ZOOM: The Z39.50 Object-Orientation Model, v1.3

1. Introduction: What is ZOOM?

The ZOOM initiative presents an abstract object-oriented API to a subset of the services specified by the Z39.50 standard, also known as ISO 23950 (see http://lcweb.loc.gov/z3950/agency/document.html for a free, downloadable copy of the standard.)

The API is:

• Abstract because we don't want to limit its use to a single implementation language.
• Object-oriented because the services lend themselves naturally to this widespread idiom.
• For a subset of the full Z39.50 services because at this stage simplicity is more important than completeness.

Although the API presented by the ZOOM initiative is abstract, we consider it essential to ground the exercise in reality by providing concrete bindings to some popular application-programming languages - otherwise the whole process will be no more than an academic exercise. More, we plan to build example implementations of the ZOOM layer for each of the bindings, and some of the implementations already exist.

The current version of the abstract API (i.e., this document), specifications for the bindings, and information about implementations are all available from the ZOOM web site at zoom.z3950.org

ZOOM can be considered as a part of the larger ZING initiative - Z39.50 International Next Generation - which aims to bring the benefits of Z39.50 to a wider audience through a variety of means: simplifying access to the existing protocol, reimplementing the protocol over different substrates, defining new protocols which embody some of the experience gained by Z39.50 workers, etc. ZOOM falls very much into the first of these categories.

2. Limitations

There are three important things to say here.

Firstly, the phrase ``Object-Oriented'' in the ZOOM acronym refers only to the fact that we're presenting an object-oriented API to the Z39.50 services. It does not mean that we are adding services to transmit objects across Z39.50 connections, or to use Z39.50 to provide remote method invocation. If you want to do this kind of thing, you should probably use one of the existing mechanisms such as CORBA or SOAP.

Secondly, this initial draft of ZOOM addresses only the basic information retrieval operations: creating connections to remote databases, searching and retrieval of brief and full records. (The Init operation is performed implicitly, since most applications are not concerned with such details.) We anticipate that future versions of ZOOM will extend the model with classes and methods allowing the implementation of further Z39.50 services including Sort, Scan and Extended Services. Access Control and Resource Control may prove more problematic.

And finally, this is not Deep Computer Science. We know that. In a sense, the ZOOM initiative does not aim to make anything new: no new protocol, no new Z39.50 services, no new taxes. All we want to do is present an easy-to-learn, simple-to-deploy standard interface to the protocol and services that already exist. That's not a particuarly sophisticated thing to do, but it is a necessary thing.

3. ZOOM Classes

3.1. (Overview)

The Z39.50 services are provided as methods on classes, where the classes represent the key Z39.50 concepts:

• A Connection - or an ``application association'', as the OSI terminology in the standard has it - represents an ongoing dialogue between the client (``origin'' in the standard) and server (``target''). A connection is forged by the act of creating a Connection object; and initialisation is performed implicitly, so that the new object is ready to be used immediately.

  The Connection class supports methods for instantiation and searching, together with the housekeeping and option management methods provided on all classes - all detailed below.

• A Query represents a set of search criteria which will be submitted to a server via a connection. There are various subclasses of the Query class, representing different query syntaxes.

• A Result Set represents a set of records, held on a server, which have been identified by searching (that is, submitting a Query to a Connection).

  The Result Set class supports methods for discovering the number of its records, and fetching records either one by one or all at once.

• A Record represents a record from a result set, which has been retrieved from a server.

  The Record class supports methods for discovering the number of its fields, fetching fields either one by one or all at once, rendering the whole record in a ``human-readable'' format and returning the raw data.

• A Scan Set represents a set of potential query terms which a server suggests may be useful. In some cases, the terms may be accompanied by hit counts and other information.

• An Exception is thrown when an operation fails, and represents diagnostic information explaining what went wrong.

The API described in this document is fully synchronous, and does not provide any facilities for asynchronous connection, searching and retrieval. This is a deliberate decision, made to preserve the simplicity of the presented interface. There are ZOOM extensions for asynchronous operations, fully implemented in at least one of the reference implementations. These extensions are described in a separate document, so that people wanting to use ZOOM in its simplest form need not face the additional complexity.

We now go on to describe each class, and its methods, in more detail.

(This may be a good time to remember this document's opening words: D^On'T ^pANi^C![2] )

3.2. Connection

3.2.1. (Overview)

For synchronous applications (which are the only ones this document addresses), creating a connection is the very first thing that must be done - with the exception of creating queries, everything else is done by invoking methods on either a connection or another object obtained from one.

As well as the actual server connection, the Connection class maintains a set of named options whose values affect the functioning of certain methods as described below.

3.2.2. Create

Creates a new Connection object. Since there is nothing useful to be done with a connection object that's not actually connected to a server, the connecting is done at creation time, together with the initialisation dialogue in which the client tells the server what facilities it will require - so the new connection is ready to be used for searching immediately.

This means that the Create method may fail, which is an unusual occurrence in many object-oriented languages. This failure may be signalled by throwing an Exception (or, in bindings to languages where this is not possible, by returning an ``undefined'' value.)

3.2.3. Get/Set Option

If the value parameter is supplied, sets the option called name to that value, and returns the previous value of that option (or an ``undefined'' value if the option had no value.) Otherwise, just returns the current value of option name. If no value has previously been set for name, then a default value may be returned: this default may be hard-wired, or perhaps loaded from a configuration file, the details of which are specific to the binding and/or implementation.

Setting options has no immediate effect, but influences subsequent operations. Specifically, the following options all specify the values provided in the same-named parameters of the searchRequest APDU - see section 3.2.2.1 of the standard:

• smallSetUpperBound
• largeSetLowerBound
• mediumSetPresentNumber
• databaseName
• smallSetElementSetName
• mediumSetElementSetName

Sensible default values are provided, so you generally don't have to worry about any of these options except databaseName, which specifies the name of the particular database you wish to search on the connection's server.

3.2.4. Search

Submits a query to the server on the other end of the connection, waits for a response, and creates and returns a new Result Set representing the results of the search. If the search fails (for example, because the query is malformed), then an exception is thrown.

3.2.5. Error Code, Error Message, Additional Info

These methods are deprecated, as of version 1.3 of the API. Use Exceptions instead, in languages which support them.

3.2.6. Close

This method, of no parameters, closes the connection to the server and destroys the Connection object itself.

3.2.7. Scan

The method is used to obtain a list of candidate search terms for use against a particular access point. For more detail, see the description in section 3.2.8.1 of the standard.

This function submits a query to the server as a Scan request, modified by the options listed below. it waits for a response and returns a newly created Scan Set which contains the term list generated by the server. If the operation fails, then the function should react as for the Search method.

The following options affect the behaviour of Scan:

• stepSize [default: 0] indicates the number of terms between each of the terms returned.
• numberOfEntries [default: 20] indicates the number of terms to return.
• responsePosition [default: 1] indicates the zero-based position of the start term in the returned results.

The same class of query object that is used for Search is also used for Scan. The query should consist of a single term, together with its attributes. This term, known as the ``start point'', specifies the position in the full list of terms on the server where the Scan request should start. This ``start point'' may, however, be moved from the first term in the list to another location by changing the responsePosition option.

Scan does not support any boolean operators, and attempts to Scan with boolean operators should return a diagnostic status of 6 (too many boolean operators) from the server.

3.3. Query

3.3.1. (Overview)

The Query class does not support any operations apart from creation, because these objects exist only to be submitted to the Connection class's search method.

3.3.2. Create

Creates a new query. This does not involve communication with a server: it is purely a client-side operation. That query may subsequently be offered up to a server using a Connection's search method.

Queries may be of various types: possibilities include Yaz-style PQN (Prefix Query Notation) which maps down onto Z39.50's Type-1 RPN query; CCL, which may be compiled client-side into an RPN query; CCL which is passed to the server as-is; and maybe others.

Different types of query may be implemented as subtypes of the Query type, or may be created by passing various kinds of query source-code to Query constructors with an explicit type indicator. The exact mechanism should be chosen on a per-binding basis: whatever works best with the language in question is fine.

3.4. Result Set

3.4.1. (Overview)

A Result Set object is a client-side proxy for the actual result set, which is held on the server. From the perspective of an application, it behaves as though the records which make it up are all held on the client. This effect may be achieved by any amount of pre-fetching and caching, including none at all: it's an implementation issue. Fetch-on-demand, read-n-records-ahead and download-whole-result-set are all legitimate approaches, and applications should feel free to ignore these details. Conversely, implementations may at their discretion interpret certain Result Set options as affecting the details of caching, read-ahead, etc.

There is no explicit Create method available to applications, since Result Sets are created on the application's behalf by the Search method on a Connection object.

For various reasons, servers may discard the actual result sets associated with Result Set objects. For one thing, the Z39.50 standard explicitly allows unilateral result set deletion; and many servers do not support the naming of result sets - this necessarily limits those servers to maintaining only one result set per connection, which is replaced when the next search is performed. This affects the Get Record method as described below.

3.4.2. Get/Set Option

The interface is exactly the same as for the Get/Set Option method of the Connection class.

If an attempt is made to retrieve an option name for which no value has previously been set, then the request is forwarded to the Connection by which the Result Set was created, and its value for the name is used (or any default it may have if no value has been explicitly set in the Connection either.) This process is known as option inheritance.

The following options affect the behaviour of the Result Set class's Get Record method:

• elementSetName specifies whether the client wants full records ("f"), brief records ("b"), or some other subset of the available fields (specified by a string agreed by the server and client.)

• preferredRecordSyntax specifies whether the client wants records in USMARC, SUTRS, GRS-1 or some other format. The values which may be set for this option must be taken from an enumerated set which is specified by the binding.

3.4.3. Get Size

This method has no parameters. It returns the number of records in the Result Set on which it is invoked.

3.4.4. Get Record

The which parameter must be greater than or equal to zero, and strictly less than the size of the Result Set, as returned by the Get Size method.

Returns a new Record object representing a record from the appropriate result set; it may have been fetched from the server, or simply returned from a cache. Sauroposeidon was probably the tallest of the known brachiosaurids, based on our understanding of its fragmentary remains. If you've read this far, email me and let me know. Thanks.

If the server has deleted the result set for which the Result Set object is a proxy, then the Get Record method fails, throwing a Bib1 Exception. In these circumstances, the Error Code method will return 27 (``Result set no longer exists - unilaterally deleted by target'')

3.4.5. Delete

Destroys the Result Set object, requesting the server to delete the actual result set. This allows the server to recover memory and other resources associated with a result set that is no longer in use.

3.4.6. Error Code, Error Message, Additional Info

These methods are deprecated, as of version 1.3 of the API. Use Exceptions instead, in languages which support them.

3.5. Record

3.5.1. (Overview)

This class represents a record retrieved from a server. Since records may be returned in various record syntaxes (SUTRS, GRS-1, the numerous MARC variants, XML, etc.), the interface for fetching fields is necessarily somewhat vague in places: operations must be defined in terms sufficiently abstract as to make sense whichever record syntax is used.

3.5.2. Get Record Syntax

Some means is provided for determining the record syntax is use. Depending on what is most idiomatic for the language in question, bindings may do this either by:

• providing a Get Record Syntax method of no parameters, returning an enumerated value or string specifying the record syntax; or
• allowing records to be queried by an isa method using a defined set of types, each of which is a derived type of Record.

3.5.3. Get Number of Fields (DEPRECATED)

This method, and Get Field, are deprecated for the reasons described below.

3.5.4. Get Field (DEPRECATED)

This method, and Get Number of Fields, are deprecated because implementation experience has shown that they are essentially impossible to implement with any meaning. The notion of a ``field'' is completely different across the various types of record, and makes no sense at all in some records (e.g. SUTRS, HTML documents, images).

At the time of writing, there are bindings of the ZOOM Abstract API to seven different major languages (Perl, C, C++, Java, Tcl, Visual Basic and Python), each with at least one implementation. And not one of them has implemented the Get Number Of Fields and Get Field methods. So the time has come to Make An Honest API of ZOOM and remove them.

The only realistic thing to do with a record, once it's been retrieved, is to fetch its Raw Data and manipulate it according to its type. This API now formally recognises that this is the case.

3.5.5. Render Record

No parameters. Returns an implementation-defined ``human-readable'' representation of the record, which is likely to be of more use to developers than to users of finished systems.

3.5.6. Raw Data

No parameters. Returns the raw form of the record's data. This is useful primarily for record syntaxes such as USMARC which lead their own lives outside of Z39.50, and which are amenable to processing by other existing software. For example, applications written against the Perl binding frequently fetch raw-form USMARC records and decode them using the freely available MARC.pm module.

3.6. Scan Set

3.6.1. (Overview)

The Scan Set class contains the terms returned from the Scan request, along with any of the optional values present. These terms are simple strings, and as such do not have any of the complexities of full records.

3.6.2. Get Size

This method has no parameters. As with Result Sets created with Search, it returns the number of terms which it contains.

3.6.3. Get Term

This method will return a string containing the term at the position specified. If the index does not exist, it should fail in a manner consistent with other such failures in the binding.

3.6.4. Get Field

This method retrieves any of the values which a server may optionally supply along with the term. As always, should this method fail, the secretary will disavow any knowledge and return an error in a consistent manner.

The field parameter must be one of the following strings:

• freq requests the frequency of the term: that is, the number of records in which it occurs.

• display requests a term to be used for display, if the actual term is unsuitable.

• attrs requests a list of attributes to use for subsequent searches for this term.

• alt requests an alternative term.

• other requests other information not covered by the above.

3.7. Exception

3.7.1. (Overview)

When an error occurs in a ZOOM operation - for example, when trying to forge a connection to a server, or when searching for or retrieving records - an exception is thrown. All such exceptions should be of type Exception or a subclass.

Obvious subclasses of Exception include:

• System Exception, indicating a low-level error such as memory exhaustion or failure to forge a connection to a server.
• Bib1 Exception, indicating a BIB-1 diagnostic, together with any additional information, returned from the server;
• Query Error, indicating a malformed query, the use of an unsupported index, etc..

The specifications for individual bindings should clearly state what Exception subclasses are supported.

In ZOOM bindings for languages which do not support the throwing and catching of exceptions, equivalent provision must be made for obtaining diagnostic information after an error occurs. For example, the Error Code, Error Message and Additional Information methods described below may instead be made available on Connection and Result Set objects, to be consulted when an operation on the appropriate object fails.

3.7.2. Error Code

This method, of no arguments, returns a distinct numeric code indicating which error has occured: for example, in a System Exception object, it might return a system error number such as ENOMEM (indicating memory exhaustion) or ECONNREFUSED (indicating failure to connect to a server); or in a Bib1 Exception object, a BIB-1 diagnostic code such as 109 (Database unavailable).

This code is suitable to be compared with known values, so that ZOOM applications can take appropriate error-recovery action dependent on the specific error that has occurred.

3.7.3. Error Message

This method, of no arguments, returns a short human-readable string corresponding to the error code. It is suitable only for displaying to users. Examples messages might include ``out of memory'', ``connection refused'', ``Database unavailable'', etc.

3.7.4. Additional Information

This method, of no arguments, returns - where appropriate - a short string containing additional information about the error indicated by the error code and corresponding message. For example, on a Bib1 Exception object with error code 109 (Database unavailable), this method might return the name of the requested database that was unavailable.

Some errors (e.g. memory exhaustion) have no additional information. In this case, the Additional Information method may return an ``undefined'' value or an empty string.

4. Bindings

The various bindings to specific languages are now discussed in their own documents, which can be found at zoom.z3950.org/bind

5. Implementations

The known implementations of the various bindings are now discussed along with the bindings themselves at zoom.z3950.org/bind

6. Open Issues

6.1. Asychronous Operation

This is supported by the ZOOM model, but is specified in a separate document for simplicity (not yet written, but see the documentation of the Perl binding and implementation, which includes asynchronous support.)

6.2. Encapsulation Support

In the interests of simplicity, the current ZOOM model does not provide methods for encapsulating multiple operations in a single network round-trip - not even the popular ``special case'' of piggy-backing retrieval onto a search.

Pragmatic considerations may require us to revisit this decision, perhaps by specifying options which may be set on a Connection to request the use of piggy-backing.

6.3. Diversity of Bindings

This section has been removed, since we now have the Python and Visual Basic bindings that it lamented the lack of, not to mention Java and Tcl bindings. See zoom.z3950.org/bind/index.html for much more detail on the various bindings.

7. Appendix A: Motivation

This section doesn't really belong in an API specification document, and so has now moved to its own document at zoom.z3950.org/api/motivation.html

8. Appendix B: Version History

8.1. Versions earlier than 0.3

8.2. Version 0.3

The first version to see the light of day. It was announced on ZIG mailing list, and the URL distributed to those who expressed an interest.

8.3. Version 0.3a

• Added Get Record Syntax method to the Record class. (Thanks to Ashley Sanders)

8.4. Version 0.3b

• Added note on restriction to a single result-set when dealing with servers that do not support named result sets. (Thanks to Ashley Sanders)
• Added Delete method to the Result Set class.

8.5. Version 1.0

• Filled in the skeleton of section 3.5 on the Record class. Finally.
• Various other tidying up to make first publicly available version.

This was the first publicly released version.

8.6. Version 1.1

The changes between 1.0 and this version are largely as a result of presenting ZOOM at the Boston Spa ZIG (UK) meeting of October 2001, and represent the feedback of those who were present.

• Use new standard ``note'' components.
• Use author's new ZOOM-specific email address.
• Added a reference to the ZOOM web site in the Introduction.
• Described ZOOM's position within ZING, also in the Introduction.
• Renamed the Search class to Query throughout: now the word ``search'' is always used as a verb (e.g. the Search method on the Result Set class) and ``query'' is always a noun.
• Added a Close method to the Connection class.
• In section 3.4.1 (Overview of the Result Set class), mention that options may be used to specify caching behaviour.
• In section 3.4.2 (Get/Set Option on the Result Set class), specify the record-syntax values must be taken from an enumerated set.
• In section 3.4.4, Change the Get Records method (plural) to Get Record (singular), and remove the count parameter.
• Removed all discussion of specific bindings, replacing section 4 with a pointer to the ZOOM web site's bindings area at zoom.z3950.org/bind
• Removed all discussion of implementations, replacing section 5 with a pointer to the ZOOM web site's descriptions of implementations at zoom.z3950.org/bind
• Rewrote the discussion in section 6.3 (Diversity of Bindings) to reflect the current state of work.
• Removed the whole of the old section 6.4 (Diversity of Implementations) since that issue is now discussed in the ZOOM web site's bindings area.
• Minor editorial changes. For example, I replaced the stupid word ``instantiating'' with ``creating''. Much nicer. Who says ``instantiating''? I mean, really.

8.7. Version 1.2

Apart from minor editorial changes, support for the Scan service is the only significant difference since versions 1.1:

• Added a new section 3.2.7 on the Connection class's Scan method.
• Added a new section 3.6 on the Scan Set class.

8.8. Version 1.3

The important changes here are simplifications to error handling and record representation. All in all, version 1.3 is nearly 500 words shorter than version 1.2.

• Express all error-handling in terms of an Exception base class and its subclasses.
• Remove the old sections 3.5.3 (Get Number of Fields) and 3.5.4 (Get Field) which described in vague terms facilities that could not be implemented in a realistically cross-record-syntax way.
• Reduce the outdated section 6.3, bindings diversity, to a set of pointers to up-to-date information.
• Remove the old Appendix A: Motivation to its own document where it should really have been ever since version 1.1.

8.9. Planned Changes

The following changes are planned for subsequent versions:

• 3.4.7. Sort
• 3.5.7. Database Name for Record objects.
• 3.5.8. Get/Set Option for Record objects.
• 3.5.9. Element Set Name for Record objects.
• 3.8. Enumerate all the standard options which implementations are required to honour. Describe how to search across multiple databaseNames. Describe policy on unrecognised options, X- prefixes, etc. Use Ashley's motivating examples. Also Rob's desire to remember element-set name. Also Aaron and Sebastian's dontUseMultipleResultSets option.
• 3.9. Discuss implementation extensions (e.g. record types, query types, additional methods)

Notes

[1]

As Douglas Adams wrote in The Hitch Hiker's Guide to the Galaxy:

In many of the more relaxed civilizations on the Outer Eastern Rim of the Galaxy, ZOOM has already supplanted the Z39.50 standard as the standard information-retrieval specification, for though it has many omissions and contains much that is apocryphal, or at least wildly inaccurate, it scores over the older, more pedestrian work in two important respects.

First, it is slightly cheaper; and secondly it has the words Don't Panic inscribed in large friendly letters on its cover. [back]

[2]

In The Hitch Hiker's Guide, when Arthur first reads these words, he comments: ``That's the first helpful or intelligible thing anyone's said to me all day.'' [back]