RFC 9290

Internet Engineering Task Force (IETF)                        T. Fossati
Request for Comments: 9290                                   Arm Limited
Category: Standards Track                                     C. Bormann
ISSN: 2070-1721                                   Universität Bremen TZI
                                                            October 2022

Concise Problem Details for Constrained Application Protocol (CoAP) APIs


   This document defines a concise "problem detail" as a way to carry
   machine-readable details of errors in a Representational State
   Transfer (REST) response to avoid the need to define new error
   response formats for REST APIs for constrained environments.  The
   format is inspired by, but intended to be more concise than, the
   problem details for HTTP APIs defined in RFC 7807.

Status of This Memo

   This is an Internet Standards Track document.

   This document is a product of the Internet Engineering Task Force
   (IETF).  It represents the consensus of the IETF community.  It has
   received public review and has been approved for publication by the
   Internet Engineering Steering Group (IESG).  Further information on
   Internet Standards is available in Section 2 of RFC 7841.

   Information about the current status of this document, any errata,
   and how to provide feedback on it may be obtained at

Copyright Notice

   Copyright (c) 2022 IETF Trust and the persons identified as the
   document authors.  All rights reserved.

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents
   (https://trustee.ietf.org/license-info) in effect on the date of
   publication of this document.  Please review these documents
   carefully, as they describe your rights and restrictions with respect
   to this document.  Code Components extracted from this document must
   include Revised BSD License text as described in Section 4.e of the
   Trust Legal Provisions and are provided without warranty as described
   in the Revised BSD License.

Table of Contents

   1.  Introduction
     1.1.  Terminology and Requirements Language
   2.  Basic Problem Details
   3.  Extending Concise Problem Details
     3.1.  Standard Problem Detail Entries
       3.1.1.  Standard Problem Detail Entry: Unprocessed CoAP Option
     3.2.  Custom Problem Detail Entries
   4.  Privacy Considerations
   5.  Security Considerations
   6.  IANA Considerations
     6.1.  Standard Problem Detail Keys Registry
     6.2.  Custom Problem Detail Keys Registry
     6.3.  Media Type
     6.4.  Content-Format
     6.5.  CBOR Tag 38
   7.  References
     7.1.  Normative References
     7.2.  Informative References
   Appendix A.  Language-Tagged Strings
     A.1.  Introduction
     A.2.  Detailed Semantics
     A.3.  Examples
   Appendix B.  Interworking with RFC 7807

   Authors' Addresses

1.  Introduction

   REST response status information such as Constrained Application
   Protocol (CoAP) response codes (Section 5.9 of [RFC7252]) is
   sometimes not sufficient to convey enough information about an error
   to be helpful.  This specification defines a simple and extensible
   framework to define Concise Binary Object Representation (CBOR)
   [STD94] data items to suit this purpose.  This framework is designed
   to be reused by REST APIs, which can identify distinct "shapes" of
   these data items specific to their needs.  Thus, API clients can be
   informed of both the high-level error class (using the response code)
   and the finer-grained details of the problem (using the vocabulary
   defined here).  This pattern of communication is illustrated in
   Figure 1.

                    .--------.          .--------.
                    |  CoAP  |          |  CoAP  |
                    | Client |          | Server |
                    '----+---'          '---+----'
                         |                  |
                         | Request          |
                         |                  | (failure)
                         |   Error Response |
                         |      with a CBOR |
                         | data item giving |
                         |  Problem Details |
                         |                  |

                Figure 1: Problem Details: Example with CoAP

   The framework presented is largely inspired by the problem details
   for HTTP APIs defined in [RFC7807].  Appendix B discusses
   applications where interworking with [RFC7807] is required.

1.1.  Terminology and Requirements Language

   The terminology from [RFC7252], [STD94], and [RFC8610] applies; in
   particular, CBOR diagnostic notation is defined in Section 8 of RFC
   8949 [STD94] and Appendix G of [RFC8610].  Readers are also expected
   to be familiar with the terminology from [RFC7807].

   In this document, the structure of data is specified in Concise Data
   Definition Language (CDDL) [RFC8610] [RFC9165].

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "OPTIONAL" in this document are to be interpreted as described in
   BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all
   capitals, as shown here.

2.  Basic Problem Details

   A Concise Problem Details data item is a CBOR data item with the
   following structure (rules named starting with tag38 are defined in
   Appendix A):

   problem-details = non-empty<{
     ? &(title: -1) => oltext
     ? &(detail: -2) => oltext
     ? &(instance: -3) => ~uri
     ? &(response-code: -4) => uint .size 1
     ? &(base-uri: -5) => ~uri
     ? &(base-lang: -6) => tag38-ltag
     ? &(base-rtl: -7) => tag38-direction

   standard-problem-detail-entries = (
     * nint => any

   custom-problem-detail-entries = (
     * (uint/~uri) => { + any => any }

   non-empty<M> = (M) .and ({ + any => any })

   oltext = text / tag38

          Figure 2: Structure of Concise Problem Details Data Item

   (Examples of elaborated Concise Problem Details data items can be
   found later in the document, e.g., Figure 3.)

   A number of problem detail entries, the Standard Problem Detail
   entries, are predefined (more predefined details can be registered,
   see Section 3.1).

   Note that, unlike [RFC7807], Concise Problem Details data items have
   no explicit "problem type".  Instead, the category (or, one could
   say, Gestalt) of the problem can be understood from the shape of the
   problem details offered.  We talk of a "problem shape" for short.

   The title (key -1):
      A short, human-readable summary of the problem shape.  Beyond the
      shape of the problem, it is not intended to summarize all the
      specific information given with the problem details.  For
      instance, the summary might include that an account does not have
      enough money for a transaction to succeed but not the detailed
      information such as the account number, how much money that
      account has, and how much would be needed.

   The detail (key -2):
      A human-readable explanation specific to this occurrence of the

   The instance (key -3):
      A URI reference that identifies the specific occurrence of the
      problem.  It may or may not yield further information if

   The response-code (key -4):
      The CoAP response code (Sections 5.9 and 12.1.2 of [RFC7252])
      generated by the origin server for this occurrence of the problem.

   The base-uri (key -5):
      The base URI (see Section 5.1 of RFC 3986 [STD66]) that should be
      used to resolve relative URI references embedded in this Concise
      Problem Details data item.

   The base-lang (key -6):
      The language-tag (tag38-ltag) that applies to the presentation of
      unadorned text strings (not using tag 38) in this Concise Problem
      Details data item; see Appendix A.

   The base-rtl (key -7):
      The writing-direction (tag38-direction) that applies to the
      presentation of unadorned text strings (not using tag 38) in this
      Concise Problem Details data item; see Appendix A.

   Both "title" and "detail" can use either an unadorned CBOR text
   string (text) or a language-tagged text string (tag38); see
   Appendix A for the definition of the latter.  Language tag and
   writing direction information for unadorned text strings is intended
   to be obtained from context; if that context needs to be saved or
   forwarded with a Concise Problem Details data item, "base-lang" and
   "base-rtl" can be used.  If no such (explicitly saved or implicit)
   context information is available, unadorned text is interpreted with
   language-tag "en" and writing-direction "false" (ltr).

   The "title" string is advisory and included to give consumers a
   shorthand for the category (problem shape) of the error encountered.

   The "detail" member, if present, ought to focus on helping the client
   correct the problem rather than giving extensive server-side
   debugging information.  Consumers SHOULD NOT parse the "detail"
   member for information; extensions (see Section 3) are more suitable
   and less error-prone ways to obtain such information.  Note that the
   "instance" URI reference may be relative; this means that it must be
   resolved relative to the representation's base URI, as per Section 5
   of RFC 3986 [STD66].

   The "response-code" member, if present, is only advisory; it conveys
   the CoAP response code used for the convenience of the consumer.
   Generators MUST use the same response code here as in the actual CoAP
   response; the latter is needed to assure that generic CoAP software
   that does not understand the problem-details format still behaves
   correctly.  Consumers can use the "response-code" member to determine
   what the original response code used by the generator was, in cases
   where it has been changed (e.g., by an intermediary or cache), and
   when message bodies persist without CoAP information (e.g., in an
   events log or analytics database).  Generic CoAP software will still
   use the CoAP response code.  To support the use case of message-body
   persistence without support by the problem-details generator, the
   entity that persists the Concise Problem Details data item can copy
   over the CoAP response code that it received on the CoAP level.  Note
   that the "response-code" value is a numeric representation of the
   actual code (see Section 3 of [RFC7252]), so it does not take the
   usual presentation form that resembles an HTTP status code: 4.04 Not
   Found is represented by the number 132.

   The "base-uri" member is usually not present in the initial request-
   response communication as it can be inferred as per Section 5.1.3 of
   RFC 3986 [STD66].  An entity that stores a Concise Problem Details
   data item or otherwise makes it available for consumers without this
   context might add in a "base-uri" member to allow those consumers to
   perform resolution of any relative URI references embedded in the
   data item.

3.  Extending Concise Problem Details

   This specification defines a generic problem-details container with
   only a minimal set of attributes to make it usable.

   It is expected that applications will extend the base format by
   defining new attributes.

   These new attributes fall into two categories: generic and
   application specific.

   Generic attributes will be allocated in the standard-problem-detail-
   entries slot according to the registration procedure defined in
   Section 3.1.

   Application-specific attributes will be allocated in the custom-
   problem-detail-entries slot according to the procedure described in
   Section 3.2.

   Consumers of a Concise Problem Details data item MUST ignore any
   Standard Problem Detail entries or Custom Problem Detail entries, or
   keys inside the Custom Problem Detail entries, that they do not
   recognize ("ignore-unknown rule"); this allows problem details to
   evolve.  When storing the data item for future use or forwarding it
   to other consumers, it is strongly RECOMMENDED to retain the
   unrecognized entries; exceptions might be when storage or forwarding
   occurs in a different format/protocol that cannot accommodate them or
   when the storage or forwarding function needs to filter out privacy-
   sensitive information and for that needs to assume unrecognized
   entries might be privacy-sensitive.

3.1.  Standard Problem Detail Entries

   Beyond the Standard Problem Detail keys defined in Figure 2,
   additional Standard Problem Detail keys can be registered for use in
   the standard-problem-detail-entries slot (see Section 6.1).

   Standard Problem Detail keys are negative integers, so they can never
   conflict with Custom Problem Detail keys defined for a specific
   application domain (which are unsigned integers or URIs.)

   In summary, the keys for Standard Problem Detail entries are in a
   global namespace that is not specific to a particular application

3.1.1.  Standard Problem Detail Entry: Unprocessed CoAP Option

   Section 2 provides a number of generally applicable Standard Problem
   Detail entries.  The present section both registers another useful
   Standard Problem Detail entry and serves as an example of a Standard
   Problem Detail Entry registration, in the registration template
   format that would be ready for registration.

   Key value:
   CDDL type:
      one-or-more<uint>, where
      one-or-more<T> = T / [ 2* T ]
   Brief description:
      Option number(s) of CoAP option(s) that were not understood
   Specification reference:
      Section 3.1.1 of RFC 9290

   The specification of the Standard Problem Detail entry referenced by
   the above registration template follows:

   The Standard Problem Detail entry unprocessed-coap-option provides
   the option number or numbers of any CoAP options present in the
   request that could not be processed by the server.

   This may be a critical option that the server is unaware of, or an
   option the server is aware of but could not process (and chose not
   to, or was not allowed to, ignore it).

   The Concise Problem Details data item including this Standard Problem
   Detail Entry can be used in fulfillment of the "SHOULD" requirement
   in Section 5.4.1 of [RFC7252].

   Several option numbers may be given in a list (in no particular
   order), without any guarantee that the list is a complete
   representation of all the problems in the request (as the server
   might have stopped processing already at one of the problematic
   options).  If an option with the given number was repeated, there is
   no indication which of the values caused the error.

   Clients need to expect to see options in the list that they did not
   send in the request; this can happen if the request traversed a proxy
   that added the option but did not act on the problem-details response
   being returned by the origin server.

   For a few special values of unprocessed CoAP options (such as Accept
   or Proxy-Uri), note that there are special response codes (4.06 Not
   Acceptable, 5.05 Proxying Not Supported, respectively) to be sent
   instead of 4.02 Bad Option.

3.2.  Custom Problem Detail Entries

   Applications may extend the Concise Problem Details data item with
   additional entries to convey additional, application-specific

   Such new entries are allocated in the custom-problem-detail-entries
   slot and carry a nested map specific to that application.  The map
   key can be either an (absolute!)  URI (under control of the entity
   defining this extension) or an unsigned integer.  Only the latter
   needs to be registered (Section 6.2).

   Within the nested map, any number of attributes can be given for a
   single extension.  The semantics of each custom attribute MUST be
   described in the documentation for the extension; for extensions that
   are registered (i.e., are identified by an unsigned int), that
   documentation goes along with the registration.

   The unsigned integer form allows a more compact representation.  In
   exchange, authors are expected to comply with the required
   registration and documentation process.  In comparison, the URI form
   is less space efficient but requires no registration.  Therefore, it
   is useful for experimenting during the development cycle and for
   applications deployed in environments where producers and consumers
   of Concise Problem Details are more tightly integrated.  (Thus, the
   URI form covers the potential need we might otherwise have for a
   "Private Use" range for the unsigned integers.)

   Note that the URI given for the extension is for identification
   purposes only and, even if dereferenceable in principle, it MUST NOT
   be dereferenced in the normal course of handling problem details
   (i.e., outside diagnostic or debugging procedures involving humans).

   Figure 3 shows an example (in CBOR diagnostic notation) of a custom
   extension using a (made-up) URI as the custom-problem-detail-entries

     / title /         -1: "title of the error",
     / detail /        -2: "detailed information about the error",
     / instance /      -3: "coaps://pd.example/FA317434",
     / response-code / -4: 128, / 4.00 /

     "tag:3gpp.org,2022-03:TS29112": {
       / cause /  0: "machine-readable error cause",
       / invalidParams / 1: [
           / param / "first parameter name",
           / reason / "must be a positive integer"
           / param / "second parameter name"
       / supportedFeatures / 2: "d34db33f"

                  Figure 3: Example Extension with URI Key

   Obviously, a Standards Development Organization (SDO) like 3GPP can
   also easily register such a Custom Problem Detail entry to receive a
   more efficient unsigned integer key; Figure 4 shows how the same
   example would look using a (made-up) registered unsigned int as the
   custom-problem-detail-entries key:

     / title /         -1: "title of the error",
     / detail /        -2: "detailed information about the error",
     / instance /      -3: "coaps://pd.example/FA317434",
     / response-code / -4: 128, / 4.00 /

     /4711 is made-up example key that is not actually registered:/
     4711: {
       / cause /  0: "machine-readable error cause",
       / invalidParams / 1: [
           / param / "first parameter name",
           / reason / "must be a positive integer"
           / param / "second parameter name"
       / supportedFeatures / 2: "d34db33f"

       Figure 4: Example Extension with Unsigned Int (Registered) Key

   In summary, the keys for the maps used inside Custom Problem Detail
   entries are defined specifically for use with the identifier of that
   Custom Problem Detail entry, the documentation of which defines these
   internal entries, typically chosen to address a given application

   When there is a need to evolve a Custom Problem Detail entry
   definition, the "ignore-unknown rule" discussed in Section 3 provides
   an easy way to include additional information.  The assumption is
   that this is done in a backward- and forward-compatible way.
   Sometimes, Custom Problem Detail entries may need to evolve in a way
   where forward compatibility by applying the "ignore-unknown rule"
   would not be appropriate: for example, when adding a "must-
   understand" member, which can only be ignored at the peril of
   misunderstanding the Concise Problem Details data item ("false
   interoperability").  In this case, a new Custom Problem Detail key
   can simply be registered for this case, keeping the old key backward
   and forward compatible.

4.  Privacy Considerations

   Problem details may unintentionally disclose information.  This can
   lead to both privacy and security problems.  See Section 5 for more
   details that apply to both domains; particular attention needs to be
   given to unintentionally disclosing Personally Identifiable
   Information (PII).

5.  Security Considerations

   Concise Problem Details can contain URIs that are not intended to be
   dereferenced (Section 3.2, Paragraph 5).  One reason is that
   dereferencing these can lead to information disclosure (tracking).
   Information disclosure can also be caused by URIs in problem details
   that _are_ intended for dereferencing, e.g., the "instance" URI.
   Implementations need to consider which component of a client should
   perform the dereferencing and which servers are trusted with serving
   them.  In any case, the security considerations of Section 7 of RFC
   3986 [STD66] apply.

   The security and privacy considerations outlined in Section 5 of
   [RFC7807] apply in full.  While these are phrased in terms of
   security considerations for new RFC 7807 problem types, they equally
   apply to the problem detail entry definitions used here (Section 3).
   In summary, both when defining new detail entries and when actually
   using them to generate a Concise Problem Details data item, care
   needs to be taken that they do not leak sensitive information.
   Entities storing or forwarding Concise Problem Details data items
   need to consider whether this leads to information being transferred
   out of the context within which access to sensitive information was
   acceptable.  See also Section 3, Paragraph 6 (the last paragraph of
   the introduction to that section).  Privacy-sensitive information in
   the problem details SHOULD NOT be obscured in ways that might lead to
   misclassification as non-sensitive (e.g., by base64-encoding).

6.  IANA Considerations

6.1.  Standard Problem Detail Keys Registry

   This specification defines a new subregistry titled "Standard Problem
   Detail Keys" in the "Constrained RESTful Environments (CoRE)
   Parameters" registry [IANA.core-parameters], with "Specification
   Required" as the Registration Procedure (Section 4.6 of [RFC8126]).

   Each entry in the registry must include:

   Key Value:
      a negative integer to be used as the value of the key

      a name that could be used in implementations for the key

   CDDL Type:
      type of the data associated with the key in CDDL notation

   Brief Description:
      a brief description

      a reference document

   Change Controller:
      (see Section 2.3 of [RFC8126])

   The designated expert is requested to assign the shortest key values
   (1+0 and 1+1 encoding) to registrations that are likely to enjoy wide
   use and can benefit from short encodings.

   To be immediately useful in CDDL and programming-language contexts, a
   name consists of a lowercase ASCII letter (a-z) and zero or more
   additional ASCII characters that are either lowercase letters,
   digits, or a hyphen-minus, i.e., it matches [a-z][-a-z0-9]*. As with
   the key values, names need to be unique.

   The specification in the reference document needs to provide a
   description of the Standard Problem Detail entry, replicating the
   CDDL description in "CDDL Type", and describing the semantics of the
   presence of this entry and the semantics of the value given with it.

   Initial entries in this subregistry are as follows:

   |Key  |Name        |CDDL Type      |Brief      |Reference|Change    |
   |Value|            |               |Description|         |Controller|
   |-1   |title       |text / tag38   |Short,     |RFC 9290 |IETF      |
   |     |            |               |human-     |         |          |
   |     |            |               |readable   |         |          |
   |     |            |               |summary of |         |          |
   |     |            |               |the problem|         |          |
   |     |            |               |shape      |         |          |
   |-2   |detail      |text / tag38   |Human-     |RFC 9290 |IETF      |
   |     |            |               |readable   |         |          |
   |     |            |               |explanation|         |          |
   |     |            |               |specific to|         |          |
   |     |            |               |this       |         |          |
   |     |            |               |occurrence |         |          |
   |     |            |               |of the     |         |          |
   |     |            |               |problem    |         |          |
   |-3   |instance    |~uri           |URI        |RFC 9290 |IETF      |
   |     |            |               |reference  |         |          |
   |     |            |               |identifying|         |          |
   |     |            |               |specific   |         |          |
   |     |            |               |occurrence |         |          |
   |     |            |               |of the     |         |          |
   |     |            |               |problem    |         |          |
   |-4   |response-   |uint .size 1   |CoAP       |RFC 9290 |IETF      |
   |     |code        |               |response   |         |          |
   |     |            |               |code       |         |          |
   |-5   |base-uri    |~uri           |Base URI   |RFC 9290 |IETF      |
   |-6   |base-lang   |tag38-ltag     |Base       |RFC 9290 |IETF      |
   |     |            |               |language   |         |          |
   |     |            |               |tag (see   |         |          |
   |     |            |               |Appendix A)|         |          |
   |-7   |base-rtl    |tag38-direction|Base       |RFC 9290 |IETF      |
   |     |            |               |writing    |         |          |
   |     |            |               |direction  |         |          |
   |     |            |               |(see       |         |          |
   |     |            |               |Appendix A)|         |          |
   |-8   |unprocessed-|one-or-        |Option     |RFC 9290,|IETF      |
   |     |coap-option |more<uint>     |number(s)  |Section  |          |
   |     |            |               |of CoAP    |3.1.1    |          |
   |     |            |               |option(s)  |         |          |
   |     |            |               |that were  |         |          |
   |     |            |               |not        |         |          |
   |     |            |               |understood |         |          |

   Table 1: Initial Entries in the Standard Problem Detail Keys Registry

6.2.  Custom Problem Detail Keys Registry

   This specification defines a new subregistry titled "Custom Problem
   Detail Keys" in the "Constrained RESTful Environments (CoRE)
   Parameters" registry [IANA.core-parameters], with as "Expert Review"
   as the Registration Procedure (Section 4.5 of [RFC8126]).

   The designated expert is instructed to attempt making the
   registration experience as close to First Come First Served as
   reasonably achievable, but checking that the reference document does
   provide a description as set out below.  (This requirement is a
   relaxed version of "Specification Required" as defined in Section 4.6
   of [RFC8126].)

   Each entry in the registry must include:

   Key Value:
      an unsigned integer to be used as the value of the key

      a name that could be used in implementations for the key

   Brief Description:
      a brief description

      a reference document that provides a description of the map,
      including a CDDL description, that describes all inside keys and

   Change Controller
      (see Section 2.3 of [RFC8126])

   The designated expert is requested to assign the shortest key values
   (1+0 and 1+1 encoding) to registrations that are likely to enjoy wide
   use and can benefit from short encodings.

   To be immediately useful in CDDL and programming-language contexts, a
   name consists of a lowercase ASCII letter (a-z) and zero or more
   additional ASCII characters that are either lowercase letters,
   digits, or a hyphen-minus, i.e., it matches [a-z][-a-z0-9]*. As with
   the key values, names need to be unique.

   Initial entries in this subregistry are as follows:

   | Key   | Name        | Brief              | Reference | Change     |
   | Value |             | Description        |           | Controller |
   | 7807  | tunnel-7807 | Carry RFC 7807     | RFC 9290, | IETF       |
   |       |             | problem details    | Appendix  |            |
   |       |             | in a Concise       | B         |            |
   |       |             | Problem Details    |           |            |
   |       |             | data item          |           |            |

     Table 2: Initial Entries in the Custom Problem Detail Key Registry

6.3.  Media Type

   IANA has added the following media type to the "Media Types" registry

   |Name                        |Template                    |Reference|
   |concise-problem-details+cbor|application/concise-problem-|RFC 9290,|
   |                            |details+cbor                |Section  |
   |                            |                            |6.3      |

     Table 3: New Media Type 'application/concise-problem-details+cbor'

   Type name:  application

   Subtype name:  concise-problem-details+cbor

   Required parameters:  N/A

   Optional parameters:  N/A

   Encoding considerations:  binary (CBOR data item)

   Security considerations:  Section 5 of RFC 9290

   Interoperability considerations:  none

   Published specification:  Section 6.3 of RFC 9290

   Applications that use this media type:  Clients and servers in the
      Internet of Things

   Fragment identifier considerations:  The syntax and semantics of
      fragment identifiers is as specified for "application/cbor".  (At
      publication of RFC 9290, there is no fragment identification
      syntax defined for "application/cbor".)

   Additional information:

      Deprecated alias names for this type:  N/A
      Magic number(s):  N/A
      File extension(s):  N/A
      Macintosh file type code(s):  N/A

   Person & email address to contact for further information:  CoRE WG
      mailing list (core@ietf.org) or IETF Applications and Real-Time
      Area (art@ietf.org)

   Intended usage:  COMMON

   Restrictions on usage:  none

   Author/Change controller:  IETF

   Provisional registration:  no

6.4.  Content-Format

   IANA has registered a Content-Format number in the "CoAP
   Content-Formats" subregistry, within the "Constrained RESTful
   Environments (CoRE) Parameters" registry [IANA.core-parameters], as

   | Media Type                   | Encoding | ID  | Reference |
   | application/concise-problem- | -        | 257 | RFC 9290  |
   | details+cbor                 |          |     |           |

                Table 4: Content-Format Registration

6.5.  CBOR Tag 38

   In the registry "CBOR Tags" [IANA.cbor-tags], IANA has registered
   CBOR tag 38.  IANA has updated the reference for CBOR tag 38 to point
   to RFC 9290, Appendix A.

7.  References

7.1.  Normative References

              IANA, "Concise Binary Object Representation (CBOR) Tags",

              IANA, "Constrained RESTful Environments (CoRE)

              IANA, "Provisional Standard Media Type Registry",

   [RFC2119]  Bradner, S., "Key words for use in RFCs to Indicate
              Requirement Levels", BCP 14, RFC 2119,
              DOI 10.17487/RFC2119, March 1997,

   [RFC4647]  Phillips, A., Ed. and M. Davis, Ed., "Matching of Language
              Tags", BCP 47, RFC 4647, DOI 10.17487/RFC4647, September
              2006, <https://www.rfc-editor.org/info/rfc4647>.

   [RFC5646]  Phillips, A., Ed. and M. Davis, Ed., "Tags for Identifying
              Languages", BCP 47, RFC 5646, DOI 10.17487/RFC5646,
              September 2009, <https://www.rfc-editor.org/info/rfc5646>.

   [RFC7252]  Shelby, Z., Hartke, K., and C. Bormann, "The Constrained
              Application Protocol (CoAP)", RFC 7252,
              DOI 10.17487/RFC7252, June 2014,

   [RFC7807]  Nottingham, M. and E. Wilde, "Problem Details for HTTP
              APIs", RFC 7807, DOI 10.17487/RFC7807, March 2016,

   [RFC8126]  Cotton, M., Leiba, B., and T. Narten, "Guidelines for
              Writing an IANA Considerations Section in RFCs", BCP 26,
              RFC 8126, DOI 10.17487/RFC8126, June 2017,

   [RFC8174]  Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC
              2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174,
              May 2017, <https://www.rfc-editor.org/info/rfc8174>.

   [RFC8610]  Birkholz, H., Vigano, C., and C. Bormann, "Concise Data
              Definition Language (CDDL): A Notational Convention to
              Express Concise Binary Object Representation (CBOR) and
              JSON Data Structures", RFC 8610, DOI 10.17487/RFC8610,
              June 2019, <https://www.rfc-editor.org/info/rfc8610>.

   [RFC9165]  Bormann, C., "Additional Control Operators for the Concise
              Data Definition Language (CDDL)", RFC 9165,
              DOI 10.17487/RFC9165, December 2021,

   [STD66]    Berners-Lee, T., Fielding, R., and L. Masinter, "Uniform
              Resource Identifier (URI): Generic Syntax", STD 66,
              RFC 3986, January 2005.


   [STD94]    Bormann, C. and P. Hoffman, "Concise Binary Object
              Representation (CBOR)", STD 94, RFC 8949, December 2020.


7.2.  Informative References

   [HTTPAPI]  Nottingham, M., Wilde, E., and S. Dalal, "Problem Details
              for HTTP APIs", Work in Progress, Internet-Draft, draft-
              ietf-httpapi-rfc7807bis-04, 5 September 2022,

   [RDF]      Cyganiak, R., Wood, D., and M. Lanthaler, "RDF 1.1
              Concepts and Abstract Syntax", W3C Recommendation, 25
              February 2014,

   [RFC4648]  Josefsson, S., "The Base16, Base32, and Base64 Data
              Encodings", RFC 4648, DOI 10.17487/RFC4648, October 2006,

   [RFC6082]  Whistler, K., Adams, G., Duerst, M., Presuhn, R., Ed., and
              J. Klensin, "Deprecating Unicode Language Tag Characters:
              RFC 2482 is Historic", RFC 6082, DOI 10.17487/RFC6082,
              November 2010, <https://www.rfc-editor.org/info/rfc6082>.

              The Unicode Consortium, "The Unicode Standard, Version
              14.0.0", Mountain View: The Unicode Consortium,
              ISBN 978-1-936213-29-0, September 2021,

              The Unicode Consortium, "Unicode Standard Annex #9 ---
              Unicode Bidirectional Algorithm", 27 August 2021,

Appendix A.  Language-Tagged Strings

   This appendix serves as the archival documentation for CBOR tag 38, a
   tag for serializing language-tagged text strings in CBOR.  The text
   of this appendix is adapted from the specification text supplied for
   its initial registration.  It has been extended to allow
   supplementing the language tag by a direction indication.

   As with any IANA-registered item, a specification that further
   updates this registration needs to update the reference column of the
   IANA registry (see Section 6.5).  Future specifications may update
   this appendix, other parts of this document, or both.  (When updating
   this appendix, keep in mind that applications beyond Concise Problem
   Details data items may adopt the tag defined here.)  Users of this
   tag are advised to consult the registry to obtain the most recent
   update for this appendix.

A.1.  Introduction

   In some cases, it is useful to specify the natural language of a text
   string.  This specification defines a tag that does just that.  One
   technology that supports language-tagged strings is the Resource
   Description Framework (RDF) [RDF].

A.2.  Detailed Semantics

   A language-tagged text string in CBOR has the tag 38 and consists of
   an array with a length of 2 or 3.

   The first element is a well-formed language tag described in BCP 47
   ([RFC5646] and [RFC4647]), represented as a UTF-8 text string (major
   type 3).

   The second element is an arbitrary UTF-8 text string (major type 3).
   Both the language tag and the arbitrary string can optionally be
   annotated with CBOR tags; this is not shown in the CDDL below.

   The optional third element, if present, represents a ternary value
   that indicates a direction, as follows:

   *  false: left-to-right direction ("ltr").  The text is expected to
      be displayed with left-to-right base direction if standalone and
      isolated with left-to-right direction (as if enclosed in LRI ...
      PDI or equivalent, see [Unicode-14.0.0-bidi]) in the context of a
      longer string or text.

   *  true: right-to-left direction ("rtl").  The text is expected to be
      displayed with right-to-left base direction if standalone and
      isolated with right-to-left direction (as if enclosed in RLI ...
      PDI or equivalent, see [Unicode-14.0.0-bidi]) in the context of a
      longer string or text.

   *  null: indicates that no indication is made about the direction
      ("auto"), enabling an internationalization library to make an
      auto-detection decision such as treating the string as if enclosed
      in FSI ... PDI or equivalent, see [Unicode-14.0.0-bidi].

   If the third element is absent, directionality context may be applied
   (e.g., base-directionality information for an entire CBOR message or
   part thereof).  If there is no directionality context applied, the
   default interpretation is the same as for null ("auto").

   In CDDL:

   tag38 = #6.38([tag38-ltag, text, ?tag38-direction])
   tag38-ltag = text .regexp "[a-zA-Z]{1,8}(-[a-zA-Z0-9]{1,8})*"
   tag38-direction = &(ltr: false, rtl: true, auto: null)

   NOTE: Language tags of any combination of case are allowed.  But
   Section 2.1.1 of [RFC5646], part of BCP 47, recommends a case
   combination for language tags that encoders that support tag 38 may
   wish to follow when generating language tags.

   Data items with tag 38 that do not meet the criteria above are not
   valid (see Section 5.3.2 of RFC 8949 [STD94]).

   NOTE: The Unicode Standard [Unicode-14.0.0] includes a set of
   characters designed for tagging text (including language tagging) in
   the range U+E0000 to U+E007F.  Although many applications, including
   RDF, do not disallow these characters in text strings, the Unicode
   Consortium has deprecated these characters and recommends annotating
   language via a higher-level protocol instead.  See the section
   "Deprecated Tag Characters" in Section 23.9 of [Unicode-14.0.0] as
   well as [RFC6082].

   NOTE: while this document references a version of Unicode that was
   recent at the time of writing, the statements made based on this
   version are expected to remain valid for future versions.

A.3.  Examples

   Examples in this section are given in CBOR diagnostic notation first
   and then as a pretty-printed hexadecimal representation of the
   encoded item.

   The following example shows how the English-language string "Hello"
   is represented.

   38(["en", "Hello"])

   D8 26               # tag(38)
      82               # array(2)
         62            # text(2)
            656E       # "en"
         65            # text(5)
            48656C6C6F # "Hello"

   The following example shows how the French-language string "Bonjour"
   is represented.

   38(["fr", "Bonjour"])

   D8 26                   # tag(38)
      82                   # array(2)
         62                # text(2)
            6672           # "fr"
         67                # text(7)
            426F6E6A6F7572 # "Bonjour"

   The following example uses right-to-left (RTL) script, which in the
   context of this specification may be rendered differently by
   different document presentation environments.  The descriptive text
   may be more reliable to follow than the necessarily device- and
   application-specific rendering.  The example shows how the Hebrew-
   language string


   is represented, where in direction of reading, the sequence of
   characters is: "ש" (HEBREW LETTER SHIN, U+05E9), "ל" (HEBREW LETTER
   FINAL MEM, U+05DD).  Note the rtl direction expressed by setting the
   third element in the array to "true".

   38(["he", "שלום", true])

   D8 26                     # tag(38)
      83                     # array(3)
         62                  # text(2)
            6865             # "he"
         68                  # text(8)
            D7A9D79CD795D79D # "שלום"
         F5                  # primitive(21)

Appendix B.  Interworking with RFC 7807

   On certain occasions, it will be necessary to carry ("tunnel")
   [RFC7807] problem details in a Concise Problem Details data item.

   This appendix defines a Custom Problem Detail entry for that purpose.
   This is assigned Custom Problem Detail key 7807 in Section 6.2.  Its
   structure is:

   tunnel-7807 = {
     ? &(type: 0) => ~uri
     ? &(status: 1) => 0..999
     * text => any

   To carry an [RFC7807] problem details JSON object in a Concise
   Problem Details data item, first convert the JSON object to CBOR as
   per Section 6.2 of RFC 8949 [STD94].  Create an empty Concise Problem
   Details data item.

   Move the values for "title", "detail", and "instance", if present,
   from the [RFC7807] problem details to the equivalent Standard Problem
   Detail entries.  Create a Custom Problem Detail entry with key 7807.
   Move the values for "type" and "status", if present, to the
   equivalent keys 0 and 1 of the Custom Problem Detail entry.  Move all
   remaining key/value pairs (additional members as per Section 3.2 of
   [RFC7807]) in the converted [RFC7807] problem details object to the
   Custom Problem Detail map unchanged.

   The inverse direction, carrying Concise Problem Details in an RFC
   7807 problem details JSON object requires the additional support
   provided by [HTTPAPI], which is planned to create the HTTP Problem
   Types Registry.  An HTTP Problem Type can then be registered that
   extracts top-level items from the Concise Problem Details data item
   in a similar way to the conversion described above and that carries
   the rest of the Concise Problem Details data item in an additional
   member via base64url encoding without padding (Section 5 of
   [RFC4648]).  Details can be defined in a separate document when the
   work on [HTTPAPI] is completed.


   The authors would like to thank Mark Nottingham and Erik Wilde, the
   authors of RFC 7807; Klaus Hartke and Jaime Jiménez, the coauthors of
   an earlier draft version of this specification; Christian Amsüss,
   Marco Tiloca, Ari Keränen, and Michael Richardson for review and
   comments on this document.  Francesca Palombini for her review (and
   support) as responsible AD, and Joel Jaeggli for his OPSDIR review,
   both of which brought significant additional considerations to this

   For Appendix A, John Cowan and Doug Ewell are also to be
   acknowledged.  The content of an earlier draft version of this
   appendix was also discussed in the "apps-discuss@ietf.org" and
   "ltru@ietf.org" mailing lists.  More recently, the authors initiated
   a discussion about the handling of writing direction information in
   conjunction with language tags.  That led to discussions within the
   W3C Internationalization Core Working Group.  The authors would like
   to acknowledge that cross-organization cooperation and particular
   contributions from John Klensin and Addison Phillips and specific
   text proposals by Martin Dürst.


   Peter Occil
   Email: poccil14@gmail.com
   URI:   http://peteroupc.github.io/CBOR/

   Peter defined CBOR tag 38, basis of Appendix A.

   Christian Amsüss
   Email: christian@amsuess.com

   Christian contributed what became Section 3.1.1.

Authors' Addresses

   Thomas Fossati
   Arm Limited
   Email: thomas.fossati@arm.com

   Carsten Bormann
   Universität Bremen TZI
   Postfach 330440
   D-28359 Bremen
   Phone: +49-421-218-63921
   Email: cabo@tzi.org