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 1.41 2007-08-20 15:34:29 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($self->{EXPLAIN});
99 if (defined($self->{DELETE})) {
100 set_delete_handler($self->{DELETE});
107 # Register packages that we will use in translated RPNs
108 package Net::Z3950::RPN::Node;
109 package Net::Z3950::APDU::Query;
110 our @ISA = qw(Net::Z3950::RPN::Node);
111 package Net::Z3950::APDU::OID;
112 package Net::Z3950::RPN::And;
113 our @ISA = qw(Net::Z3950::RPN::Node);
114 package Net::Z3950::RPN::Or;
115 our @ISA = qw(Net::Z3950::RPN::Node);
116 package Net::Z3950::RPN::AndNot;
117 our @ISA = qw(Net::Z3950::RPN::Node);
118 package Net::Z3950::RPN::Term;
119 our @ISA = qw(Net::Z3950::RPN::Node);
120 package Net::Z3950::RPN::RSID;
121 our @ISA = qw(Net::Z3950::RPN::Node);
122 package Net::Z3950::RPN::Attributes;
123 package Net::Z3950::RPN::Attribute;
126 # Utility method for re-rendering Type-1 query back down to PQF
127 package Net::Z3950::RPN::Node;
131 my $class = ref $this;
133 if ($class eq "Net::Z3950::APDU::Query") {
135 my $set = $this->{attributeSet};
136 $res .= "\@attrset $set " if defined $set;
137 return $res . $this->{query}->toPQF();
138 } elsif ($class eq "Net::Z3950::RPN::Or") {
139 return '@or ' . $this->[0]->toPQF() . ' ' . $this->[1]->toPQF();
140 } elsif ($class eq "Net::Z3950::RPN::And") {
141 return '@and ' . $this->[0]->toPQF() . ' ' . $this->[1]->toPQF();
142 } elsif ($class eq "Net::Z3950::RPN::AndNot") {
143 return '@not ' . $this->[0]->toPQF() . ' ' . $this->[1]->toPQF();
144 } elsif ($class eq "Net::Z3950::RPN::RSID") {
145 return '@set ' . $this->{id};
146 } elsif ($class ne "Net::Z3950::RPN::Term") {
147 die "unknown PQF node-type '$class'";
151 foreach my $attr (@{ $this->{attributes} }) {
153 my $set = $attr->{attributeSet};
154 $res .= "$set " if defined $set;
155 $res .= $attr->{attributeType} . "=" . $attr->{attributeValue} . " ";
158 return $res . $this->{term};
162 # Must revert to original package for Autoloader's benefit
163 package Net::Z3950::SimpleServer;
166 # Autoload methods go after =cut, and are processed by the autosplit program.
170 # Below is the stub of documentation for your module. You better edit it!
174 Net::Z3950::SimpleServer - Simple Perl API for building Z39.50 servers.
178 use Net::Z3950::SimpleServer;
180 sub my_search_handler {
183 my $set_id = $args->{SETNAME};
184 my @database_list = @{ $args->{DATABASES} };
185 my $query = $args->{QUERY};
187 ## Perform the query on the specified set of databases
188 ## and return the number of hits:
190 $args->{HITS} = $hits;
193 sub my_fetch_handler { # Get a record for the user
196 my $set_id = $args->{SETNAME};
198 my $record = fetch_a_record($args->{OFFSET});
200 $args->{RECORD} = $record;
201 if (number_of_hits() == $args->{OFFSET}) { ## Last record in set?
208 ## Register custom event handlers:
209 my $z = new Net::Z3950::SimpleServer(GHANDLE = $someObject,
210 INIT => \&my_init_handler,
211 CLOSE => \&my_close_handler,
212 SEARCH => \&my_search_handler,
213 FETCH => \&my_fetch_handler);
216 $z->launch_server("ztest.pl", @ARGV);
220 The SimpleServer module is a tool for constructing Z39.50 "Information
221 Retrieval" servers in Perl. The module is easy to use, but it
222 does help to have an understanding of the Z39.50 query
223 structure and the construction of structured retrieval records.
225 Z39.50 is a network protocol for searching remote databases and
226 retrieving the results in the form of structured "records". It is widely
227 used in libraries around the world, as well as in the US Federal Government.
228 In addition, it is generally useful whenever you wish to integrate a number
229 of different database systems around a shared, asbtract data model.
231 The model of the module is simple: It implements a "generic" Z39.50
232 server, which invokes callback functions supplied by you to search
233 for content in your database. You can use any tools available in
234 Perl to supply the content, including modules like DBI and
237 The server will take care of managing the network connections for
238 you, and it will spawn a new process (or thread, in some
239 environments) whenever a new connection is received.
241 The programmer can specify subroutines to take care of the following type
247 - Fetching of records
248 - Scan request (browsing)
249 - Closing down connection
251 Note that only the Search and Fetch handler functions are required.
252 The module can supply default responses to the other on its own.
254 After the launching of the server, all control is given away from
255 the Perl script to the server. The server calls the registered
256 subroutines to field incoming requests from Z39.50 clients.
258 A reference to an anonymous hash is passed to each handler. Some of
259 the entries of these hashes are to be considered input and others
262 The Perl programmer specifies the event handlers for the server by
263 means of the SimpleServer object constructor
265 my $z = new Net::Z3950::SimpleServer(
266 INIT => \&my_init_handler,
267 CLOSE => \&my_close_handler,
268 SEARCH => \&my_search_handler,
269 PRESENT => \&my_present_handler,
270 SCAN => \&my_scan_handler,
271 FETCH => \&my_fetch_handler,
272 EXPLAIN => \&my_explain_handler,
273 DELETE => \&my_delete_handler);
275 In addition, the arguments to the constructor may include GHANDLE, a
276 global handle which is made available to each invocation of every
277 callback function. This is typically a reference to either a hash or
280 If you want your SimpleServer to start a thread (threaded mode) to
281 handle each incoming Z39.50 request instead of forking a process
282 (forking mode), you need to register the handlers by symbol rather
283 than by code reference. Thus, in threaded mode, you will need to
284 register your handlers this way:
286 my $z = new Net::Z3950::SimpleServer(
287 INIT => "my_package::my_init_handler",
288 CLOSE => "my_package::my_close_handler",
292 where my_package is the Perl package in which your handler is
295 After the custom event handlers are declared, the server is launched
296 by means of the method
298 $z->launch_server("MyServer.pl", @ARGV);
300 Notice, the first argument should be the name of your server
301 script (for logging purposes), while the rest of the arguments
302 are documented in the YAZ toolkit manual: The section on
303 application invocation: <http://www.indexdata.dk/yaz/yaz-7.php>
305 In particular, you need to use the -T switch to start your SimpleServer
310 The init handler is called whenever a Z39.50 client is attempting
311 to logon to the server. The exchange of parameters between the
312 server and the handler is carried out via an anonymous hash reached
317 The argument hash passed to the init handler has the form
320 ## Response parameters:
322 IMP_ID => "", ## Z39.50 Implementation ID
323 IMP_NAME => "", ## Z39.50 Implementation name
324 IMP_VER => "", ## Z39.50 Implementation version
325 ERR_CODE => 0, ## Error code, cnf. Z39.50 manual
326 ERR_STR => "", ## Error string (additional info.)
327 USER => "xxx" ## If Z39.50 authentication is used,
328 ## this member contains user name
329 PASS => "yyy" ## Under same conditions, this member
330 ## contains the password in clear text
331 GHANDLE => $obj ## Global handler specified at creation
332 HANDLE => undef ## Handler of Perl data structure
335 The HANDLE member can be used to store any scalar value which will then
336 be provided as input to all subsequent calls (ie. for searching, record
337 retrieval, etc.). A common use of the handle is to store a reference to
338 a hash which may then be used to store session-specific parameters.
339 If you have any session-specific information (such as a list of
340 result sets or a handle to a back-end search engine of some sort),
341 it is always best to store them in a private session structure -
342 rather than leaving them in global variables in your script.
344 The Implementation ID, name and version are only really used by Z39.50
345 client developers to see what kind of server they're dealing with.
346 Filling these in is optional.
348 The ERR_CODE should be left at 0 (the default value) if you wish to
349 accept the connection. Any other value is interpreted as a failure
350 and the client will be shown the door, with the code and the
351 associated additional information, ERR_STR returned.
353 =head2 Search handler
355 Similarly, the search handler is called with a reference to an anony-
356 mous hash. The structure is the following:
359 ## Request parameters:
361 GHANDLE => $obj ## Global handler specified at creation
362 HANDLE => ref, ## Your session reference.
363 SETNAME => "id", ## ID of the result set
364 REPL_SET => 0, ## Replace set if already existing?
365 DATABASES => ["xxx"], ## Reference to a list of data-
367 QUERY => "query", ## The query expression
368 RPN => $obj, ## Reference to a Net::Z3950::APDU::Query
370 ## Response parameters:
372 ERR_CODE => 0, ## Error code (0=Succesful search)
373 ERR_STR => "", ## Error string
374 HITS => 0 ## Number of matches
377 Note that a search which finds 0 hits is considered successful in
378 Z39.50 terms - you should only set the ERR_CODE to a non-zero value
379 if there was a problem processing the request. The Z39.50 standard
380 provides a comprehensive list of standard diagnostic codes, and you
381 should use these whenever possible.
383 The QUERY is a tree-structure of terms combined by operators, the
384 terms being qualified by lists of attributes. The query is presented
385 to the search function in the Prefix Query Format (PQF) which is
386 used in many applications based on the YAZ toolkit. The full grammar
387 is described in the YAZ manual.
389 The following are all examples of valid queries in the PQF.
395 @or "dylan" "zimmerman"
399 @or @and bob dylan @set Result-1
401 @and @attr 1=1 "bob dylan" @attr 1=4 "slow train coming"
403 @attrset @attr 4=1 @attr 1=4 "self portrait"
405 You will need to write a recursive function or something similar to
406 parse incoming query expressions, and this is usually where a lot of
407 the work in writing a database-backend happens. Fortunately, you don't
408 need to support anymore functionality than you want to. For instance,
409 it is perfectly legal to not accept boolean operators, but you SHOULD
410 try to return good error codes if you run into something you can't or
413 A more convenient alternative to the QUERY member may be the RPN
414 member, which is a reference to a Net::Z3950::APDU::Query object
415 representing the RPN query tree. The structure of that object is
416 supposed to be self-documenting, but here's a brief summary of what
423 C<Net::Z3950::APDU::Query> is a hash with two fields:
429 =item C<attributeSet>
431 Optional. If present, it is a reference to a
432 C<Net::Z3950::APDU::OID>. This is a string of dot-separated integers
433 representing the OID of the query's top-level attribute set.
437 Mandatory: a refererence to the RPN tree itself.
443 Each node of the tree is an object of one of the following types:
449 =item C<Net::Z3950::RPN::And>
451 =item C<Net::Z3950::RPN::Or>
453 =item C<Net::Z3950::RPN::AndNot>
455 These three classes are all arrays of two elements, each of which is a
456 node of one of the above types.
458 =item C<Net::Z3950::RPN::Term>
460 See below for details.
462 =item C<Net::Z3950::RPN::RSID>
464 A reference to a result-set ID indicating a previous search. The ID
465 of the result-set is in the C<id> element.
475 C<Net::Z3950::RPN::Term> is a hash with two fields:
483 A string containing the search term itself.
487 A reference to a C<Net::Z3950::RPN::Attributes> object.
493 C<Net::Z3950::RPN::Attributes> is an array of references to
494 C<Net::Z3950::RPN::Attribute> objects. (Note the plural/singular
499 C<Net::Z3950::RPN::Attribute> is a hash with three elements:
505 =item C<attributeSet>
507 Optional. If present, it is dot-separated OID string, as above.
509 =item C<attributeType>
511 An integer indicating the type of the attribute - for example, under
512 the BIB-1 attribute set, type 1 indicates a ``use'' attribute, type 2
513 a ``relation'' attribute, etc.
515 =item C<attributeValue>
517 An integer or string indicating the value of the attribute - for example, under
518 BIB-1, if the attribute type is 1, then value 4 indictates a title
519 search and 7 indictates an ISBN search; but if the attribute type is
520 2, then value 4 indicates a ``greater than or equal'' search, and 102
521 indicates a relevance match.
527 All of these classes except C<Attributes> and C<Attribute> are
528 subclasses of the abstract class C<Net::Z3950::RPN::Node>. That class
529 has a single method, C<toPQF()>, which may be used to turn an RPN
530 tree, or part of one, back into a textual prefix query.
532 Note that, apart to C<toPQF()>, none of these classes have any methods at
533 all: the blessing into classes is largely just a documentation thing
534 so that, for example, if you do
536 { use Data::Dumper; print Dumper($args->{RPN}) }
538 you get something fairly human-readable. But of course, the type
539 distinction between the three different kinds of boolean node is
542 By adding your own methods to these classes (building what I call
543 ``augmented classes''), you can easily build code that walks the tree
544 of the incoming RPN. Take a look at C<samples/render-search.pl> for a
545 sample implementation of such an augmented classes technique.
548 =head2 Present handler
550 The presence of a present handler in a SimpleServer front-end is optional.
551 Each time a client wishes to retrieve records, the present service is
552 called. The present service allows the origin to request a certain number
553 of records retrieved from a given result set.
554 When the present handler is called, the front-end server should prepare a
555 result set for fetching. In practice, this means to get access to the
556 data from the backend database and store the data in a temporary fashion
557 for fast and efficient fetching. The present handler does *not* fetch
558 anything. This task is taken care of by the fetch handler, which will be
559 called the correct number of times by the YAZ library. More about this
561 If no present handler is implemented in the front-end, the YAZ toolkit
562 will take care of a minimum of preparations itself. This default present
563 handler is sufficient in many situations, where only a small amount of
564 records are expected to be retrieved. If on the other hand, large result
565 sets are likely to occur, the implementation of a reasonable present
566 handler can gain performance significantly.
568 The informations exchanged between client and present handle are:
571 ## Client/server request:
573 GHANDLE => $obj ## Global handler specified at creation
574 HANDLE => ref, ## Reference to datastructure
575 SETNAME => "id", ## Result set ID
576 START => xxx, ## Start position
577 COMP => "", ## Desired record composition
578 NUMBER => yyy, ## Number of requested records
581 ## Respons parameters:
583 HITS => zzz, ## Number of returned records
584 ERR_CODE => 0, ## Error code
585 ERR_STR => "" ## Error message
591 The fetch handler is asked to retrieve a SINGLE record from a given
592 result set (the front-end server will automatically call the fetch
593 handler as many times as required).
595 The parameters exchanged between the server and the fetch handler are
598 ## Client/server request:
600 GHANDLE => $obj ## Global handler specified at creation
601 HANDLE => ref ## Reference to data structure
602 SETNAME => "id" ## ID of the requested result set
603 OFFSET => nnn ## Record offset number
604 REQ_FORM => "n.m.k.l"## Client requested format OID
605 COMP => "xyz" ## Formatting instructions
606 SCHEMA => "abc" ## Requested schema, if any
610 RECORD => "" ## Record string
611 BASENAME => "" ## Origin of returned record
612 LAST => 0 ## Last record in set?
613 ERR_CODE => 0 ## Error code
614 ERR_STR => "" ## Error string
615 SUR_FLAG => 0 ## Surrogate diagnostic flag
616 REP_FORM => "n.m.k.l"## Provided format OID
617 SCHEMA => "abc" ## Provided schema, if any
620 The REP_FORM value has by default the REQ_FORM value but can be set to
621 something different if the handler desires. The BASENAME value should
622 contain the name of the database from where the returned record originates.
623 The ERR_CODE and ERR_STR works the same way they do in the search
624 handler. If there is an error condition, the SUR_FLAG is used to
625 indicate whether the error condition pertains to the record currently
626 being retrieved, or whether it pertains to the operation as a whole
627 (eg. the client has specified a result set which does not exist.)
629 If you need to return USMARC records, you might want to have a look at
630 the MARC module on CPAN, if you don't already have a way of generating
633 NOTE: The record offset is 1-indexed - 1 is the offset of the first
638 A full featured Z39.50 server should support scan (or in some literature
639 browse). The client specifies a starting term of the scan, and the server
640 should return an ordered list of specified length consisting of terms
641 actually occurring in the data base. Each of these terms should be close
642 to or equal to the term originally specified. The quality of scan compared
643 to simple search is a guarantee of hits. It is simply like browsing through
644 an index of a book, you always find something! The parameters exchanged are
649 GHANDLE => $obj, ## Global handler specified at creation
650 HANDLE => $ref, ## Reference to data structure
651 DATABASES => ["xxx"], ## Reference to a list of data-
653 TERM => 'start', ## The start term
654 RPN => $obj, ## Reference to a Net::Z3950::RPN::Term
656 NUMBER => xx, ## Number of requested terms
657 POS => yy, ## Position of starting point
658 ## within returned list
659 STEP => 0, ## Step size
663 ERR_CODE => 0, ## Error code
664 ERR_STR => '', ## Diagnostic message
665 NUMBER => zz, ## Number of returned terms
666 STATUS => $status, ## ScanSuccess/ScanFailure
667 ENTRIES => $entries ## Referenced list of terms
670 where the term list is returned by reference in the scalar $entries, which
671 should point at a data structure of this kind,
677 { TERM => 'energy density',
680 { TERM => 'energy flow',
688 The $status flag is only meaningful after a successful scan, and
689 should be assigned one of two values:
691 Net::Z3950::SimpleServer::ScanSuccess Full success (default)
692 Net::Z3950::SimpleServer::ScanPartial Fewer terms returned than requested
694 The STEP member contains the requested number of entries in the term-list
695 between two adjacent entries in the response.
697 A better alternative to the TERM member is the the RPN
698 member, which is a reference to a Net::Z3950::RPN::Term object
699 representing the scan cloause. The structure of that object is the
700 same as for Term objects included as part of the RPN tree passed to
701 search handlers. This is more useful than the simple TERM because it
702 includes attributes (e.g. access points associated with the term),
703 which are discarded by the TERM element.
707 The argument hash recieved by the close handler has two elements only:
712 GHANDLE => $obj ## Global handler specified at creation
713 HANDLE => ref ## Reference to data structure
716 What ever data structure the HANDLE value points at goes out of scope
717 after this call. If you need to close down a connection to your server
718 or something similar, this is the place to do it.
720 =head2 Delete handler
722 The argument hash recieved by the delete handler has the following elements:
726 GHANDLE => $obj, ## Global handler specified at creation
727 HANDLE => ref, ## Reference to data structure
728 SETNAME => "id", ## Result set ID
731 STATUS => 0 ## Deletion status
734 The SETNAME element of the argument hash may or may not be defined.
735 If it is, then SETNAME is the name of a result set to be deleted; if
736 not, then all result-sets associated with the current session should
737 be deleted. In either case, the callback function should report on
738 success or failure by setting the STATUS element either to zero, on
739 success, or to an integer from 1 to 10, to indicate one of the ten
740 possible failure codes described in section 3.2.4.1.4 of the Z39.50
742 http://www.loc.gov/z3950/agency/markup/05.html#Delete-list-statuses1
744 =head2 Support for SRU and SRW
746 Since release 1.0, SimpleServer includes support for serving the SRU
747 and SRW protocols as well as Z39.50. These ``web-friendly'' protocols
748 enable similar functionality to that of Z39.50, but by means of rich
749 URLs in the case of SRU, and a SOAP-based web-service in the case of
750 SRW. These protocols are described at
751 http://www.loc.gov/sru
753 In order to serve these protocols from a SimpleServer-based
754 application, it is necessary to launch the application with a YAZ
755 Generic Frontend Server (GFS) configuration file, which can be
756 specified using the command-line argument C<-f> I<filename>. A
757 minimal configuration file looks like this:
761 <cql2rpn>pqf.properties</cql2rpn>
765 This file specifies only that C<pqf.properties> should be used to
766 translate the CQL queries of SRU and SRW into corresponding Z39.50
767 Type-1 queries. For more information about YAZ GFS configuration,
768 including how to specify an Explain record, see the I<Virtual Hosts>
769 section of the YAZ manual at
770 http://indexdata.com/yaz/doc/server.vhosts.tkl
772 The mapping of CQL queries into Z39.50 Type-1 queries is specified by
773 a file that indicates which BIB-1 attributes should be generated for
774 each CQL index, relation, modifiers, etc. A typical section of this
775 file looks like this:
778 index.dc.subject = 1=21
779 index.dc.creator = 1=1003
783 This file specifies the BIB-1 access points (type=1) for the Dublin
784 Core indexes C<title>, C<subject> and C<creator>, and the BIB-1
785 relations (type=2) corresponding to the CQL relations C<E<lt>> and
786 C<E<lt>=>. For more information about the format of this file, see
787 the I<CQL> section of the YAZ manual at
788 http://indexdata.com/yaz/doc/tools.tkl#tools.cql
790 The YAZ distribution include a sample CQL-to-PQF mapping configuration
791 file called C<pqf.properties>; this is sufficient for many
792 applications, and a good base to work from for most others.
794 If a SimpleServer-based application is run without this SRU-specific
795 configuration, it can still serve SRU; however, CQL queries will not
796 be translated, but passed straight through to the search-handler
797 function, as the C<CQL> member of the parameters hash. It is then the
798 responsibility of the back-end application to parse and handle the CQL
799 query, which is most easily done using Ed Summers' fine C<CQL::Parser>
800 module, available from CPAN at
801 http://search.cpan.org/~esummers/CQL-Parser/
805 Anders Sønderberg (sondberg@indexdata.dk),
806 Sebastian Hammer (quinn@indexdata.dk),
807 Mike Taylor (indexdata.com).
811 Any Perl module which is useful for accessing the database of your