2 ## Copyright (c) 2000-2006, Index Data.
4 ## Permission to use, copy, modify, distribute, and sell this software and
5 ## its documentation, in whole or in part, for any purpose, is hereby granted,
8 ## 1. This copyright and permission notice appear in all copies of the
9 ## software and its documentation. Notices of copyright or attribution
10 ## which appear at the beginning of any file must remain unchanged.
12 ## 2. The name of Index Data or the individual authors may not be used to
13 ## endorse or promote products derived from this software without specific
14 ## prior written permission.
16 ## THIS SOFTWARE IS PROVIDED "AS IS" AND WITHOUT WARRANTY OF ANY KIND,
17 ## EXPRESS, IMPLIED, OR OTHERWISE, INCLUDING WITHOUT LIMITATION, ANY
18 ## WARRANTY OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
19 ## IN NO EVENT SHALL INDEX DATA BE LIABLE FOR ANY SPECIAL, INCIDENTAL,
20 ## INDIRECT OR CONSEQUENTIAL DAMAGES OF ANY KIND, OR ANY DAMAGES
21 ## WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER OR
22 ## NOT ADVISED OF THE POSSIBILITY OF DAMAGE, AND ON ANY THEORY OF
23 ## LIABILITY, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE
28 ## $Id: SimpleServer.pm,v 1.40 2007-08-20 11:06:09 mike Exp $
30 package Net::Z3950::SimpleServer;
33 use vars qw($VERSION @ISA @EXPORT @EXPORT_OK);
40 @ISA = qw(Exporter AutoLoader DynaLoader);
44 bootstrap Net::Z3950::SimpleServer $VERSION;
46 # Preloaded methods go here.
56 carp "SimpleServer.pm: WARNING: Multithreaded server unsupported";
60 croak "SimpleServer.pm: ERROR: Unspecified search handler" unless defined($self->{SEARCH});
61 croak "SimpleServer.pm: ERROR: Unspecified fetch handler" unless defined($self->{FETCH});
72 ### This modal internal interface, in which we set a bunch of
73 # globals and then call start_server(), is asking for
74 # trouble. Instead, we should just pass the $self object
75 # as a parameter into start_server().
76 if (defined($self->{GHANDLE})) {
77 set_ghandle($self->{GHANDLE});
79 if (defined($self->{INIT})) {
80 set_init_handler($self->{INIT});
82 set_search_handler($self->{SEARCH});
83 set_fetch_handler($self->{FETCH});
84 if (defined($self->{CLOSE})) {
85 set_close_handler($self->{CLOSE});
87 if (defined($self->{PRESENT})) {
88 set_present_handler($self->{PRESENT});
90 if (defined($self->{SCAN})) {
91 set_scan_handler($self->{SCAN});
93 if (defined($self->{SORT})) {
94 set_sort_handler($self->{SORT});
96 if (defined($self->{EXPLAIN})) {
97 set_explain_handler($self->{EXPLAIN});
104 # Register packages that we will use in translated RPNs
105 package Net::Z3950::RPN::Node;
106 package Net::Z3950::APDU::Query;
107 our @ISA = qw(Net::Z3950::RPN::Node);
108 package Net::Z3950::APDU::OID;
109 package Net::Z3950::RPN::And;
110 our @ISA = qw(Net::Z3950::RPN::Node);
111 package Net::Z3950::RPN::Or;
112 our @ISA = qw(Net::Z3950::RPN::Node);
113 package Net::Z3950::RPN::AndNot;
114 our @ISA = qw(Net::Z3950::RPN::Node);
115 package Net::Z3950::RPN::Term;
116 our @ISA = qw(Net::Z3950::RPN::Node);
117 package Net::Z3950::RPN::RSID;
118 our @ISA = qw(Net::Z3950::RPN::Node);
119 package Net::Z3950::RPN::Attributes;
120 package Net::Z3950::RPN::Attribute;
123 # Utility method for re-rendering Type-1 query back down to PQF
124 package Net::Z3950::RPN::Node;
128 my $class = ref $this;
130 if ($class eq "Net::Z3950::APDU::Query") {
132 my $set = $this->{attributeSet};
133 $res .= "\@attrset $set " if defined $set;
134 return $res . $this->{query}->toPQF();
135 } elsif ($class eq "Net::Z3950::RPN::Or") {
136 return '@or ' . $this->[0]->toPQF() . ' ' . $this->[1]->toPQF();
137 } elsif ($class eq "Net::Z3950::RPN::And") {
138 return '@and ' . $this->[0]->toPQF() . ' ' . $this->[1]->toPQF();
139 } elsif ($class eq "Net::Z3950::RPN::AndNot") {
140 return '@not ' . $this->[0]->toPQF() . ' ' . $this->[1]->toPQF();
141 } elsif ($class eq "Net::Z3950::RPN::RSID") {
142 return '@set ' . $this->{id};
143 } elsif ($class ne "Net::Z3950::RPN::Term") {
144 die "unknown PQF node-type '$class'";
148 foreach my $attr (@{ $this->{attributes} }) {
150 my $set = $attr->{attributeSet};
151 $res .= "$set " if defined $set;
152 $res .= $attr->{attributeType} . "=" . $attr->{attributeValue} . " ";
155 return $res . $this->{term};
159 # Must revert to original package for Autoloader's benefit
160 package Net::Z3950::SimpleServer;
163 # Autoload methods go after =cut, and are processed by the autosplit program.
167 # Below is the stub of documentation for your module. You better edit it!
171 Net::Z3950::SimpleServer - Simple Perl API for building Z39.50 servers.
175 use Net::Z3950::SimpleServer;
177 sub my_search_handler {
180 my $set_id = $args->{SETNAME};
181 my @database_list = @{ $args->{DATABASES} };
182 my $query = $args->{QUERY};
184 ## Perform the query on the specified set of databases
185 ## and return the number of hits:
187 $args->{HITS} = $hits;
190 sub my_fetch_handler { # Get a record for the user
193 my $set_id = $args->{SETNAME};
195 my $record = fetch_a_record($args->{OFFSET});
197 $args->{RECORD} = $record;
198 if (number_of_hits() == $args->{OFFSET}) { ## Last record in set?
205 ## Register custom event handlers:
206 my $z = new Net::Z3950::SimpleServer(GHANDLE = $someObject,
207 INIT => \&my_init_handler,
208 CLOSE => \&my_close_handler,
209 SEARCH => \&my_search_handler,
210 FETCH => \&my_fetch_handler);
213 $z->launch_server("ztest.pl", @ARGV);
217 The SimpleServer module is a tool for constructing Z39.50 "Information
218 Retrieval" servers in Perl. The module is easy to use, but it
219 does help to have an understanding of the Z39.50 query
220 structure and the construction of structured retrieval records.
222 Z39.50 is a network protocol for searching remote databases and
223 retrieving the results in the form of structured "records". It is widely
224 used in libraries around the world, as well as in the US Federal Government.
225 In addition, it is generally useful whenever you wish to integrate a number
226 of different database systems around a shared, asbtract data model.
228 The model of the module is simple: It implements a "generic" Z39.50
229 server, which invokes callback functions supplied by you to search
230 for content in your database. You can use any tools available in
231 Perl to supply the content, including modules like DBI and
234 The server will take care of managing the network connections for
235 you, and it will spawn a new process (or thread, in some
236 environments) whenever a new connection is received.
238 The programmer can specify subroutines to take care of the following type
244 - Fetching of records
245 - Scan request (browsing)
246 - Closing down connection
248 Note that only the Search and Fetch handler functions are required.
249 The module can supply default responses to the other on its own.
251 After the launching of the server, all control is given away from
252 the Perl script to the server. The server calls the registered
253 subroutines to field incoming requests from Z39.50 clients.
255 A reference to an anonymous hash is passed to each handler. Some of
256 the entries of these hashes are to be considered input and others
259 The Perl programmer specifies the event handlers for the server by
260 means of the SimpleServer object constructor
262 my $z = new Net::Z3950::SimpleServer(
263 INIT => \&my_init_handler,
264 CLOSE => \&my_close_handler,
265 SEARCH => \&my_search_handler,
266 PRESENT => \&my_present_handler,
267 SCAN => \&my_scan_handler,
268 FETCH => \&my_fetch_handler,
269 EXPLAIN => \&my_explain_handler);
271 In addition, the arguments to the constructor may include GHANDLE, a
272 global handle which is made available to each invocation of every
273 callback function. This is typically a reference to either a hash or
276 If you want your SimpleServer to start a thread (threaded mode) to
277 handle each incoming Z39.50 request instead of forking a process
278 (forking mode), you need to register the handlers by symbol rather
279 than by code reference. Thus, in threaded mode, you will need to
280 register your handlers this way:
282 my $z = new Net::Z3950::SimpleServer(
283 INIT => "my_package::my_init_handler",
284 CLOSE => "my_package::my_close_handler",
288 where my_package is the Perl package in which your handler is
291 After the custom event handlers are declared, the server is launched
292 by means of the method
294 $z->launch_server("MyServer.pl", @ARGV);
296 Notice, the first argument should be the name of your server
297 script (for logging purposes), while the rest of the arguments
298 are documented in the YAZ toolkit manual: The section on
299 application invocation: <http://www.indexdata.dk/yaz/yaz-7.php>
301 In particular, you need to use the -T switch to start your SimpleServer
306 The init handler is called whenever a Z39.50 client is attempting
307 to logon to the server. The exchange of parameters between the
308 server and the handler is carried out via an anonymous hash reached
313 The argument hash passed to the init handler has the form
316 ## Response parameters:
318 IMP_ID => "", ## Z39.50 Implementation ID
319 IMP_NAME => "", ## Z39.50 Implementation name
320 IMP_VER => "", ## Z39.50 Implementation version
321 ERR_CODE => 0, ## Error code, cnf. Z39.50 manual
322 ERR_STR => "", ## Error string (additional info.)
323 USER => "xxx" ## If Z39.50 authentication is used,
324 ## this member contains user name
325 PASS => "yyy" ## Under same conditions, this member
326 ## contains the password in clear text
327 GHANDLE => $obj ## Global handler specified at creation
328 HANDLE => undef ## Handler of Perl data structure
331 The HANDLE member can be used to store any scalar value which will then
332 be provided as input to all subsequent calls (ie. for searching, record
333 retrieval, etc.). A common use of the handle is to store a reference to
334 a hash which may then be used to store session-specific parameters.
335 If you have any session-specific information (such as a list of
336 result sets or a handle to a back-end search engine of some sort),
337 it is always best to store them in a private session structure -
338 rather than leaving them in global variables in your script.
340 The Implementation ID, name and version are only really used by Z39.50
341 client developers to see what kind of server they're dealing with.
342 Filling these in is optional.
344 The ERR_CODE should be left at 0 (the default value) if you wish to
345 accept the connection. Any other value is interpreted as a failure
346 and the client will be shown the door, with the code and the
347 associated additional information, ERR_STR returned.
349 =head2 Search handler
351 Similarly, the search handler is called with a reference to an anony-
352 mous hash. The structure is the following:
355 ## Request parameters:
357 GHANDLE => $obj ## Global handler specified at creation
358 HANDLE => ref, ## Your session reference.
359 SETNAME => "id", ## ID of the result set
360 REPL_SET => 0, ## Replace set if already existing?
361 DATABASES => ["xxx"], ## Reference to a list of data-
363 QUERY => "query", ## The query expression
364 RPN => $obj, ## Reference to a Net::Z3950::APDU::Query
366 ## Response parameters:
368 ERR_CODE => 0, ## Error code (0=Succesful search)
369 ERR_STR => "", ## Error string
370 HITS => 0 ## Number of matches
373 Note that a search which finds 0 hits is considered successful in
374 Z39.50 terms - you should only set the ERR_CODE to a non-zero value
375 if there was a problem processing the request. The Z39.50 standard
376 provides a comprehensive list of standard diagnostic codes, and you
377 should use these whenever possible.
379 The QUERY is a tree-structure of terms combined by operators, the
380 terms being qualified by lists of attributes. The query is presented
381 to the search function in the Prefix Query Format (PQF) which is
382 used in many applications based on the YAZ toolkit. The full grammar
383 is described in the YAZ manual.
385 The following are all examples of valid queries in the PQF.
391 @or "dylan" "zimmerman"
395 @or @and bob dylan @set Result-1
397 @and @attr 1=1 "bob dylan" @attr 1=4 "slow train coming"
399 @attrset @attr 4=1 @attr 1=4 "self portrait"
401 You will need to write a recursive function or something similar to
402 parse incoming query expressions, and this is usually where a lot of
403 the work in writing a database-backend happens. Fortunately, you don't
404 need to support anymore functionality than you want to. For instance,
405 it is perfectly legal to not accept boolean operators, but you SHOULD
406 try to return good error codes if you run into something you can't or
409 A more convenient alternative to the QUERY member may be the RPN
410 member, which is a reference to a Net::Z3950::APDU::Query object
411 representing the RPN query tree. The structure of that object is
412 supposed to be self-documenting, but here's a brief summary of what
419 C<Net::Z3950::APDU::Query> is a hash with two fields:
425 =item C<attributeSet>
427 Optional. If present, it is a reference to a
428 C<Net::Z3950::APDU::OID>. This is a string of dot-separated integers
429 representing the OID of the query's top-level attribute set.
433 Mandatory: a refererence to the RPN tree itself.
439 Each node of the tree is an object of one of the following types:
445 =item C<Net::Z3950::RPN::And>
447 =item C<Net::Z3950::RPN::Or>
449 =item C<Net::Z3950::RPN::AndNot>
451 These three classes are all arrays of two elements, each of which is a
452 node of one of the above types.
454 =item C<Net::Z3950::RPN::Term>
456 See below for details.
458 =item C<Net::Z3950::RPN::RSID>
460 A reference to a result-set ID indicating a previous search. The ID
461 of the result-set is in the C<id> element.
471 C<Net::Z3950::RPN::Term> is a hash with two fields:
479 A string containing the search term itself.
483 A reference to a C<Net::Z3950::RPN::Attributes> object.
489 C<Net::Z3950::RPN::Attributes> is an array of references to
490 C<Net::Z3950::RPN::Attribute> objects. (Note the plural/singular
495 C<Net::Z3950::RPN::Attribute> is a hash with three elements:
501 =item C<attributeSet>
503 Optional. If present, it is dot-separated OID string, as above.
505 =item C<attributeType>
507 An integer indicating the type of the attribute - for example, under
508 the BIB-1 attribute set, type 1 indicates a ``use'' attribute, type 2
509 a ``relation'' attribute, etc.
511 =item C<attributeValue>
513 An integer or string indicating the value of the attribute - for example, under
514 BIB-1, if the attribute type is 1, then value 4 indictates a title
515 search and 7 indictates an ISBN search; but if the attribute type is
516 2, then value 4 indicates a ``greater than or equal'' search, and 102
517 indicates a relevance match.
523 All of these classes except C<Attributes> and C<Attribute> are
524 subclasses of the abstract class C<Net::Z3950::RPN::Node>. That class
525 has a single method, C<toPQF()>, which may be used to turn an RPN
526 tree, or part of one, back into a textual prefix query.
528 Note that, apart to C<toPQF()>, none of these classes have any methods at
529 all: the blessing into classes is largely just a documentation thing
530 so that, for example, if you do
532 { use Data::Dumper; print Dumper($args->{RPN}) }
534 you get something fairly human-readable. But of course, the type
535 distinction between the three different kinds of boolean node is
538 By adding your own methods to these classes (building what I call
539 ``augmented classes''), you can easily build code that walks the tree
540 of the incoming RPN. Take a look at C<samples/render-search.pl> for a
541 sample implementation of such an augmented classes technique.
544 =head2 Present handler
546 The presence of a present handler in a SimpleServer front-end is optional.
547 Each time a client wishes to retrieve records, the present service is
548 called. The present service allows the origin to request a certain number
549 of records retrieved from a given result set.
550 When the present handler is called, the front-end server should prepare a
551 result set for fetching. In practice, this means to get access to the
552 data from the backend database and store the data in a temporary fashion
553 for fast and efficient fetching. The present handler does *not* fetch
554 anything. This task is taken care of by the fetch handler, which will be
555 called the correct number of times by the YAZ library. More about this
557 If no present handler is implemented in the front-end, the YAZ toolkit
558 will take care of a minimum of preparations itself. This default present
559 handler is sufficient in many situations, where only a small amount of
560 records are expected to be retrieved. If on the other hand, large result
561 sets are likely to occur, the implementation of a reasonable present
562 handler can gain performance significantly.
564 The informations exchanged between client and present handle are:
567 ## Client/server request:
569 GHANDLE => $obj ## Global handler specified at creation
570 HANDLE => ref, ## Reference to datastructure
571 SETNAME => "id", ## Result set ID
572 START => xxx, ## Start position
573 COMP => "", ## Desired record composition
574 NUMBER => yyy, ## Number of requested records
577 ## Respons parameters:
579 HITS => zzz, ## Number of returned records
580 ERR_CODE => 0, ## Error code
581 ERR_STR => "" ## Error message
587 The fetch handler is asked to retrieve a SINGLE record from a given
588 result set (the front-end server will automatically call the fetch
589 handler as many times as required).
591 The parameters exchanged between the server and the fetch handler are
594 ## Client/server request:
596 GHANDLE => $obj ## Global handler specified at creation
597 HANDLE => ref ## Reference to data structure
598 SETNAME => "id" ## ID of the requested result set
599 OFFSET => nnn ## Record offset number
600 REQ_FORM => "n.m.k.l"## Client requested format OID
601 COMP => "xyz" ## Formatting instructions
602 SCHEMA => "abc" ## Requested schema, if any
606 RECORD => "" ## Record string
607 BASENAME => "" ## Origin of returned record
608 LAST => 0 ## Last record in set?
609 ERR_CODE => 0 ## Error code
610 ERR_STR => "" ## Error string
611 SUR_FLAG => 0 ## Surrogate diagnostic flag
612 REP_FORM => "n.m.k.l"## Provided format OID
613 SCHEMA => "abc" ## Provided schema, if any
616 The REP_FORM value has by default the REQ_FORM value but can be set to
617 something different if the handler desires. The BASENAME value should
618 contain the name of the database from where the returned record originates.
619 The ERR_CODE and ERR_STR works the same way they do in the search
620 handler. If there is an error condition, the SUR_FLAG is used to
621 indicate whether the error condition pertains to the record currently
622 being retrieved, or whether it pertains to the operation as a whole
623 (eg. the client has specified a result set which does not exist.)
625 If you need to return USMARC records, you might want to have a look at
626 the MARC module on CPAN, if you don't already have a way of generating
629 NOTE: The record offset is 1-indexed - 1 is the offset of the first
634 A full featured Z39.50 server should support scan (or in some literature
635 browse). The client specifies a starting term of the scan, and the server
636 should return an ordered list of specified length consisting of terms
637 actually occurring in the data base. Each of these terms should be close
638 to or equal to the term originally specified. The quality of scan compared
639 to simple search is a guarantee of hits. It is simply like browsing through
640 an index of a book, you always find something! The parameters exchanged are
645 GHANDLE => $obj, ## Global handler specified at creation
646 HANDLE => $ref, ## Reference to data structure
647 DATABASES => ["xxx"], ## Reference to a list of data-
649 TERM => 'start', ## The start term
650 RPN => $obj, ## Reference to a Net::Z3950::RPN::Term
652 NUMBER => xx, ## Number of requested terms
653 POS => yy, ## Position of starting point
654 ## within returned list
655 STEP => 0, ## Step size
659 ERR_CODE => 0, ## Error code
660 ERR_STR => '', ## Diagnostic message
661 NUMBER => zz, ## Number of returned terms
662 STATUS => $status, ## ScanSuccess/ScanFailure
663 ENTRIES => $entries ## Referenced list of terms
666 where the term list is returned by reference in the scalar $entries, which
667 should point at a data structure of this kind,
673 { TERM => 'energy density',
676 { TERM => 'energy flow',
684 The $status flag is only meaningful after a successful scan, and
685 should be assigned one of two values:
687 Net::Z3950::SimpleServer::ScanSuccess Full success (default)
688 Net::Z3950::SimpleServer::ScanPartial Fewer terms returned than requested
690 The STEP member contains the requested number of entries in the term-list
691 between two adjacent entries in the response.
693 A better alternative to the TERM member is the the RPN
694 member, which is a reference to a Net::Z3950::RPN::Term object
695 representing the scan cloause. The structure of that object is the
696 same as for Term objects included as part of the RPN tree passed to
697 search handlers. This is more useful than the simple TERM because it
698 includes attributes (e.g. access points associated with the term),
699 which are discarded by the TERM element.
703 The argument hash recieved by the close handler has two elements only:
708 GHANDLE => $obj ## Global handler specified at creation
709 HANDLE => ref ## Reference to data structure
712 What ever data structure the HANDLE value points at goes out of scope
713 after this call. If you need to close down a connection to your server
714 or something similar, this is the place to do it.
716 =head2 Support for SRU and SRW
718 Since release 1.0, SimpleServer includes support for serving the SRU
719 and SRW protocols as well as Z39.50. These ``web-friendly'' protocols
720 enable similar functionality to that of Z39.50, but by means of rich
721 URLs in the case of SRU, and a SOAP-based web-service in the case of
722 SRW. These protocols are described at
723 http://www.loc.gov/sru
725 In order to serve these protocols from a SimpleServer-based
726 application, it is necessary to launch the application with a YAZ
727 Generic Frontend Server (GFS) configuration file, which can be
728 specified using the command-line argument C<-f> I<filename>. A
729 minimal configuration file looks like this:
733 <cql2rpn>pqf.properties</cql2rpn>
737 This file specifies only that C<pqf.properties> should be used to
738 translate the CQL queries of SRU and SRW into corresponding Z39.50
739 Type-1 queries. For more information about YAZ GFS configuration,
740 including how to specify an Explain record, see the I<Virtual Hosts>
741 section of the YAZ manual at
742 http://indexdata.com/yaz/doc/server.vhosts.tkl
744 The mapping of CQL queries into Z39.50 Type-1 queries is specified by
745 a file that indicates which BIB-1 attributes should be generated for
746 each CQL index, relation, modifiers, etc. A typical section of this
747 file looks like this:
750 index.dc.subject = 1=21
751 index.dc.creator = 1=1003
755 This file specifies the BIB-1 access points (type=1) for the Dublin
756 Core indexes C<title>, C<subject> and C<creator>, and the BIB-1
757 relations (type=2) corresponding to the CQL relations C<E<lt>> and
758 C<E<lt>=>. For more information about the format of this file, see
759 the I<CQL> section of the YAZ manual at
760 http://indexdata.com/yaz/doc/tools.tkl#tools.cql
762 The YAZ distribution include a sample CQL-to-PQF mapping configuration
763 file called C<pqf.properties>; this is sufficient for many
764 applications, and a good base to work from for most others.
766 If a SimpleServer-based application is run without this SRU-specific
767 configuration, it can still serve SRU; however, CQL queries will not
768 be translated, but passed straight through to the search-handler
769 function, as the C<CQL> member of the parameters hash. It is then the
770 responsibility of the back-end application to parse and handle the CQL
771 query, which is most easily done using Ed Summers' fine C<CQL::Parser>
772 module, available from CPAN at
773 http://search.cpan.org/~esummers/CQL-Parser/
777 Anders Sønderberg (sondberg@indexdata.dk),
778 Sebastian Hammer (quinn@indexdata.dk),
779 Mike Taylor (indexdata.com).
783 Any Perl module which is useful for accessing the database of your