1 # $Id: Session.pm,v 1.22 2004-09-09 15:23:07 heikki Exp $
3 # Zebra perl API header
4 # =============================================================================
5 package IDZebra::Session;
14 use IDZebra::Logger qw(:flags :calls);
15 use IDZebra::Resultset;
16 use IDZebra::ScanList;
17 use IDZebra::RetrievalRecord;
19 our $VERSION = do { my @r = (q$Revision: 1.22 $ =~ /\d+/g); sprintf "%d."."%02d" x $#r, @r };
20 our @ISA = qw(IDZebra::Logger Exporter);
21 our @EXPORT = qw (TRANS_RW TRANS_RO);
24 use constant TRANS_RW => 1;
25 use constant TRANS_RO => 0;
28 # -----------------------------------------------------------------------------
29 # Class constructors, destructor
30 # -----------------------------------------------------------------------------
32 my ($proto, %args) = @_;
33 my $class = ref($proto) || $proto;
35 $self->{args} = \%args;
37 bless ($self, $class);
38 $self->{cql_ct} = undef;
39 $self->{cql_mapfile} = "";
42 $self->{databases} = {};
46 my ($self, %args) = @_;
49 unless (defined($self->{zs})) {
50 if (defined($args{'configFile'})) {
51 $self->{zs} = IDZebra::start($args{'configFile'});
53 $self->{zs} = IDZebra::start("zebra.cfg");
60 if (defined($self->{zs})) {
61 IDZebra::stop($self->{zs}) if ($self->{zs});
68 my ($proto,%args) = @_;
71 if (ref($proto)) { $self = $proto; } else {
72 $self = $proto->new(%args);
76 %args = %{$self->{args}};
79 $self->start_service(%args);
81 unless (defined($self->{zs})) {
82 croak ("Falied to open zebra service");
85 unless (defined($self->{zh})) {
86 $self->{zh}=IDZebra::open($self->{zs});
89 # Reset result set counter
92 # This is needed in order to somehow initialize the service
93 $self->databases("Default");
95 # ADAM: group call deleted
96 # Load the default configuration
97 # $self->group(%args);
99 # ADAM: Set group resource instead
100 if (defined($args{groupName})) {
101 IDZebra::set_resource($self->{zh}, "group", $args{groupName});
105 my $shadow = defined($args{shadow}) ? $args{shadow} : 0;
106 $self->shadow($shadow);
108 $self->{odr_input} = IDZebra::odr_createmem($IDZebra::ODR_DECODE);
109 $self->{odr_output} = IDZebra::odr_createmem($IDZebra::ODR_ENCODE);
116 unless (defined($self->{zh})) {
117 croak ("Zebra session is not opened");
127 # Delete all resulsets
128 my $r = IDZebra::deleteResultSet($self->{zh},
129 1, #Z_DeleteRequest_all,
133 while (IDZebra::trans_no($self->{zh}) > 0) {
134 logf (LOG_WARN,"Explicitly closing transaction with session");
138 IDZebra::close($self->{zh});
142 if ($self->{odr_input}) {
143 IDZebra::odr_reset($self->{odr_input});
144 IDZebra::odr_destroy($self->{odr_input});
145 $self->{odr_input} = undef;
148 if ($self->{odr_output}) {
149 IDZebra::odr_reset($self->{odr_output});
150 IDZebra::odr_destroy($self->{odr_output});
151 $self->{odr_output} = undef;
159 logf (LOG_LOG,"DESTROY $self");
162 if (defined ($self->{cql_ct})) {
163 IDZebra::cql_transform_close($self->{cql_ct});
167 # -----------------------------------------------------------------------------
168 # Record group selection This is a bit nasty... but used at many places
169 # -----------------------------------------------------------------------------
171 # ADAM: All these group functions have been disabled.
173 my ($self,%args) = @_;
176 $self->{rg} = $self->_makeRecordGroup(%args);
177 $self->_selectRecordGroup($self->{rg});
182 sub selectRecordGroup_deleted {
183 my ($self, $groupName) = @_;
185 $self->{rg} = $self->_getRecordGroup($groupName);
186 $self->_selectRecordGroup($self->{rg});
189 sub _displayRecordGroup_deleted {
190 my ($self, $rg) = @_;
191 print STDERR "-----\n";
192 foreach my $key qw (groupName
203 print STDERR "$key:",$rg->{$key},"\n";
207 sub _cloneRecordGroup_deleted {
208 my ($self, $orig) = @_;
209 my $rg = IDZebra::recordGroup->new();
210 my $r = IDZebra::init_recordGroup($rg);
211 foreach my $key qw (groupName
223 $rg->{$key} = $orig->{$key} if ($orig->{$key});
228 sub _getRecordGroup_deleted {
229 my ($self, $groupName, $ext) = @_;
230 my $rg = IDZebra::recordGroup->new();
231 my $r = IDZebra::init_recordGroup($rg);
232 $rg->{groupName} = $groupName if ($groupName ne "");
233 $ext = "" unless ($ext);
234 $r = IDZebra::res_get_recordGroup($self->{zh}, $rg, $ext);
238 sub _makeRecordGroup_deleted {
239 my ($self, %args) = @_;
242 my @keys = keys(%args);
243 unless ($#keys >= 0) {
244 return ($self->{rg});
247 if ($args{groupName}) {
248 $rg = $self->_getRecordGroup($args{groupName});
250 $rg = $self->_cloneRecordGroup($self->{rg});
252 $self->_setRecordGroupOptions($rg, %args);
256 sub _setRecordGroupOptions_deleted {
257 my ($self, $rg, %args) = @_;
259 foreach my $key qw (databaseName
270 if (defined ($args{$key})) {
271 $rg->{$key} = $args{$key};
275 sub _selectRecordGroup_deleted {
276 my ($self, $rg) = @_;
278 my $r = IDZebra::set_group($self->{zh}, $rg);
280 unless ($dbName = $rg->{databaseName}) {
283 unless ($self->databases($dbName)) {
284 croak("Fatal error selecting database $dbName");
287 # -----------------------------------------------------------------------------
288 # Selecting databases for search (and also for updating - internally)
289 # -----------------------------------------------------------------------------
291 my ($self, @databases) = @_;
296 return (keys(%{$self->{databases}}));
301 foreach my $db (@databases) {
303 next if ($self->{databases}{$db});
307 foreach my $db (keys (%{$self->{databases}})) {
308 $changed++ unless ($tmp{$db});
313 delete ($self->{databases});
314 foreach my $db (@databases) {
315 $self->{databases}{$db}++;
318 if (IDZebra::select_databases($self->{zh},
322 "Could not select database(s) %s errCode=%d",
323 join(",",@databases),
327 logf(LOG_LOG,"Database(s) selected: %s",join(",",@databases));
330 return (keys(%{$self->{databases}}));
333 # -----------------------------------------------------------------------------
335 # -----------------------------------------------------------------------------
338 return(IDZebra::errCode($self->{zh}));
343 return(IDZebra::errString($self->{zh}));
348 return(IDZebra::errAdd($self->{zh}));
351 # -----------------------------------------------------------------------------
353 # -----------------------------------------------------------------------------
356 $m = TRANS_RW unless (defined ($m));
357 if (my $err = IDZebra::begin_trans($self->{zh},$m)) {
358 if ($self->errCode == 2) {
359 croak ("TRANS_RW not allowed within TRANS_RO");
361 croak("Error starting transaction; code:".
362 $self->errCode . " message: " . $self->errString);
370 my $stat = IDZebra::ZebraTransactionStatus->new();
371 IDZebra::end_trans($self->{zh}, $stat);
376 my ($self, $value) = @_;
379 $value = 0 unless (defined($value));
380 my $r =IDZebra::set_shadow_enable($self->{zh},$value);
382 return (IDZebra::get_shadow_enable($self->{zh}));
389 return(IDZebra::commit($self->{zh}));
393 # -----------------------------------------------------------------------------
394 # We don't really need that...
395 # -----------------------------------------------------------------------------
397 my ($self, $name) = @_;
398 if ($name !~/^(input|output)$/) {
399 croak("Undefined ODR '$name'");
401 IDZebra::odr_reset($self->{"odr_$name"});
404 # -----------------------------------------------------------------------------
406 # -----------------------------------------------------------------------------
410 return(IDZebra::init($self->{zh}));
416 return(IDZebra::compact($self->{zh}));
420 my ($self, %args) = @_;
422 # ADAM: Set group resource
423 if (defined($args{groupName})) {
424 IDZebra::set_resource($self->{zh}, "group", $args{groupName});
427 # my $rg = $self->_update_args(%args); deleted
428 # $self->_selectRecordGroup($rg); deleted
430 IDZebra::repository_update($self->{zh}, $args{path});
431 # $self->_selectRecordGroup($self->{rg}); deleted
436 my ($self, %args) = @_;
438 # ADAM: Set group resource
439 if (defined($args{groupName})) {
440 IDZebra::set_resource($self->{zh}, "group", $args{groupName});
443 # my $rg = $self->_update_args(%args); deleted
444 # $self->_selectRecordGroup($rg); deleted
446 IDZebra::repository_delete($self->{zh}, $args{path});
448 # $self->_selectRecordGroup($self->{rg});
453 my ($self, %args) = @_;
455 # ADAM: Set group resource
456 if (defined($args{groupName})) {
457 IDZebra::set_resource($self->{zh}, "group", $args{groupName});
460 # my $rg = $self->_update_args(%args);
461 # $self->_selectRecordGroup($rg);
464 IDZebra::repository_show($self->{zh});
465 $self->_selectRecordGroup($self->{rg});
469 sub _update_args_deleted {
470 my ($self, %args) = @_;
471 my $rg = $self->_makeRecordGroup(%args);
472 $self->_selectRecordGroup($rg);
476 # -----------------------------------------------------------------------------
478 # -----------------------------------------------------------------------------
480 my ($self, %args) = @_;
482 $args{sysno}=0; # make sure we don't overwrite any records
483 my @args = $self->_record_update_args(%args);
484 my @ret = IDZebra::insert_record($self->{zh}, @args);
485 return @ret; # returns ($status, $sysno)
489 my ($self, %args) = @_;
491 my @args = $self->_record_update_args(%args);
492 my @ret = IDZebra::update_record($self->{zh}, @args);
493 return @ret; # ($status, $sysno)
497 my ($self, %args) = @_;
499 my @args = $self->_record_update_args(%args);
500 my $stat = IDZebra::delete_record($self->{zh}, @args);
504 sub _record_update_args {
505 my ($self, %args) = @_;
506 my $sysno = $args{sysno} ? $args{sysno} : 0;
507 my $match = $args{match} ? $args{match} : "";
508 my $rectype = $args{recordType} ? $args{recordType} : "";
509 my $fname = $args{file} ? $args{file} : "<no file>";
510 my $force = $args{force} ? $args{force} : 0;
517 elsif ($args{file}) {
518 CORE::open (F, $args{file}) || warn ("Cannot open $args{file}");
519 $buff = join('',(<F>));
522 my $len = length($buff);
524 delete ($args{sysno});
525 delete ($args{match});
526 delete ($args{recordType});
527 delete ($args{file});
528 delete ($args{data});
529 delete ($args{force});
531 # ADAM: recordGroup removed ...
532 # my $rg = $self->_makeRecordGroup(%args);
534 # If no record type is given, then try to find it out from the
535 # file extension; deleted
537 # if (my ($ext) = $fname =~ /\.(\w+)$/) {
538 # my $rg2 = $self->_getRecordGroup($rg->{groupName},$ext);
539 # $rectype = $rg2->{recordType};
543 # $rg->{databaseName} = "Default" unless ($rg->{databaseName});
548 # ADAM: set group resource
549 if (defined($args{groupName})) {
550 IDZebra::set_resource($self->{zh}, "group", $args{groupName});
553 # ADAM: rg no longer part of vector..
554 return ($rectype, $sysno, $match, $fname, $buff, $len, $force);
557 # -----------------------------------------------------------------------------
560 my ($self,$mapfile) = @_;
562 if ($self->{cql_mapfile} ne $mapfile) {
563 unless (-f $mapfile) {
564 croak("Cannot find $mapfile");
566 if (defined ($self->{cql_ct})) {
567 IDZebra::cql_transform_close($self->{cql_ct});
569 $self->{cql_ct} = IDZebra::cql_transform_open_fname($mapfile);
570 $self->{cql_mapfile} = $mapfile;
573 return ($self->{cql_mapfile});
577 my ($self, $cqlquery) = @_;
578 unless (defined($self->{cql_ct})) {
579 croak("CQL map file is not specified yet.");
581 my $res = "\0" x 2048;
582 my $r = IDZebra::cql2pqf($self->{cql_ct}, $cqlquery, $res, 2048);
584 # carp ("Error transforming CQL query: '$cqlquery', status:$r");
591 # -----------------------------------------------------------------------------
593 # -----------------------------------------------------------------------------
595 my ($self, %args) = @_;
599 if ($args{cqlmap}) { $self->cqlmap($args{cqlmap}); }
607 ($query, $cqlstat) = $self->cql2pqf($args{cql});
609 croak ("Failed to transform query: '$args{cql}', ".
610 "status: ($cqlstat)");
614 croak ("No query given to search");
619 if ($args{databases}) {
620 @origdbs = $self->databases;
621 $self->databases(@{$args{databases}});
625 my $rsname = $args{rsname} ? $args{rsname} : $self->_new_setname;
627 my $rs = $self->_search_pqf($query, $rsname);
629 if ($args{databases}) {
630 $self->databases(@origdbs);
635 carp("Sort skipped due to search error: ".
638 $rs->sort($args{sort});
647 return ("set_".$self->{rscount}++);
651 my ($self, $query, $setname) = @_;
656 my $res = IDZebra::search_PQF($self->{zh},
661 my $rs = IDZebra::Resultset->new($self,
664 recordCount => $hits,
665 errCode => $self->errCode,
666 errString => $self->errString);
670 # -----------------------------------------------------------------------------
673 # Sorting of multiple result sets is not supported by zebra...
674 # -----------------------------------------------------------------------------
677 my ($self, $sortspec, $setname, @sets) = @_;
682 croak ("Sorting/merging of multiple resultsets is not supported now");
687 foreach my $rs (@sets) {
688 push (@setnames, $rs->{name});
689 $count += $rs->{recordCount}; # is this really sure ??? It doesn't
693 my $status = IDZebra::sort($self->{zh},
699 my $errCode = $self->errCode;
700 my $errString = $self->errString;
702 logf (LOG_LOG, "Sort status $setname: %d, errCode: %d, errString: %s",
703 $status, $errCode, $errString);
705 if ($status || $errCode) {$count = 0;}
707 my $rs = IDZebra::Resultset->new($self,
709 recordCount => $count,
711 errString => $errString);
715 # -----------------------------------------------------------------------------
717 # -----------------------------------------------------------------------------
719 my ($self, %args) = @_;
723 unless ($args{expression}) {
724 croak ("No scan expression given");
727 my $sl = IDZebra::ScanList->new($self,%args);
732 # ============================================================================
738 IDZebra::Session - A Zebra database server session for update and retrieval
742 $sess = IDZebra::Session->new(configFile => 'demo/zebra.cfg');
745 $sess = IDZebra::Session->open(configFile => 'demo/zebra.cfg',
746 groupName => 'demo1');
748 $sess->group(groupName => 'demo2');
754 $sess->update(path => 'lib');
756 my $s1=$sess->update_record(data => $rec1,
757 recordType => 'grs.perl.pod',
758 groupName => "demo1",
761 my $stat = $sess->end_trans;
763 $sess->databases('demo1','demo2');
765 my $rs1 = $sess->search(cqlmap => 'demo/cql.map',
766 cql => 'dc.title=IDZebra',
767 databases => [qw(demo1 demo2)]);
772 Zebra is a high-performance, general-purpose structured text indexing and retrieval engine. It reads structured records in a variety of input formats (eg. email, XML, MARC) and allows access to them through exact boolean search expressions and relevance-ranked free-text queries.
774 Zebra supports large databases (more than ten gigabytes of data, tens of millions of records). It supports incremental, safe database updates on live systems. You can access data stored in Zebra using a variety of Index Data tools (eg. YAZ and PHP/YAZ) as well as commercial and freeware Z39.50 clients and toolkits.
776 =head1 OPENING AND CLOSING A ZEBRA SESSIONS
778 For the time beeing only local database services are supported, the same way as calling zebraidx or zebrasrv from the command shell. In order to open a local Zebra database, with a specific configuration file, use
780 $sess = IDZebra::Session->new(configFile => 'demo/zebra.cfg');
785 $sess = IDZebra::Session->open(configFile => 'demo/zebra.cfg');
787 where $sess is going to be the object representing a Zebra Session. Whenever this variable gets out of scope, the session is closed, together with all active transactions, etc... Anyway, if you'd like to close the session, just say:
792 - close all transactions
793 - destroy all result sets and scan lists
796 Note, that if I<shadow registers> are enabled, the changes will not be committed automatically.
798 In the future different database access methods are going to be available,
801 $sess = IDZebra::Session->open(server => 'ostrich.technomat.hu:9999');
803 You can also use the B<record group> arguments described below directly when calling the constructor, or the open method:
805 $sess = IDZebra::Session->open(configFile => 'demo/zebra.cfg',
806 groupName => 'demo');
811 If you manage different sets of records that share common characteristics, you can organize the configuration settings for each type into "groups". See the Zebra manual on the configuration file (zebra.cfg).
813 For each open session a default record group is assigned. You can configure it in the constructor, or by the B<group> method:
815 $sess->group(groupName => ..., ...)
817 The following options are available:
823 This will select the named record group, and load the corresponding settings from the configuration file. All subsequent values will overwrite those...
825 =item B<databaseName>
827 The name of the (logical) database the updated records will belong to.
831 This path is used for directory updates (B<update>, B<delete> methods);
835 This option determines how to identify your records. See I<Zebra manual: Locating Records>
839 The record type used for indexing.
841 =item B<flagStoreData>
843 Specifies whether the records should be stored internally in the Zebra system files. If you want to maintain the raw records yourself, this option should be false (0). If you want Zebra to take care of the records for you, it should be true(1).
845 =item B<flagStoreKeys>
847 Specifies whether key information should be saved for a given group of records. If you plan to update/delete this type of records later this should be specified as 1; otherwise it should be 0 (default), to save register space.
853 =item B<fileVerboseLimit>
855 Skip log messages, when doing a directory update, and the specified number of files are processed...
857 =item B<databaseNamePath>
861 =item B<explainDatabase>
863 The name of the explain database to be used
867 Follow links when doing directory update.
871 You can use the same parameters calling all update methods.
873 =head1 TRANSACTIONS (READ / WRITE LOCKS)
875 A transaction is a block of record update (insert / modify / delete) or retrieval procedures. So, all call to such function will implicitly start a transaction, unless one is already started by
881 $sess->begin_trans(TRANS_RW)
883 (these two are equivalents). The effect of this call is a kind of lock: if you call is a write lock is put on the registers, so other processes trying to update the database will be blocked. If there is already an RW (Read-Write) transaction opened by another process, the I<begin_trans> call will be blocked.
887 $sess->begin_trans(TRANS_RO),
889 if you would like to put on a "read lock". This one is B<deprecated>, as while you have explicitly opened a transaction for read, you can't open another one for update. For example:
891 $sess->begin_trans(TRANS_RO);
892 $sess->begin_tran(TRANS_RW); # invalid, die here
898 $sess->begin_tran(TRANS_RW);
899 $sess->begin_trans(TRANS_RO);
903 is valid, but probably useless. Note again, that for each retrieval call, an RO transaction is opened. I<TRANS_RW> and I<TRANS_RO> are exported by default by IDZebra::Session.pm.
905 For multiple per-record I<updates> it's efficient to start transactions explicitly: otherwise registers (system files, vocabularies, etc..) are updated one by one. After finishing all requested updates, use
907 $stat = $sess->end_trans;
909 The return value is a ZebraTransactionStatus object, containing the following members as a hash reference:
911 $stat->{processed} # Number of records processed
912 $stat->{updated} # Number of records processed
913 $stat->{deleted} # Number of records processed
914 $stat->{inserted} # Number of records processed
915 $stat->{stime} # System time used
916 $stat->{utime} # User time used
918 Normally, if the perl code dies due to some runtime error, or the session is closed, then the API attempts to close all pending transactions.
920 =head1 THE SHADOW REGISTERS
922 The Zebra server supports updating of the index structures. That is, you can add, modify, or remove records from databases managed by Zebra without rebuilding the entire index. Since this process involves modifying structured files with various references between blocks of data in the files, the update process is inherently sensitive to system crashes, or to process interruptions: Anything but a successfully completed update process will leave the register files in an unknown state, and you will essentially have no recourse but to re-index everything, or to restore the register files from a backup medium. Further, while the update process is active, users cannot be allowed to access the system, as the contents of the register files may change unpredictably.
924 You can solve these problems by enabling the shadow register system in Zebra. During the updating procedure, zebraidx will temporarily write changes to the involved files in a set of "shadow files", without modifying the files that are accessed by the active server processes. If the update procedure is interrupted by a system crash or a signal, you simply repeat the procedure - the register files have not been changed or damaged, and the partially written shadow files are automatically deleted before the new updating procedure commences.
926 At the end of the updating procedure (or in a separate operation, if you so desire), the system enters a "commit mode". First, any active server processes are forced to access those blocks that have been changed from the shadow files rather than from the main register files; the unmodified blocks are still accessed at their normal location (the shadow files are not a complete copy of the register files - they only contain those parts that have actually been modified). If the commit process is interrupted at any point during the commit process, the server processes will continue to access the shadow files until you can repeat the commit procedure and complete the writing of data to the main register files. You can perform multiple update operations to the registers before you commit the changes to the system files, or you can execute the commit operation at the end of each update operation. When the commit phase has completed successfully, any running server processes are instructed to switch their operations to the new, operational register, and the temporary shadow files are deleted.
928 By default, (in the API !) the use of shadow registers is disabled. If zebra is configured that way (there is a "shadow" entry in zebra.cfg), then the shadow system can be enabled by calling:
936 If shadow system is enabled, then you have to commit changes you did, by calling:
940 Note, that you can also determine shadow usage in the session constructor:
942 $sess = IDZebra::Session->open(configFile => 'demo/zebra.cfg',
945 Changes to I<shadow> will not have effect, within a I<transaction> (ie.: a transaction is started either with shadow enabled or disabled). For more details, read Zebra documentation: I<Safe Updating - Using Shadow Registers>.
949 There are two ways to update data in a Zebra database using the perl API. You can update an entire directory structure just the way it's done by zebraidx:
951 $sess->update(path => 'lib');
953 This will update the database with the files in directory "lib", according to the current record group settings.
957 This will update the database with the files, specified by the default record group setting. I<path> has to be specified there...
959 $sess->update(groupName => 'demo1',
962 Update the database with files in "lib" according to the settings of group "demo1"
964 $sess->delete(groupName => 'demo1',
967 Delete the records derived from the files in directory "lib", according to the "demo1" group settings. Sounds complex? Read zebra documentation about identifying records.
969 You can also update records one by one, even directly from the memory:
971 $sysno = $sess->update_record(data => $rec1,
972 recordType => 'grs.perl.pod',
973 groupName => "demo1");
975 This will update the database with the given record buffer. Note, that in this case recordType is explicitly specified, as there is no filename given, and for the demo1 group, no default record type is specified. The return value is the system assigned id of the record.
977 You can also index a single file:
979 $sysno = $sess->update_record(file => "lib/IDZebra/Data1.pm");
981 Or, provide a buffer, and a filename (where filename will only be used to identify the record, if configured that way, and possibly to find out it's record type):
983 $sysno = $sess->update_record(data => $rec1,
984 file => "lib/IDZebra/Data1.pm");
986 And some crazy stuff:
988 $sysno = $sess->delete_record(sysno => $sysno);
990 where sysno in itself is sufficient to identify the record
992 $sysno = $sess->delete_record(data => $rec1,
993 recordType => 'grs.perl.pod',
994 groupName => "demo1");
996 This case the record is extracted, and if already exists, located in the database, then deleted...
998 $sysno = $sess->update_record(data => $rec1,
1000 recordType => 'grs.perl.pod',
1001 groupName => "demo1");
1003 Don't try this at home! This case, the record identifier string (which is normally generated according to the rules set in I<recordId> member of the record group, or in the I<recordId> parameter) is provided directly.... Looks much better this way:
1005 $sysno = $sess->update_record(data => $rec1,
1006 databaseName => 'books',
1007 recordId => '(bib1,ISBN)',
1008 recordType => 'grs.perl.pod',
1010 flagStoreKeys => 1);
1012 You can notice, that it's not necessary to define a record group in zebra.cfg: you can do it "on the fly" in your code.
1014 B<Important:> Note, that one record can be updated only once within a transaction - all subsequent updates are skipped. If you'd like to override this feature, use the I<force=E<gt>1> flag:
1016 $sysno = $sess->update_record(data => $rec1,
1017 recordType => 'grs.perl.pod',
1018 groupName => "demo1",
1021 If you don't like to update the record, if it alerady exists, use the I<insert_record> method:
1023 $sysno = $sess->insert_record(data => $rec1,
1024 recordType => 'grs.perl.pod',
1025 groupName => "demo1");
1027 In this case, sysno will be -1, if the record could not be added, because there was already one in the database, with the same record identifier (generated according to the I<recordId> setting).
1029 =head1 DATABASE SELECTION
1031 Within a zebra repository you can define logical databases. You can either do this by record groups, or by providing the databaseName argument for update methods. For each record the database name it belongs to is stored.
1033 For searching, you can select databases by calling:
1035 $sess->databases('db1','db2');
1037 This will not do anything if the given and only the given databases are already selected. You can get the list of the actually selected databases, by calling:
1039 @dblist = $sess->databases();
1043 It's nice to be able to store data in your repository... But it's useful to reach it as well. So this is how to do searching:
1045 $rs = $sess->search(databases => [qw(demo1,demo2)], # optional
1046 pqf => '@attr 1=4 computer');
1048 This is going to execute a search in databases demo1 and demo2, for title 'com,puter'. This is a PQF (Prefix Query Format) search, see YAZ documentation for details. The database selection is optional: if it's provided, the given list of databases is selected for this particular search, then the original selection is restored.
1050 =head2 CCL searching
1052 Not all users enjoy typing in prefix query structures and numerical attribute values, even in a minimalistic test client. In the library world, the more intuitive Common Command Language (or ISO 8777) has enjoyed some popularity - especially before the widespread availability of graphical interfaces. It is still useful in applications where you for some reason or other need to provide a symbolic language for expressing boolean query structures.
1054 The CCL searching is not currently supported by this API.
1056 =head2 CQL searching
1058 CQL - Common Query Language - was defined for the SRW protocol. In many ways CQL has a similar syntax to CCL. The objective of CQL is different. Where CCL aims to be an end-user language, CQL is the protocol query language for SRW.
1060 In order to map CQL queries to Zebra internal search structures, you have to define a mapping, the way it is described in YAZ documentation: I<Specification of CQL to RPN mapping>. The mapping is interpreted by the method:
1062 $sess->cqlmap($mapfile);
1064 Or, you can directly provide the I<mapfile> parameter for the search:
1066 $rs = $sess->search(cqlmap => 'demo/cql.map',
1067 cql => 'dc.title=IDZebra');
1069 As you see, CQL searching is so simple: just give the query in the I<cql> parameter.
1073 If you'd like the search results to be sorted, use the I<sort> parameter:
1075 $rs = $sess->search(cql => 'IDZebra',
1078 Note, that B<currently> this is (almost) equivalent to
1080 $rs = $sess->search(cql => 'IDZebra');
1081 $rs->sort('1=4 ia');
1083 but in the further versions of Zebra and this API a single phase search and sort will take place, optimizing performance. For more details on sorting, see I<IDZebra::ResultSet> manpage.
1087 As you have seen, the result of the search request is a I<Resultset> object.
1088 It contains number of hits, and search status, and can be used to sort and retrieve the resulting records.
1090 $count = $rs->count;
1092 printf ("RS Status is %d (%s)\n", $rs->errCode, $rs->errString);
1094 I<$rs-E<gt>errCode> is 0, if there were no errors during search. Read the I<IDZebra::Resultset> manpage for more details.
1098 Zebra supports scanning index values. The result of the
1100 $sl = $sess->scan(expression => "a");
1102 call is an I<IDZebra::ScanList> object, what you can use to list the values. The scan expression has to be provided in a PQF like format. Examples:
1104 B< a> (scan trough words of "default", "Any" indexes)
1107 B< @attr 1=1016 a> (same effect)
1110 B< @attr 1=4 @attr 6=2 a> (scan trough titles as phrases)
1112 An illegal scan expression will cause your code to die. If you'd like to select databases just for the scan call, you can optionally use the I<databases> parameter:
1114 $sl = $sess->scan(expression => "a",
1115 databases => [qw(demo1 demo2)]);
1117 You can use the I<IDZebra::ScanList> object returned by the i<scan> method, to reach the result. Check I<IDZebra::ScanList> manpage for more details.
1119 =head1 SESSION STATUS AND ERRORS
1121 Most of the API calls causes die, if an error occures. You avoid this, by using eval {} blocks. The following methods are available to get the status of Zebra service:
1127 The Zebra provided error code... (for the result of the last call);
1131 Error string corresponding to the message
1135 Additional information for the status
1139 This functionality may change, see TODO.
1141 =head1 LOGGING AND MISC. FUNCTIONS
1143 Zebra provides logging facility for the internal events, and also for application developers trough the API. See manpage I<IDZebra::Logger> for details.
1147 =item B<IDZebra::LogFile($filename)>
1149 Will set the output file for logging. By default it's STDERR;
1151 =item B<IDZebra::LogLevel(15)>
1153 Set log level. 0 for no logs. See IDZebra::Logger for usable flags.
1157 Some other functions
1161 =item B<$sess-E<gt>init>
1163 Initialize, and clean registers. This will remove all data!
1165 =item B<$sess-E<gt>compact>
1167 Compact the registers (? does this work)
1169 =item B<$sess-E<gt>show>
1171 Doesn't have too much meaning. Don't try :)
1179 =item B<Clean up error handling>
1181 By default all zebra errors should cause die. (such situations could be avoided by using eval {}), and then check for errCode, errString... An optional flag or package variable should be introduced to override this, and skip zebra errors, to let the user decide what to do.
1183 =item B<Make the package self-distributable>
1185 Build and link with installed header and library files
1189 Test shadow system, unicode...
1193 Cleanup, arrange, remove redundancy
1203 Peter Popovics, pop@technomat.hu
1207 Zebra documentation, Zebra::ResultSet, Zebra::ScanList, Zebra::Logger manpages