Z39.50 interface

Introduction

Z39.50 is a standard (ISO 23950 and ANSI/NISO Z39.50) that allows different information retrieval systems to communicate behind the scenes. You can get full details from the Z39.50 Maintenance Agency pages on the Library of Congress Web site: http://lcweb.loc.gov/z3950/agency/

Quick Connection Details

The most vital information you need to know to get your origin/client connected to the Copac Z39.50 target is:

Address: z3950.copac.ac.uk
Port: 210
Database name: COPAC
Record syntax: XML, GRS-1 or SUTRS
No username or password are required.

Full Configuration Details

Accessing the target

The target is only accessible through TCP and lives at Internet address z3950.copac.ac.uk on port 210.

You must set the database name to COPAC. If the database name is set to "default" or anything else then nothing will work! This is the most common cause of problems with accessing Copac through Z39.50.

No username or password should be specified. Specifying a username and password is the second most common cause of problems. Don't do it!

Supported parts of Z39.50

Init, Search, Present, Named Result Sets, Delete Result Set, Sort, Trigger Resource Control and Close are supported. Records are available in XML, GRS-1, or SUTRS formats. Earlier queries can be back-referenced in the current query. Both version 2 and version 3 of the standard are supported.

Record syntaxes

XML, GRS-1 and SUTRS are supported. There are no plans to deliver MARC records as part of the Copac service. The XML records conform to the MODS Schema.

Record Character Set

The records use the Unicode character set in UTF-8 encoding.

Element set names

Records can be retrieved in either Brief (B) or Full (F) form. The default is Full.

Other ESNs available for SUTRS records only are:

REF
This will present records in a form suitable for loading into a personal bibliographic reference manager such as Papyrus.
PRINTF
This format is the same as the format "Print format (Full)" available from the Download option of the Copac web interface.
PRINTB
This format is the same as the format "Print format (Brief)" available from the Download option of the Copac web interface.

Supported attributes

Only the BIB-1 attribute set is supported for searching Copac. Now there is a small amount of Explain implemented we support Exp-1 for searching the Explain database.

Use attributes

Personal name (1)
A name in the author or subject fields. When used with the Key structure attribute it should be Surname,initials, eg Doyle,AC
Corporate name (2) and Conference name (3)
Look for a corporation or conference as an author or subject of a work.
Title (4)
Searches the title index. When used with the structure attribute Key the term is assumed to be a derived key of the form 3,2,2,1 eg. lor,of,th,r
ISBN (7)
We index all ISBNs as 13 digit ISBNs. If your search term looks like a 10 digit ISBN the target will convert it to a 13 digit ISBN before permorming the search. Hence, you should be able to use 10 or 13 digit ISBNs for your search. As from 28/3/2007 this Use attribute will no longer find ISSNs.
ISSN (8)
Searches our ISSN index. As from 28/3/2007 this Use attribute will no longer find ISBNs.
Subject heading (21)
This can be used for keyword searching on subject headings.
Date (30 or 31)
Searches the date index. The relational operators can be used for date range searching. Performance of date range searching is now as good as it is going to get.
Language (54)
Search for works by language. Use language codes as Fre (French) or Ger (German) etc. You can get language name to code conversions from the MARC code list for languages.
Institution (56)
Search by the contributor code.
Place publication (59)
Search for the place of publication.
Note (63)
Keyword search of the notes index.
Author-title(1000)
This is a derived key index and needs the first four letters of the author's surname and the first four letters of the title separated by punctuation, eg Shak,Henr
Author (1003)
Any author name, be it personal, corporate or conference.
Personal Author (1004)
Look for a author. Can be used with the Key structure attribute, when the search term should be of the form Surname, Initials, eg Dylan,B
Corporate Author (1005) and Conference Author (1006)
Look for a corporation or conference as an author of a work.
Identifier-standard (1007)
Look for standard numbers such as ISBN or ISSN
Personal Subject (1009)
Look for a work about a person. When used with the Key structure attribute the search term should be of the form Surname,Initials, eg OBrian,P
Any (1016) and Server-choice (1017)
Searches for the given terms in all the indexes.
Publisher (1018)
Keyword search of the publisher name index.
Local number (12), Number DB (52) and Doc ID (1032)
Search for the Copac document IDs.
Map-scale (1024)
Just enter a number eg. 500000
Material Type (1031)
If you want to select all the records we have identified as periodicals then search for the string "periodical" against this Use attribute.

The initial search for periodicals will be slow, but any subsequent searches should be much faster as the target now caches searches such a this one.

Fuzzy-scale (5001)
Just enter a number eg. 500000. The fuzzy scale search will find matching scales plus other similar scales. Most useful when searching for early maps where the original scale may have been converted into slightly different ratios in different records.

Relational attributes

Less than (1), Less than or equal (2), Equal (3), Greater or equal (4), Greater than (5).

Attributes 1,2 4, and 5 can only be used with the Date (30) Use attribute.

Position attributes

Any position in field (3).

Structure attributes

Phrase (1), Word (2), Key (3) and Word List (6).

If structure is Phrase then the words will be searched for in the order given in the term. Therefore, it probably isn't a good idea to search for an author with structure as Phrase unless you know how we store our author names in the database.

If structure is Word or Word List (treated exactly the same) then order isn't preserved.

Note: When using the Key structure attribute, each part of the derived key must be separated with punctuation. All the punctuation must be supplied or the search will fail. The following are all valid:

    Lor,of,th,r
    Pri-an-pr- 
    Oli,,,
    Muc/ad/ab/n
    Levi-Lex
    Forster,EM
  

Truncation attributes

Right Truncation (1), Do Not Truncate (100), Process # in search term (101).

Truncation can be used with any of the supported structure attributes. If Right truncation is specified each word in the term will be right truncated. When using the 101 type truncation, the # (hash) can be anywhere in the term except the first character. This enables you to do limited pattern matching.

Using truncation with terms less than 5 characters in length is very likely to cause overflow in the BRS/Search search engine with resulting loss of records from your result set.

You might find it useful to set up your attribute combinations so that the 101 truncation attribute is always sent. Then all you have to do is type a # (hash) where you want truncation in your search terms. You incur no overhead by send a 101 truncation attribute and not specifying any truncation with a # (hash).

Completeness attributes

Incomplete Subfield (1), Complete Subfield (3).

Complex attributes

Complex attributes are not supported.

Only Type-1 queries are supported. The boolean operators AND, OR and NOT are supported as is Proximity. Proximity operators must be configured as shown in the table below.

Relation Less than or equal
Unit Word
Distance Up to 99
Ordered True or false (ie ordered or unordered)
Exclusion True or false (ie exclusive or not :-)

Sort

Database, Generic and keyword sort are supported. The only valid keyword sort is COPAC1 which will sort on title and descending date.

Only case insensitive and ordinary ascending and descending sorts are supported.

Merging

Results may be merged using sort. If you asked for a merge, the target includes the result count in the OtherInformation field in the format: Result-count: 132

Supported attributes

Many of the Use attributes supported for searching can be specified for sorting. A sort on the "Server Choice" or "Any" Use attribute is the the same as sorting by title.

Diagnostic Messages

The diagnostic messages sent by the target are detailed on the Z39.50 diagnostics page.

Explain

Explain has been implemented and should provide enough information for an origin to configure itself.

Implementation details

The underlying database software is OpenText LiveLink Discovery Server with access to the Copac union catalogue. The records themselves are supplied by members of the RLUK libraries, with the addition of records for special collection materials from some other libraries. A full list of the libraries contributing data to Copac is available on our 'Libraries' page. The records accessible to the general public are not available in any MARC format.

Date range searching

Date range searching has been implemented as LiveLink limit commands which significantly improves the speed of searching. Several date limits can be joined together with the boolean AND and OR operators.

Date ranges are quite speedy, especially if the rest of your query terms didn't gather too many hits.

EndNote and Reference Manager

EndNote and Reference Manager are Z39.50 clients that allow you to access Copac. If you have either of these clients and they aren't configured for Copac you can download the EndNote connection file for Copac and the Reference Manager filter file for Copac from our FAQ page.

Contact information

Any comments concerning the operation of the Z39.50 target should be sent to copac@mimas.ac.uk

Share This Page

  • Send to Facebook
  • Tweet this!
  • Email
  • Print