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' => '245a,245b',
66 'biblionumber => '999c',
68 'publisher' => '260b',
71 'subject' => '600a, 650a',
74 'host-item' => '952a,952c',\" where variable='NoZebraIndexes'");
75 %index = GetNoZebraIndexes();
80 print "***********************************\n";
81 print "***** building BIBLIO indexes *****\n";
82 print "***********************************\n";
84 $sth=$dbh->prepare("select biblionumber from biblioitems order by biblionumber $limit");
88 while (my ($biblionumber) = $sth->fetchrow) {
93 $record = GetMarcBiblio($biblionumber);
96 print " There was some pb getting biblionumber : ".$biblionumber."\n";
100 # get title of the record (to store the 10 first letters with the index)
101 my ($titletag,$titlesubfield) = GetMarcFromKohaField('biblio.title');
102 my $title = lc($record->subfield($titletag,$titlesubfield));
104 # remove blancks comma (that could cause problem when decoding the string for CQL retrieval) and regexp specific values
105 $title =~ s/ |,|;|\[|\]|\(|\)|\*|-|'|=//g;
106 # limit to 10 char, should be enough, and limit the DB size
107 $title = substr($title,0,10);
109 foreach my $field ($record->fields()) {
111 next if $field->tag <10;
112 foreach my $subfield ($field->subfields()) {
113 my $tag = $field->tag();
114 my $subfieldcode = $subfield->[0];
116 # check each index to see if the subfield is stored somewhere
117 # otherwise, store it in __RAW__ index
118 foreach my $key (keys %index) {
119 if ($index{$key} =~ /$tag\*/ or $index{$key} =~ /$tag$subfieldcode/) {
121 my $line= lc $subfield->[1];
122 # remove meaningless value in the field...
123 $line =~ s/-|\.|\?|,|;|!|'|\(|\)|\[|\]|{|}|"|<|>|&|\+|\*|\/|=|:/ /g;
124 # ... and split in words
125 foreach (split / /,$line) {
126 next unless $_; # skip empty values (multiple spaces)
127 # remove any accented char
128 # if the entry is already here, improve weight
129 if ($result{$key}->{"$_"} =~ /$biblionumber,$title\-(\d);/) {
131 $result{$key}->{"$_"} =~ s/$biblionumber,$title\-(\d);//;
132 $result{$key}->{"$_"} .= "$biblionumber,$title-$weight;";
133 # otherwise, create it, with weight=1
135 $result{$key}->{"$_"}.="$biblionumber,$title-1;";
140 # the subfield is not indexed, store it in __RAW__ index anyway
142 my $line= lc $subfield->[1];
143 $line =~ s/-|\.|\?|,|;|!|'|\(|\)|\[|\]|{|}|"|<|>|&|\+|\*|\/|=/ /g;
144 foreach (split / /,$line) {
146 # warn $record->as_formatted."$_ =>".$title;
147 if ($result{__RAW__}->{"$_"} =~ /$biblionumber,$title\-(\d);/) {
150 $result{__RAW__}->{"$_"} =~ s/$biblionumber,$title\-(\d);//;
151 $result{__RAW__}->{"$_"} .= "$biblionumber,$title-$weight;";
153 $result{__RAW__}->{"$_"}.="$biblionumber,$title-1;";
160 print "\nInserting records...\n";
162 my $sth = $dbh->prepare("INSERT INTO nozebra (server,indexname,value,biblionumbers) VALUES ('biblioserver',?,?,?)");
163 foreach my $key (keys %result) {
164 foreach my $index (keys %{$result{$key}}) {
165 if (length($result{$key}->{$index}) > 1000000) {
166 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";
170 $sth->execute($key,$index,$result{$key}->{$index});
173 print "\nbiblios done\n";
175 print "\n***********************************\n";
176 print "***** building AUTHORITIES indexes *****\n";
177 print "***********************************\n";
180 $sth=$dbh->prepare("select authid from auth_header order by authid $limit");
184 while (my ($authid) = $sth->fetchrow) {
189 $record = GetAuthority($authid);
192 print " There was some pb getting authnumber : ".$authid."\n";
197 # for authorities, the "title" is the $a mainentry
198 my $authref = C4::AuthoritiesMarc::GetAuthType($record->subfield(152,'b'));
200 warn "ERROR : authtype undefined for ".$record->as_formatted unless $authref;
201 my $title = $record->subfield($authref->{auth_tag_to_report},'a');
202 $index{'mainmainentry'}= $authref->{'auth_tag_to_report'}.'a';
203 $index{'mainentry'} = $authref->{'auth_tag_to_report'}.'*';
204 $index{'auth_type'} = '152b';
206 # remove blancks comma (that could cause problem when decoding the string for CQL retrieval) and regexp specific values
207 $title =~ s/ |,|;|\[|\]|\(|\)|\*|-|'|=//g;
208 $title = quotemeta $title;
209 # limit to 10 char, should be enough, and limit the DB size
210 $title = substr($title,0,10);
212 foreach my $field ($record->fields()) {
214 next if $field->tag <10;
215 foreach my $subfield ($field->subfields()) {
216 my $tag = $field->tag();
217 my $subfieldcode = $subfield->[0];
219 # check each index to see if the subfield is stored somewhere
220 # otherwise, store it in __RAW__ index
221 foreach my $key (keys %index) {
222 if ($index{$key} =~ /$tag\*/ or $index{$key} =~ /$tag$subfieldcode/) {
224 my $line= lc $subfield->[1];
225 # remove meaningless value in the field...
226 $line =~ s/-|\.|\?|,|;|!|'|\(|\)|\[|\]|{|}|"|<|>|&|\+|\*|\/|=|:/ /g;
227 # ... and split in words
228 foreach (split / /,$line) {
229 next unless $_; # skip empty values (multiple spaces)
230 # if the entry is already here, improve weight
231 if ($result{$key}->{"$_"} =~ /$authid,$title\-(\d);/) {
233 $result{$key}->{"$_"} =~ s/$authid,$title\-(\d);//;
234 $result{$key}->{"$_"} .= "$authid,$title-$weight;";
235 # otherwise, create it, with weight=1
237 $result{$key}->{"$_"}.="$authid,$title-1;";
242 # the subfield is not indexed, store it in __RAW__ index anyway
244 my $line= lc $subfield->[1];
245 $line =~ s/-|\.|\?|,|;|!|'|\(|\)|\[|\]|{|}|"|<|>|&|\+|\*|\/|=/ /g;
246 foreach (split / /,$line) {
248 # warn $record->as_formatted."$_ =>".$title;
249 if ($result{__RAW__}->{"$_"} =~ /$authid,$title\-(\d);/) {
252 $result{__RAW__}->{"$_"} =~ s/$authid,$title\-(\d);//;
253 $result{__RAW__}->{"$_"} .= "$authid,$title-$weight;";
255 $result{__RAW__}->{"$_"}.="$authid,$title-1;";
262 print "\nInserting...\n";
264 my $sth = $dbh->prepare("INSERT INTO nozebra (server,indexname,value,biblionumbers) VALUES ('authorityserver',?,?,?)");
265 foreach my $key (keys %result) {
266 foreach my $index (keys %{$result{$key}}) {
267 if (length($result{$key}->{$index}) > 1000000) {
268 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";
272 $sth->execute($key,$index,$result{$key}->{$index});
275 print "\nauthorities done\n";