Koha/misc/migration_tools/rebuild_zebra.pl

#!/usr/bin/perl

use strict;
#use warnings; FIXME - Bug 2505

use C4::Context;
use Getopt::Long;
use File::Temp qw/ tempdir /;
use File::Path;
use C4::Biblio;
use C4::AuthoritiesMarc;
use C4::Items;
use Koha::RecordProcessor;
use XML::LibXML;

# 
# script that checks zebradir structure & create directories & mandatory files if needed
#
#

$|=1; # flushes output
# If the cron job starts us in an unreadable dir, we will break without
# this.
chdir $ENV{HOME} if (!(-r '.'));
my $directory;
my $nosanitize;
my $skip_export;
my $keep_export;
my $skip_index;
my $reset;
my $biblios;
my $authorities;
my $noxml;
my $noshadow;
my $do_munge;
my $want_help;
my $as_xml;
my $process_zebraqueue;
my $do_not_clear_zebraqueue;
my $length;
my $where;
my $offset;
my $verbose_logging = 0;
my $zebraidx_log_opt = " -v none,fatal,warn ";
my $result = GetOptions(
    'd:s'           => \$directory,
    'r|reset'       => \$reset,
    's'             => \$skip_export,
    'k'             => \$keep_export,
    'I|skip-index'    => \$skip_index,
    'nosanitize'    => \$nosanitize,
    'b'             => \$biblios,
    'noxml'         => \$noxml,
    'w'             => \$noshadow,
    'munge-config'  => \$do_munge,
    'a'             => \$authorities,
    'h|help'        => \$want_help,
	'x'				=> \$as_xml,
    'y'             => \$do_not_clear_zebraqueue,
    'z'             => \$process_zebraqueue,
    'where:s'        => \$where,
    'length:i'        => \$length,
    'offset:i'      => \$offset,
    'v+'             => \$verbose_logging,
);


if (not $result or $want_help) {
    print_usage();
    exit 0;
}

if (not $biblios and not $authorities) {
    my $msg = "Must specify -b or -a to reindex bibs or authorities\n";
    $msg   .= "Please do '$0 --help' to see usage.\n";
    die $msg;
}

if ( !$as_xml and $nosanitize ) {
    my $msg = "Cannot specify both -no_xml and -nosanitize\n";
    $msg   .= "Please do '$0 --help' to see usage.\n";
    die $msg;
}

if ($process_zebraqueue and ($skip_export or $reset)) {
    my $msg = "Cannot specify -r or -s if -z is specified\n";
    $msg   .= "Please do '$0 --help' to see usage.\n";
    die $msg;
}

if ($process_zebraqueue and $do_not_clear_zebraqueue) {
    my $msg = "Cannot specify both -y and -z\n";
    $msg   .= "Please do '$0 --help' to see usage.\n";
    die $msg;
}

if ($reset) {
    $noshadow = 1;
}

if ($noshadow) {
    $noshadow = ' -n ';
}

#  -v is for verbose, which seems backwards here because of how logging is set
#    on the CLI of zebraidx.  It works this way.  The default is to not log much
if ($verbose_logging >= 2) {
    $zebraidx_log_opt = '-v none,fatal,warn,all';
}

my $use_tempdir = 0;
unless ($directory) {
    $use_tempdir = 1;
    $directory = tempdir(CLEANUP => ($keep_export ? 0 : 1));
} 


my $biblioserverdir = C4::Context->zebraconfig('biblioserver')->{directory};
my $authorityserverdir = C4::Context->zebraconfig('authorityserver')->{directory};

my $kohadir = C4::Context->config('intranetdir');
my $bib_index_mode = C4::Context->config('zebra_bib_index_mode') || 'grs1';
my $auth_index_mode = C4::Context->config('zebra_auth_index_mode') || 'dom';

my $dbh = C4::Context->dbh;
my ($biblionumbertagfield,$biblionumbertagsubfield) = &GetMarcFromKohaField("biblio.biblionumber","");
my ($biblioitemnumbertagfield,$biblioitemnumbertagsubfield) = &GetMarcFromKohaField("biblioitems.biblioitemnumber","");

if ( $verbose_logging ) {
    print "Zebra configuration information\n";
    print "================================\n";
    print "Zebra biblio directory      = $biblioserverdir\n";
    print "Zebra authorities directory = $authorityserverdir\n";
    print "Koha directory              = $kohadir\n";
    print "BIBLIONUMBER in :     $biblionumbertagfield\$$biblionumbertagsubfield\n";
    print "BIBLIOITEMNUMBER in : $biblioitemnumbertagfield\$$biblioitemnumbertagsubfield\n";
    print "================================\n";
}

if ($do_munge) {
    munge_config();
}

my $tester = XML::LibXML->new();

if ($authorities) {
    index_records('authority', $directory, $skip_export, $skip_index, $process_zebraqueue, $as_xml, $noxml, $nosanitize, $do_not_clear_zebraqueue, $verbose_logging, $zebraidx_log_opt, $authorityserverdir);
} else {
    print "skipping authorities\n" if ( $verbose_logging );
}

if ($biblios) {
    index_records('biblio', $directory, $skip_export, $skip_index, $process_zebraqueue, $as_xml, $noxml, $nosanitize, $do_not_clear_zebraqueue, $verbose_logging, $zebraidx_log_opt, $biblioserverdir);
} else {
    print "skipping biblios\n" if ( $verbose_logging );
}


if ( $verbose_logging ) {
    print "====================\n";
    print "CLEANING\n";
    print "====================\n";
}
if ($keep_export) {
    print "NOTHING cleaned : the export $directory has been kept.\n";
    print "You can re-run this script with the -s ";
    if ($use_tempdir) {
        print " and -d $directory parameters";
    } else {
        print "parameter";
    }
    print "\n";
    print "if you just want to rebuild zebra after changing the record.abs\n";
    print "or another zebra config file\n";
} else {
    unless ($use_tempdir) {
        # if we're using a temporary directory
        # created by File::Temp, it will be removed
        # automatically.
        rmtree($directory, 0, 1);
        print "directory $directory deleted\n";
    }
}

# This checks to see if the zebra directories exist under the provided path.
# If they don't, then zebra is likely to spit the dummy. This returns true
# if the directories had to be created, false otherwise.
sub check_zebra_dirs {
	my ($base) = shift() . '/';
	my $needed_repairing = 0;
	my @dirs = ( '', 'key', 'register', 'shadow', 'tmp' );
	foreach my $dir (@dirs) {
		my $bdir = $base . $dir;
        if (! -d $bdir) {
        	$needed_repairing = 1;
        	mkdir $bdir || die "Unable to create '$bdir': $!\n";
        	print "$0: needed to create '$bdir'\n";
        }
    }
    return $needed_repairing;
}	# ----------  end of subroutine check_zebra_dirs  ----------

sub index_records {
    my ($record_type, $directory, $skip_export, $skip_index, $process_zebraqueue, $as_xml, $noxml, $nosanitize, $do_not_clear_zebraqueue, $verbose_logging, $zebraidx_log_opt, $server_dir) = @_;

    my $num_records_exported = 0;
    my $records_deleted;
    my $need_reset = check_zebra_dirs($server_dir);
    if ($need_reset) {
    	print "$0: found broken zebra server directories: forcing a rebuild\n";
    	$reset = 1;
    }
    if ($skip_export && $verbose_logging) {
        print "====================\n";
        print "SKIPPING $record_type export\n";
        print "====================\n";
    } else {
        if ( $verbose_logging ) {
            print "====================\n";
            print "exporting $record_type\n";
            print "====================\n";
        }
        mkdir "$directory" unless (-d $directory);
        mkdir "$directory/$record_type" unless (-d "$directory/$record_type");
        if ($process_zebraqueue) {
            my $entries = select_zebraqueue_records($record_type, 'deleted');
            mkdir "$directory/del_$record_type" unless (-d "$directory/del_$record_type");
            $records_deleted = generate_deleted_marc_records($record_type, $entries, "$directory/del_$record_type", $as_xml);
            mark_zebraqueue_batch_done($entries);
            $entries = select_zebraqueue_records($record_type, 'updated');
            mkdir "$directory/upd_$record_type" unless (-d "$directory/upd_$record_type");
            $num_records_exported = export_marc_records_from_list($record_type, 
                                                                  $entries, "$directory/upd_$record_type", $as_xml, $noxml, $records_deleted);
            mark_zebraqueue_batch_done($entries);
        } else {
            my $sth = select_all_records($record_type);
            $num_records_exported = export_marc_records_from_sth($record_type, $sth, "$directory/$record_type", $as_xml, $noxml, $nosanitize);
            unless ($do_not_clear_zebraqueue) {
                mark_all_zebraqueue_done($record_type);
            }
        }
    }

    #
    # and reindexing everything
    #
    if ($skip_index) {
        if ($verbose_logging) {
            print "====================\n";
            print "SKIPPING $record_type indexing\n";
            print "====================\n";
        }
    } else {
        if ( $verbose_logging ) {
            print "====================\n";
            print "REINDEXING zebra\n";
            print "====================\n";
        }
        my $record_fmt = ($as_xml) ? 'marcxml' : 'iso2709' ;
        if ($process_zebraqueue) {
            do_indexing($record_type, 'adelete', "$directory/del_$record_type", $reset, $noshadow, $record_fmt, $zebraidx_log_opt)
                if %$records_deleted;
            do_indexing($record_type, 'update', "$directory/upd_$record_type", $reset, $noshadow, $record_fmt, $zebraidx_log_opt)
                if $num_records_exported;
        } else {
            do_indexing($record_type, 'update', "$directory/$record_type", $reset, $noshadow, $record_fmt, $zebraidx_log_opt)
                if ($num_records_exported or $skip_export);
        }
    }
}


sub select_zebraqueue_records {
    my ($record_type, $update_type) = @_;

    my $server = ($record_type eq 'biblio') ? 'biblioserver' : 'authorityserver';
    my $op = ($update_type eq 'deleted') ? 'recordDelete' : 'specialUpdate';

    my $sth = $dbh->prepare("SELECT id, biblio_auth_number 
                             FROM zebraqueue
                             WHERE server = ?
                             AND   operation = ?
                             AND   done = 0
                             ORDER BY id DESC");
    $sth->execute($server, $op);
    my $entries = $sth->fetchall_arrayref({});
}

sub mark_all_zebraqueue_done {
    my ($record_type) = @_;

    my $server = ($record_type eq 'biblio') ? 'biblioserver' : 'authorityserver';

    my $sth = $dbh->prepare("UPDATE zebraqueue SET done = 1
                             WHERE server = ?
                             AND done = 0");
    $sth->execute($server);
}

sub mark_zebraqueue_batch_done {
    my ($entries) = @_;

    $dbh->{AutoCommit} = 0;
    my $sth = $dbh->prepare("UPDATE zebraqueue SET done = 1 WHERE id = ?");
    $dbh->commit();
    foreach my $id (map { $_->{id} } @$entries) {
        $sth->execute($id);
    }
    $dbh->{AutoCommit} = 1;
}

sub select_all_records {
    my $record_type = shift;
    return ($record_type eq 'biblio') ? select_all_biblios() : select_all_authorities();
}

sub select_all_authorities {
    my $strsth=qq{SELECT authid FROM auth_header};
    $strsth.=qq{ WHERE $where } if ($where);
    $strsth.=qq{ LIMIT $length } if ($length && !$offset);
    $strsth.=qq{ LIMIT $offset,$length } if ($length && $offset);
    my $sth = $dbh->prepare($strsth);
    $sth->execute();
    return $sth;
}

sub select_all_biblios {
    my $strsth = qq{ SELECT biblionumber FROM biblioitems };
    $strsth.=qq{ WHERE $where } if ($where);
    $strsth.=qq{ LIMIT $length } if ($length && !$offset);
    $strsth.=qq{ LIMIT $offset,$length } if ($offset);
    my $sth = $dbh->prepare($strsth);
    $sth->execute();
    return $sth;
}

sub include_xml_wrapper {
    my $as_xml = shift;
    my $record_type = shift;

    return 0 unless $as_xml;
    return 1 if $record_type eq 'biblio' and $bib_index_mode eq 'dom';
    return 1 if $record_type eq 'authority' and $auth_index_mode eq 'dom';
    return 0;

}

sub export_marc_records_from_sth {
    my ($record_type, $sth, $directory, $as_xml, $noxml, $nosanitize) = @_;

    my $num_exported = 0;
    open my $fh, '>:encoding(UTF-8) ', "$directory/exported_records" or die $!;
    if (include_xml_wrapper($as_xml, $record_type)) {
        # include XML declaration and root element
        print {$fh} '<?xml version="1.0" encoding="UTF-8"?><collection>';
    }
    my $i = 0;
    my ( $itemtag, $itemsubfield ) = GetMarcFromKohaField("items.itemnumber",'');
    while (my ($record_number) = $sth->fetchrow_array) {
        print "." if ( $verbose_logging );
        print "\r$i" unless ($i++ %100 or !$verbose_logging);
        if ( $nosanitize ) {
            my $marcxml = $record_type eq 'biblio'
                          ? GetXmlBiblio( $record_number )
                          : GetAuthorityXML( $record_number );
            if ($record_type eq 'biblio'){
                my @items = GetItemsInfo($record_number);
                if (@items){
                    my $record = MARC::Record->new;
                    $record->encoding('UTF-8');
                    my @itemsrecord;
                    foreach my $item (@items){
                        my $record = Item2Marc($item, $record_number);                        
                        push @itemsrecord, $record->field($itemtag);
                    }
                    $record->insert_fields_ordered(@itemsrecord);
                    my $itemsxml = $record->as_xml_record();
                    $marcxml =
                        substr($marcxml, 0, length($marcxml)-10) .
                        substr($itemsxml, index($itemsxml, "</leader>\n", 0) + 10);
                }
            }
            # extra test to ensure that result is valid XML; otherwise
            # Zebra won't parse it in DOM mode
            eval {
                my $doc = $tester->parse_string($marcxml);
            };
            if ($@) {
                warn "Error exporting record $record_number ($record_type): $@\n";
                next;
            }
            if ( $marcxml ) {
                $marcxml =~ s!<\?xml version="1.0" encoding="UTF-8"\?>\n!!;
                print {$fh} $marcxml;
                $num_exported++;
            }
            next;
        }
        my ($marc) = get_corrected_marc_record($record_type, $record_number, $noxml);
        if (defined $marc) {
            eval {
                my $rec;
                if ($as_xml) {
                    $rec = $marc->as_xml_record(C4::Context->preference('marcflavour'));
                    eval {
                        my $doc = $tester->parse_string($rec);
                    };
                    if ($@) {
                        die "invalid XML: $@";
                    }
                    $rec =~ s!<\?xml version="1.0" encoding="UTF-8"\?>\n!!;
                } else {
                    $rec = $marc->as_usmarc();
                }
                print {$fh} $rec;
                $num_exported++;
            };
            if ($@) {
                warn "Error exporting record $record_number ($record_type) ".($noxml ? "not XML" : "XML");
                warn "... specific error is $@" if $verbose_logging;
            }
        }
    }
    print "\nRecords exported: $num_exported\n" if ( $verbose_logging );
    print {$fh} '</collection>' if (include_xml_wrapper($as_xml, $record_type));
    close $fh;
    return $num_exported;
}

sub export_marc_records_from_list {
    my ($record_type, $entries, $directory, $as_xml, $noxml, $records_deleted) = @_;

    my $num_exported = 0;
    open my $fh, '>:encoding(UTF-8)', "$directory/exported_records" or die $!;
    if (include_xml_wrapper($as_xml, $record_type)) {
        # include XML declaration and root element
        print {$fh} '<?xml version="1.0" encoding="UTF-8"?><collection>';
    }
    my $i = 0;

    # Skip any deleted records. We check for this anyway, but this reduces error spam
    my %found = %$records_deleted;
    foreach my $record_number ( map { $_->{biblio_auth_number} }
                                grep { !$found{ $_->{biblio_auth_number} }++ }
                                @$entries ) {
        print "." if ( $verbose_logging );
        print "\r$i" unless ($i++ %100 or !$verbose_logging);
        my ($marc) = get_corrected_marc_record($record_type, $record_number, $noxml);
        if (defined $marc) {
            eval {
                my $rec;
                if ($as_xml) {
                    $rec = $marc->as_xml_record(C4::Context->preference('marcflavour'));
                    $rec =~ s!<\?xml version="1.0" encoding="UTF-8"\?>\n!!;
                } else {
                    $rec = $marc->as_usmarc();
                }
                print {$fh} $rec;
                $num_exported++;
            };
            if ($@) {
              warn "Error exporting record $record_number ($record_type) ".($noxml ? "not XML" : "XML");
            }
            $num_exported++;
        }
    }
    print "\nRecords exported: $num_exported\n" if ( $verbose_logging );
    print {$fh} '</collection>' if (include_xml_wrapper($as_xml, $record_type));
    close $fh;
    return $num_exported;
}

sub generate_deleted_marc_records {
    my ($record_type, $entries, $directory, $as_xml) = @_;

    my $records_deleted = {};
    open my $fh, '>:encoding(UTF-8)', "$directory/exported_records" or die $!;
    if (include_xml_wrapper($as_xml, $record_type)) {
        # include XML declaration and root element
        print {$fh} '<?xml version="1.0" encoding="UTF-8"?><collection>';
    }
    my $i = 0;
    foreach my $record_number (map { $_->{biblio_auth_number} } @$entries ) {
        print "\r$i" unless ($i++ %100 or !$verbose_logging);
        print "." if ( $verbose_logging );

        my $marc = MARC::Record->new();
        if ($record_type eq 'biblio') {
            fix_biblio_ids($marc, $record_number, $record_number);
        } else {
            fix_authority_id($marc, $record_number);
        }
        if (C4::Context->preference("marcflavour") eq "UNIMARC") {
            fix_unimarc_100($marc);
        }

        my $rec;
        if ($as_xml) {
            $rec = $marc->as_xml_record(C4::Context->preference('marcflavour'));
            $rec =~ s!<\?xml version="1.0" encoding="UTF-8"\?>\n!!;
        } else {
            $rec = $marc->as_usmarc();
        }
        print {$fh} $rec;

        $records_deleted->{$record_number} = 1;
    }
    print "\nRecords exported: $i\n" if ( $verbose_logging );
    print {$fh} '</collection>' if (include_xml_wrapper($as_xml, $record_type));
    close $fh;
    return $records_deleted;
    

}

sub get_corrected_marc_record {
    my ($record_type, $record_number, $noxml) = @_;

    my $marc = get_raw_marc_record($record_type, $record_number, $noxml); 

    if (defined $marc) {
        fix_leader($marc);
        if ($record_type eq 'authority') {
            fix_authority_id($marc, $record_number);
        } elsif ($record_type eq 'biblio' && C4::Context->preference('IncludeSeeFromInSearches')) {
            my $normalizer = Koha::RecordProcessor->new( { filters => 'EmbedSeeFromHeadings' } );
            $marc = $normalizer->process($marc);
        }
        if (C4::Context->preference("marcflavour") eq "UNIMARC") {
            fix_unimarc_100($marc);
        }
    }

    return $marc;
}

sub get_raw_marc_record {
    my ($record_type, $record_number, $noxml) = @_;
  
    my $marc; 
    if ($record_type eq 'biblio') {
        if ($noxml) {
            my $fetch_sth = $dbh->prepare_cached("SELECT marc FROM biblioitems WHERE biblionumber = ?");
            $fetch_sth->execute($record_number);
            if (my ($blob) = $fetch_sth->fetchrow_array) {
                $marc = MARC::Record->new_from_usmarc($blob);
                unless ($marc) {
                    warn "error creating MARC::Record from $blob";
                }
            }
            # failure to find a bib is not a problem -
            # a delete could have been done before
            # trying to process a record update

            $fetch_sth->finish();
            return unless $marc;
        } else {
            eval { $marc = GetMarcBiblio($record_number, 1); };
            if ($@ || !$marc) {
                # here we do warn since catching an exception
                # means that the bib was found but failed
                # to be parsed
                warn "error retrieving biblio $record_number";
                return;
            }
        }
    } else {
        eval { $marc = GetAuthority($record_number); };
        if ($@) {
            warn "error retrieving authority $record_number";
            return;
        }
    }
    return $marc;
}

sub fix_leader {
    # FIXME - this routine is suspect
    # It blanks the Leader/00-05 and Leader/12-16 to
    # force them to be recalculated correct when
    # the $marc->as_usmarc() or $marc->as_xml() is called.
    # But why is this necessary?  It would be a serious bug
    # in MARC::Record (definitely) and MARC::File::XML (arguably) 
    # if they are emitting incorrect leader values.
    my $marc = shift;

    my $leader = $marc->leader;
    substr($leader,  0, 5) = '     ';
    substr($leader, 10, 7) = '22     ';
    $marc->leader(substr($leader, 0, 24));
}

sub fix_biblio_ids {
    # FIXME - it is essential to ensure that the biblionumber is present,
    #         otherwise, Zebra will choke on the record.  However, this
    #         logic belongs in the relevant C4::Biblio APIs.
    my $marc = shift;
    my $biblionumber = shift;
    my $biblioitemnumber;
    if (@_) {
        $biblioitemnumber = shift;
    } else {    
        my $sth = $dbh->prepare(
            "SELECT biblioitemnumber FROM biblioitems WHERE biblionumber=?");
        $sth->execute($biblionumber);
        ($biblioitemnumber) = $sth->fetchrow_array;
        $sth->finish;
        unless ($biblioitemnumber) {
            warn "failed to get biblioitemnumber for biblio $biblionumber";
            return 0;
        }
    }

    # FIXME - this is cheating on two levels
    # 1. C4::Biblio::_koha_marc_update_bib_ids is meant to be an internal function
    # 2. Making sure that the biblionumber and biblioitemnumber are correct and
    #    present in the MARC::Record object ought to be part of GetMarcBiblio.
    #
    # On the other hand, this better for now than what rebuild_zebra.pl used to
    # do, which was duplicate the code for inserting the biblionumber 
    # and biblioitemnumber
    C4::Biblio::_koha_marc_update_bib_ids($marc, '', $biblionumber, $biblioitemnumber);

    return 1;
}

sub fix_authority_id {
    # FIXME - as with fix_biblio_ids, the authid must be present
    #         for Zebra's sake.  However, this really belongs
    #         in C4::AuthoritiesMarc.
    my ($marc, $authid) = @_;
    unless ($marc->field('001') and $marc->field('001')->data() eq $authid){
        $marc->delete_field($marc->field('001'));
        $marc->insert_fields_ordered(MARC::Field->new('001',$authid));
    }
}

sub fix_unimarc_100 {
    # FIXME - again, if this is necessary, it belongs in C4::AuthoritiesMarc.
    my $marc = shift;

    my $string;
    if ( length($marc->subfield( 100, "a" )) == 36 ) {
        $string = $marc->subfield( 100, "a" );
        my $f100 = $marc->field(100);
        $marc->delete_field($f100);
    }
    else {
        $string = POSIX::strftime( "%Y%m%d", localtime );
        $string =~ s/\-//g;
        $string = sprintf( "%-*s", 35, $string );
    }
    substr( $string, 22, 6, "frey50" );
    unless ( length($marc->subfield( 100, "a" )) == 36 ) {
        $marc->delete_field($marc->field(100));
        $marc->insert_grouped_field(MARC::Field->new( 100, "", "", "a" => $string ));
    }
}

sub do_indexing {
    my ($record_type, $op, $record_dir, $reset_index, $noshadow, $record_format, $zebraidx_log_opt) = @_;

    my $zebra_server  = ($record_type eq 'biblio') ? 'biblioserver' : 'authorityserver';
    my $zebra_db_name = ($record_type eq 'biblio') ? 'biblios' : 'authorities';
    my $zebra_config  = C4::Context->zebraconfig($zebra_server)->{'config'};
    my $zebra_db_dir  = C4::Context->zebraconfig($zebra_server)->{'directory'};

    system("zebraidx -c $zebra_config $zebraidx_log_opt -g $record_format -d $zebra_db_name init") if $reset_index;
    system("zebraidx -c $zebra_config $zebraidx_log_opt $noshadow -g $record_format -d $zebra_db_name $op $record_dir");
    system("zebraidx -c $zebra_config $zebraidx_log_opt -g $record_format -d $zebra_db_name commit") unless $noshadow;

}

sub print_usage {
    print <<_USAGE_;
$0: reindex MARC bibs and/or authorities in Zebra.

Use this batch job to reindex all biblio or authority
records in your Koha database.

Parameters:
    -b                      index bibliographic records

    -a                      index authority records

    -z                      select only updated and deleted
                            records marked in the zebraqueue
                            table.  Cannot be used with -r
                            or -s.

    -r                      clear Zebra index before
                            adding records to index. Implies -w.

    -d                      Temporary directory for indexing.
                            If not specified, one is automatically
                            created.  The export directory
                            is automatically deleted unless
                            you supply the -k switch.

    -k                      Do not delete export directory.

    -s                      Skip export.  Used if you have
                            already exported the records 
                            in a previous run.

    -noxml                  index from ISO MARC blob
                            instead of MARC XML.  This
                            option is recommended only
                            for advanced user.

    -x                      export and index as xml instead of is02709 (biblios only).
                            use this if you might have records > 99,999 chars,
							
    -nosanitize             export biblio/authority records directly from DB marcxml
                            field without sanitizing records. It speed up
                            dump process but could fail if DB contains badly
                            encoded records. Works only with -x,

    -w                      skip shadow indexing for this batch

    -y                      do NOT clear zebraqueue after indexing; normally,
                            after doing batch indexing, zebraqueue should be
                            marked done for the affected record type(s) so that
                            a running zebraqueue_daemon doesn't try to reindex
                            the same records - specify -y to override this.  
                            Cannot be used with -z.

    -v                      increase the amount of logging.  Normally only 
                            warnings and errors from the indexing are shown.
                            Use log level 2 (-v -v) to include all Zebra logs.

    --length   1234         how many biblio you want to export
    --offset 1243           offset you want to start to
                                example: --offset 500 --length=500 will result in a LIMIT 500,1000 (exporting 1000 records, starting by the 500th one)
                                note that the numbers are NOT related to biblionumber, that's the intended behaviour.
    --where                 let you specify a WHERE query, like itemtype='BOOK'
                            or something like that

    --munge-config          Deprecated option to try
                            to fix Zebra config files.
    --help or -h            show this message.
_USAGE_
}

# FIXME: the following routines are deprecated and 
# will be removed once it is determined whether
# a script to fix Zebra configuration files is 
# actually needed.
sub munge_config {
#
# creating zebra-biblios.cfg depending on system
#

# getting zebraidx directory
my $zebraidxdir;
foreach (qw(/usr/local/bin/zebraidx
        /opt/bin/zebraidx
        /usr/bin/zebraidx
        )) {
    if ( -f $_ ) {
        $zebraidxdir=$_;
    }
}

unless ($zebraidxdir) {
    print qq|
    ERROR: could not find zebraidx directory
    ERROR: Either zebra is not installed,
    ERROR: or it's in a directory I don't checked.
    ERROR: do a which zebraidx and edit this file to add the result you get
|;
    exit;
}
$zebraidxdir =~ s/\/bin\/.*//;
print "Info : zebra is in $zebraidxdir \n";

# getting modules directory
my $modulesdir;
foreach (qw(/usr/local/lib/idzebra-2.0/modules/mod-grs-xml.so
            /usr/local/lib/idzebra/modules/mod-grs-xml.so
            /usr/lib/idzebra/modules/mod-grs-xml.so
            /usr/lib/idzebra-2.0/modules/mod-grs-xml.so
        )) {
    if ( -f $_ ) {
        $modulesdir=$_;
    }
}

unless ($modulesdir) {
    print qq|
    ERROR: could not find mod-grs-xml.so directory
    ERROR: Either zebra is not properly compiled (libxml2 is not setup and you don t have mod-grs-xml.so,
    ERROR: or it's in a directory I don't checked.
    ERROR: find where mod-grs-xml.so is and edit this file to add the result you get
|;
    exit;
}
$modulesdir =~ s/\/modules\/.*//;
print "Info: zebra modules dir : $modulesdir\n";

# getting tab directory
my $tabdir;
foreach (qw(/usr/local/share/idzebra/tab/explain.att
            /usr/local/share/idzebra-2.0/tab/explain.att
            /usr/share/idzebra/tab/explain.att
            /usr/share/idzebra-2.0/tab/explain.att
        )) {
    if ( -f $_ ) {
        $tabdir=$_;
    }
}

unless ($tabdir) {
    print qq|
    ERROR: could not find explain.att directory
    ERROR: Either zebra is not properly compiled,
    ERROR: or it's in a directory I don't checked.
    ERROR: find where explain.att is and edit this file to add the result you get
|;
    exit;
}
$tabdir =~ s/\/tab\/.*//;
print "Info: tab dir : $tabdir\n";

#
# AUTHORITIES creating directory structure
#
my $created_dir_or_file = 0;
if ($authorities) {
    if ( $verbose_logging ) {
        print "====================\n";
        print "checking directories & files for authorities\n";
        print "====================\n";
    }
    unless (-d "$authorityserverdir") {
        system("mkdir -p $authorityserverdir");
        print "Info: created $authorityserverdir\n";
        $created_dir_or_file++;
    }
    unless (-d "$authorityserverdir/lock") {
        mkdir "$authorityserverdir/lock";
        print "Info: created $authorityserverdir/lock\n";
        $created_dir_or_file++;
    }
    unless (-d "$authorityserverdir/register") {
        mkdir "$authorityserverdir/register";
        print "Info: created $authorityserverdir/register\n";
        $created_dir_or_file++;
    }
    unless (-d "$authorityserverdir/shadow") {
        mkdir "$authorityserverdir/shadow";
        print "Info: created $authorityserverdir/shadow\n";
        $created_dir_or_file++;
    }
    unless (-d "$authorityserverdir/tab") {
        mkdir "$authorityserverdir/tab";
        print "Info: created $authorityserverdir/tab\n";
        $created_dir_or_file++;
    }
    unless (-d "$authorityserverdir/key") {
        mkdir "$authorityserverdir/key";
        print "Info: created $authorityserverdir/key\n";
        $created_dir_or_file++;
    }
    
    unless (-d "$authorityserverdir/etc") {
        mkdir "$authorityserverdir/etc";
        print "Info: created $authorityserverdir/etc\n";
        $created_dir_or_file++;
    }
    
    #
    # AUTHORITIES : copying mandatory files
    #
    # the record model, depending on marc flavour
    unless (-f "$authorityserverdir/tab/record.abs") {
        if (C4::Context->preference("marcflavour") eq "UNIMARC") {
            system("cp -f $kohadir/etc/zebradb/marc_defs/unimarc/authorities/record.abs $authorityserverdir/tab/record.abs");
            print "Info: copied record.abs for UNIMARC\n";
        } else {
            system("cp -f $kohadir/etc/zebradb/marc_defs/marc21/authorities/record.abs $authorityserverdir/tab/record.abs");
            print "Info: copied record.abs for USMARC\n";
        }
        $created_dir_or_file++;
    }
    unless (-f "$authorityserverdir/tab/sort-string-utf.chr") {
        system("cp -f $kohadir/etc/zebradb/lang_defs/fr/sort-string-utf.chr $authorityserverdir/tab/sort-string-utf.chr");
        print "Info: copied sort-string-utf.chr\n";
        $created_dir_or_file++;
    }
    unless (-f "$authorityserverdir/tab/word-phrase-utf.chr") {
        system("cp -f $kohadir/etc/zebradb/lang_defs/fr/sort-string-utf.chr $authorityserverdir/tab/word-phrase-utf.chr");
        print "Info: copied word-phase-utf.chr\n";
        $created_dir_or_file++;
    }
    unless (-f "$authorityserverdir/tab/auth1.att") {
        system("cp -f $kohadir/etc/zebradb/authorities/etc/bib1.att $authorityserverdir/tab/auth1.att");
        print "Info: copied auth1.att\n";
        $created_dir_or_file++;
    }
    unless (-f "$authorityserverdir/tab/default.idx") {
        system("cp -f $kohadir/etc/zebradb/etc/default.idx $authorityserverdir/tab/default.idx");
        print "Info: copied default.idx\n";
        $created_dir_or_file++;
    }
    
    unless (-f "$authorityserverdir/etc/ccl.properties") {
#         system("cp -f $kohadir/etc/zebradb/ccl.properties ".C4::Context->zebraconfig('authorityserver')->{ccl2rpn});
        system("cp -f $kohadir/etc/zebradb/ccl.properties $authorityserverdir/etc/ccl.properties");
        print "Info: copied ccl.properties\n";
        $created_dir_or_file++;
    }
    unless (-f "$authorityserverdir/etc/pqf.properties") {
#         system("cp -f $kohadir/etc/zebradb/pqf.properties ".C4::Context->zebraconfig('authorityserver')->{ccl2rpn});
        system("cp -f $kohadir/etc/zebradb/pqf.properties $authorityserverdir/etc/pqf.properties");
        print "Info: copied pqf.properties\n";
        $created_dir_or_file++;
    }
    
    #
    # AUTHORITIES : copying mandatory files
    #
    unless (-f C4::Context->zebraconfig('authorityserver')->{config}) {
    open my $zd, '>:encoding(UTF-8)' ,C4::Context->zebraconfig('authorityserver')->{config};
    print {$zd} "
# generated by KOHA/misc/migration_tools/rebuild_zebra.pl 
profilePath:\${srcdir:-.}:$authorityserverdir/tab/:$tabdir/tab/:\${srcdir:-.}/tab/

encoding: UTF-8
# Files that describe the attribute sets supported.
attset: auth1.att
attset: explain.att
attset: gils.att

modulePath:$modulesdir/modules/
# Specify record type
iso2709.recordType:grs.marcxml.record
recordType:grs.xml
recordId: (auth1,Local-Number)
storeKeys:1
storeData:1


# Lock File Area
lockDir: $authorityserverdir/lock
perm.anonymous:r
perm.kohaadmin:rw
register: $authorityserverdir/register:4G
shadow: $authorityserverdir/shadow:4G

# Temp File area for result sets
setTmpDir: $authorityserverdir/tmp

# Temp File area for index program
keyTmpDir: $authorityserverdir/key

# Approx. Memory usage during indexing
memMax: 40M
rank:rank-1
    ";
        print "Info: creating zebra-authorities.cfg\n";
        $created_dir_or_file++;
    }
    
    if ($created_dir_or_file) {
        print "Info: created : $created_dir_or_file directories & files\n";
    } else {
        print "Info: file & directories OK\n";
    }
    
}
if ($biblios) {
    if ( $verbose_logging ) {
        print "====================\n";
        print "checking directories & files for biblios\n";
        print "====================\n";
    }

    #
    # BIBLIOS : creating directory structure
    #
    unless (-d "$biblioserverdir") {
        system("mkdir -p $biblioserverdir");
        print "Info: created $biblioserverdir\n";
        $created_dir_or_file++;
    }
    unless (-d "$biblioserverdir/lock") {
        mkdir "$biblioserverdir/lock";
        print "Info: created $biblioserverdir/lock\n";
        $created_dir_or_file++;
    }
    unless (-d "$biblioserverdir/register") {
        mkdir "$biblioserverdir/register";
        print "Info: created $biblioserverdir/register\n";
        $created_dir_or_file++;
    }
    unless (-d "$biblioserverdir/shadow") {
        mkdir "$biblioserverdir/shadow";
        print "Info: created $biblioserverdir/shadow\n";
        $created_dir_or_file++;
    }
    unless (-d "$biblioserverdir/tab") {
        mkdir "$biblioserverdir/tab";
        print "Info: created $biblioserverdir/tab\n";
        $created_dir_or_file++;
    }
    unless (-d "$biblioserverdir/key") {
        mkdir "$biblioserverdir/key";
        print "Info: created $biblioserverdir/key\n";
        $created_dir_or_file++;
    }
    unless (-d "$biblioserverdir/etc") {
        mkdir "$biblioserverdir/etc";
        print "Info: created $biblioserverdir/etc\n";
        $created_dir_or_file++;
    }
    
    #
    # BIBLIOS : copying mandatory files
    #
    # the record model, depending on marc flavour
    unless (-f "$biblioserverdir/tab/record.abs") {
        if (C4::Context->preference("marcflavour") eq "UNIMARC") {
            system("cp -f $kohadir/etc/zebradb/marc_defs/unimarc/biblios/record.abs $biblioserverdir/tab/record.abs");
            print "Info: copied record.abs for UNIMARC\n";
        } else {
            system("cp -f $kohadir/etc/zebradb/marc_defs/marc21/biblios/record.abs $biblioserverdir/tab/record.abs");
            print "Info: copied record.abs for USMARC\n";
        }
        $created_dir_or_file++;
    }
    unless (-f "$biblioserverdir/tab/sort-string-utf.chr") {
        system("cp -f $kohadir/etc/zebradb/lang_defs/fr/sort-string-utf.chr $biblioserverdir/tab/sort-string-utf.chr");
        print "Info: copied sort-string-utf.chr\n";
        $created_dir_or_file++;
    }
    unless (-f "$biblioserverdir/tab/word-phrase-utf.chr") {
        system("cp -f $kohadir/etc/zebradb/lang_defs/fr/sort-string-utf.chr $biblioserverdir/tab/word-phrase-utf.chr");
        print "Info: copied word-phase-utf.chr\n";
        $created_dir_or_file++;
    }
    unless (-f "$biblioserverdir/tab/bib1.att") {
        system("cp -f $kohadir/etc/zebradb/biblios/etc/bib1.att $biblioserverdir/tab/bib1.att");
        print "Info: copied bib1.att\n";
        $created_dir_or_file++;
    }
    unless (-f "$biblioserverdir/tab/default.idx") {
        system("cp -f $kohadir/etc/zebradb/etc/default.idx $biblioserverdir/tab/default.idx");
        print "Info: copied default.idx\n";
        $created_dir_or_file++;
    }
    unless (-f "$biblioserverdir/etc/ccl.properties") {
#         system("cp -f $kohadir/etc/zebradb/ccl.properties ".C4::Context->zebraconfig('biblioserver')->{ccl2rpn});
        system("cp -f $kohadir/etc/zebradb/ccl.properties $biblioserverdir/etc/ccl.properties");
        print "Info: copied ccl.properties\n";
        $created_dir_or_file++;
    }
    unless (-f "$biblioserverdir/etc/pqf.properties") {
#         system("cp -f $kohadir/etc/zebradb/pqf.properties ".C4::Context->zebraconfig('biblioserver')->{ccl2rpn});
        system("cp -f $kohadir/etc/zebradb/pqf.properties $biblioserverdir/etc/pqf.properties");
        print "Info: copied pqf.properties\n";
        $created_dir_or_file++;
    }
    
    #
    # BIBLIOS : copying mandatory files
    #
    unless (-f C4::Context->zebraconfig('biblioserver')->{config}) {
    open my $zd, '>:encoding(UTF-8)', C4::Context->zebraconfig('biblioserver')->{config};
    print {$zd} "
# generated by KOHA/misc/migrtion_tools/rebuild_zebra.pl 
profilePath:\${srcdir:-.}:$biblioserverdir/tab/:$tabdir/tab/:\${srcdir:-.}/tab/

encoding: UTF-8
# Files that describe the attribute sets supported.
attset:bib1.att
attset:explain.att
attset:gils.att

modulePath:$modulesdir/modules/
# Specify record type
iso2709.recordType:grs.marcxml.record
recordType:grs.xml
recordId: (bib1,Local-Number)
storeKeys:1
storeData:1


# Lock File Area
lockDir: $biblioserverdir/lock
perm.anonymous:r
perm.kohaadmin:rw
register: $biblioserverdir/register:4G
shadow: $biblioserverdir/shadow:4G

# Temp File area for result sets
setTmpDir: $biblioserverdir/tmp

# Temp File area for index program
keyTmpDir: $biblioserverdir/key

# Approx. Memory usage during indexing
memMax: 40M
rank:rank-1
    ";
        print "Info: creating zebra-biblios.cfg\n";
        $created_dir_or_file++;
    }
    
    if ($created_dir_or_file) {
        print "Info: created : $created_dir_or_file directories & files\n";
    } else {
        print "Info: file & directories OK\n";
    }
    
}
}