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";
32 's' => \$sysprefs, # rebuild 'NoZebraIndexes' syspref
35 $directory = "export" unless $directory;
36 my $dbh=C4::Context->dbh;
37 $dbh->do("update systempreferences set value=1 where variable='NoZebra'");
39 $dbh->do("truncate nozebra");
41 my %index = GetNoZebraIndexes();
43 if (!%index || $sysprefs ) {
44 if (C4::Context->preference('marcflavour') eq 'UNIMARC') {
45 $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',
46 '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',
49 'biblionumber' =>'0909',
52 'publisher' => '210c',
54 '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',
55 'Koha-Auth-Number' => '6009,6019,6029,6039,6049,6059,6069,6109,7009,7019,7029,7109,7119,7129',
56 'subject' => '600*,601*,606*,610*',
58 'host-item' => '995a,995c',\" where variable='NoZebraIndexes'");
59 %index = GetNoZebraIndexes();
60 } elsif (C4::Context->preference('marcflavour') eq 'MARC21') {
61 $dbh->do("UPDATE systempreferences SET value=\"
62 'title' => '130a,210a,222a,240a,243a,245a,245b,246a,246b,247a,247b,250a,250b,440a,830a',
63 'author' => '100a,100b,100c,100d,110a,111a,111b,111c,111d,245c,700a,710a,711a,800a,810a,811a',
67 'biblionumber => '999c',
69 'publisher' => '260b',
71 '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',
72 'subject' => '600*,610*,611*,630*,650*,651*,653*,654*,655*,662*,690*',
76 'an' => '6009,6109,6119',
77 'series' => 440*,490*,
78 'host-item' => '952a,952c',
79 \"WHERE variable='NoZebraIndexes'");
81 %index = GetNoZebraIndexes();
86 print "***********************************\n";
87 print "***** building BIBLIO indexes *****\n";
88 print "***********************************\n";
90 $sth=$dbh->prepare("select biblionumber from biblioitems order by biblionumber $limit");
94 while (my ($biblionumber) = $sth->fetchrow) {
99 $record = GetMarcBiblio($biblionumber);
102 print " There was some pb getting biblionumber : ".$biblionumber."\n";
106 # get title of the record (to store the 10 first letters with the index)
107 my ($titletag,$titlesubfield) = GetMarcFromKohaField('biblio.title');
108 my $title = lc($record->subfield($titletag,$titlesubfield));
110 # remove blancks comma (that could cause problem when decoding the string for CQL retrieval) and regexp specific values
111 $title =~ s/ |,|;|\[|\]|\(|\)|\*|-|'|=//g;
112 # limit to 10 char, should be enough, and limit the DB size
113 $title = substr($title,0,10);
115 foreach my $field ($record->fields()) {
117 next if $field->tag <10;
118 foreach my $subfield ($field->subfields()) {
119 my $tag = $field->tag();
120 my $subfieldcode = $subfield->[0];
122 # check each index to see if the subfield is stored somewhere
123 # otherwise, store it in __RAW__ index
124 foreach my $key (keys %index) {
125 if ($index{$key} =~ /$tag\*/ or $index{$key} =~ /$tag$subfieldcode/) {
127 my $line= lc $subfield->[1];
128 # remove meaningless value in the field...
129 $line =~ s/-|\.|\?|,|;|!|'|\(|\)|\[|\]|{|}|"|<|>|&|\+|\*|\/|=|:/ /g;
130 # ... and split in words
131 foreach (split / /,$line) {
132 next unless $_; # skip empty values (multiple spaces)
133 # remove any accented char
134 # if the entry is already here, improve weight
135 if ($result{$key}->{"$_"} =~ /$biblionumber,\Q$title\E\-(\d);/) {
137 $result{$key}->{"$_"} =~ s/$biblionumber,\Q$title\E\-(\d);//;
138 $result{$key}->{"$_"} .= "$biblionumber,$title-$weight;";
139 # otherwise, create it, with weight=1
141 $result{$key}->{"$_"}.="$biblionumber,$title-1;";
146 # the subfield is not indexed, store it in __RAW__ index anyway
148 my $line= lc $subfield->[1];
149 $line =~ s/-|\.|\?|,|;|!|'|\(|\)|\[|\]|{|}|"|<|>|&|\+|\*|\/|=/ /g;
150 foreach (split / /,$line) {
152 # warn $record->as_formatted."$_ =>".$title;
153 if ($result{__RAW__}->{"$_"} =~ /$biblionumber,\Q$title\E\-(\d);/) {
156 $result{__RAW__}->{"$_"} =~ s/$biblionumber,\Q$title\E\-(\d);//;
157 $result{__RAW__}->{"$_"} .= "$biblionumber,$title-$weight;";
159 $result{__RAW__}->{"$_"}.="$biblionumber,$title-1;";
166 print "\nInserting records...\n";
170 $dbh->{AutoCommit} = 0;
172 $sth = $dbh->prepare("INSERT INTO nozebra (server,indexname,value,biblionumbers) VALUES ('biblioserver',?,?,?)");
173 foreach my $key (keys %result) {
174 foreach my $index (keys %{$result{$key}}) {
175 if (length($result{$key}->{$index}) > 1000000) {
176 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";
180 $sth->execute($key,$index,$result{$key}->{$index});
181 $dbh->commit() if (0 == $i % $commitnum);
184 print "\nbiblios done\n";
186 print "\n***********************************\n";
187 print "***** building AUTHORITIES indexes *****\n";
188 print "***********************************\n";
190 $sth=$dbh->prepare("select authid from auth_header order by authid $limit");
194 while (my ($authid) = $sth->fetchrow) {
199 $record = GetAuthority($authid);
202 print " There was some pb getting authnumber : ".$authid."\n";
207 # for authorities, the "title" is the $a mainentry
208 my $authref = C4::AuthoritiesMarc::GetAuthType(C4::AuthoritiesMarc::GetAuthTypeCode($authid));
210 warn "ERROR : authtype undefined for ".$record->as_formatted unless $authref;
211 my $title = $record->subfield($authref->{auth_tag_to_report},'a');
212 $index{'mainmainentry'}= $authref->{'auth_tag_to_report'}.'a';
213 $index{'mainentry'} = $authref->{'auth_tag_to_report'}.'*';
214 $index{'auth_type'} = '152b';
216 # remove blancks comma (that could cause problem when decoding the string for CQL retrieval) and regexp specific values
217 $title =~ s/ |,|;|\[|\]|\(|\)|\*|-|'|=//g;
218 $title = quotemeta $title;
219 # limit to 10 char, should be enough, and limit the DB size
220 $title = substr($title,0,10);
222 foreach my $field ($record->fields()) {
224 next if $field->tag <10;
225 foreach my $subfield ($field->subfields()) {
226 my $tag = $field->tag();
227 my $subfieldcode = $subfield->[0];
229 # check each index to see if the subfield is stored somewhere
230 # otherwise, store it in __RAW__ index
231 foreach my $key (keys %index) {
232 if ($index{$key} =~ /$tag\*/ or $index{$key} =~ /$tag$subfieldcode/) {
234 my $line= lc $subfield->[1];
235 # remove meaningless value in the field...
236 $line =~ s/-|\.|\?|,|;|!|'|\(|\)|\[|\]|{|}|"|<|>|&|\+|\*|\/|=|:/ /g;
237 # ... and split in words
238 foreach (split / /,$line) {
239 next unless $_; # skip empty values (multiple spaces)
240 # if the entry is already here, improve weight
241 if ($result{$key}->{"$_"} =~ /$authid,$title\-(\d);/) {
243 $result{$key}->{"$_"} =~ s/$authid,$title\-(\d);//;
244 $result{$key}->{"$_"} .= "$authid,$title-$weight;";
245 # otherwise, create it, with weight=1
247 $result{$key}->{"$_"}.="$authid,$title-1;";
252 # the subfield is not indexed, store it in __RAW__ index anyway
254 my $line= lc $subfield->[1];
255 $line =~ s/-|\.|\?|,|;|!|'|\(|\)|\[|\]|{|}|"|<|>|&|\+|\*|\/|=/ /g;
256 foreach (split / /,$line) {
258 # warn $record->as_formatted."$_ =>".$title;
259 if ($result{__RAW__}->{"$_"} =~ /$authid,$title\-(\d);/) {
262 $result{__RAW__}->{"$_"} =~ s/$authid,$title\-(\d);//;
263 $result{__RAW__}->{"$_"} .= "$authid,$title-$weight;";
265 $result{__RAW__}->{"$_"}.="$authid,$title-1;";
272 print "\nInserting...\n";
274 $sth = $dbh->prepare("INSERT INTO nozebra (server,indexname,value,biblionumbers) VALUES ('authorityserver',?,?,?)");
275 foreach my $key (keys %result) {
276 foreach my $index (keys %{$result{$key}}) {
277 if (length($result{$key}->{$index}) > 1000000) {
278 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";
282 $sth->execute($key,$index,$result{$key}->{$index});
285 print "\nauthorities done\n";