6 use C4::AuthoritiesMarc;
10 # script that fills the nozebra table
14 $|=1; # flushes output
16 # limit for database dumping
17 my $limit;# = "LIMIT 100";
33 $directory = "export" unless $directory;
34 my $dbh=C4::Context->dbh;
35 $dbh->do("update systempreferences set value=1 where variable='NoZebra'");
36 $dbh->do("CREATE TABLE `nozebra` (
37 `server` varchar(20) NOT NULL,
38 `indexname` varchar(40) NOT NULL,
39 `value` varchar(250) NOT NULL,
40 `biblionumbers` longtext NOT NULL,
41 KEY `indexname` (`server`,`indexname`),
42 KEY `value` (`server`,`value`))
43 ENGINE=InnoDB DEFAULT CHARSET=utf8");
45 $dbh->do("truncate nozebra");
47 my %index = GetNoZebraIndexes();
50 if (C4::Context->preference('marcflavour') eq 'UNIMARC') {
51 $dbh->do("UPDATE systempreferences SET value=\"'title' => '200a,200c,200d,200e,225a,225d,225e,225f,225h,225i,225v,500*,501*,503*,510*,512*,513*,514*,515*,516*,517*,518*,519*,520*,530*,531*,532*,540*,541*,545*,604t,610t,605a',
52 'author' =>'200f,600a,601a,604a,700a,700b,700c,700d,700a,701b,701c,701d,702a,702b,702c,702d,710a,710b,710c,710d,711a,711b,711c,711d,712a,712b,712c,712d',
55 'biblionumber' =>'0909',
58 'publisher' => '210c',
60 'note' => '300a,301a,302a,303a,304a,305a,306az,307a,308a,309a,310a,311a,312a,313a,314a,315a,316a,317a,318a,319a,320a,321a,322a,323a,324a,325a,326a,327a,328a,330a,332a,333a,336a,337a,345a',
61 'Koha-Auth-Number' => '6009,6019,6029,6039,6049,6059,6069,6109,7009,7019,7029,7109,7119,7129',
62 'subject' => '600*,601*,606*,610*',
64 'host-item' => '995a,995c',\" where variable='NoZebraIndexes'");
65 %index = GetNoZebraIndexes();
67 # build a MARC21 default index file
72 print "***********************************\n";
73 print "***** building BIBLIO indexes *****\n";
74 print "***********************************\n";
76 $sth=$dbh->prepare("select biblionumber from biblioitems order by biblionumber $limit");
80 while (my ($biblionumber) = $sth->fetchrow) {
85 $record = GetMarcBiblio($biblionumber);
88 print " There was some pb getting biblionumber : ".$biblionumber."\n";
91 # get title of the record (to store the 10 first letters with the index)
92 my ($titletag,$titlesubfield) = GetMarcFromKohaField('biblio.title');
93 my $title = lc($record->subfield($titletag,$titlesubfield));
95 # remove blancks comma (that could cause problem when decoding the string for CQL retrieval) and regexp specific values
96 $title =~ s/ |,|;|\[|\]|\(|\)|\*|-|'|=//g;
97 $title = quotemeta $title;
98 # limit to 10 char, should be enough, and limit the DB size
99 $title = substr($title,0,10);
101 foreach my $field ($record->fields()) {
103 next if $field->tag <10;
104 foreach my $subfield ($field->subfields()) {
105 my $tag = $field->tag();
106 my $subfieldcode = $subfield->[0];
108 # check each index to see if the subfield is stored somewhere
109 # otherwise, store it in __RAW__ index
110 foreach my $key (keys %index) {
111 if ($index{$key} =~ /$tag\*/ or $index{$key} =~ /$tag$subfieldcode/) {
113 my $line= lc $subfield->[1];
114 # remove meaningless value in the field...
115 $line =~ s/-|\.|\?|,|;|!|'|\(|\)|\[|\]|{|}|"|<|>|&|\+|\*|\/|=|:/ /g;
116 # ... and split in words
117 foreach (split / /,$line) {
118 next unless $_; # skip empty values (multiple spaces)
119 # remove any accented char
120 # if the entry is already here, improve weight
121 if ($result{$key}->{"$_"} =~ /$biblionumber,$title\-(\d);/) {
123 $result{$key}->{"$_"} =~ s/$biblionumber,$title\-(\d);//;
124 $result{$key}->{"$_"} .= "$biblionumber,$title-$weight;";
125 # otherwise, create it, with weight=1
127 $result{$key}->{"$_"}.="$biblionumber,$title-1;";
132 # the subfield is not indexed, store it in __RAW__ index anyway
134 my $line= lc $subfield->[1];
135 $line =~ s/-|\.|\?|,|;|!|'|\(|\)|\[|\]|{|}|"|<|>|&|\+|\*|\/|=/ /g;
136 foreach (split / /,$line) {
138 # warn $record->as_formatted."$_ =>".$title;
139 if ($result{__RAW__}->{"$_"} =~ /$biblionumber,$title\-(\d);/) {
142 $result{__RAW__}->{"$_"} =~ s/$biblionumber,$title\-(\d);//;
143 $result{__RAW__}->{"$_"} .= "$biblionumber,$title-$weight;";
145 $result{__RAW__}->{"$_"}.="$biblionumber,$title-1;";
152 print "\nInserting records...\n";
154 my $sth = $dbh->prepare("INSERT INTO nozebra (server,indexname,value,biblionumbers) VALUES ('biblioserver',?,?,?)");
155 foreach my $key (keys %result) {
156 foreach my $index (keys %{$result{$key}}) {
157 if (length($result{$key}->{$index}) > 1000000) {
158 print "very long index (".length($result{$key}->{$index}).")for $key / $index. update mySQL config file if you have an error just after this warning (max_paquet_size parameter)\n";
162 $sth->execute($key,$index,$result{$key}->{$index});
165 print "\nbiblios done\n";
167 print "\n***********************************\n";
168 print "***** building AUTHORITIES indexes *****\n";
169 print "***********************************\n";
172 $sth=$dbh->prepare("select authid from auth_header order by authid $limit");
176 while (my ($authid) = $sth->fetchrow) {
181 $record = GetAuthority($authid);
184 print " There was some pb getting authnumber : ".$authid."\n";
189 # for authorities, the "title" is the $a mainentry
190 my $authref = C4::AuthoritiesMarc::GetAuthType($record->subfield(152,'b'));
192 warn "ERROR : authtype undefined for ".$record->as_formatted unless $authref;
193 my $title = $record->subfield($authref->{auth_tag_to_report},'a');
194 $index{'mainmainentry'}= $authref->{'auth_tag_to_report'}.'a';
195 $index{'mainentry'} = $authref->{'auth_tag_to_report'}.'*';
196 $index{'auth_type'} = '152b';
198 # remove blancks comma (that could cause problem when decoding the string for CQL retrieval) and regexp specific values
199 $title =~ s/ |,|;|\[|\]|\(|\)|\*|-|'|=//g;
200 $title = quotemeta $title;
201 # limit to 10 char, should be enough, and limit the DB size
202 $title = substr($title,0,10);
204 foreach my $field ($record->fields()) {
206 next if $field->tag <10;
207 foreach my $subfield ($field->subfields()) {
208 my $tag = $field->tag();
209 my $subfieldcode = $subfield->[0];
211 # check each index to see if the subfield is stored somewhere
212 # otherwise, store it in __RAW__ index
213 foreach my $key (keys %index) {
214 if ($index{$key} =~ /$tag\*/ or $index{$key} =~ /$tag$subfieldcode/) {
216 my $line= lc $subfield->[1];
217 # remove meaningless value in the field...
218 $line =~ s/-|\.|\?|,|;|!|'|\(|\)|\[|\]|{|}|"|<|>|&|\+|\*|\/|=|:/ /g;
219 # ... and split in words
220 foreach (split / /,$line) {
221 next unless $_; # skip empty values (multiple spaces)
222 # if the entry is already here, improve weight
223 if ($result{$key}->{"$_"} =~ /$authid,$title\-(\d);/) {
225 $result{$key}->{"$_"} =~ s/$authid,$title\-(\d);//;
226 $result{$key}->{"$_"} .= "$authid,$title-$weight;";
227 # otherwise, create it, with weight=1
229 $result{$key}->{"$_"}.="$authid,$title-1;";
234 # the subfield is not indexed, store it in __RAW__ index anyway
236 my $line= lc $subfield->[1];
237 $line =~ s/-|\.|\?|,|;|!|'|\(|\)|\[|\]|{|}|"|<|>|&|\+|\*|\/|=/ /g;
238 foreach (split / /,$line) {
240 # warn $record->as_formatted."$_ =>".$title;
241 if ($result{__RAW__}->{"$_"} =~ /$authid,$title\-(\d);/) {
244 $result{__RAW__}->{"$_"} =~ s/$authid,$title\-(\d);//;
245 $result{__RAW__}->{"$_"} .= "$authid,$title-$weight;";
247 $result{__RAW__}->{"$_"}.="$authid,$title-1;";
254 print "\nInserting...\n";
256 my $sth = $dbh->prepare("INSERT INTO nozebra (server,indexname,value,biblionumbers) VALUES ('authorityserver',?,?,?)");
257 foreach my $key (keys %result) {
258 foreach my $index (keys %{$result{$key}}) {
259 if (length($result{$key}->{$index}) > 1000000) {
260 print "very long index (".length($result{$key}->{$index}).")for $key / $index. update mySQL config file if you have an error just after this warning (max_paquet_size parameter)\n";
264 $sth->execute($key,$index,$result{$key}->{$index});
267 print "\nauthorities done\n";