidx and didx for lucene)
        # and certainly never during the infodb pass of buildcol. The latter can end up duplicating Coordinate/CoordShort/GPSMapOverlayLabel
        # when rebuilding with the online doc editor, as that runs incremental-rebuild which then calls basebuilder::reconstruct_doc_objs_metadata()
        # on all docs NOT being incrementally rebuilt. That call would get meta from the infodb and use it to reconstruct the doc objects of docs
        # NOT being incrementally built. If the Coord and Label meta were written to the infodb, they would then be loaded back in memory when the
        # collection is incrementally rebuilt for those docs that don't need incremental processing. Then this function would once again add the
        # same meta into the infodb, thus duplicating what goes into the infodb. Hence, don't do all the following if doc::processor_mode =~ "infodb",
        # or anything other than a text mode.

        # Note that for incremental rebuilding, the text pass can be called textreindex for instance (and infodb pass can be incinfodb).
        # So don't check for exact string match

        if($doc::cmd_line_mode eq "buildcol" && $doc::processor_mode =~ m/^text/) { # currently known text processor_modes:
                                                                                    # text, textreindex, possibly textdelete (see ArchivesInfPlugin.pm::read() for last 2).
                        # OR: ..."buildcol" && $doc::processor_mode !~ m/infodb$/) # if dummy pass important
        
        ###print STDERR "GPS.mapOverlay before decoding, val = " . $value . "\n";
        
        # TODO: html decode?
        $value =~ s@&#091;@[@g;
        $value =~ s@&#093;@]@g;
        $value =~ s@&quot;@"@g;
        ###print STDERR "GPS.mapOverlay after decoding, val = " . $value . "\n";

        my $json_array = decode_json $value;
        
        foreach my $shape (@$json_array) {

            # Put each available shape description/label into this section's metadata with GPSMapOverlayLabel as metaname.
            # Just as for Coordinate meta, don't need to know which shape a label belongs too. This is just so each label
            # will be indexed, and therefore can be searched.
            
            my $description = $shape->{"description"};
            if($description) {
            push (@{$section_ptr->{'metadata'}}, ["GPSMapOverlayLabel", $description]);
            ###print STDERR "@@@@############################################ Just added description meta: " . $description . "\n";
            }
            
            my $type = $shape->{"type"};
            ###print STDERR "Shape type : " . $type . "\n";
        
            if($type eq "circle") {
                ###print STDERR "Found a circle:\n" . &printShape($json, $shape);
        
                # work out bounding box
                # want the inverse of this useful page:
                # https://stackoverflow.com/questions/639695/how-to-convert-latitude-or-longitude-to-meters
                # https://www.geodatasource.com/developers/javascript               
                
                # Dr Bainbridge wants us to use: https://gis.stackexchange.com/questions/5821/calculating-latitude-longitude-x-miles-from-point
                # Since for N,E,S,W the following works out the same as above link, we're using the rule of thumb in the following:
                # https://gis.stackexchange.com/questions/2951/algorithm-for-offsetting-a-latitude-longitude-by-some-amount-of-meters
                # which states
                # "If your displacements aren't too great (less than a few kilometers) and you're not right at the poles,
                # use the quick and dirty estimate that 111,111 meters (111.111 km) in the y direction is 1 degree
                # (of latitude) and 111,111 * cos(latitude) meters in the x direction is 1 degree (of longitude)."
                my $centre_lat = $shape->{"center"}->{"lat"};
                my $centre_lng = $shape->{"center"}->{"lng"};
                my $radius = $shape->{"radius"}; # in metres!

                ###print STDERR "@@@ circle centre: ($centre_lat, $centre_lng), radius: $radius\n";

                my $lat_north = $centre_lat + ($radius/111111);
                my $lat_south = $centre_lat - ($radius/111111);
                
                ###print STDERR "### lat_north:  $lat_north\n"; 
                ###print STDERR "### lat_south:  $lat_south\n";

                # our latitude and longitude values are in degrees. But cos and sin etc in perl and generally all prog languages
                # all expect their angle to be in radians. So need to convert from degree to radians before we can take the cos of it.
                my $centre_lat_radians = $self->degreesToRadians($centre_lat);
                my $cos_in_radians = cos($centre_lat_radians);              
                ###print STDERR "cos $centre_lat_radians " . cos($centre_lat_radians) . "\n";
                my $lng_east = $centre_lng + ($radius/(111111 * $cos_in_radians));
                my $lng_west = $centre_lng - ($radius/(111111 * $cos_in_radians));
                ###print STDERR "### lng_east  $lng_east\n"; 
                ###print STDERR "### lng_west  $lng_west\n";

                my $cos_lat = cos($centre_lat);             
                ###print STDERR "cos $centre_lat is $cos_lat\n";

                $self->processCoordinate($section, $lat_north, $lng_east);
                $self->processCoordinate($section, $lat_south, $lng_east);
                $self->processCoordinate($section, $lat_south, $lng_west);
                $self->processCoordinate($section, $lat_north, $lng_west);
            
            }
            elsif ($type eq "marker") {
                ###print STDERR "@@ MARKER FOUND WITH LAT: " . $shape->{"position"}->{"lat"} . "\n";
                ###print STDERR "@@ MARKER FOUND WITH LNG: " . $shape->{"position"}->{"lng"} . "\n";
                $self->processCoordinate($section, $shape->{"position"}->{"lat"}, $shape->{"position"}->{"lng"});               
            }
            elsif ($type eq "polyline" || $type eq "polygon") {
                my $path_array = $shape->{"path"};
                foreach my $position (@$path_array) {                                   
                    $self->processCoordinate($section, $position->{"lat"}, $position->{"lng"});
                }
            }
            elsif ($type eq "rectangle") {
        
                my $bounds = $shape->{"bounds"};
                $self->processCoordinate($section, $bounds->{"north"}, $bounds->{"east"});
                $self->processCoordinate($section, $bounds->{"south"}, $bounds->{"east"});
                $self->processCoordinate($section, $bounds->{"south"}, $bounds->{"west"});
                $self->processCoordinate($section, $bounds->{"north"}, $bounds->{"west"});
            }   
        
        } # end for on each shape in GPS.mapOverlay
        } # end if(buildcol and text pass)
    } # end GPS.mapOverlay meta

    push (@{$section_ptr->{'metadata'}}, [$field, $value]);
}

# https://en.wikipedia.org/wiki/Radian
sub degreesToRadians
{
    my $self = shift (@_);
    my ($degrees) = @_;

    return $degrees * pi /180; # returns radians
}

sub radiansToDegrees
{
    my $self = shift (@_);
    my ($radians) = @_;

    return $radians * 180 / pi; # returns degrees
}

# Call as:
# my $json = JSON->new->allow_nonref;
# &printAllShapes($json, $json_array);
sub printAllShapes {
    my ($json, $json_array) = @_; 
    
    
    #my $pretty_print_shape = $json->pretty->encode( $json_array->[0] ); 
    foreach my $shape (@$json_array) {
        my $pretty_print_shape = $json->pretty->encode( $shape ); 
        print STDERR "Shape: $pretty_print_shape\n";
    }
    
}

# For the (lat, lng) coordinate given,
# attaches Coordinate and multiple CoordShort (different precision level) metadata to the doc object
sub processCoordinate {
    my $self = shift (@_);
    my ($section, $latitude, $longitude) = @_;

    my $section_ptr = $self->_lookup_section($section);

    my $lat_direction = ($latitude =~ m/^-/) ? "S" : "N";
    my $lng_direction = ($longitude =~ m/^-/) ? "W" : "E";
    
    # have to store (lat, lng) in pairs, when there are so many coords to store
    #push (@{$section_ptr->{'metadata'}}, ["Latitude", $latitude]);
    #push (@{$section_ptr->{'metadata'}}, ["Longitude", $longitude]);

    push (@{$section_ptr->{'metadata'}}, ["Coordinate", "$latitude $longitude"]); # "$latitude$lat_direction $longitude$lng_direction"

    my ($latBeforeDec, $latAfterDec);
    my ($lngBeforeDec, $lngAfterDec);

    if($latitude !~ m/\./) {
        $latBeforeDec = $latitude;
        $latAfterDec = "";
    } else {
        ($latBeforeDec, $latAfterDec) = ($latitude =~ m/^-?([0-9]+)\.([0-9]+)$/);
    }
    if($longitude !~ m/\./) {
        $lngBeforeDec = $longitude;
        $lngAfterDec = "";
    } else {
        ($lngBeforeDec, $lngAfterDec) = ($longitude =~ m/^-?([0-9]+)\.([0-9]+)$/);
    }   
    

    my $name = "CoordShort";
    push (@{$section_ptr->{'metadata'}}, [$name, "$latBeforeDec$lat_direction $lngBeforeDec$lng_direction"]);
        
    for(my $i = 2; $i <= 4; $i++)
    {
        my $latDecPlaces = (length($latAfterDec) >= $i) ? substr($latAfterDec, 0, $i) : "";         
        my $lngDecPlaces = (length($lngAfterDec) >= $i) ? substr($lngAfterDec, 0, $i) : "";

        push (@{$section_ptr->{'metadata'}}, [$name, 
                $latBeforeDec . $lat_direction. $latDecPlaces . " " . $lngBeforeDec . $lng_direction. $lngDecPlaces]);
        
    }
        
    #Only add the metadata if it has not already been added
    #my $metaMap = $self->get_metadata_hashmap($section); ### TODO: metaMap not used. Unnecesssary step? (Called function has no apparent side-effects.)
}


# methods for dealing with text

# returns the text for a section
sub get_text {
    my $self = shift (@_);
    my ($section) = @_;

    my $section_ptr = $self->_lookup_section($section);
    if (!defined $section_ptr) {
    print STDERR "doc::get_text couldn't find section " .
        "$section\n";
    return "";
    }

    return $section_ptr->{'text'};
}

# returns the (utf-8 encoded) length of the text for a section
sub get_text_length {
    my $self = shift (@_);
    my ($section) = @_;

    my $section_ptr = $self->_lookup_section($section);
    if (!defined $section_ptr) {
    print STDERR "doc::get_text_length couldn't find section " .
        "$section\n";
    return 0;
    }

    return length ($section_ptr->{'text'});
}

# returns the total length for all the sections
sub get_total_text_length {
    my $self = shift (@_);

    my $section = $self->get_top_section(); 
    my $length = 0;
    while (defined $section) {
    $length += $self->get_text_length($section);
    $section = $self->get_next_section($section);
    }
    return $length;
}

sub delete_text {
    my $self = shift (@_);
    my ($section) = @_;

    my $section_ptr = $self->_lookup_section($section);
    if (!defined $section_ptr) {
    print STDERR "doc::delete_text couldn't find section " .
        "$section\n";
    return;
    }

    $section_ptr->{'text'} = "";
}

# add_text assumes the text is in (extended) ascii form. For
# text which has been already converted to the UTF-8 format
# use add_utf8_text.
sub add_text {
    my $self = shift (@_);
    my ($section, $text) = @_;

    # convert the text to UTF-8 encoded unicode characters
    # and add the text
    $self->add_utf8_text($section, &unicode::ascii2utf8(\$text));
}


# add_utf8_text assumes the text to be added has already
# been converted to the UTF-8 encoding. For ascii text use
# add_text
# Pass by value version (internally calls pass by ref version
# to avoid code duplication)
sub add_utf8_text {
    my $self = shift (@_);
    my ($section, $text) = @_;
    
    $self->add_utf8_textref($section, \$text);
}

# Pass by reference version, used by GreenstoneSQLPlugin for fulltext
sub add_utf8_textref {
    my $self = shift (@_);
    my ($section, $text_ref) = @_;

    my $section_ptr = $self->_lookup_section($section);
    if (!defined $section_ptr) {
    print STDERR "doc::add_utf8_textref couldn't find section " .
        "$section\n";
    return;
    }

    $section_ptr->{'text'} .= $$text_ref;
}

# returns the Source meta, which is the utf8 filename generated.
# Added a separate method here for convenience
sub get_source {
    my $self = shift (@_);
    return $self->get_metadata_element ($self->get_top_section(), "Source");
}

# returns the SourceFile meta, which is the url reference to the URL-encoded
# version of Source (the utf8 filename). Added a separate method here for convenience
sub get_sourcefile {
    my $self = shift (@_);
    return $self->get_metadata_element ($self->get_top_section(), "SourceFile");
}

# Get the actual name of the assocfile, a url-encoded string derived from SourceFile.
# The SourceFile meta is the (escaped) url reference to the url-encoded assocfile.
sub get_assocfile_from_sourcefile {
    my $self = shift (@_);
    
    # get the SourceFile meta, which is a *URL* to a file on the filesystem
    my $top_section = $self->get_top_section();
    my $source_file = $self->get_metadata_element($top_section, "SourceFile");

    # get the actual filename as it exists on the filesystem which this url refers to
    $source_file = &unicode::url_to_filename($source_file); 
    my ($assocfilename) = $source_file =~ /([^\\\/]+)$/;
    return $assocfilename;
}

# methods for dealing with associated files

# a file is associated with a document, NOT a section.
# if section is defined it is noted in the data structure
# only so that files associated from a particular section
# may be removed later (using delete_section_assoc_files)
sub associate_file {
    my $self = shift (@_);
    my ($real_filename, $assoc_filename, $mime_type, $section) = @_;
    $mime_type = &ghtml::guess_mime_type ($real_filename) unless defined $mime_type;

    # remove all associated files with the same name
    $self->delete_assoc_file ($assoc_filename);

    # Too harsh a requirement
    # Definitely get HTML docs, for example, with some missing
    # support files
#    if (!&util::fd_exists($real_filename)) {
#   print STDERR "****** doc::associate_file(): Failed to find the file $real_filename\n";
#   exit -1;
#    }

#    print STDERR "**** is the following a UTF8 rep of *real* filename?\n   $real_filename\n";
#    print STDERR "****##### so, ensure it is before storing?!?!?\n";
##    my $utf8_filename = Encode::encode("utf8",$filename);

    push (@{$self->{'associated_files'}}, 
      [$real_filename, $assoc_filename, $mime_type, $section]);
}

# returns a list of associated files in the form
#   [[real_filename, assoc_filename, mimetype], ...]
sub get_assoc_files {
    my $self = shift (@_);

    return $self->{'associated_files'};
}

# the following two methods used to keep track of original associated files
# for incremental building. eg a txt file used by an item file does not end
# up as an assoc file for the doc.xml, but it needs to be recorded as a source
# file for incremental build
sub associate_source_file {
    my $self = shift (@_);
    my ($full_filename) = @_;

    push (@{$self->{'source_assoc_files'}}, $full_filename);

}

sub get_source_assoc_files {
    my $self = shift (@_);

    return $self->{'source_assoc_files'};
 

}
sub metadata_file {
    my $self = shift (@_);
    my ($real_filename, $filename) = @_;
    
    push (@{$self->{'metadata_files'}}, 
      [$real_filename, $filename]);
}

# used for writing out the archiveinf-doc info database, to list all the metadata files
sub get_meta_files {
    my $self = shift (@_);

    return $self->{'metadata_files'};
}

sub delete_section_assoc_files {
    my $self = shift (@_);
    my ($section) = @_;

    my $i=0;
    while ($i < scalar (@{$self->{'associated_files'}})) {
    if (defined $self->{'associated_files'}->[$i]->[3] &&
        $self->{'associated_files'}->[$i]->[3] eq $section) {
        splice (@{$self->{'associated_files'}}, $i, 1);
    } else {
        $i++;
    }
    }
}

sub delete_assoc_file {
    my $self = shift (@_);
    my ($assoc_filename) = @_;

    my $i=0;
    while ($i < scalar (@{$self->{'associated_files'}})) {
    if ($self->{'associated_files'}->[$i]->[1] eq $assoc_filename) {
        splice (@{$self->{'associated_files'}}, $i, 1);
    } else {
        $i++;
    }
    }
}

sub reset_nextsection_ptr {
    my $self = shift (@_);
    my ($section) = @_;
    
    my $section_ptr = $self->_lookup_section($section);
    $section_ptr->{'next_subsection'} = 1;
}

1;