Network Working Group                                            M. Wahl
Request for Comments: 2596                  Innosoft International, Inc.
Category: Standards Track                                       T. Howes
                                           Netscape Communications Corp.
                                                                May 1999


                     Use of Language Codes in LDAP


Status of this Memo

   This document specifies an Internet standards track protocol for the
   Internet community, and requests discussion and suggestions for
   improvements.  Please refer to the current edition of the "Internet
   Official Protocol Standards" (STD 1) for the standardization state
   and status of this protocol.  Distribution of this memo is unlimited.

Copyright Notice

   Copyright (C) The Internet Society (1999).  All Rights Reserved.

1. Abstract

   The Lightweight Directory Access Protocol [1] provides a means for
   clients to interrogate and modify information stored in a distributed
   directory system.  The information in the directory is maintained as
   attributes [2] of entries.  Most of these attributes have syntaxes
   which are human-readable strings, and it is desirable to be able to
   indicate the natural language associated with attribute values.

   This document describes how language codes [3] are carried in LDAP
   and are to be interpreted by LDAP servers.  All implementations MUST
   be prepared to accept language codes in the LDAP protocols.  Servers
   may or may not be capable of storing attributes with language codes
   in the directory.  This document does not specify how to determine
   whether particular attributes can or cannot have language codes.

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED",  "MAY", and "OPTIONAL" in this
   document are to be interpreted as described in RFC 2119 [4].

2. Language Codes

   Section 2 of RFC 1766 [3] describes the language code format which is
   used in LDAP.  Briefly, it is a string of ASCII alphabetic characters
   and hyphens.  Examples include "fr", "en-US" and "ja-JP".

   Language codes are case insensitive.  For example, the language code
   "en-us" is the same as "EN-US" and "en-US".

   Implementations MUST NOT otherwise interpret the structure of the
   code when comparing two codes, and MUST treat them as simply strings
   of characters. Client and server implementations MUST allow any
   arbitrary string which follows the patterns given in RFC 1766 to be
   used as a language code.

3. Use of Language Codes in LDAP

   This section describes how LDAP implementations MUST interpret
   language codes in performing operations.

   In general, an attribute with a language code is to be treated as a
   subtype of the attribute without a language code.  If a server does
   not support storing language codes with attribute values in the DIT,
   then it MUST always treat an attribute with a language code as an
   unrecognized attribute.

3.1. Attribute Description

   An attribute consists of a type, a list of options for that type, and
   a set of one or more values.  In LDAP, the type and the options are
   combined into the AttributeDescription, defined in section 4.1.5 of
   [1]. This is represented as an attribute type name and a possibly-
   empty list of options.  One of these options associates a natural
   language with values for that attribute.

        language-option = "lang-" lang-code

        lang-code = printable-ascii ; a code as defined in RFC 1766

   Multiple language options may be present on a particular value.

   The language code has no effect on the character set encoding for
   string representations of DirectoryString syntax values; the UTF-8
   representation of UniversalString (ISO 10646) is always used.

   Examples of valid AttributeDescription:
        givenName;lang-en-US
        CN;lang-ja

   In LDAP and in examples in this document, a directory attribute is
   represented as an AttributeDescription with a list of values.  Note
   that the data could be stored in the LDAP server in a different
   representation.

3.2. Distinguished Names and Relative Distinguished Names

   No attribute description options are permitted in Distinguished Names
   or Relative Distinguished Names.  Thus language codes MUST NOT be
   used in forming DNs.

3.3. Search Filter

   If a language code is present in an AttributeDescription in a search
   filter, then only attribute values in the directory which match the
   base attribute type or its subtype, the language code and the
   assertion value match this filter.

   Thus for example a filter of an equality match of type "name;lang-
   en-US" and assertion value "Billy Ray", against the following
   directory entry

   objectclass: top                     DOES NOT MATCH (wrong type)
   objectclass: person                  DOES NOT MATCH (wrong type)
   name;lang-EN-US: Billy Ray           MATCHES
   name;lang-EN-US: Billy Bob           DOES NOT MATCH (wrong value)
   CN;lang-en-us: Billy Ray                MATCHES
   CN;lang-EN-US;dynamic: Billy Ray     MATCHES
   CN;lang-en;dynamic: Billy Ray        DOES NOT MATCH (differing lang-)
   name: Billy Ray                      DOES NOT MATCH (no lang-)
   SN: Ray                              DOES NOT MATCH (wrong value)

   (Note that "CN" and "SN" are subtypes of "name".)

   Client implementors should however note that providing a language
   code in a search filter AttributeDescription will often filter out
   desirable values where the language code does not match exactly.  For
   example, the filter (name;lang-en=Billy Ray) does NOT match the
   attribute "name;lang-en-US: Billy Ray".

   If the server does not support storing language codes with attribute
   values in the DIT, then any filter which includes a language code
   will always fail to match, as it is an unrecognized attribute type.
   No error would be returned because of this; a presence filter would
   evaluate to FALSE and all other forms to Undefined.

   If no language code is specified in the search filter, then only the
   base attribute type and the assertion value need match the value in
   the directory.

   Thus for example a filter of an equality match of type "name" and
   assertion value "Billy Ray", against the following directory entry

   objectclass: top                     DOES NOT MATCH (wrong type)
   objectclass: person                  DOES NOT MATCH (wrong type)
   name;lang-EN-US: Billy Ray           MATCHES
   name;lang-EN-US: Billy Bob           DOES NOT MATCH (wrong value)
   CN;lang-EN-US;dynamic: Billy Ray     MATCHES
   CN;lang-en;dynamic: Billy Ray        MATCHES
   name: Billy Ray                      MATCHES
   SN: Ray                              DOES NOT MATCH (wrong value)

   Thus in general, clients SHOULD NOT use the language code option in
   AttributeDescription fields in search filters.

3.4. Compare

   A language code can be present in an AttributeDescription used in a
   compare request AttributeValueAssertion.  This is to be treated by
   servers the same as the use of language codes in a search filter with
   an equality match, as described in the previous section.  If there is
   no attribute in the entry with the same subtype and language code,
   the noSuchAttributeType error will be returned.

   Thus for example a compare request of type "name" and assertion value
   "Johann", against an entry with all the following directory entry

   objectclass: top
   objectclass: person
   givenName;lang-de-DE: Johann
   CN: Johann Sibelius
   SN: Sibelius

   will cause the server to return compareTrue.

   However, if the client issued a compare request of type "name;lang-
   de" and assertion value "Johann" against the above entry, the request
   would fail with the noSuchAttributeType error.

   If the server does not support storing language codes with attribute
   values in the DIT, then any comparison which includes a language code
   will always fail to locate an attribute type, and noSuchAttributeType
   will be returned.

   Thus in general, clients SHOULD NOT use the language code option in
   AttributeDescription fields in the compare request.

3.5. Requested Attributes in Search

   Clients MAY provide language codes in AttributeDescription in the
   requested attribute list in a search request.

   If a language code is provided in an attribute description, then only
   attribute values in a directory entry which have the same language
   code as that provided are to be returned. Thus if a client requests
   an attribute "description;lang-en", the server MUST NOT return values
   of an attribute "description" or "description;lang-fr".

   Clients MAY provide in the attribute list multiple
   AttributeDescription which have the same base attribute type but
   different options. For example a client MAY provide both "name;lang-
   en" and "name;lang-fr", and this would permit an attribute with
   either language code to be returned.  Note there would be no need to
   provide both "name" and "name;lang-en" since all subtypes of name
   would match "name".

   If a server does not support storing language codes with attribute
   values in the DIT, then any attribute descriptions in the list which
   include language codes are to be ignored, just as if they were
   unknown attribute types.

   If a request is made specifying all attributes or an attribute is
   requested without providing a language code, then all attribute
   values regardless of their language code are returned.

   For example, if the client requests a "description" attribute, and a
   matching entry contains

   objectclass: top
   objectclass: organization
   O: Software GmbH
   description: software
   description;lang-en: software products
   description;lang-de: Softwareprodukte
   postalAddress: Berlin 8001 Germany
   postalAddress;lang-de: Berlin 8001 Deutschland

   The server will return:

   description: software
   description;lang-en: software products
   description;lang-de: Softwareprodukte

3.6. Add Operation

   Clients MAY provide language codes in AttributeDescription in
   attributes of a new entry to be created, subject to the limitation
   that the client MUST NOT use language codes in the attribute value or
   values which form the RDN of the entry.

   A client MAY provide multiple attributes with the same attribute type
   and value, so long as each attribute has a different language code,
   and at most one attribute does not have a language code option.

   Servers which support storing language codes in the DIT MUST allow
   any attribute it recognizes that has the Directory String syntax to
   have a language option associated with it. Servers SHOULD allow
   language options to be associated with other attributes.

   For example, the following is a legal request.

   objectclass: top
   objectclass: person
   objectclass: residentialPerson
   name: John Smith
   CN: John Smith
   CN;lang-en: John Smith
   SN: Smith
   streetAddress: 1 University Street
   streetAddress;lang-en: 1 University Street
   streetAddress;lang-fr: 1 rue Universite
   houseIdentifier;lang-fr: 9e etage

   If a server does not support storing language codes with attribute
   values in the DIT, then it MUST treat an AttributeDescription with a
   language code as an unrecognized attribute. If the server forbids the
   addition of unrecognized attributes then it MUST fail the add request
   with the appropriate result code.

3.7. Modify Operation

   A client MAY provide a language code in an AttributeDescription as
   part of a modification element in the modify operation.

   Attribute types and language codes MUST match exactly against values
   stored in the directory.  For example, if the modification is a
   "delete", then if the stored values to be deleted have a language
   code, the language code MUST be provided in the modify operation, and
   if the stored values to be deleted do not have a language code, then
   no language code is to be provided.

   If the server does not support storing language codes with attribute
   values in the DIT, then it MUST treat an AttributeDescription with a
   language code as an unrecognized attribute, and MUST fail the request
   with an appropriate result code.

3.8. Diagnostic Messages

   Servers SHOULD use only printable ASCII characters in the
   errorMessage field, as not all clients will be able to display the
   full range of Unicode.

4. Differences from X.500(1997)

   X.500(1997) defines a different mechanism, contexts, as the means of
   representing language tags.  This section summarizes the major
   differences in approach.

   a) An X.500 operation which has specified a language code on a value
      matches a value in the directory without a language code.
   b) LDAP references RFC 1766, which allows for IANA registration of
      new tags.
   c) LDAP does not allow language codes in distinguished names.
   d) X.500 describes subschema administration procedures to allow
      language codes to be associated with particular attributes types.

5. Security Considerations

   There are no known security considerations for this document.  See
   the security considerations sections of [1] and [2] for security
   considerations of LDAP in general.

6. Acknowledgements

   This document is a product of the IETF ASID and LDAPEXT working
   groups.  Martin Duerst provided many valuable comments on an earlier
   version of this document.

7. Bibliography

   [1] Wahl, M., Howes, T. and S. Kille, "Lightweight Directory Access
       Protocol (v3)", RFC 2251, December 1997.

   [2] Wahl, M., Coulbeck, A., Howes, T. and S. Kille, "Lightweight
       X.500 Directory Access Protocol Attribute Syntax Definitions",
       RFC 2252, December 1997.

   [3] Alvestrand, H.,"Tags for the Identification of Languages", RFC
       1766, March 1995.

   [4] Bradner, S., "Key words for use in RFCs to Indicate Requirement
       Levels", BCP 14, RFC 2119, March 1997.

8. Authors' Addresses

   Mark Wahl
   Innosoft International, Inc.
   8911 Capital of Texas Hwy Suite 4140
   Austin, TX 78759 USA

   EMail:  M.Wahl@innosoft.com


   Tim Howes
   Netscape Communications Corp.
   501 E. Middlefield Rd
   Mountain View, CA 94043 USA

   Phone:  +1 650 937-3419
   EMail:   howes@netscape.com

Full Copyright Statement

   Copyright (C) The Internet Society (1999).  All Rights Reserved.

   This document and translations of it may be copied and furnished to
   others, and derivative works that comment on or otherwise explain it
   or assist in its implementation may be prepared, copied, published
   and distributed, in whole or in part, without restriction of any
   kind, provided that the above copyright notice and this paragraph are
   included on all such copies and derivative works.  However, this
   document itself may not be modified in any way, such as by removing
   the copyright notice or references to the Internet Society or other
   Internet organizations, except as needed for the purpose of
   developing Internet standards in which case the procedures for
   copyrights defined in the Internet Standards process must be
   followed, or as required to translate it into languages other than
   English.

   The limited permissions granted above are perpetual and will not be
   revoked by the Internet Society or its successors or assigns.

   This document and the information contained herein is provided on an
   "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING
   TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
   BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
   HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF
   MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Acknowledgement

   Funding for the RFC Editor function is currently provided by the
   Internet Society.