use C4::AuthoritiesMarc;
use strict;
+#use warnings; FIXME - Bug 2505
#
# script that fills the nozebra table
#
# limit for database dumping
my $limit;# = "LIMIT 100";
my $directory;
-my $skip_export;
-my $keep_export;
-my $reset;
-my $biblios;
+#my $skip_export;
+#my $keep_export;
+#my $reset;
+#my $biblios;
my $authorities;
my $sysprefs;
-GetOptions(
- 'd:s' => \$directory,
- 'reset' => \$reset,
- 's' => \$skip_export,
- 'k' => \$keep_export,
- 'b' => \$biblios,
- 'a' => \$authorities,
- 's' => \$sysprefs, # rebuild 'NoZebraIndexes' syspref
- );
+my $commit;
+my $want_help;
+
+my $result = GetOptions(
+ 'd:s' => \$directory,
+# 'reset' => \$reset,
+# 's' => \$skip_export, # Not used and conflicts with 's' option some lines below for sysprefs!!!
+# 'k' => \$keep_export,
+# 'b' => \$biblios,
+# 'a' => \$authorities,
+ 's' => \$sysprefs, # rebuild 'NoZebraIndexes' syspref
+ 'h|help' => \$want_help,
+ 'commit:f' => \$commit,
+ );
+
+if (not $result or $want_help) {
+ print_usage();
+ exit 0;
+}
+
+
+sub print_usage {
+ print <<_USAGE_;
+$0: reindex MARC bibs and authorities if NOT using Zebra ("NoZebra").
+
+Use this batch job to reindex all biblio and authority
+records in your Koha database. This job is useful
+only if you are NOT using Zebra ('NoZebra'); if you are
+using the 'Zebra'mode, this job should NOT be used.
+
+Parameters:
+ -d Temporary directory for indexing.
+ If not specified, one is automatically
+ created. The export directory
+ is automatically deleted unless
+ you supply the -k switch.
+
+ -s Rebuild "NoZebraIndexes" System Preference
+
+ --help or -h show this message.
+_USAGE_
+} # END of print_usage sub
+
+
+my $commitnum = 1000;
+$commitnum = $commit if ($commit) ;
$directory = "export" unless $directory;
my $dbh=C4::Context->dbh;
'host-item' => '995a,995c',\" where variable='NoZebraIndexes'");
%index = GetNoZebraIndexes();
} elsif (C4::Context->preference('marcflavour') eq 'MARC21') {
- $dbh->do("UPDATE systempreferences SET value=\"
- 'title' => '245a,245b',
- 'author' => '100a',
- 'isbn' => '020a',
- 'issn' => '022a',
- 'biblionumber => '999c',
- 'itemtype' => '942c',
- 'publisher' => '260b',
- 'date' => '260c',
- 'note' => '500a',
- 'subject' => '600a, 650a',
- 'dewey' => '082',
- 'bc' => '952p',
- 'host-item' => '952a,952c',\" where variable='NoZebraIndexes'");
+ $dbh->do("UPDATE systempreferences SET value=\"
+'title' => '130a,210a,222a,240a,243a,245a,245b,246a,246b,247a,247b,250a,250b,440a,830a',
+'author' => '100a,100b,100c,100d,110a,111a,111b,111c,111d,245c,700a,710a,711a,800a,810a,811a',
+'isbn' => '020a',
+'issn' => '022a',
+'lccn' => '010a',
+'biblionumber => '999c',
+'itemtype' => '942c',
+'publisher' => '260b',
+'date' => '260c',
+'note' => '500a, 501a,504a,505a,508a,511a,518a,520a,521a,522a,524a,526a,530a,533a,538a,541a,546a,555a,556a,562a,563a,583a,585a,582a',
+'subject' => '600*,610*,611*,630*,650*,651*,653*,654*,655*,662*,690*',
+'dewey' => '082',
+'bc' => '952p',
+'callnum' => '952o',
+'an' => '6009,6109,6119',
+'series' => 440*,490*,
+'host-item' => '9529
+'shelf' => '952c',
+'collection' => '9528',
+\"WHERE variable='NoZebraIndexes'");
+
%index = GetNoZebraIndexes();
}
}
$|=1;
+$dbh->{AutoCommit} = 0;
+
print "***********************************\n";
print "***** building BIBLIO indexes *****\n";
print "***********************************\n";
+
my $sth;
$sth=$dbh->prepare("select biblionumber from biblioitems order by biblionumber $limit");
$sth->execute();
}
next unless $record;
# get title of the record (to store the 10 first letters with the index)
- my ($titletag,$titlesubfield) = GetMarcFromKohaField('biblio.title');
+ my ($titletag,$titlesubfield) = GetMarcFromKohaField('biblio.title', '');
my $title = lc($record->subfield($titletag,$titlesubfield));
# remove blancks comma (that could cause problem when decoding the string for CQL retrieval) and regexp specific values
- $title =~ s/ |,|;|\[|\]|\(|\)|\*|-|'|=//g;
+ $title =~ s/ |\.|,|;|\[|\]|\(|\)|\*|-|'|=|://g;
# limit to 10 char, should be enough, and limit the DB size
$title = substr($title,0,10);
#parse each field
# check each index to see if the subfield is stored somewhere
# otherwise, store it in __RAW__ index
foreach my $key (keys %index) {
- if ($index{$key} =~ /$tag\*/ or $index{$key} =~ /$tag$subfieldcode/) {
+ if ($index{$key} =~ /\Q$tag\E\*/ or $index{$key} =~ /\Q$tag$subfieldcode\E/) {
$indexed=1;
my $line= lc $subfield->[1];
# remove meaningless value in the field...
next unless $_; # skip empty values (multiple spaces)
# remove any accented char
# if the entry is already here, improve weight
- if ($result{$key}->{"$_"} =~ /$biblionumber,$title\-(\d);/) {
+ if ($result{$key}->{"$_"} =~ /$biblionumber,\Q$title\E\-(\d);/) {
my $weight=$1+1;
- $result{$key}->{"$_"} =~ s/$biblionumber,$title\-(\d);//;
+ $result{$key}->{"$_"} =~ s/$biblionumber,\Q$title\E\-(\d);//;
$result{$key}->{"$_"} .= "$biblionumber,$title-$weight;";
# otherwise, create it, with weight=1
} else {
foreach (split / /,$line) {
next unless $_;
# warn $record->as_formatted."$_ =>".$title;
- if ($result{__RAW__}->{"$_"} =~ /$biblionumber,$title\-(\d);/) {
+ if ($result{__RAW__}->{"$_"} =~ /$biblionumber,\Q$title\E\-(\d);/) {
my $weight=$1+1;
# $weight++;
- $result{__RAW__}->{"$_"} =~ s/$biblionumber,$title\-(\d);//;
+ $result{__RAW__}->{"$_"} =~ s/$biblionumber,\Q$title\E\-(\d);//;
$result{__RAW__}->{"$_"} .= "$biblionumber,$title-$weight;";
} else {
$result{__RAW__}->{"$_"}.="$biblionumber,$title-1;";
}
}
}
+
+
print "\nInserting records...\n";
$i=0;
-my $sth = $dbh->prepare("INSERT INTO nozebra (server,indexname,value,biblionumbers) VALUES ('biblioserver',?,?,?)");
+
+my $commitnum = 100;
+$dbh->{AutoCommit} = 0;
+
+$sth = $dbh->prepare("INSERT INTO nozebra (server,indexname,value,biblionumbers) VALUES ('biblioserver',?,?,?)");
foreach my $key (keys %result) {
foreach my $index (keys %{$result{$key}}) {
if (length($result{$key}->{$index}) > 1000000) {
print "\r$i";
$i++;
$sth->execute($key,$index,$result{$key}->{$index});
+ $dbh->commit() if (0 == $i % $commitnum);
}
+ $dbh->commit() if (0 == $i % $commitnum);
}
+$dbh->commit;
+
+
print "\nbiblios done\n";
print "\n***********************************\n";
print "***** building AUTHORITIES indexes *****\n";
print "***********************************\n";
-my $sth;
$sth=$dbh->prepare("select authid from auth_header order by authid $limit");
$sth->execute();
-my $i=0;
-my %result;
+$i=0;
+%result = ();
while (my ($authid) = $sth->fetchrow) {
$i++;
print "\r$i";
my %index;
# for authorities, the "title" is the $a mainentry
- my $authref = C4::AuthoritiesMarc::GetAuthType($record->subfield(152,'b'));
+ my $authref = C4::AuthoritiesMarc::GetAuthType(C4::AuthoritiesMarc::GetAuthTypeCode($authid));
warn "ERROR : authtype undefined for ".$record->as_formatted unless $authref;
my $title = $record->subfield($authref->{auth_tag_to_report},'a');
$index{'auth_type'} = '152b';
# remove blancks comma (that could cause problem when decoding the string for CQL retrieval) and regexp specific values
- $title =~ s/ |,|;|\[|\]|\(|\)|\*|-|'|=//g;
+ $title =~ s/ |\.|,|;|\[|\]|\(|\)|\*|-|'|:|=//g;
$title = quotemeta $title;
# limit to 10 char, should be enough, and limit the DB size
$title = substr($title,0,10);
}
}
}
+
+
+
print "\nInserting...\n";
$i=0;
-my $sth = $dbh->prepare("INSERT INTO nozebra (server,indexname,value,biblionumbers) VALUES ('authorityserver',?,?,?)");
+
+my $commitnum = 100;
+$dbh->{AutoCommit} = 0;
+$sth = $dbh->prepare("INSERT INTO nozebra (server,indexname,value,biblionumbers) VALUES ('authorityserver',?,?,?)");
foreach my $key (keys %result) {
foreach my $index (keys %{$result{$key}}) {
if (length($result{$key}->{$index}) > 1000000) {
print "\r$i";
$i++;
$sth->execute($key,$index,$result{$key}->{$index});
+ $dbh->commit() if (0 == $i % $commitnum);
}
+ $dbh->commit() if (0 == $i % $commitnum);
}
+$dbh->commit;
print "\nauthorities done\n";