RFC 4910






Network Working Group                                            S. Legg
Request for Comments: 4910                                       eB2Bcom
Category: Experimental                                         D. Prager
                                                               July 2007


                  Robust XML Encoding Rules (RXER) for
                  Abstract Syntax Notation One (ASN.1)

Status of This Memo



   This memo defines an Experimental Protocol for the Internet
   community.  It does not specify an Internet standard of any kind.
   Discussion and suggestions for improvement are requested.
   Distribution of this memo is unlimited.

Copyright Notice



   Copyright (C) The IETF Trust (2007).

Abstract



   This document defines a set of Abstract Syntax Notation One (ASN.1)
   encoding rules, called the Robust XML Encoding Rules or RXER, that
   produce an Extensible Markup Language (XML) representation for values
   of any given ASN.1 data type.  Rules for producing a canonical RXER
   encoding are also defined.
























Legg & Prager                 Experimental                      [Page 1]

RFC 4910               Robust XML Encoding Rules               July 2007


Table of Contents



   1. Introduction ....................................................3
   2. Conventions .....................................................4
   3. Definitions .....................................................5
   4. Additional Basic Types ..........................................6
      4.1. The Markup Type ............................................6
           4.1.1. Self-Containment ....................................9
           4.1.2. Normalization for Canonical Encoding Rules .........12
      4.2. The AnyURI Type ...........................................13
      4.3. The NCName Type ...........................................14
      4.4. The Name Type .............................................14
      4.5. The QName Type ............................................14
   5. Expanded Names for ASN.1 Types .................................15
   6. Encoding Rules .................................................17
      6.1. Identifiers ...............................................19
      6.2. Component Encodings .......................................20
           6.2.1. Referenced Components ..............................20
           6.2.2. Element Components .................................20
                  6.2.2.1. Namespace Properties for Elements .........22
                  6.2.2.2. Namespace Prefixes for Element Names ......24
           6.2.3. Attribute Components ...............................25
                  6.2.3.1. Namespace Prefixes for Attribute Names ....26
           6.2.4. Unencapsulated Components ..........................26
           6.2.5. Examples ...........................................27
      6.3. Standalone Encodings ......................................28
      6.4. Embedded ASN.1 Values .....................................28
      6.5. Type Referencing Notations ................................32
      6.6. TypeWithConstraint, SEQUENCE OF Type, and SET OF Type .....33
      6.7. Character Data Translations ...............................34
           6.7.1. Restricted Character String Types ..................35
           6.7.2. BIT STRING .........................................36
           6.7.3. BOOLEAN ............................................38
           6.7.4. ENUMERATED .........................................38
           6.7.5. GeneralizedTime ....................................39
           6.7.6. INTEGER ............................................41
           6.7.7. NULL ...............................................42
           6.7.8. ObjectDescriptor ...................................43
           6.7.9. OBJECT IDENTIFIER and RELATIVE-OID .................43
           6.7.10. OCTET STRING ......................................43
           6.7.11. QName .............................................44
                  6.7.11.1. Namespace Prefixes for Qualified Names ...44
           6.7.12. REAL ..............................................45
           6.7.13. UTCTime ...........................................46
           6.7.14. CHOICE as UNION ...................................47
           6.7.15. SEQUENCE OF as LIST ...............................50
      6.8. Combining Types ...........................................50
           6.8.1. CHARACTER STRING ...................................51



Legg & Prager                 Experimental                      [Page 2]

RFC 4910               Robust XML Encoding Rules               July 2007


           6.8.2. CHOICE .............................................51
           6.8.3. EMBEDDED PDV .......................................52
           6.8.4. EXTERNAL ...........................................52
           6.8.5. INSTANCE OF ........................................52
           6.8.6. SEQUENCE and SET ...................................52
           6.8.7. SEQUENCE OF and SET OF .............................54
           6.8.8. Extensible Combining Types .........................55
                  6.8.8.1. Unknown Elements in Extensions ............55
                  6.8.8.2. Unknown Attributes in Extensions ..........59
      6.9. Open Type .................................................60
      6.10. Markup ...................................................61
      6.11. Namespace Prefixes for CRXER .............................63
      6.12. Serialization ............................................65
           6.12.1. Non-Canonical Serialization .......................65
           6.12.2. Canonical Serialization ...........................68
           6.12.3. Unicode Normalization in XML Version 1.1 ..........70
      6.13. Syntax-Based Canonicalization ............................70
   7. Transfer Syntax Identifiers ....................................71
      7.1. RXER Transfer Syntax ......................................71
      7.2. CRXER Transfer Syntax .....................................71
   8. Relationship to XER ............................................71
   9. Security Considerations ........................................73
   10. Acknowledgements ..............................................74
   11. IANA Considerations ...........................................75
   12. References ....................................................75
      12.1. Normative References .....................................75
      12.2. Informative References ...................................77
   Appendix A. Additional Basic Definitions Module ...................78

1.  Introduction



   This document defines a set of Abstract Syntax Notation One (ASN.1)
   [X.680] encoding rules, called the Robust XML Encoding Rules or RXER,
   that produce an Extensible Markup Language (XML) [XML10][XML11]
   representation of ASN.1 values of any given ASN.1 type.

   An ASN.1 value is regarded as analogous to the content and attributes
   of an XML element, or in some cases, just an XML attribute value.
   The RXER encoding of an ASN.1 value is the well-formed and valid
   content and attributes of an element, or an attribute value, in an
   XML document [XML10][XML11] conforming to XML namespaces
   [XMLNS10][XMLNS11].  Simple ASN.1 data types such as PrintableString,
   INTEGER, and BOOLEAN define character data content or attribute
   values, while the ASN.1 combining types (i.e., SET, SEQUENCE, SET OF,
   SEQUENCE OF, and CHOICE) define element content and attributes.  The
   attribute and child element names are generally provided by the
   identifiers of the components in combining type definitions, i.e.,
   elements and attributes correspond to the NamedType notation.



Legg & Prager                 Experimental                      [Page 3]

RFC 4910               Robust XML Encoding Rules               July 2007


   RXER leaves some formatting details to the discretion of the encoder,
   so there is not a single unique RXER encoding for an ASN.1 value.
   However, this document also defines a restriction of RXER, called the
   Canonical Robust XML Encoding Rules (CRXER), which does produce a
   single unique encoding for an ASN.1 value.  Obviously, the CRXER
   encoding of a value is also a valid RXER encoding of that value.  The
   restrictions on RXER to produce the CRXER encoding are interspersed
   with the description of the rules for RXER.

   Note that "ASN.1 value" does not mean a Basic Encoding Rules (BER)
   [X.690] encoding.  The ASN.1 value is an abstract concept that is
   independent of any particular encoding.  BER is just one possible way
   to encode an ASN.1 value.  This document defines an alternative way
   to encode an ASN.1 value.

   A separate document [RXEREI] defines encoding instructions [X.680-1]
   that may be used in an ASN.1 specification to modify how values are
   encoded in RXER, for example, to encode a component of a combining
   ASN.1 type as an attribute rather than as a child element.  A
   pre-existing ASN.1 specification will not have RXER encoding
   instructions, so any mention of encoding instructions in this
   document can be ignored when dealing with such specifications.
   Encoding instructions for other encoding rules have no effect on RXER
   encodings.

2.  Conventions



   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", and "MAY" in this document are
   to be interpreted as described in BCP 14, RFC 2119 [BCP14].  The key
   word "OPTIONAL" is exclusively used with its ASN.1 meaning.

   A reference to an ASN.1 production [X.680] (e.g., Type, NamedType) is
   a reference to the text in an ASN.1 specification corresponding to
   that production.

   The specification of RXER makes use of definitions from the XML
   Information Set (Infoset) [INFOSET].  In particular, information item
   property names follow the Infoset convention of being shown in square
   brackets, e.g., [local name].  Literal values of Infoset properties
   are enclosed in double quotes; however, the double quotes are not
   part of the property values.  In the sections that follow,
   "information item" will be abbreviated to "item", e.g., "element
   information item" is abbreviated to "element item".  The term
   "element" or "attribute" (without the "item") is referring to an
   element or attribute in an XML document, rather than an information
   item.




Legg & Prager                 Experimental                      [Page 4]

RFC 4910               Robust XML Encoding Rules               July 2007


   Literal character strings to be used in an RXER encoding appear
   within double quotes; however, the double quotes are not part of the
   literal value and do not appear in the encoding.

   This document uses the namespace prefix [XMLNS10][XMLNS11] "asnx:" to
   stand for the namespace name "urn:ietf:params:xml:ns:asnx", uses the
   namespace prefix "xs:" to stand for the namespace name
   "http://www.w3.org/2001/XMLSchema", and uses the namespace prefix
   "xsi:" to stand for the namespace name
   "http://www.w3.org/2001/XMLSchema-instance".  However, in practice,
   any valid namespace prefixes are permitted in non-canonical RXER
   encodings (namespace prefixes are deterministically generated for
   CRXER).

   The encoding instructions [X.680-1] referenced by name in this
   specification are encoding instructions for RXER [RXEREI].

   Throughout this document, references to the Markup, AnyURI, NCName,
   Name, and QName ASN.1 types are references to the types described in
   Section 4 and consolidated in the AdditionalBasicDefinitions module
   in Appendix A.  Any provisions associated with the reference do not
   apply to types defined in other ASN.1 modules that happen to have
   these same names.

   Code points for characters [UCS][UNICODE] are expressed using the
   Unicode convention U+n, where n is four to six hexadecimal digits,
   e.g., the space character is U+0020.

3.  Definitions



   Definition (white space character): A white space character is a
   space (U+0020), tab (U+0009), carriage return (U+000D), or line feed
   (U+000A) character.

   Definition (white space):  White space is a sequence of one or more
   white space characters.

   Definition (line break):  A line break is any sequence of characters
   that is normalized to a line feed by XML End-of-Line Handling
   [XML10][XML11].

   Definition (serialized white space): Serialized white space is a
   sequence of one or more white space characters and/or line breaks.

   Definition (declaring the default namespace):  A namespace
   declaration attribute item is declaring the default namespace if the
   [prefix] of the attribute item has no value, the [local name] of the
   attribute item is "xmlns" and the [normalized value] is not empty.



Legg & Prager                 Experimental                      [Page 5]

RFC 4910               Robust XML Encoding Rules               July 2007


   Definition (undeclaring the default namespace):  A namespace
   declaration attribute item is undeclaring the default namespace if
   the [prefix] of the attribute item has no value, the [local name] of
   the attribute item is "xmlns" and the [normalized value] is empty
   (i.e., xmlns="").

   Definition (canonical namespace prefix): A canonical namespace prefix
   is an NCName [XMLNS10] beginning with the letter 'n' (U+006E)
   followed by a non-negative number string.  A non-negative number
   string is either the digit character '0' (U+0030), or a non-zero
   decimal digit character (U+0031-U+0039) followed by zero, one, or
   more of the decimal digit characters '0' to '9' (U+0030-U+0039).

   For convenience, a CHOICE type where the ChoiceType is subject to a
   UNION encoding instruction will be referred to as a UNION type, and a
   SEQUENCE OF type where the SequenceOfType is subject to a LIST
   encoding instruction will be referred to as a LIST type.

4.  Additional Basic Types



   This section defines an ASN.1 type for representing markup in
   abstract values, as well as basic types that are useful in encoding
   instructions [RXEREI] and other related specifications [ASN.X].

   The ASN.1 definitions in this section are consolidated in the
   AdditionalBasicDefinitions ASN.1 module in Appendix A.

4.1.  The Markup Type



   A value of the Markup ASN.1 type holds the [prefix], [attributes],
   [namespace attributes], and [children] of an element item, i.e., the
   content and attributes of an element.

   RXER has special provisions for encoding values of the Markup type
   (see Section 6.10).  For other encoding rules, a value of the Markup
   type is encoded according to the following ASN.1 type definition
   (with AUTOMATIC TAGS):

      Markup ::= CHOICE {
          text    SEQUENCE {
              prolog      UTF8String (SIZE(1..MAX)) OPTIONAL,
              prefix      NCName OPTIONAL,
              attributes  UTF8String (SIZE(1..MAX)) OPTIONAL,
              content     UTF8String (SIZE(1..MAX)) OPTIONAL
          }
      }





Legg & Prager                 Experimental                      [Page 6]

RFC 4910               Robust XML Encoding Rules               July 2007


   The text alternative of the Markup CHOICE type provides for the
   [prefix], [attributes], [namespace attributes], and [children] of an
   element item to be represented as serialized XML using the UTF-8
   character encoding [UTF-8].

      Aside: The CHOICE allows for one or more alternative compact
      representations of the content and attributes of an element to be
      supported in a future specification.

   With respect to some element item whose content and attributes are
   represented by a value of the text alternative of the Markup type:

   (1) the prolog component of the value contains text that, after line
       break normalization, conforms to the XML prolog production
       [XML10][XML11],

   (2) the prefix component is absent if the [prefix] of the element
       item has no value; otherwise, the prefix component contains the
       [prefix] of the element item,

   (3) the attributes component of the value contains an XML
       serialization of the [attributes] and [namespace attributes] of
       the element item, if any, with each attribute separated from the
       next by serialized white space, and

   (4) the content component is absent if the [children] property of the
       element item is empty; otherwise, the content component of the
       value contains an XML serialization of the [children] of the
       element item.

   All the components of a value of the Markup type MUST use the same
   version of XML, either version 1.0 [XML10] or version 1.1 [XML11].
   If XML version 1.1 is used, then the prolog component MUST be present
   and MUST have an XMLDecl for version 1.1.  If the prolog component is
   absent, then XML version 1.0 is assumed.

   If the prefix component is present, then there MUST be a namespace
   declaration attribute in the attributes component that defines that
   namespace prefix (since an element whose content and attributes are
   described by a value of Markup is required to be self-contained; see
   Section 4.1.1).

   Note that the prefix component is critically related to the NamedType
   that has Markup as its type.  If a Markup value is extracted from one
   enclosing abstract value and embedded in another enclosing abstract
   value (i.e., becomes associated with a different NamedType), then the
   prefix may no longer be appropriate, in which case it will need to be
   revised.  It may also be necessary to add another namespace



Legg & Prager                 Experimental                      [Page 7]

RFC 4910               Robust XML Encoding Rules               July 2007


   declaration attribute to the attributes component so as to declare a
   new namespace prefix.

   Leading and/or trailing serialized white space is permitted in the
   attributes component.  A value of the attributes component consisting
   only of serialized white space (i.e., no actual attributes) is
   permitted.

   The attributes and content components MAY contain entity references
   [XML10][XML11].  If any entity references are used (other than
   references to the predefined entities), then the prolog component
   MUST be present and MUST contain entity declarations for those
   entities in the internal or external subset of the document type
   definition.

   Example

      Given the following ASN.1 module:

         MyModule DEFINITIONS
         AUTOMATIC TAGS ::= BEGIN

         Message ::= SEQUENCE {
             messageType   INTEGER,
             messageValue  Markup
         }

         ENCODING-CONTROL RXER

             TARGET-NAMESPACE "http://example.com/ns/MyModule"

             COMPONENT message Message
                 -- a top-level NamedType

         END

      consider the following XML document:

         <?xml version='1.0'?>
         <!DOCTYPE message [
             <!ENTITY TRUE 'true'>
         ]>
         <message>
          <messageType>1</messageType>
          <messageValue xmlns:ns="http://www.example.com/ABD"
                        ns:foo="1" bar="0">
           <this>&TRUE;</this>
           <that/>



Legg & Prager                 Experimental                      [Page 8]

RFC 4910               Robust XML Encoding Rules               July 2007


          </messageValue>
         </message>

      A Markup value corresponding to the content and attributes of the
      <messageValue> element is, in ASN.1 value notation [X.680] (where
      "lf" represents the line feed character):

         text:{
             prolog     { "<?xml version='1.0'?>", lf,
                          "<!DOCTYPE message [", lf,
                          "    <!ENTITY TRUE 'true'>", lf,
                          "]>", lf },
             attributes { " xmlns:ns=""http://www.example.com/ABD""",
                          lf,
                          "               ns:foo=""1"" bar=""0""" },
             content    { lf,
                          "  <this>&TRUE;</this>", lf,
                          "  <that/>", lf, " " }
         }

      The following Markup value is an equivalent representation of the
      content and attributes of the <messageValue> element:

         text:{
             attributes {
                          "bar=""0"" ns:foo=""1"" ",
                          "xmlns:ns=""http://www.example.com/ABD""" },
             content    { lf,
                          "  <this>true</this>", lf,
                          "  <that/>", lf, " " }
         }

   By itself, the Markup ASN.1 type imposes no data type restriction on
   the markup contained by its values and is therefore analogous to the
   XML Schema anyType [XSD1].

   There is no ASN.1 basic notation that can directly impose the
   constraint that the markup represented by a value of the Markup type
   must conform to the markup allowed by a specific type definition.
   However, certain encoding instructions (i.e., the reference encoding
   instructions [RXEREI]) have been defined to have this effect.

4.1.1.  Self-Containment



   An element, its attributes and its content, including descendent
   elements, may contain qualified names [XMLNS10][XMLNS11] as the names
   of elements and attributes, in the values of attributes, and as
   character data content of elements.  The binding between namespace



Legg & Prager                 Experimental                      [Page 9]

RFC 4910               Robust XML Encoding Rules               July 2007


   prefix and namespace name for these qualified names is potentially
   determined by the namespace declaration attributes of ancestor
   elements (which in the Infoset representation are inherited as
   namespace items in the [in-scope namespaces]).

   In the absence of complete knowledge of the data type of an element
   item whose content and attributes are described by a value of the
   Markup type, it is not possible to determine with absolute certainty
   which of the namespace items inherited from the [in-scope namespaces]
   of the [parent] element item are significant in interpreting the
   Markup value.  The safe and easy option would be to assume that all
   the namespace items from the [in-scope namespaces] of the [parent]
   element item are significant and need to be retained within the
   Markup value.  When the Markup value is re-encoded, any of the
   retained namespace items that do not appear in the
   [in-scope namespaces] of the enclosing element item in the new
   encoding could be made to appear by outputting corresponding
   namespace declaration attribute items in the [namespace attributes]
   of the enclosing element item.

   From the perspective of the receiver of the new encoding, this
   enlarges the set of attribute items in the [namespace attributes]
   represented by the Markup value.

   In addition, there is no guarantee that the sender of the new
   encoding has recreated the original namespace declaration attributes
   on the ancestor elements, so the [in-scope namespaces] of the
   enclosing element item is likely to have new namespace declarations
   that the receiver will retain and pass on in the
   [namespace attributes] when it in turn re-encodes the Markup value.

   This unbounded growth in the set of attribute items in the
   [namespace attributes] defeats any attempt to produce a canonical
   encoding.

   The principle of self-containment is introduced to avoid this
   problem.  An element item (the subject element item) is
   self-contained if the constraints of Namespaces in XML 1.0 [XMLNS10]
   are satisfied (i.e., that prefixes are properly declared) and none of
   the following bindings are determined by a namespace declaration
   attribute item in the [namespace attributes] of an ancestor element
   item of the subject element item:

   (1) the binding between the [prefix] and [namespace name] of the
       subject element item,

   (2) the binding between the [prefix] and [namespace name] of any
       descendant element item of the subject element item,



Legg & Prager                 Experimental                     [Page 10]

RFC 4910               Robust XML Encoding Rules               July 2007


   (3) the binding between the [prefix] and [namespace name] of any
       attribute item in the [attributes] of the subject element item or
       the [attributes] of any descendant element item of the subject
       element item,

   (4) the binding between the namespace prefix and namespace name of
       any qualified name in the [normalized value] of any attribute
       item in the [attributes] of the subject element item or the
       [attributes] of any descendant element item of the subject
       element item, or

   (5) the binding between the namespace prefix and namespace name of
       any qualified name represented by a series of character items
       (ignoring processing instruction and comment items) in the
       [children] of the subject element item or the [children] of any
       descendant element item of the subject element item.

      Aside: If an element is self-contained, then separating the
      element from its parent does not change the semantic
      interpretation of its name and any names in its content and
      attributes.

   A supposedly self-contained element in a received RXER encoding that
   is in fact not self-contained SHALL be treated as an ASN.1 constraint
   violation.

      Aside: ASN.1 does not require an encoding with a constraint
      violation to be immediately rejected; however, the constraint
      violation must be reported at some point, possibly in a separate
      validation step.

   Implementors should note that an RXER decoder will be able to detect
   some, but not all, violations of self-containment.  For example, it
   can detect element and attribute names that depend on namespace
   declarations appearing in the ancestors of a supposedly
   self-contained element.  Similarly, where type information is
   available, it can detect qualified names in character data that
   depend on the namespace declarations of ancestor elements.  However,
   type information is not always available, so some qualified names
   will escape constraint checking.  Thus, the onus is on the creator of
   the original encoding to ensure that element items required to be
   self-contained really are completely self-contained.

   An element item whose content and attributes are described by a value
   of the Markup type MUST be self-contained.






Legg & Prager                 Experimental                     [Page 11]

RFC 4910               Robust XML Encoding Rules               July 2007


      Aside: The procedures in Section 6 take account of the
      requirements for self-containment so that an RXER encoder
      following these procedures will not create violations of
      self-containment.

4.1.2.  Normalization for Canonical Encoding Rules



   Implementations are given some latitude in how the content and
   attributes of an element are represented as an abstract value of the
   Markup type, in part because an Infoset can have different equivalent
   serializations.  For example, the order of attributes and the amount
   and kind of white space characters between attributes are irrelevant
   to the Infoset representation.  The content can also include one or
   more elements corresponding to an ASN.1 top-level NamedType or having
   a data type that is an ASN.1 type.  It is only necessary to preserve
   the abstract value for such elements, and a particular abstract value
   can have different Infoset representations.

   These two characteristics mean that when an RXER encoded value of the
   Markup type is decoded, the components of the recovered Markup value
   may not be exactly the same, character for character, as the original
   value that was encoded, though the recovered value will be
   semantically equivalent.

   However, canonical ASN.1 encoding rules such as the Distinguished
   Encoding Rules (DER) and the Canonical Encoding Rules (CER) [X.690],
   which encode Markup values according to the ASN.1 definition of the
   Markup type, depend on character-for-character preservation of string
   values.  This requirement can be accommodated if values of the Markup
   type are normalized when they are encoded according to a set of
   canonical encoding rules.

      Aside: The RXER encoding and decoding of a Markup value might
      change the character string components of the value from the
      perspective of BER, but there will be a single, repeatable
      encoding for DER.

   A value of the Markup type will appear as the content and attributes
   of an element in an RXER encoding.  When the value is encoded using a
   set of ASN.1 canonical encoding rules other than CRXER, the
   components of the text alternative of the value MUST be normalized as
   follows, by reference to the element as it would appear in a CRXER
   encoding:

   (1) The value of the prolog component SHALL be the XMLDecl
       <?xml version="1.1"?> with no other leading or trailing
       characters.




Legg & Prager                 Experimental                     [Page 12]

RFC 4910               Robust XML Encoding Rules               July 2007


   (2) If the element's name is unprefixed in the CRXER encoding, then
       the prefix component SHALL be absent; otherwise, the value of the
       prefix component SHALL be the prefix of the element's name in the
       CRXER encoding.

   (3) Take the character string representing the element's attributes,
       including namespace declarations, in the CRXER encoding.  If the
       first attribute is a namespace declaration that undeclares the
       default namespace (i.e., xmlns=""), then remove it.  Remove any
       leading space characters.  If the resulting character string is
       empty, then the attributes component SHALL be absent; otherwise,
       the value of the attributes component SHALL be the resulting
       character string.

          Aside: Note that the attributes of an element can change if an
          RXER encoding is re-encoded in CRXER.

   (4) If the element has no characters between the start-tag and
       end-tag [XML11] in the CRXER encoding, then the content component
       SHALL be absent; otherwise, the value of the content component
       SHALL be identical to the character string in the CRXER encoding
       bounded by the element's start-tag and end-tag.

      Aside: A consequence of invoking the CRXER encoding is that any
      nested element corresponding to an ASN.1 top-level NamedType, or
      indeed the element itself, will be normalized according to its
      ASN.1 value rather than its Infoset representation.  Likewise for
      an element whose data type is an ASN.1 type.  Section 6.4
      describes how these situations can arise.

      Aside: It is only through values of the Markup type that
      processing instructions and comments can appear in CRXER
      encodings.

   If an application uses DER, but has no knowledge of RXER, then it
   will not know to normalize values of the Markup type.  If RXER is
   deployed into an environment containing such applications, then
   Markup values SHOULD be normalized, even when encoding using
   non-canonical encoding rules.

4.2.  The AnyURI Type



   A value of the AnyURI ASN.1 type is a character string conforming to
   the format of a Uniform Resource Identifier (URI) [URI].

      AnyURI ::= UTF8String (CONSTRAINED BY
                  { -- conforms to the format of a URI -- })




Legg & Prager                 Experimental                     [Page 13]

RFC 4910               Robust XML Encoding Rules               July 2007


4.3.  The NCName Type



   A value of the NCName ASN.1 type is a character string conforming to
   the NCName production of Namespaces in XML 1.0 [XMLNS10].

      NCName ::= UTF8String (CONSTRAINED BY
                     { -- conforms to the NCName production of
                       -- Namespaces in XML 1.0 -- })

      Aside: The NCName production for Namespaces in XML 1.1 [XMLNS11]
      allows a wider range of characters than the NCName production for
      Namespaces in XML 1.0.  The NCName type for ASN.1 is currently
      restricted to the characters allowed by Namespaces in XML 1.0,
      though this may change in a future specification of RXER.

4.4.  The Name Type



   A value of the Name ASN.1 type is a character string conforming to
   the Name production of XML version 1.0 [XML10].

      Name ::= UTF8String (CONSTRAINED BY
                     { -- conforms to the Name production of XML -- })

4.5.  The QName Type



   A value of the QName ASN.1 type describes an expanded name [XMLNS10],
   which appears as a qualified name [XMLNS10] in an RXER encoding.

   RXER has special provisions for encoding values of the QName type
   (see Section 6.7.11).  For other encoding rules, a value of the Qname
   type is encoded according to the following ASN.1 type definition
   (with AUTOMATIC TAGS):

      QName ::= SEQUENCE {
          namespace-name  AnyURI OPTIONAL,
          local-name      NCName
      }

   The namespace-name component holds the namespace name of the expanded
   name.  If the namespace name of the expanded name has no value, then
   the namespace-name component is absent.

      Aside: A namespace name can be associated with ASN.1 types and
      top-level NamedType instances by using the TARGET-NAMESPACE
      encoding instruction.

   The local-name component holds the local name of the expanded name.




Legg & Prager                 Experimental                     [Page 14]

RFC 4910               Robust XML Encoding Rules               July 2007


5.  Expanded Names for ASN.1 Types



   A TypeAssignment in ASN.1 associates a typereference with a Type.
   For RXER and Abstract Syntax Notation X (ASN.X) [ASN.X], a
   TypeAssignment is also regarded as associating an expanded name
   [XMLNS10] with the Type.  The local name of the expanded name is the
   typereference on the left-hand side of the TypeAssignment.  If the
   target namespace [RXEREI] of the ASN.1 module in which the
   TypeAssignment is defined is not absent, then the namespace name of
   the expanded name is that target namespace; otherwise, the namespace
   name of the expanded name has no value.

   A Type that is a BuiltinType or ReferencedType that is one of the
   productions in Table 1 is regarded as a reference to a built-in ASN.1
   type.  These built-in types also have expanded names.  In each case,
   the local name of the expanded name is as indicated in Table 1, and
   the namespace name of the expanded name is
   "urn:ietf:params:xml:ns:asnx".

































Legg & Prager                 Experimental                     [Page 15]

RFC 4910               Robust XML Encoding Rules               July 2007


   Table 1: Local Names for Built-in Types

      +------------------------------------+-------------------+
      | ASN.1 Production                   | Local Name        |
      +====================================+===================+
      | BitStringType                      |                   |
      |    without a NamedBitList          | BIT-STRING        |
      +------------------------------------+-------------------+
      | BooleanType                        | BOOLEAN           |
      +------------------------------------+-------------------+
      | CharacterStringType                |                   |
      |    RestrictedCharacterStringType   |                   |
      |       BMPString                    | BMPString         |
      |       GeneralString                | GeneralString     |
      |       GraphicString                | GraphicString     |
      |       IA5String                    | IA5String         |
      |       ISO646String                 | ISO646String      |
      |       NumericString                | NumericString     |
      |       PrintableString              | PrintableString   |
      |       TeletexString                | TeletexString     |
      |       T61String                    | T61String         |
      |       UniversalString              | UniversalString   |
      |       UTF8String                   | UTF8String        |
      |       VideotexString               | VideotexString    |
      |       VisibleString                | VisibleString     |
      |    UnrestrictedCharacterStringType | CHARACTER-STRING  |
      +------------------------------------+-------------------+
      | EmbeddedPDVType                    | EMBEDDED-PDV      |
      | ExternalType                       | EXTERNAL          |
      +------------------------------------+-------------------+
      | IntegerType                        |                   |
      |    without a NamedNumberList       | INTEGER           |
      +------------------------------------+-------------------+
      | NullType                           | NULL              |
      | ObjectIdentifierType               | OBJECT-IDENTIFIER |
      | OctetStringType                    | OCTET-STRING      |
      | RealType                           | REAL              |
      | RelativeOIDType                    | RELATIVE-OID      |
      +------------------------------------+-------------------+
      | UsefulType                         |                   |
      |    GeneralizedTime                 | GeneralizedTime   |
      |    UTCTime                         | UTCTime           |
      |    ObjectDescriptor                | ObjectDescriptor  |
      +------------------------------------+-------------------+







Legg & Prager                 Experimental                     [Page 16]

RFC 4910               Robust XML Encoding Rules               July 2007


   When the expanded name for an ASN.1 type is used in an RXER encoding,
   it appears as a qualified name [XMLNS10][XMLNS11].  The namespace
   prefix for the qualified name is determined according to
   Section 6.7.11.1.

   If a compatible XML Schema translation of an ASN.1 specification is
   provided (see Section 6.4), then that schema SHOULD associate the
   same expanded name with the XML Schema translation of an ASN.1 type.

   Definition (namespace-qualified reference): An ASN.1 Type is a
   namespace-qualified reference if one of the following applies:

   (1) the Type is a typereference (not a DummyReference) or an
       ExternalTypeReference in a DefinedType in a ReferencedType, the
       ASN.1 module in which the referenced type is defined has a
       TARGET-NAMESPACE encoding instruction, the referenced type is not
       directly or indirectly an open type [X.681], and the referenced
       type is not directly or indirectly the Markup type (Section 4.1),
       or

   (2) the Type is a BuiltinType or ReferencedType that is one of the
       productions in Table 1.

   The type definition referenced by a namespace-qualified reference
   will have an expanded name with a value for the namespace name.

6.  Encoding Rules



   With respect to RXER, ASN.1 abstract values are uniformly regarded as
   analogous to the content and attributes of an element, or just an
   attribute value, not complete elements or attributes in their own
   right.  Elements and attributes in an RXER encoding are defined by
   ASN.1 NamedType notation.  Since elements are the fundamental
   discrete structures of an XML document, the notion of a NamedType
   having a value that can be encoded is useful for descriptive purposes
   (particularly for describing the RXER encoding of values of the ASN.1
   combining types).  There is no conceptual basis in X.680 [X.680] for
   talking about the value of a NamedType, or its encoding, so the
   terminology is introduced here.

   Definition (value of a NamedType):  An abstract value of the Type in
   a NamedType is also a value of that NamedType.  The RXER encoding of
   the value of a NamedType is the RXER encoding of the abstract value
   of the Type encapsulated according to the definition of that
   NamedType.






Legg & Prager                 Experimental                     [Page 17]

RFC 4910               Robust XML Encoding Rules               July 2007


   This document does not refer to a value of a NamedType as being an
   abstract value so as to remain consistent with X.680.  An abstract
   value is exclusively a value of an ASN.1 type.

   A complete ASN.1 encoding is traditionally the encoding of an
   abstract value, but it is more natural to think of an XML document as
   being the RXER encoding of a value of a NamedType (because an XML
   document has a single root element that contains all the other
   elements and attributes).  The ASN.1 basic notation does not allow a
   NamedType to appear on its own, outside of an enclosing combining
   type.  That is, the basic notation does not have a concept analogous
   to a global element or attribute definition.  However, an ASN.1
   specification may use an RXER encoding control section [RXEREI] to
   define global elements and attributes using the NamedType notation.
   A NamedType that is not contained in an ASN.1 type definition is
   called a top-level NamedType [RXEREI].  Thus, an RXER encoding would
   typically be described as the encoding of a value of a top-level
   NamedType.

   Section 6.2 describes how a value of a NamedType is encoded.
   Section 6.3 defines an alternative method for encoding the document
   element of an XML document when a top-level NamedType is not
   specified.  Section 6.4 describes how the encodings of ASN.1 values
   can be embedded in an XML document where the other parts of the
   document are validated by an XML Schema.

   The RXER encoding of an abstract value, or the encoding of a value of
   a NamedType, is described as a translation into a synthetic Infoset,
   which is then serialized as XML.  This separation has been chosen for
   descriptive convenience and is not intended to impose any particular
   architecture on RXER implementations.  An RXER encoder is free to
   encode an ASN.1 value directly to XML provided the result is
   equivalent to following the two stage procedure described in this
   document.

   The process of translating an abstract value into an Infoset is
   described as producing either:

   (1) a string of characters that either becomes part of the
       [normalized value] of an attribute item or becomes character
       items among the [children] of an enclosing element item, or

   (2) a collection of zero or more attribute items contributing to the
       [attributes] of an enclosing element item, plus a series of zero
       or more character, element, processing instruction (PI), or
       comment items contributing to the [children] of the enclosing
       element item.




Legg & Prager                 Experimental                     [Page 18]

RFC 4910               Robust XML Encoding Rules               July 2007


   NamedType notation in the ASN.1 specification controls whether the
   translation of an abstract value is encapsulated in an element item
   or in an attribute item.

   Sections 6.5 to 6.10 describe the translation of abstract values into
   an Infoset for each of the ASN.1 type notations.

   Section 6.11 describes post-processing of namespace prefixes for
   CRXER encodings.

   Section 6.12 specifies how the Infoset translation is serialized as
   XML.

   This specification assumes that the COMPONENTS OF transformation
   specified in X.680, Clause 24.4 [X.680] has already been applied to
   all relevant types.

   Examples of RXER encodings in the following sections use a <value>
   start-tag and </value> end-tag to hold attributes and delimit the
   content.  These start-tags and end-tags are for illustration only and
   are not part of the encoding of an abstract value.  In normal use,
   the name of the enclosing element is provided by the context of the
   type of the abstract value, e.g., a NamedType in an enclosing
   SEQUENCE type.

   An RXER decoder is a conforming XML processor [XML10][XML11].

6.1.  Identifiers



   An identifier, as defined in ASN.1 notation (Clause 11.3 of X.680
   [X.680]), is a character string that begins with a Latin lowercase
   letter (U+0061-U+007A) and is followed by zero, one or more Latin
   letters (U+0041-U+005A, U+0061-U+007A), decimal digits (U+0030-
   U+0039), and hyphens (U+002D).  A hyphen is not permitted to be the
   last character, and a hyphen is not permitted to be followed by
   another hyphen.  The case of letters in an identifier is always
   significant.

   ASN.1 identifiers are used for the [local name] of attribute and
   element items, and may also appear in the character data content of
   elements or the values of attributes.  RXER encoding instructions can
   be used to substitute an NCName [XMLNS10] for an identifier.









Legg & Prager                 Experimental                     [Page 19]

RFC 4910               Robust XML Encoding Rules               July 2007


6.2.  Component Encodings



   The translation of the value of a NamedType is the translation of the
   abstract value of the Type of the NamedType encapsulated according to
   the definition of that NamedType.  This section specifies the form of
   this encapsulation.

6.2.1.  Referenced Components



   A value of a NamedType that is subject to a COMPONENT-REF encoding
   instruction is translated as a value of the top-level NamedType
   referenced by the encoding instruction.

6.2.2.  Element Components



   A value of a NamedType that is not subject to an ATTRIBUTE,
   ATTRIBUTE-REF, GROUP, or SIMPLE-CONTENT encoding instruction is
   translated as an element item, either as a child element item added
   to the [children] of the enclosing element item or as the document
   element item added to the [children] and [document element] of the
   document item.  If the element item is a child element item, then the
   [parent] is the enclosing element item; otherwise, the [parent] is
   the document item.

   The [local name] of the element item is the local name of the
   expanded name of the NamedType (see [RXEREI]).

      Aside: If there are no NAME, ATTRIBUTE-REF, COMPONENT-REF,
      ELEMENT-REF, or REF-AS-ELEMENT encoding instructions, then the
      local name of the expanded name of a NamedType is the same as the
      identifier of the NamedType.

   If the namespace name of the expanded name has no value, then the
   [namespace name] of the element item has no value (i.e., the
   element's name is not namespace qualified); otherwise, the
   [namespace name] is the namespace name of the expanded name.

   If the type of the NamedType is directly or indirectly the Markup
   type, then the [in-scope namespaces] and [namespace attributes] of
   the element item are constructed as specified in Section 6.10;
   otherwise, the [in-scope namespaces] and [namespace attributes] of
   the element item are constructed as specified in Section 6.2.2.1.

   If the [namespace name] of the element item has no value, then the
   [prefix] of the element item has no value; else if the type of the
   NamedType is not directly or indirectly the Markup type, then the





Legg & Prager                 Experimental                     [Page 20]

RFC 4910               Robust XML Encoding Rules               July 2007


   [prefix] of the element item is determined as specified in
   Section 6.2.2.2; otherwise, the [prefix] is determined by the Markup
   value as specified in Section 6.10.

   The element item becomes the enclosing element item for the
   translation of the value of the Type of the NamedType.

   For a non-canonical RXER encoding, if the type of the NamedType is
   not directly or indirectly the Markup type, then PI and comment items
   MAY be added to the [children] of the element item (before or after
   any other items).  The element item becomes the [parent] for each PI
   and comment item.  These particular PI and comment items in a
   received RXER encoding MAY be discarded by an application.

      Aside: There is no provision for representing comments and PIs in
      ASN.1 abstract values of types other than the Markup type.  These
      items will be lost if the abstract value is re-encoded using a
      different set of encoding rules.

   For a non-canonical RXER encoding, an attribute item with the
   [local name] "type" and the [namespace name]
   "http://www.w3.org/2001/XMLSchema-instance" (i.e., xsi:type [XSD1])
   SHOULD be added to the [attributes] of the element item if the
   corresponding NamedType is subject to a TYPE-AS-VERSION encoding
   instruction and MAY be added to the [attributes] of the element item
   if the Type of the corresponding NamedType is a namespace-qualified
   reference (see Section 5).  The [prefix] of this attribute item is
   determined as specified in Section 6.2.3.1.  The [normalized value]
   of this attribute item is a qualified name for the expanded name of
   the referenced type, with the namespace prefix determined as
   specified in Section 6.7.11.1.  The element item is the
   [owner element] for the attribute item.

      Aside: Where a compatible XML Schema translation of the ASN.1
      specification has been provided, the xsi:type attribute indicates
      to an XML Schema validator which type definition it should use for
      validating the RXER encoding.

      Aside: An xsi:type attribute is generally not permitted in a CRXER
      encoding.  Section 6.4 describes some circumstances where it is
      required in a CRXER encoding.  An xsi:type attribute might also
      appear in a CRXER encoding if it is contained in a value of the
      Markup type.

   For a non-canonical RXER encoding, if the type of the NamedType is
   not directly or indirectly the Markup type, then attribute items with
   the [local name] "schemaLocation" or "noNamespaceSchemaLocation" and
   the [namespace name] "http://www.w3.org/2001/XMLSchema-instance"



Legg & Prager                 Experimental                     [Page 21]

RFC 4910               Robust XML Encoding Rules               July 2007


   [XSD1] MAY be added to the [attributes] of the element item.  The
   [prefix] for each of these attribute items is determined as specified
   in Section 6.2.3.1.  The [normalized value] of these attribute items
   MUST reference a compatible XML Schema translation of the ASN.1
   specification.  The element item is the [owner element] for the
   attribute items.

6.2.2.1.  Namespace Properties for Elements



   This section describes how the [in-scope namespaces] and
   [namespace attributes] of an element item are constructed when the
   content and attributes of the element item are not described by a
   value of the Markup type (otherwise, see Section 6.10).

   The [in-scope namespaces] property of the element item initially
   contains only the mandatory namespace item for the "xml" prefix
   [INFOSET].

   For a CRXER encoding, if the element item is not the
   [document element] of the document item and the [in-scope namespaces]
   property of the element item's [parent] contains a namespace item for
   the default namespace, then a namespace declaration attribute item
   that undeclares the default namespace (see Section 3) SHALL be added
   to the element item's [namespace attributes].

   Definition (default namespace restricted): With respect to an element
   item, the default namespace is restricted if:

   (1) the [namespace name] of the element item has no value (i.e., the
       element's name is not namespace qualified), or

   (2) the element item is the enclosing element item for a value of the
       UNION type where the member attribute will be required (see
       Section 6.7.14), or

   (3) the element item is the enclosing element item for a value of the
       QName type where the namespace-name component is absent (see
       Section 6.7.11).  This includes the case where the translation of
       the QName value is contained in the [normalized value] of an
       attribute item in the [attributes] of the element item.

   For a non-canonical RXER encoding, if the element item is not the
   [document element] of the document item and the [in-scope namespaces]
   property of the element item's [parent] contains a namespace item for
   the default namespace, then either:

   (1) that item is copied to the [in-scope namespaces] of the element
       item, or



Legg & Prager                 Experimental                     [Page 22]

RFC 4910               Robust XML Encoding Rules               July 2007


   (2) a namespace declaration attribute item that declares the default
       namespace is added to the element item's [namespace attributes]
       (the namespace name is the encoder's choice), and an equivalent
       namespace item is added to the [in-scope namespaces] of the
       element item, or

   (3) a namespace declaration attribute item that undeclares the
       default namespace is added to the element item's
       [namespace attributes].

   Options (1) and (2) SHALL NOT be used if the default namespace is
   restricted with respect to the element item.

   For a CRXER encoding, if the element item is not the
   [document element] of the document item and the element item is not
   required to be self-contained, then all the namespace items in the
   [in-scope namespaces] of the [parent], excluding the namespace item
   for the "xml" prefix and any namespace item for the default
   namespace, are copied to the [in-scope namespaces] of the element
   item.

   For a non-canonical RXER encoding, if the element item is not the
   [document element] of the document item and the element item is not
   required to be self-contained, then any subset (including none or
   all) of the namespace items in the [in-scope namespaces] of the
   [parent], excluding certain items, is copied to the
   [in-scope namespaces] of the element item.  The excluded items that
   MUST NOT be copied are:  the namespace item for the "xml" prefix, any
   namespace item for the default namespace, and any namespace item that
   matches the [prefix], but not the [namespace name], of a namespace
   item retained for the re-encoding of an unknown attribute item (see
   Section 6.8.8) or an unknown alternative of a UNION (see
   Section 6.7.14).

      Aside: The descriptive approach used by this document only allows
      a namespace prefix to be used by a new namespace item if it is not
      currently used by another namespace item in the
      [in-scope namespaces].  By not inheriting a namespace item, the
      prefix of that namespace is again available for reuse without fear
      of breaking an existing dependency on the prefix.

   Element items that are required to be self-contained inherit none of
   the namespace items in the [in-scope namespaces] of the [parent].

   Any namespace item that is retained for the re-encoding of an unknown
   attribute item (Section 6.8.8) or an unknown alternative of a UNION
   (Section 6.7.14) and which is not in the [in-scope namespaces] of the
   element item MUST be added to the [in-scope namespaces].  An



Legg & Prager                 Experimental                     [Page 23]

RFC 4910               Robust XML Encoding Rules               July 2007


   equivalent namespace declaration attribute item MUST be added to the
   [namespace attributes] of the element item.

   Definition (unused namespace prefix): A namespace prefix is unused if
   it does not match the [prefix] of any namespace item in the
   [in-scope namespaces] of the element item.

   For a non-canonical RXER encoding, if the type of the NamedType is
   not directly or indirectly the Markup type, then additional namespace
   declaration attribute items for currently unused namespace prefixes
   MAY be added to the [namespace attributes] of the element item.  An
   equivalent namespace item MUST be added to the [in-scope namespaces]
   of the element item for each additional namespace declaration
   attribute item.

   For a non-canonical RXER encoding, if the type of the NamedType is
   not directly or indirectly the Markup type, and the
   [in-scope namespaces] property of the element item does not contain a
   namespace item for the default namespace, and the default namespace
   is not restricted with respect to the element item, then a namespace
   declaration attribute item for the default namespace MAY be added to
   the [namespace attributes] of the element item, in which case an
   equivalent namespace item MUST be added to the [in-scope namespaces]
   of the element item.

   Whenever a namespace declaration attribute item is added to an
   element item's [namespace attributes], the [owner element] of the
   attribute item is set to the element item.

6.2.2.2.  Namespace Prefixes for Element Names



   This section describes how the [prefix] of an element item is
   determined when the element item has a value for its [namespace name]
   and the content and attributes of the element item are not described
   by a value of the Markup type (otherwise, see Section 6.10).

   For a CRXER encoding, if the [namespace name] of the element item has
   a value, then the [prefix] of the element item is any unused
   non-canonical namespace prefix unless the [in-scope namespaces]
   property of the element item contains a namespace item with the same
   [namespace name] as the element item.  In that case, the [prefix] of
   that namespace item SHALL be used as the [prefix] of the element
   item.

      Aside: These prefixes will be rewritten to canonical namespace
      prefixes during the final step in producing the Infoset
      translation (see Section 6.11).  Canonical namespace prefixes are
      not used here in the first instance because canonicalization



Legg & Prager                 Experimental                     [Page 24]

RFC 4910               Robust XML Encoding Rules               July 2007


      depends on knowing the final [namespace attributes] produced by
      encoding the abstract value of the type of the NamedType.  If an
      implementation looks ahead to determine this final set prior to
      translating the abstract value, then it can assign the appropriate
      canonical namespace prefix in this step and skip the rewriting
      step.

   For a non-canonical RXER encoding, if the [namespace name] has a
   value, then the [prefix] of the element item is any unused namespace
   prefix unless the [in-scope namespaces] property of the element item
   contains a namespace item with the same [namespace name] as the
   element item.  In that case, the [prefix] of that namespace item MAY
   be used as the [prefix] of the element item.  Note that the [prefix]
   of a namespace item for the default namespace has no value.

   If the [prefix] of the element item is an unused namespace prefix,
   then a namespace declaration attribute item associating the namespace
   prefix with the namespace name MUST be added to the
   [namespace attributes] of the element item, and a corresponding
   namespace item MUST be added to the [in-scope namespaces] of the
   element item.

      Aside: The [local name] of the namespace declaration attribute
      item is the same as the [prefix] of the element item, the
      [namespace name] of the attribute item is
      "http://www.w3.org/2000/xmlns/", and the [normalized value] of the
      attribute item is the same as the [namespace name] of the element
      item.  The namespace item has the same [prefix] and
      [namespace name] as the element item.

6.2.3.  Attribute Components



   A value of a NamedType subject to an ATTRIBUTE or ATTRIBUTE-REF
   encoding instruction is translated as an attribute item added to the
   [attributes] of the enclosing element item (which becomes the
   [owner element] of the attribute item).

   The [local name] of the attribute item is the local name of the
   expanded name of the NamedType (see [RXEREI]).

   If the namespace name of the expanded name has no value, then the
   [namespace name] of the attribute item has no value; otherwise, the
   [namespace name] is the namespace name of the expanded name.

   If the [namespace name] has a value, then the [prefix] of the
   attribute item is determined as specified in Section 6.2.3.1;
   otherwise, the [prefix] of the attribute item has no value.




Legg & Prager                 Experimental                     [Page 25]

RFC 4910               Robust XML Encoding Rules               July 2007


   The [normalized value] of the attribute item is the translation of
   the value of the Type of the NamedType.

   For completeness, the [specified] property is set to true, the
   [attribute type] has no value, and the value of the [references]
   property is set to unknown.

6.2.3.1.  Namespace Prefixes for Attribute Names



   This section applies when an attribute item with a value for its
   [namespace name] is added to the [attributes] of an element item.

   For a CRXER encoding, the [prefix] of the attribute item is any
   unused non-canonical namespace prefix unless the
   [in-scope namespaces] property of the [owner element] contains a
   namespace item with a value for the [prefix] (i.e., is not a
   namespace item for the default namespace) and the same
   [namespace name] as the attribute item.  In that case, the [prefix]
   of that namespace item SHALL be used as the [prefix] of the attribute
   item.

   For a non-canonical RXER encoding, the [prefix] of the attribute item
   is any unused namespace prefix unless the [in-scope namespaces]
   property of the [owner element] contains a namespace item with a
   value for the [prefix] and the same [namespace name] as the attribute
   item.  In that case, the [prefix] of that namespace item MAY be used
   as the [prefix] of the attribute item.

   If the [prefix] of the attribute item is an unused namespace prefix,
   then a namespace declaration attribute item associating the namespace
   prefix with the namespace name MUST be added to the
   [namespace attributes] of the [owner element], and a corresponding
   namespace item MUST be added to the [in-scope namespaces] of the
   [owner element].

6.2.4.  Unencapsulated Components



   A value of a NamedType subject to a GROUP or SIMPLE-CONTENT encoding
   instruction is translated as the value of the Type of the NamedType,
   i.e., without encapsulation in an element item or attribute item.
   Consequently, the enclosing element item for the translation of the
   value of the NamedType is also the enclosing element item for the
   translation of the value of the Type of the NamedType.








Legg & Prager                 Experimental                     [Page 26]

RFC 4910               Robust XML Encoding Rules               July 2007


6.2.5.  Examples



   Consider this type definition:

      CHOICE {
          one    [0] BOOLEAN,
          two    [1] [RXER:ATTRIBUTE] INTEGER,
          three  [2] [RXER:NAME AS "THREE"] OBJECT IDENTIFIER,
          four   [3] [RXER:ATTRIBUTE-REF {
                         namespace-name "http://www.example.com",
                         local-name     "foo" }] UTF8String,
          five   [4] [RXER:ELEMENT-REF {
                         namespace-name "http://www.example.com",
                         local-name     "bar" }] Markup,
          six    [5] [RXER:GROUP] SEQUENCE {
              seven  [0] [RXER:ATTRIBUTE] INTEGER,
              eight  [1] INTEGER
          }
      }

   The content and attributes of each of the following <value> elements
   are the RXER encoding of a value of the above type:

      <value>
       <one>true</one>
      </value>

      <value two="100"/>

      <value>
       <THREE>2.5.4.3</THREE>
      </value>

      <value xmlns:ex="http://www.example.com"
             ex:foo="a string"/>

      <value>
       <ex:bar xmlns:ex="http://www.example.com">another string</ex:bar>
      </value>

      <value seven="200">
       <eight>300</eight>
      </value>








Legg & Prager                 Experimental                     [Page 27]

RFC 4910               Robust XML Encoding Rules               July 2007


6.3.  Standalone Encodings



   A typical RXER encoding is the encoding of a value of a nominated
   top-level NamedType.  An abstract value MAY be encoded as an XML
   document without nominating an explicit top-level NamedType by
   invoking a Standalone RXER Encoding or Standalone CRXER Encoding.

   In a Standalone RXER Encoding or Standalone CRXER Encoding, the
   abstract value is encoded as the value of a notional NamedType where
   the identifier of the NamedType is "value" and the Type of the
   NamedType is the type of the abstract value.  The NamedType is
   assumed to be subject to no encoding instructions.

      Aside: Thus, the element item corresponding to the document
      element will have the [local name] "value" and no value for the
      [namespace name] and [prefix].

   If RXER is chosen as the transfer syntax in an EMBEDDED PDV value,
   then the data-value OCTET STRING SHALL contain a Standalone RXER
   encoding.

   If CRXER is chosen as the transfer syntax in an EMBEDDED PDV value,
   then the data-value OCTET STRING SHALL contain a Standalone CRXER
   encoding.

   If RXER is chosen as the transfer syntax in an EXTERNAL value, then
   the octet-aligned OCTET STRING or arbitrary BIT STRING SHALL contain
   a Standalone RXER encoding.

   If CRXER is chosen as the transfer syntax in an EXTERNAL value, then
   the octet-aligned OCTET STRING or arbitrary BIT STRING SHALL contain
   a Standalone CRXER encoding.

6.4.  Embedded ASN.1 Values



   The reference encoding instructions [RXEREI] allow XML Schema
   definitions to be referenced from an ASN.1 specification.  It is also
   possible to reference an ASN.1 type or top-level NamedType from an
   XML Schema definition or from an information item validated by an
   XML Schema wildcard.  The manner in which an XML Schema definition
   references an ASN.1 type or top-level NamedType has an effect on the
   CRXER encoding of a value of the type or top-level NamedType.

   This section also applies to XML Schema definitions that validate
   information items that are contained in a value of the Markup type.






Legg & Prager                 Experimental                     [Page 28]

RFC 4910               Robust XML Encoding Rules               July 2007


      Aside: So the document element of an XML document might be
      described by an XML Schema definition that at some point
      references an ASN.1 definition that uses a reference encoding
      instruction to reference another XML Schema definition that then
      references another ASN.1 definition, and so on.

   In each of the following cases, an element or attribute item is only
   permitted to be, or to encapsulate, an RXER Infoset translation of an
   ASN.1 value if an XML Schema element declaration or ASN.1 NamedType
   is known for the [parent] element item ([owner element] in the case
   of an attribute declaration), for the [parent] of the [parent]
   element item, and so on, to the document element of the XML document.
   This condition is not satisfied by a NamedType where the Type is
   directly or indirectly the Markup type and the NamedType is not
   subject to a reference encoding instruction.

      Aside: An element declaration becomes known for an element item
      through assessment [XSD1].  A NamedType becomes known for an
      element item through decoding.

      Aside: If an XML Schema element declaration or ASN.1 NamedType is
      not known for an element item, then the type of the element item
      and the type of every nested element item are treated as unknown.
      Although an xsi:type attribute definitively identifies the type of
      an element item even if an element declaration for the element
      item is not known, this attribute is generally optional in an RXER
      encoding and so cannot be relied upon when seen in isolation from
      an element declaration.  Although only top-level NamedType
      instances can have namespace-qualified names in the current RXER
      specification, a future version may allow nested NamedType
      instances to also have namespace-qualified names, in which case it
      will not necessarily be possible to distinguish a nested NamedType
      from a top-level NamedType without knowledge of the type of the
      [parent] element item.

   An ASN.1 type with an expanded name (Section 5) MAY be referenced by
   the type attribute of an XML Schema element declaration.  The
   reference takes the form of a qualified name for the expanded name.
   An element item validated by such an element declaration encapsulates
   the Infoset translation of an abstract value of the ASN.1 type.  The
   [namespace name] and [local name] of the element item are determined
   by the XML Schema element declaration.  The remaining properties are
   determined according to RXER.  The element item MUST be
   self-contained for a CRXER encoding.

      Aside: The element item is not required to be self-contained for a
      non-canonical RXER encoding.




Legg & Prager                 Experimental                     [Page 29]

RFC 4910               Robust XML Encoding Rules               July 2007


   A top-level NamedType MAY be referenced by the ref attribute of an
   XML Schema element declaration if the NamedType is not subject to an
   ATTRIBUTE encoding instruction.  The reference takes the form of a
   qualified name for the expanded name of the top-level NamedType
   [RXEREI].  An element item validated by such an element declaration
   is the Infoset translation of a value of the referenced top-level
   NamedType.  All the properties of the element item are determined
   according to RXER.  The element item MUST be self-contained for a
   CRXER encoding.

   A top-level NamedType MAY be referenced by the ref attribute of an
   XML Schema attribute declaration if the NamedType is subject to an
   ATTRIBUTE encoding instruction and the definition of the type of the
   NamedType does not depend on the QName type in any way.  An attribute
   item validated by such an attribute declaration is the Infoset
   translation of a value of the referenced top-level NamedType, except
   that whatever valid [prefix] is initially chosen for the attribute
   item MUST be preserved in any re-encoding.  The remaining properties
   of the attribute item are determined according to RXER.

      Aside: The exclusion of the QName type means that the attribute
      value is not dependent upon any namespace declarations of its
      parent element item.

   An element item that is validated by an XML Schema element
   declaration that has the ur-type (i.e., anyType) as its type
   definition MAY encapsulate the Infoset translation of a value of an
   ASN.1 type with an expanded name.  The [namespace name] and
   [local name] of the element item are determined by the XML Schema
   element declaration.  The remaining properties of the element item
   are determined according to RXER.  The [attributes] of the element
   item SHALL contain an attribute item with the [local name] "type" and
   the [namespace name] "http://www.w3.org/2001/XMLSchema-instance"
   (i.e., an xsi:type attribute).  The [prefix] of this attribute item
   is determined as specified in Section 6.2.3.1.  The
   [normalized value] of this attribute item is a qualified name for the
   expanded name of the ASN.1 type, with the namespace prefix determined
   as specified in Section 6.7.11.1.  The element item MUST be
   self-contained for a CRXER encoding.

   An element item that is validated by an XML Schema wildcard (i.e.,
   <xs:any/>) MAY be the Infoset translation of a value of a top-level
   NamedType that is not subject to an ATTRIBUTE encoding instruction
   and comes from an ASN.1 module with a target namespace [RXEREI] that
   satisfies the namespace constraint of the wildcard.  All the
   properties of the element item are determined according to RXER.  The
   element item MUST be self-contained for a CRXER encoding.




Legg & Prager                 Experimental                     [Page 30]

RFC 4910               Robust XML Encoding Rules               July 2007


   An attribute item that is validated by an XML Schema wildcard (i.e.,
   <xs:anyAttribute/>) MAY be the Infoset translation of a value of a
   top-level NamedType if the NamedType is subject to an ATTRIBUTE
   encoding instruction, comes from an ASN.1 module with a target
   namespace that satisfies the namespace constraint of the wildcard,
   and has a type that does not depend on the QName type in any way.
   Whatever valid [prefix] is initially chosen for the attribute item
   MUST be preserved in any re-encoding.  The remaining properties of
   the attribute item are determined according to RXER.

   No other mechanisms for referencing an ASN.1 type or top-level
   NamedType from a different XML schema language are supported in this
   version of RXER.  In particular, this excludes an ASN.1 type being
   used as the base type in an XML Schema derivation by extension or
   restriction, as a member type for an XML Schema union type, as an
   item type for an XML Schema list type, or as the type in an
   XML Schema attribute declaration.

   A fully conformant RXER implementation will understand both ASN.1 and
   XML Schema and will recognize the transitions between information
   items controlled by ASN.1 definitions and those controlled by
   XML Schema definitions.  However, a purely XML Schema validator used
   to assess the validity of an RXER encoding will perceive any
   reference to an ASN.1 type or top-level NamedType as an unresolved
   reference.  In order to enable such assessment, it is desirable to
   provide an XML Schema translation of the ASN.1 definitions being
   referenced from an XML Schema.  Although XML Schema and ASN.1 are
   broadly similar, they each have unique features that cannot be
   adequately expressed in the other language, so a semantically
   equivalent translation is not possible in the general case.
   Fortunately, to simply achieve successful assessment it is sufficient
   for the XML Schema translation of an ASN.1 specification to be
   compatible with that ASN.1 specification.  That is, the XML Schema
   translation MUST be constructed such that every correct RXER encoding
   is assessed as valid.  Although not ideal, it is acceptable for the
   XML Schema to assess some incorrect RXER encodings as also being
   valid (a conformant RXER decoder will, of course, reject such an
   encoding).

   The simplest compatible XML Schema translation of an ASN.1 module is
   one in which every type is equivalent to the XML Schema ur-type.  For
   example, given an ASN.1 type with the reference name MyType, a
   sufficient compatible XML Schema type definition is:








Legg & Prager                 Experimental                     [Page 31]

RFC 4910               Robust XML Encoding Rules               July 2007


      <xs:complexType name="MyType" mixed="true">
       <xs:sequence>
        <xs:any processContents="lax"
                 minOccurs="0" maxOccurs="unbounded"/>
       </xs:sequence>
       <xs:anyAttribute processContents="lax"/>
      </xs:complexType>

          OR

      <xs:complexType name="MyType">
       <xs:complexContent>
        <xs:extension base="xs:anyType"/>
       </xs:complexContent>
      </xs:complexType>

      Aside: Because of the possible presence of an asnx:context
      attribute (Section 6.8.8.1), it is easiest to assume that all
      ASN.1 types translate into XML Schema complex types.

   Given an ASN.1 top-level NamedType that is not subject to an
   ATTRIBUTE encoding instruction and has the reference name myElement,
   a sufficient compatible XML Schema element declaration is:

      <xs:element name="myElement"/>

   Given an ASN.1 top-level NamedType that is subject to an ATTRIBUTE
   encoding instruction and has the reference name myAttribute, a
   sufficient compatible XML Schema attribute declaration is:

      <xs:attribute name="myAttribute"/>

   An application specification that mixes ASN.1 and XML Schema is free
   to provide a stricter translation of its ASN.1 definitions; however,
   a more thorough treatment for translating an ASN.1 module into an
   XML Schema is out of scope for this document.

6.5.  Type Referencing Notations



   A value of a type with a defined type name is translated according to
   the type definition on the right-hand side of the type assignment for
   the type name.

   A value of a type denoted by the use of a parameterized type with
   actual parameters is translated according to the parameterized type
   with the DummyReferences [X.683] substituted with the actual
   parameters.




Legg & Prager                 Experimental                     [Page 32]

RFC 4910               Robust XML Encoding Rules               July 2007


   A value of a constrained type is translated as a value of the type
   without the constraint.  See X.680 [X.680] and X.682 [X.682] for the
   details of ASN.1 constraint notation.

   A prefixed type [X.680-1] associates an encoding instruction with a
   type.  A value of a prefixed type is translated as a value of the
   type without the prefix.

      Aside: This does not mean that RXER encoding instructions are
      ignored.  It is simply easier to describe their effects in
      relation to specific built-in types, rather than as the
      translation of a value of a prefixed type.

   A tagged type is a special case of a prefixed type.  A value of a
   tagged type is translated as a value of the type without the tag.
   ASN.1 tags do not appear in the XML encodings defined by this
   document.

   A value of a fixed type denoted by an ObjectClassFieldType is
   translated according to that fixed type (see Section 6.9 for the case
   of an ObjectClassFieldType denoting an open type).

   A value of a selection type is translated according to the type
   referenced by the selection type.  Note that component encoding
   instructions are not inherited by the type referenced by a selection
   type [RXEREI].

   A value of a type described by TypeFromObject notation [X.681] is
   translated according to the denoted type.

   A value of a type described by ValueSetFromObjects notation [X.681]
   is translated according to the governing type.

6.6.  TypeWithConstraint, SEQUENCE OF Type, and SET OF Type



   For the purposes of this document, a TypeWithConstraint is treated as
   if it were the parent type [X.680] (either a SEQUENCE OF or SET OF
   type).

   For example,

      SEQUENCE SIZE(1..MAX) OF SomeType

         is treated like

      SEQUENCE OF SomeType





Legg & Prager                 Experimental                     [Page 33]

RFC 4910               Robust XML Encoding Rules               July 2007


   Additionally, a "SEQUENCE OF Type" (including the case where it is
   the parent type for a TypeWithConstraint) is treated as if it were a
   "SEQUENCE OF NamedType", where the identifier of the NamedType is
   assumed to be "item".  Similarly, a "SET OF Type" (including the case
   where it is the parent type for a TypeWithConstraint) is treated as
   if it were a "SET OF NamedType", where the identifier of the
   NamedType is assumed to be "item".

   For example,

      SEQUENCE SIZE(1..MAX) OF SomeType

         is ultimately treated like

      SEQUENCE OF item SomeType

6.7.  Character Data Translations



   For the majority of ASN.1 built-in types, encodings of values of
   those types never have element content.  The encoding of a value of
   an ASN.1 combining type (except a UNION or LIST type) typically has
   element content.

   For those types that do not produce element content, the translation
   of an abstract value is described as a character string of ISO 10646
   characters [UCS].  This character data translation will be destined
   to become either part of the [normalized value] of an attribute item,
   or a series of character items in the [children] of an element item
   (which becomes the [parent] for the character items).  The case that
   applies is determined in accordance with Section 6.2.

   For a non-canonical RXER encoding, if the type of the abstract value
   is not directly or indirectly a restricted character string type, the
   NULL type, or a UNION type, then leading and/or trailing white space
   characters MAY be added to the character data translation.

      Aside: White space characters are significant in the encoding of a
      value of a restricted character string type, and a restricted
      character string type can be a member type of a UNION type.  The
      encoding of a NULL value produces no character data.

      Aside: Optional white space characters are not permitted in a
      CRXER encoding.

   For a non-canonical RXER encoding, if the type of the abstract value
   is directly or indirectly the AnyURI, NCName, or Name type, then
   leading and trailing white space characters MAY be added to the
   character data translation.



Legg & Prager                 Experimental                     [Page 34]

RFC 4910               Robust XML Encoding Rules               July 2007


      Aside: These types are indirectly a restricted character string
      type (UTF8String); however, their definitions exclude white space
      characters, so any white space characters appearing in an encoding
      are not part of the abstract value and can be safely ignored.
      This exception does not apply to other subtypes of a restricted
      character string type that happen to exclude white space
      characters.

6.7.1.  Restricted Character String Types



   The character data translation of a value of a restricted character
   string type is the sequence of characters in the string.

   Depending on the ASN.1 string type, and an application's internal
   representation of that string type, a character may need to be
   translated to or from the equivalent ISO 10646 character code [UCS].
   The NumericString, PrintableString, IA5String, VisibleString
   (ISO646String), BMPString, UniversalString, and UTF8String character
   encodings use the same character codes as ISO 10646.  For the
   remaining string types (GeneralString, GraphicString, TeletexString,
   T61String, and VideotexString), see X.680 [X.680].

   The null character (U+0000) is not a legal character for XML.  It is
   omitted from the character data translation of a string value.

   Certain other control characters are legal for XML version 1.1, but
   not for version 1.0.  If any string value contains these characters,
   then the RXER encoding must use XML version 1.1 (see Section 6.12).

   All white space characters in the RXER encoding of a value of a
   restricted character string type (excluding the AnyURI, NCName, and
   Name subtypes) are significant, i.e., part of the abstract value.

   Examples

      The content of each of the following <value> elements is the RXER
      encoding of an IA5String value:

         <value> Don't run with scissors! </value>

         <value>Markup (e.g., <value>) has to be escaped.</value>

         <value>Markup (e.g., <![CDATA[<value>]]>)
         has to be escaped. </value>







Legg & Prager                 Experimental                     [Page 35]

RFC 4910               Robust XML Encoding Rules               July 2007


6.7.2.  BIT STRING



   The character data translation of a value of the BIT STRING type is
   either a binary digit string, a hexadecimal digit string, or a list
   of bit names.

   A binary digit string is a sequence of zero, one, or more of the
   binary digit characters '0' and '1' (i.e., U+0030 and U+0031).  Each
   bit in the BIT STRING value is encoded as a binary digit in order
   from the first bit to the last bit.

   For a non-canonical RXER encoding, if the BIT STRING type has a
   NamedBitList, then trailing zero bits MAY be omitted from a binary
   digit string.

   A hexadecimal digit string is permitted if and only if the number of
   bits in the BIT STRING value is zero or a multiple of eight and the
   character data translation is destined for the [children] of an
   element item.

   A hexadecimal digit string is a sequence of zero, one, or more pairs
   of the hexadecimal digit characters '0'-'9', 'A'-'F', and 'a'-'f'
   (i.e., U+0030-U+0039, U+0041-U+0046 and U+0061-U+0066).  Each group
   of eight bits in the BIT STRING value is encoded as a pair of
   hexadecimal digits where the first bit is the most significant.  An
   odd number of hexadecimal digits is not permitted.  The characters
   'a'-'f' (i.e., U+0061-U+0066) SHALL NOT be used in the CRXER encoding
   of a BIT STRING value.  If a hexadecimal digit string is used, then
   the enclosing element's [attributes] MUST contain an attribute item
   with the [local name] "format", the [namespace name]
   "urn:ietf:params:xml:ns:asnx", and the [normalized value] "hex"
   (i.e., asnx:format="hex").  The [prefix] of the attribute item is
   determined as specified in Section 6.2.3.1.

      Aside: The hexadecimal digit string is intended to conform to the
      lexical representation of the XML Schema [XSD2] hexBinary data
      type.

   For a non-canonical RXER encoding, if the preconditions for using a
   hexadecimal digit string are satisfied, then a hexadecimal digit
   string MAY be used.

   A list of bit names is permitted if and only if the BIT STRING type
   has a NamedBitList and each '1' bit in the BIT STRING value has a
   corresponding identifier in the NamedBitList.

      Aside: ASN.1 does not require that an identifier be assigned for
      every bit.



Legg & Prager                 Experimental                     [Page 36]

RFC 4910               Robust XML Encoding Rules               July 2007


   A list of bit names is a sequence of names for the '1' bits in the
   BIT STRING value, in any order, each separated from the next by at
   least one white space character.  If the BitStringType is not subject
   to a VALUES encoding instruction, then each '1' bit in the BIT STRING
   value is represented by its corresponding identifier from the
   NamedBitList.  If the BitStringType is subject to a VALUES encoding
   instruction, then each '1' bit in the BIT STRING value is represented
   by the replacement name [RXEREI] for its corresponding identifier.

   For a CRXER encoding, if the BIT STRING type has a NamedBitList, then
   a binary digit string MUST be used, and trailing zero bits MUST be
   omitted from the binary digit string; else if the number of bits in
   the BIT STRING value is greater than or equal to 64, and the
   preconditions for using a hexadecimal digit string are satisfied,
   then a hexadecimal digit string MUST be used; otherwise, a binary
   digit string MUST be used.

      Aside: Because the asnx:format attribute adds an overhead to a
      hexadecimal encoding (including a namespace declaration for the
      "asnx" prefix), a bit string of less than 64 bits is more
      compactly encoded as a binary digit string.

   Examples

      Consider this type definition:

         BIT STRING { black(0), red(1), orange(2), yellow(3),
             green(4), blue(5), indigo(6), violet(7) }

      The content and attributes of each of the following <value>
      elements are an RXER encoding of the same abstract value:

         <value>  green violet  orange</value>

         <value> 001<!--Orange-->01001 </value>

         <value xmlns:asnx="urn:ietf:params:xml:ns:asnx"
                asnx:format="hex">
          29
         </value>

         <value>00101001</value>

      The final case contains the CRXER encoding of the abstract value.







Legg & Prager                 Experimental                     [Page 37]

RFC 4910               Robust XML Encoding Rules               July 2007


6.7.3.  BOOLEAN



   For a non-canonical RXER encoding, the character data translation of
   the BOOLEAN value TRUE is the string "true" or "1", at the encoder's
   discretion.  For a CRXER encoding, the character data translation of
   the BOOLEAN value TRUE is the string "true".

   For a non-canonical RXER encoding, the character data translation of
   the BOOLEAN value FALSE is the string "false" or "0", at the
   encoder's discretion.  For a CRXER encoding, the character data
   translation of the BOOLEAN value FALSE is the string "false".

      Aside: The RXER encoding of BOOLEAN values is intended to conform
      to the lexical representation of the XML Schema [XSD2] boolean
      data type.

   Examples

      The content of each of the following <value> elements is the RXER
      encoding of a BOOLEAN value:

         <value>1</value>

         <value>
             false
         </value>

         <value> fal<!-- a pesky comment -->se </value>

6.7.4.  ENUMERATED



   The character data translation of a value of an ENUMERATED type where
   the EnumeratedType is not subject to a VALUES encoding instruction is
   the identifier corresponding to the actual value.

   Examples

      Consider this type definition:

         ENUMERATED { sunday, monday, tuesday,
             wednesday, thursday, friday, saturday }

      The content of both of the following <value> elements is the RXER
      encoding of a value of the above type:







Legg & Prager                 Experimental                     [Page 38]

RFC 4910               Robust XML Encoding Rules               July 2007


         <value>monday</value>

         <value>
             thursday
         </value>

   The character data translation of a value of an ENUMERATED type where
   the EnumeratedType is subject to a VALUES encoding instruction is the
   replacement name [RXEREI] for the identifier corresponding to the
   actual value.

   Examples

      Consider this type definition:

         [RXER:VALUES ALL CAPITALIZED,
                 sunday AS "SUNDAY", saturday AS "SATURDAY"]
             ENUMERATED { sunday, monday, tuesday,
                 wednesday, thursday, friday, saturday }

      The content of each of the following <value> elements is the RXER
      encoding of a value of the above type:

         <value>SUNDAY</value>

         <value>
             Monday
         </value>

         <value> Tuesday </value>

6.7.5.  GeneralizedTime



   The character data translation of a value of the GeneralizedTime type
   is a date, the letter 'T' (U+0054), a time of day, optional
   fractional seconds, and an optional time zone.

   The date is two decimal digits representing the century, followed by
   two decimal digits representing the year, a hyphen ('-', U+002D), two
   decimal digits representing the month, a hyphen ('-', U+002D), and
   two decimal digits representing the day.

   The time of day is two decimal digits representing the hour, followed
   by a colon (':', U+003A), two decimal digits representing the
   minutes, a colon (':', U+003A), and two decimal digits representing
   the seconds.

   Note that the hours value "24" is disallowed [X.680].



Legg & Prager                 Experimental                     [Page 39]

RFC 4910               Robust XML Encoding Rules               July 2007


   A GeneralizedTime value with fractional hours or minutes is first
   converted to the equivalent time with whole minutes and seconds and,
   if necessary, fractional seconds.

   The minutes are encoded as "00" if the GeneralizedTime value omits
   minutes.  The seconds are encoded as "00" if the GeneralizedTime
   value omits seconds.

   The fractional seconds part is a full stop ('.', U+002E) followed by
   zero, one, or more decimal digits (U+0030-U+0039).  For a CRXER
   encoding, trailing zero digits (U+0030) in the fractional seconds
   SHALL be omitted, and the full stop SHALL be omitted if there are no
   following digits.

   The time zone, if present, is either the letter 'Z' (U+005A) to
   indicate Coordinated Universal Time, a plus sign ('+', U+002B)
   followed by a time zone differential, or a minus sign ('-', U+002D)
   followed by a time zone differential.

   A time zone differential indicates the difference between local time
   (the time specified by the preceding date and time of day) and
   Coordinated Universal Time.  Coordinated Universal Time can be
   calculated from the local time by subtracting the differential.

   For a CRXER encoding, a GeneralizedTime value with a time zone
   differential SHALL be encoded as the equivalent Coordinated Universal
   Time, i.e., the time zone will be "Z".

   A local time GeneralizedTime value is not converted to Coordinated
   Universal Time for a CRXER encoding.  Other canonical ASN.1 encoding
   rules specify that local times must be encoded as Coordinated
   Universal Time but do not specify a method to convert a local time to
   a Coordinated Universal Time.  Consequently, canonicalization of
   local time values is unreliable and applications SHOULD NOT use local
   time.

   A time zone differential is encoded as two decimal digits
   representing hours, a colon (':', U+003A), and two decimal digits
   representing minutes.  The minutes are encoded as "00" if the
   GeneralizedTime value omits minutes from the time zone differential.

      Aside: The RXER encoding of GeneralizedTime values is intended to
      conform to the lexical representation of the XML Schema [XSD2]
      dateTime data type.







Legg & Prager                 Experimental                     [Page 40]

RFC 4910               Robust XML Encoding Rules               July 2007


   Examples

      The content of each of the following <value> elements is the RXER
      encoding of a GeneralizedTime value:

         <value>2004-06-15T12:00:00Z</value>

         <value> 2004-06-15T02:00:00+10:00 </value>

         <value>
             2004-06-15T12:00:00.5
         </value>

6.7.6.  INTEGER



   For a CRXER encoding, the character data translation of a value of an
   IntegerType is a canonical number string representing the integer
   value.

   A canonical number string is either the digit character '0' (U+0030),
   or an optional minus sign ('-', U+002D) followed by a non-zero
   decimal digit character (U+0031-U+0039) followed by zero, one, or
   more of the decimal digit characters '0' to '9' (U+0030-U+0039).

   For a non-canonical RXER encoding, the character data translation of
   a value of the IntegerType without a NamedNumberList is a number
   string representing the integer value.

   A number string is a sequence of one or more of the decimal digit
   characters '0' to '9' (U+0030-U+0039), with an optional leading sign,
   either '+' (U+002B) or '-' (U+002D).  Leading zero digits are
   permitted in a number string for a non-canonical RXER encoding.

      Aside: The RXER encoding of values of the IntegerType without a
      NamedNumberList is intended to conform to the lexical
      representation of the XML Schema [XSD2] integer data type.

   For a non-canonical RXER encoding, if the IntegerType has a
   NamedNumberList, and the NamedNumberList defines an identifier for
   the actual value, and the IntegerType is not subject to a VALUES
   encoding instruction, then the character data translation of the
   value is either a number string or the identifier.

   Examples

      Consider this type definition:

         INTEGER { zero(0), one(1) }



Legg & Prager                 Experimental                     [Page 41]

RFC 4910               Robust XML Encoding Rules               July 2007


      The content of each of the following <value> elements is the RXER
      encoding of a value of the above type:

         <value>0</value>

         <value> zero </value>

         <value> 2 <!-- This number doesn't have a name. --> </value>

         <value>00167</value>

   For a non-canonical RXER encoding, if the IntegerType is subject to a
   VALUES encoding instruction (it necessarily must have a
   NamedNumberList) and the NamedNumberList defines an identifier for
   the actual value, then the character data translation of the value is
   either a number string or the replacement name [RXEREI] for the
   identifier.

   Examples

      Consider this type definition:

         [RXER:VALUES ALL UPPERCASED] INTEGER { zero(0), one(1) }

      The content of both of the following <value> elements is the RXER
      encoding of a value of the above type:

         <value>0</value>

         <value> ZERO </value>

6.7.7.  NULL



   The character data translation of a value of the NULL type is an
   empty character string.

   Examples

      <value/>

      <value><!-- Comments don't matter. --></value>

      <value></value>

      The final case is the CRXER encoding.






Legg & Prager                 Experimental                     [Page 42]

RFC 4910               Robust XML Encoding Rules               July 2007


6.7.8.  ObjectDescriptor



   A value of the ObjectDescriptor type is translated according to the
   GraphicString type.

6.7.9.  OBJECT IDENTIFIER and RELATIVE-OID



   The character data translation of a value of the OBJECT IDENTIFIER or
   RELATIVE-OID type is a full stop ('.', U+002E) separated list of the
   object identifier components of the value.

   Each object identifier component is translated as a non-negative
   number string.  A non-negative number string is either the digit
   character '0' (U+0030), or a non-zero decimal digit character
   (U+0031-U+0039) followed by zero, one, or more of the decimal digit
   characters '0' to '9' (U+0030-U+0039).

   Examples

      The content of each of the following <value> elements is the RXER
      encoding of an OBJECT IDENTIFIER value:

         <value>2.5.6.0</value>

         <value>
             2.5.4.10
         </value>

         <value> 2.5.4.3 <!-- commonName --> </value>

6.7.10.  OCTET STRING



   The character data translation of a value of the OCTET STRING type is
   the hexadecimal digit string representation of the octets.

   The octets are encoded in order from the first octet to the last
   octet.  Each octet is encoded as a pair of the hexadecimal digit
   characters '0'-'9', 'A'-'F', and 'a'-'f' (i.e., U+0030-U+0039,
   U+0041-U+0046, and U+0061-U+0066) where the first digit in the pair
   corresponds to the four most significant bits of the octet.  An odd
   number of hexadecimal digits is not permitted.  The characters 'a'-
   'f' (i.e., U+0061-U+0066) SHALL NOT be used in the CRXER encoding of
   an OCTET STRING value.

      Aside: The RXER encoding of OCTET STRING values is intended to
      conform to the lexical representation of the XML Schema [XSD2]
      hexBinary data type.




Legg & Prager                 Experimental                     [Page 43]

RFC 4910               Robust XML Encoding Rules               July 2007


   Examples

      The content of each of the following <value> elements is the RXER
      encoding of an OCTET STRING value:

         <value>27F69A0300</value>

         <value>
             efA03bFF
         </value>

6.7.11.  QName



   The character data translation of a value of the QName type
   (Section 4.5) is a qualified name conforming to the QName production
   of Namespaces in XML 1.0 [XMLNS10].

   The local part (i.e., LocalPart) of the qualified name SHALL be the
   value of the local-name component of the QName value.

   If the namespace-name component of the QName value is absent, then
   the namespace prefix (i.e., Prefix) of the qualified name SHALL be
   absent; otherwise, the namespace prefix is determined as specified in
   Section 6.7.11.1 using the value of the namespace-name component of
   the QName value as the namespace name.

6.7.11.1.  Namespace Prefixes for Qualified Names



   This section describes how the namespace prefix of a qualified name
   is determined given the namespace name to which the namespace prefix
   must map.

   For a CRXER encoding, the namespace prefix of the qualified name is
   any unused non-canonical namespace prefix unless the
   [in-scope namespaces] property of the enclosing element item contains
   a namespace item with a [namespace name] that matches the namespace
   name.  In that case, the [prefix] of that namespace item SHALL be
   used as the namespace prefix of the qualified name.

      Aside: If the qualified name appears in the [normalized value] of
      an attribute item, then the enclosing element item is the
      [owner element] for that attribute item.

   For a non-canonical RXER encoding, the namespace prefix of the
   qualified name is any unused namespace prefix unless the
   [in-scope namespaces] property of the enclosing element item contains
   a namespace item with the same [namespace name] as the element item.
   In that case, the [prefix] of that namespace item MAY be used as the



Legg & Prager                 Experimental                     [Page 44]

RFC 4910               Robust XML Encoding Rules               July 2007


   namespace prefix of the qualified name.  Note that the [prefix] of a
   namespace item for the default namespace has no value.

   If the namespace prefix of the qualified name is an unused namespace
   prefix, then a namespace declaration attribute item associating the
   namespace prefix with the namespace name MUST be added to the
   [namespace attributes] of the enclosing element item, and a
   corresponding namespace item MUST be added to the
   [in-scope namespaces] of the enclosing element item.

6.7.12.  REAL



   The character data translation of a value of the REAL type is the
   character string "0" if the value is positive zero, the character
   string "-0" if the value is negative zero, the character string "INF"
   if the value is positive infinity, the character string "-INF" if the
   value is negative infinity, the character string "NaN" if the value
   is not a number, or a real number otherwise.

   A real number is the mantissa followed by either the character 'E'
   (U+0045) or 'e' (U+0065) and the exponent.  The character 'e' SHALL
   NOT
be used for a CRXER encoding.  If the exponent is zero, then the
   'E' or 'e' and exponent MAY be omitted for a non-canonical RXER
   encoding.

   The mantissa is a decimal number with an optional leading sign,
   either '+' (U+002B) or '-' (U+002D).  A decimal number is a sequence
   of one or more of the decimal digit characters '0' to '9'
   (U+0030-U+0039) optionally partitioned by a single full stop
   character ('.', U+002E) representing the decimal point.  Multiple
   leading zero digits are permitted for a non-canonical RXER encoding.

   The exponent is encoded as a number string (see Section 6.7.6).

      Aside: The RXER encoding of REAL values is intended to be
      compatible with the lexical representation of the XML Schema
      [XSD2] double data type, but allows real values outside the set
      permitted by double.

   For a CRXER encoding:

   (1) The real number MUST be normalized so that the mantissa has a
       single non-zero digit immediately to the left of the decimal
       point.

   (2) Leading zero digits SHALL NOT be used.





Legg & Prager                 Experimental                     [Page 45]

RFC 4910               Robust XML Encoding Rules               July 2007


   (3) A leading plus sign SHALL NOT be used in the mantissa or the
       exponent.

   (4) The fractional part of the mantissa (i.e., that part following
       the decimal point) MUST have at least one digit (which may be
       '0') and MUST NOT have any trailing zeroes after the first digit.

   (5) The exponent SHALL be present and SHALL be a canonical number
       string (see Section 6.7.6).

   Examples

      The content of each of the following <value> elements is the RXER
      encoding of a REAL value:

         <value>3.14159<!-- pi --></value>

         <value> 1.0e6 </value>

         <value> INF </value>

         <value>
             -01e-06
         </value>

6.7.13.  UTCTime



   The character data translation of a value of the UTCTime type is a
   date, the letter 'T' (U+0054), a time of day, and a time zone.

   The date is two decimal digits representing the year (no century), a
   hyphen ('-', U+002D), two decimal digits representing the month, a
   hyphen ('-', U+002D), and two decimal digits representing the day.

   The time of day is two decimal digits representing the hour, followed
   by a colon (':', U+003A), two decimal digits representing the
   minutes, a colon (':', U+003A), and two decimal digits representing
   the seconds.

   Note that the hours value "24" is disallowed [X.680].

   The seconds are encoded as "00" if the UTCTime value omits seconds.

   The time zone is either the letter 'Z' (U+005A) to indicate
   Coordinated Universal Time, a plus sign ('+', U+002B) followed by a
   time zone differential, or a minus sign ('-', U+002D) followed by a
   time zone differential.




Legg & Prager                 Experimental                     [Page 46]

RFC 4910               Robust XML Encoding Rules               July 2007


   A time zone differential indicates the difference between local time
   (the time specified by the preceding date and time of day) and
   Coordinated Universal Time.  Coordinated Universal Time can be
   calculated from the local time by subtracting the differential.

   For a CRXER encoding, a UTCTime value with a time zone differential
   SHALL be encoded as the equivalent Coordinated Universal Time, i.e.,
   the time zone will be "Z".

   A time zone differential is encoded as two decimal digits
   representing hours, a colon (':', U+003A), and two decimal digits
   representing minutes.

6.7.14.  CHOICE as UNION



   The chosen alternative of a value of a UNION type corresponds to some
   NamedType in the UNION type definition (a ChoiceType).

   The character data translation of a value of a UNION type is the
   character data translation of the value of the type of the chosen
   alternative, i.e., without any kind of encapsulation.

   Leading and trailing white space characters are not permitted to be
   added to the character data translation of a value of a UNION type
   (see Section 6.7); however, this does not preclude such white space
   being added to the character data translation of the value of the
   chosen alternative.

   The character data translation of a value of a UNION type is
   necessarily destined for the [children] of an enclosing element item.

      Aside: This is because the ATTRIBUTE encoding instruction cannot
      be applied to a NamedType with a type that is a UNION type.

   The chosen alternative can be identified by a member attribute item,
   i.e., an attribute item with the [local name] "member" and
   [namespace name] "urn:ietf:params:xml:ns:asnx", added to the
   [attributes] of the enclosing element item.  The [prefix] of this
   attribute item is determined as specified in Section 6.2.3.1.  The
   [normalized value] of the attribute item is a qualified name for the
   expanded name of the NamedType (see [RXEREI]) corresponding to the
   chosen alternative.

      Aside: It is not possible to associate a namespace name with a
      NamedType in a UNION type using the current specification for RXER
      encoding instructions.  Consequently, the [normalized value] of
      the member attribute item will always contain a qualified name
      without a namespace prefix.



Legg & Prager                 Experimental                     [Page 47]

RFC 4910               Robust XML Encoding Rules               July 2007


   For a CRXER encoding, the member attribute item MUST be used, and the
   [normalized value] of the attribute item MUST be the CRXER
   translation of the QName value equal to the expanded name.

   In the absence of a member attribute item, an RXER decoder MUST
   determine the chosen alternative by considering the alternatives of
   the choice in the order prescribed below and accepting the first
   alternative for which the encoding is valid.

   If the UNION encoding instruction has a PrecedenceList, then the
   alternatives of the ChoiceType referenced by the PrecedenceList are
   considered in the order identified by that PrecedenceList, then the
   remaining alternatives are considered in the order of their
   definition in the ChoiceType.  If the UNION encoding instruction does
   not have a PrecedenceList, then all the alternatives of the
   ChoiceType are considered in the order of their definition in the
   ChoiceType.

   A non-canonical RXER encoder MUST use the member attribute item if an
   RXER decoder would determine the chosen alternative to be something
   other than the actual chosen alternative of the CHOICE value being
   translated; otherwise, the member attribute item MAY be used.

   Examples

      Consider this type definition:

         [RXER:UNION PRECEDENCE serialNumber] CHOICE {
             name          [0] IA5String,
             serialNumber  [1] INTEGER
         }

      In the absence of a member attribute, an RXER decoder would first
      consider whether the received encoding was a valid serialNumber
      (an INTEGER) before considering whether it was a valid name (an
      IA5String).

      The content and attributes of each of the following <value>
      elements are the RXER encoding of a value of the above type:

         <value>Bob</value>

         <value xmlns:asnx="urn:ietf:params:xml:ns:asnx"
                asnx:member="name">Alice</value>

         <value>
          <!-- Don't have a name for this one! --> 344
         </value>



Legg & Prager                 Experimental                     [Page 48]

RFC 4910               Robust XML Encoding Rules               July 2007


         <value xmlns:asnx="urn:ietf:params:xml:ns:asnx"
                asnx:member="name"><!-- A strange name. -->100</value>

      The member attribute is required in the final case to prevent the
      value being interpreted as a serialNumber.

   If the UNION (i.e., CHOICE) type is extensible [X.680], then an
   application MUST accept and be prepared to re-encode (using the same
   encoding rules) any unknown extension in received encoded values of
   the type.  An unknown extension in a value of a UNION type (an
   unknown alternative) takes the form of an unknown name in the
   [normalized value] of the member attribute and/or character data in
   the [children] of the enclosing element item that do not conform to
   any of the known alternatives.

   To enable re-encoding of an unknown alternative, it is necessary to
   retain the [normalized value] of the member attribute, if present,
   and the [children] property of the enclosing element item.

   The character data for an unknown alternative may contain qualified
   names that depend on the [in-scope namespaces] of the enclosing
   element item for their interpretation.  Therefore, semantically
   faithful re-encoding of an unknown alternative may require
   reproduction of at least some part of the [in-scope namespaces] of
   the enclosing element item.  The problem is deciding which of the
   namespace items are actually needed.  In the absence of type
   information, it is not possible to discern whether anything that
   syntactically resembles a qualified name in the character data of the
   enclosing element item actually is a qualified name.  The simplest
   approach is to retain all the namespace items from the
   [in-scope namespaces] of the enclosing element item and output them
   as namespace declaration attribute items in the
   [namespace attributes] of the enclosing element item when re-encoding
   the unknown alternative.  At best, an application can omit the
   namespace items that do not define the namespace prefix of any
   potential qualified name.

   An application MUST retain the namespace items in the
   [in-scope namespaces] of the enclosing element item that define the
   namespace prefixes of all the potential qualified names in the
   [children] of the enclosing element item.  Other namespace items in
   the [in-scope namespaces] of the enclosing element item MAY be
   retained.  The effect of these retained namespace items on the
   [namespace attributes] and [in-scope namespaces] of the enclosing
   element item when re-encoding is considered in Section 6.2.2.1.






Legg & Prager                 Experimental                     [Page 49]

RFC 4910               Robust XML Encoding Rules               July 2007


      Aside: The context attribute (Section 6.8.8) is not added to the
      [attributes] of the enclosing element item when re-encoding an
      unknown alternative since the type of a NamedType in a UNION type
      cannot be the Markup type.

6.7.15.  SEQUENCE OF as LIST



   The character data translation of a value of a LIST type (a
   SEQUENCE OF NamedType) is the concatenation of the character data
   translations of the component values, i.e., the abstract values of
   the type of the NamedType, each separated from the next by at least
   one white space character.  For a CRXER encoding, separating white
   space MUST be exactly one space character (U+0020).

   Example

      Consider this type definition:

         [LIST] SEQUENCE OF timeStamp GeneralizedTime

      The content of the following <value> element is the RXER encoding
      of a value of the above type:

         <value>
             2004-06-15T12:14:56Z
             2004-06-15T12:18:13Z
             2004-06-15T01:00:25Z
         </value>

6.8.  Combining Types



   The encoding of a value of an ASN.1 combining type (except a UNION or
   LIST type) typically has element content.

   The Infoset translation of a value of a specific ASN.1 combining type
   (excluding a UNION or LIST type) contains zero or more attribute
   items to be added to the [attributes] of the enclosing element item
   and zero or more element items to be added to the [children] of the
   enclosing element item.  These translations are described in Sections
   6.8.1 to 6.8.7.

   For a non-canonical RXER encoding, white space character items MAY be
   added to the [children] of the enclosing element item (before or
   after any other items).







Legg & Prager                 Experimental                     [Page 50]

RFC 4910               Robust XML Encoding Rules               July 2007


   For a CRXER encoding, a character item with the [character code]
   U+000A (a line feed) MUST be inserted immediately before each element
   item in the [children] of the enclosing element item.  No other white
   space character items are permitted to be added to the [children] of
   the enclosing element item.

      Aside: Without the single line feed character before each child
      element, a typical CRXER encoding would be a single, very long
      line.

6.8.1.  CHARACTER STRING



   A value of the unrestricted CHARACTER STRING type is translated
   according to the corresponding SEQUENCE type defined in Clause 40.5
   of X.680 [X.680].

6.8.2.  CHOICE



   The chosen alternative of a value of a CHOICE type corresponds to,
   and is a value of (see Section 6), some NamedType in the CHOICE type
   definition.

   The translation of a value of a CHOICE type other than the Markup
   type or a UNION type (see Section 6.7.14) is the translation of the
   value of the NamedType corresponding to the actual chosen
   alternative.

   Examples

      Consider this type definition:

         CHOICE {
             name          [0] IA5String,
             serialNumber  [1] INTEGER
         }

      The content of each of the following <value> elements is the RXER
      encoding of a value of the above type:

         <value><name>Bob</name></value>

         <value>
          <name>Alice</name>
         </value>







Legg & Prager                 Experimental                     [Page 51]

RFC 4910               Robust XML Encoding Rules               July 2007


         <value>
          <!-- Don't have a name for this one! -->
          <serialNumber>
           344
          </serialNumber>
         </value>

         <value>
          <!-- A strange name. -->
          <name>100</name>
         </value>

   If the CHOICE type is extensible [X.680], then an application MUST
   accept, and be prepared to re-encode (in RXER), any attribute item or
   child element item with a name that is not recognized (see
   Section 6.8.8).

6.8.3.  EMBEDDED PDV



   A value of the EMBEDDED PDV type is translated according to the
   corresponding SEQUENCE type defined in Clause 33.5 of X.680 [X.680].

6.8.4.  EXTERNAL



   A value of the EXTERNAL type is translated according to the
   corresponding SEQUENCE type defined in Clause 8.18.1 of X.690
   [X.690].

6.8.5.  INSTANCE OF



   A value of the INSTANCE OF type is translated according to the
   corresponding SEQUENCE type defined in Annex C of X.681 [X.681].

6.8.6.  SEQUENCE and SET



   Each component value of a value of a SEQUENCE or SET type corresponds
   to, and is a value of (see Section 6), some NamedType in the SEQUENCE
   or SET type definition.

   A value of a SEQUENCE or SET type, other than the QName type
   (Section 4.5), is translated by translating in turn each component
   value actually present in the SEQUENCE or SET value and adding the
   resulting attribute items and/or element items to the [attributes]
   and/or [children] of the enclosing element item.  Attribute items may
   be added to the [attributes] of the enclosing element item in any
   order.  Element items resulting from the translation of component





Legg & Prager                 Experimental                     [Page 52]

RFC 4910               Robust XML Encoding Rules               July 2007


   values MUST be appended to the [children] of the enclosing element
   item in the order of the component values' corresponding NamedType
   definitions in the SEQUENCE or SET type definition.

      Aside: In the case of the SET type, this is a deliberate departure
      from BER [X.690], where the components of a SET can be encoded in
      any order.

   If a DEFAULT value is defined for a NamedType and the value of the
   NamedType is the same as the DEFAULT value, then the translation of
   the value of the NamedType SHALL be omitted for a CRXER encoding and
   MAY be omitted for a non-canonical RXER encoding.

   Examples

      Consider this type definition:

         SEQUENCE {
             name        [0] IA5String OPTIONAL,
             partNumber  [1] INTEGER,
             quantity    [2] INTEGER DEFAULT 0
         }

      The content of each of the following <value> elements is the RXER
      encoding of a value of the above type:

         <value>
          <partNumber>23</partNumber>
          <!-- The quantity defaults to zero. -->
         </value>

         <value>
          <name>chisel</name>
          <partNumber> 37 </partNumber>
          <quantity> 0 </quantity>
         </value>

         <value>
          <!-- The name component is optional. -->
          <partNumber>1543</partNumber>
          <quantity>29</quantity>
         </value>

   If the SEQUENCE or SET type is extensible [X.680], then an
   application MUST accept, and be prepared to re-encode (in RXER), any
   attribute item or child element item with a name that is not
   recognized (see Section 6.8.8).




Legg & Prager                 Experimental                     [Page 53]

RFC 4910               Robust XML Encoding Rules               July 2007


6.8.7.  SEQUENCE OF and SET OF



   Each component value of a value of a type that is a SET OF NamedType
   or a SEQUENCE OF NamedType corresponds to, and is a value of (see
   Section 6), the NamedType in the type definition.

   A value of a type that is a SET OF NamedType, or a
   SEQUENCE OF NamedType other than a LIST type (see Section 6.7.15), is
   translated by adding the translation of each value of the NamedType
   to the [children] of the enclosing element item.

      Aside: An ATTRIBUTE encoding instruction cannot appear in the
      component type for a SEQUENCE OF or SET OF type, so there are no
      attribute items to add to the [attributes] of the enclosing
      element item.

   If the type is a SEQUENCE OF NamedType, then the values of the
   NamedType are translated in the order in which they appear in the
   value of the type.

   For a non-canonical RXER encoding, if the type is a SET OF NamedType,
   then the values of the NamedType may be translated in any order.

   For a CRXER encoding, if the type is a SET OF NamedType, then the
   values of the NamedType MUST be translated in ascending order where
   the order is determined by comparing the octets of their CRXER
   encodings (which will be UTF-8 encoded character strings; see
   Section 6.12.2).  A shorter encoding is ordered before a longer
   encoding that is identical up to the length of the shorter encoding.

   Examples

      Consider this type definition:

         SEQUENCE OF timeStamp GeneralizedTime

      The content of the following <value> element is the RXER encoding
      of a value of the above type:

         <value>
             <timeStamp>2004-06-15T12:14:56Z</timeStamp>
             <timeStamp>2004-06-15T12:18:13Z</timeStamp>
             <timeStamp>
                 2004-06-15T01:00:25Z
             </timeStamp>
         </value>





Legg & Prager                 Experimental                     [Page 54]

RFC 4910               Robust XML Encoding Rules               July 2007


      Consider this type definition (also see Section 6.6):

         SEQUENCE OF INTEGER

      The content of the following <value> element is the RXER encoding
      of a value of the above type:

         <value>
          <item>12</item>
          <item>
           9
          </item>
          <item> 7 <!-- A prime number. --></item>
         </value>

6.8.8.  Extensible Combining Types



   An application must accept and be prepared to re-encode (using the
   same encoding rules) any unknown extension appearing in the encoding
   of a value of an extensible CHOICE, SEQUENCE, or SET type.  An
   unknown extension in a value of an extensible combining type (except
   UNION types) takes the form of unknown element and/or attribute
   items.  Section 6.8.8.1 describes the processing of unknown element
   items and Section 6.8.8.2 describes the processing of unknown
   attribute items.

   An application cannot produce a canonical encoding if an abstract
   value contains unknown extensions.  However, the method for
   re-encoding unknown extensions does not prevent a receiving
   application with knowledge of the extension from producing the
   correct canonical encoding.

6.8.8.1.  Unknown Elements in Extensions



   To enable re-encoding of an unknown element item it is necessary to
   retain the [prefix], [local name], [attributes],
   [namespace attributes], and [children] properties of the element
   item.

   Definition (inherited namespace item):  An inherited namespace item
   is a namespace item in the [in-scope namespaces] of an element item
   for which there is no corresponding namespace declaration attribute
   item in the [namespace attributes] of the element item.

   The content and attributes of an unknown element item may contain
   qualified names whose interpretation depends on inherited namespace
   items.  Semantically faithful re-encoding of the unknown item may
   require reproduction of at least some of the inherited namespace



Legg & Prager                 Experimental                     [Page 55]

RFC 4910               Robust XML Encoding Rules               July 2007


   items.  The problem is deciding which of the inherited namespace
   items are actually needed.  Qualified names as the names of elements
   and attributes are easily recognized, but in the absence of type
   information it is not possible to discern whether anything that
   syntactically resembles a qualified name in the value of an attribute
   or the character data of an element actually is a qualified name.

   The simplest approach is to retain all the inherited namespace items
   and output corresponding namespace declaration attribute items in the
   [namespace attributes] of the unknown element item when re-encoding
   the element item.  At best, an application can omit the inherited
   namespace items that do not define the namespace prefix of any
   definite or potential qualified name, though this requires examining
   the content and attributes of the unknown extension.

   Regardless of how clever an implementation tries to be, adding any
   namespace declaration attribute items to an unknown element item is
   harmful to canonicalization if the ASN.1 type for the element item
   turns out to be the Markup type.  To counter this problem, a special
   attribute is used to identify the namespace declaration attribute
   items added to an unknown element item so that they can be removed
   later, if it proves necessary.

   If the outermost element item in an unknown extension does not have
   an attribute item with the [local name] "context" and
   [namespace name] "urn:ietf:params:xml:ns:asnx" in its [attributes],
   then namespace declaration attribute items corresponding to the
   inherited namespace items that define the namespace prefixes of all
   the definite and potential qualified names in the content and
   attributes of the element item MUST be added to the retained
   [namespace attributes].  Other inherited namespace items MAY be added
   to the retained [namespace attributes].

   If there are one or more of these added namespace declaration
   attribute items, then an attribute item with the [local name]
   "context" and [namespace name] "urn:ietf:params:xml:ns:asnx" MUST be
   added to the retained [attributes].

   The [prefix] of the context attribute item is any namespace prefix
   that does not match the [local name] of any namespace declaration
   attribute item in the [namespace attributes] unless the
   [namespace attributes] property contains a namespace declaration
   attribute item with a non-empty [prefix] and a [normalized value] of
   "urn:ietf:params:xml:ns:asnx".  In that case, the [local name] of
   that namespace declaration attribute item MAY be used as the [prefix]
   of the context attribute item.





Legg & Prager                 Experimental                     [Page 56]

RFC 4910               Robust XML Encoding Rules               July 2007


   If the [prefix] of the context attribute item does not match the
   [local name] of any namespace declaration attribute item, then an
   attribute item with the [prefix] "xmlns", [namespace name]
   "urn:ietf:params:xml:ns:asnx", and [local name] equal to the [prefix]
   of the context attribute item MUST be added to the retained
   [namespace attributes] of the element item.

   The [normalized value] of the context attribute is the white-space-
   separated unordered list of the [local names] of the added namespace
   declaration attribute items (i.e., a list of the namespace prefixes),
   including any namespace declaration attribute item added to define
   the [prefix] of the context attribute.  Note that the [local name]
   for a namespace declaration attribute item declaring the default
   namespace is "xmlns".

      Aside: A receiver that knows about the extension will use the
      context attribute to strip out the added namespace declaration
      attributes if the type of the associated NamedType is the Markup
      type (Section 6.10), and will discard the context attribute
      otherwise.  A receiver that does not know about the extension will
      re-encode the extension as is.

   Adding the required namespace declaration attribute items to an
   element item effectively makes the element item self-contained.  A
   received encoding has an encoding error if it contains an element
   item that is not self-contained but has a context attribute item in
   its [attributes].

   An RXER encoder MUST NOT add the context attribute item to an element
   item corresponding to a NamedType that is known to it.

   An RXER decoder MUST accept the context attribute item on an element
   item corresponding to a NamedType that does not appear to be an
   extension.

      Aside: It is not uncommon for extension markers to be neglected in
      specifications traditionally using only BER, since extension
      markers do not alter BER encodings.  Consequently, it is not
      immediately obvious in later versions of the specification which
      instances of NamedType belong to extensions of the original base
      specification.

   Example

      Suppose there are three applications, A, B, and C.  Suppose that
      Application A uses the first edition of an ASN.1 specification
      containing the following type definition:




Legg & Prager                 Experimental                     [Page 57]

RFC 4910               Robust XML Encoding Rules               July 2007


         MyType ::= SEQUENCE {
                 field1  INTEGER,  -- present in first edition
                 ...
         }

      Suppose that Application B uses the second edition of the ASN.1
      specification:

         MyType ::= SEQUENCE {
                 field1  INTEGER,  -- present in first edition
                 ...,
                 field2  QName     -- added in second edition
         }

      Suppose that Application C uses the third edition of the ASN.1
      specification:

         MyType ::= SEQUENCE {
                 field1  INTEGER,  -- present in first edition
                 ...,
                 field2  QName,    -- added in second edition
                 field3  Markup    -- added in third edition
         }

      Application C produces the following RXER encoding and sends it to
      Application B:

         <value xmlns:p2="http://example.com/ns2">
          <field1> 100 </field1>
          <field2> p2:foobar </field2>
          <field3 xmlns:p1="http://example.com/ns1"> p1:foobar </field3>
         </value>

      Application B doesn't know about <field3>, so it adds the
      asnx:context attribute to <field3> when it re-encodes the abstract
      value to send to Application A:















Legg & Prager                 Experimental                     [Page 58]

RFC 4910               Robust XML Encoding Rules               July 2007


         <value xmlns:p1="http://example.com/ns2">
          <!-- Application B knows the white space in field1 and
               field2 is optional and discards it. -->
          <field1>100</field1>
          <field2>p1:foobar</field2>
          <!-- Application B doesn't know about field3
               so it leaves the character data alone. -->
          <field3 asnx:context="asnx p2"
                  xmlns:asnx="urn:ietf:params:xml:ns:asnx"
                  xmlns:p1="http://example.com/ns1"
                  xmlns:p2="http://example.com/ns2"> p1:foobar </field3>
         </value>

      Application A doesn't know about <field2> and <field3>, so it adds
      the asnx:context attribute to <field2> and leaves <field3> alone
      when it re-encodes the abstract value:

         <value>
          <!-- Application A knows about field1 and chooses
               to add some white space. -->
          <field1> 100 </field1>
          <!-- Application A doesn't know about field2 or field3
               so it leaves the character data alone. -->
          <field2 asnx:context="asnx p1"
                  xmlns:asnx="urn:ietf:params:xml:ns:asnx"
                  xmlns:p1="http://example.com/ns2">p1:foobar</field2>
          <field3 asnx:context="asnx p2"
                  xmlns:asnx="urn:ietf:params:xml:ns:asnx"
                  xmlns:p1="http://example.com/ns1"
                  xmlns:p2="http://example.com/ns2"> p1:foobar </field3>
         </value>

      If Application C receives this final encoding, it has sufficient
      information to discard the asnx:context, xmlns:asnx, and xmlns:p2
      attributes from the received Markup value of <field3> to recover
      the original value.  Application C knows about <field2>, so it
      uses the namespace declaration for p1 when decoding the QName
      value and ignores the other declarations.

6.8.8.2.  Unknown Attributes in Extensions



   To enable re-encoding of an unknown attribute item it is necessary to
   retain at least the [local name], [namespace name], and
   [normalized value] properties of the attribute item.

   The [normalized value] of an unknown attribute item may contain
   qualified names whose interpretation depends on the
   [in-scope namespaces] of the [owner element].  Semantically faithful



Legg & Prager                 Experimental                     [Page 59]

RFC 4910               Robust XML Encoding Rules               July 2007


   re-encoding of the unknown attribute item may require reproduction of
   at least some part of the [in-scope namespaces].  In the absence of
   type information, it is not possible to discern whether anything that
   syntactically resembles a qualified name in the [normalized value] of
   an unknown attribute item actually is a qualified name.

   The simplest approach is to retain all the namespace items of the
   [in-scope namespaces] and output corresponding namespace declaration
   attribute items in the [namespace attributes] of the [owner element]
   when re-encoding the extension.  At best, an application can omit the
   namespace items that do not define the namespace prefix of any
   potential qualified name in the [normalized value].

   An application MUST retain the namespace items in the
   [in-scope namespaces] of the [owner element] that define the
   namespace prefixes of all the potential qualified names in the
   [normalized value] of the unknown attribute item.  Other namespace
   items in the [in-scope namespaces] of the [owner element] MAY be
   retained.

      Aside: If the enclosing element item has more than one unknown
      attribute item, then it is sufficient to save the union of the
      retained namespace items with the element item, rather than saving
      the retained namespace items with each unknown attribute item.

   When the unknown attribute item is re-encoded, the retained namespace
   items affect the [namespace attributes] and [in-scope namespaces] of
   the enclosing element item as specified in Section 6.2.2.1, and the
   [prefix] of the attribute item is determined as specified in
   Section 6.2.3.1.

      Aside: The context attribute is not added to the [attributes] of
      the [owner element] when re-encoding an unknown attribute item
      because the type of a NamedType subject to an ATTRIBUTE or
      ATTRIBUTE-REF encoding instruction cannot be the Markup type.

6.9.  Open Type



   A value of an open type denoted by an ObjectClassFieldType [X.681] is
   translated according to the specific Type of the value.

   If the specific Type of the value is directly or indirectly the
   Markup type, then the enclosing element item MUST be self-contained.

   For a non-canonical RXER encoding, if the translation of the value
   does not generate an attribute item with the [local name] "type" and
   the [namespace name] "http://www.w3.org/2001/XMLSchema-instance"
   (i.e., xsi:type) and the specific Type of the value is a



Legg & Prager                 Experimental                     [Page 60]

RFC 4910               Robust XML Encoding Rules               July 2007


   namespace-qualified reference (Section 5), then an attribute item
   with the [local name] "type" and the [namespace name]
   "http://www.w3.org/2001/XMLSchema-instance" (i.e., xsi:type) MAY be
   added to the [attributes] of the enclosing element item.  The
   [normalized value] of this attribute item is a qualified name for the
   expanded name of the referenced type, with the namespace prefix
   determined as specified in Section 6.7.11.1.

      Aside: The xsi:type attribute is added by RXER encoders for the
      benefit of XML Schema validators.  This attribute tells an
      XML Schema validator which type definition in a compatible
      XML Schema translation of the ASN.1 specification it should use
      for validating the content and attributes of the enclosing
      element.  For an RXER decoder, the actual type in an open type
      value is generally determined by an associated component relation
      constraint [X.682], so the xsi:type attribute can be ignored.

   Example

      The content and attributes of the following <value> element are
      the RXER encoding of an open type value containing a BOOLEAN
      value:

         <value xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
                xmlns:asnx="urn:ietf:params:xml:ns:asnx"
                xsi:type="asnx:BOOLEAN"> true </value>

   If the ObjectClassFieldType denoting an open type is not constrained
   by a TableConstraint, or is constrained by a TableConstraint where
   the constraining object set is extensible, then an application MUST
   accept and be prepared to re-encode (using the same encoding rules)
   any value of the open type where the specific Type of the value is
   unknown.  In such cases, the enclosing element item is treated like
   an unknown element item in the value of an extensible combining ASN.1
   type (see Section 6.8.8.1).

6.10.  Markup



   Conceptually, a value of the Markup type holds the [prefix],
   [attributes], [namespace attributes], and [children] of an element
   item.  The Infoset translation of a value of the Markup type
   initially simply sets the [prefix], [attributes],
   [namespace attributes], and [children] of the enclosing element item
   to the corresponding properties represented by the Markup value.

   Recall that the enclosing element item for the translation of a
   Markup value is required to be self-contained (Section 4.1.1).




Legg & Prager                 Experimental                     [Page 61]

RFC 4910               Robust XML Encoding Rules               July 2007


   If the enclosing element item is not the [document element] of the
   document item, and the [in-scope namespaces] property of the
   enclosing element item's [parent] contains a namespace item for the
   default namespace, and the [namespace attributes] property
   represented by the Markup value does not contain a namespace item
   declaring or undeclaring the default namespace, then a namespace
   declaration attribute item that undeclares the default namespace
   SHALL be added to the enclosing element item's
   [namespace attributes].

   It is not necessary to populate the [in-scope namespaces] of the
   enclosing element item for encoding purposes (though it may be
   warranted for other purposes).

   An element item nested in the [children] is potentially the Infoset
   translation of a value of a top-level NamedType (as allowed by
   Section 6.4), and the entire Markup value can represent the content
   and attributes of an element item that is the translation of a value
   of a top-level NamedType.

      Aside: The latter case arises when an ELEMENT-REF encoding
      instruction references a top-level NamedType.

   The content and attributes of an element item nested in the
   [children] of a Markup value are potentially the Infoset translation
   of an abstract value of an ASN.1 type (as allowed by Section 6.4),
   and the entire Markup value can represent the translation of a single
   abstract value.

      Aside: The latter case arises when a TYPE-REF encoding instruction
      references an ASN.1 type.

   For a non-canonical RXER encoding, any element item, at any level of
   nesting (including the enclosing element item itself), that
   corresponds to the value of a top-level NamedType MAY be replaced
   with any valid translation of that value.

   For a non-canonical RXER encoding, any element item, at any level of
   nesting (including the enclosing element item itself), with content
   and attributes that correspond to an abstract value of an ASN.1 type
   MAY have that content and those attributes replaced with any valid
   translation of that abstract value.  If the content and attributes
   are replaced, then the [prefix], [in-scope namespaces], and
   [namespace attributes] of the element item are constructed as
   specified in Sections 6.2.2.1 and 6.2.2.2.  The enclosing element
   item for the Markup value is still required to be self-contained.





Legg & Prager                 Experimental                     [Page 62]

RFC 4910               Robust XML Encoding Rules               July 2007


      Aside: Insofar as a Markup value represents ASN.1 abstract values,
      it is sufficient for the RXER encoding of the Markup value to
      preserve the abstract values rather than preserve the exact
      Infoset representation.

   For a CRXER encoding, any element item, at any level of nesting
   (including the enclosing element item itself), that corresponds to a
   value of a top-level NamedType MUST be replaced with the CRXER
   translation of that value.

   For a CRXER encoding, any element item, at any level of nesting
   (including the enclosing element item itself), with content and
   attributes that correspond to an abstract value of an ASN.1 type MUST
   have that content and those attributes replaced with the CRXER
   translation of that abstract value.  The [prefix],
   [in-scope namespaces], and [namespace attributes] of the element item
   are constructed as specified in Sections 6.2.2.1 and 6.2.2.2.

   If the [attributes] property of the enclosing element item from a
   received RXER encoding contains an attribute item with the
   [local name] "context" and [namespace name]
   "urn:ietf:params:xml:ns:asnx" (i.e., asnx:context), then this
   attribute item MUST be omitted from the [attributes] represented by
   the Markup value, and each namespace declaration attribute item with
   a [local name] matching an NCName in the [normalized value] of the
   attribute item MUST be omitted from the [namespace attributes]
   represented by the Markup value.

6.11.  Namespace Prefixes for CRXER



   The final step in translating the value of a top-level NamedType for
   a CRXER encoding, or an abstract value for a Standalone CRXER
   Encoding, is the replacement of the arbitrarily chosen namespace
   prefixes with algorithmically determined canonical namespace
   prefixes.  This procedure for prefix replacement applies to each
   element item where the [namespace attributes] have been constructed
   according to Section 6.2.2.1.  This includes any element item
   corresponding to a value of a top-level NamedType, or with content
   and attributes that correspond to an abstract value of an ASN.1 type,
   that is nested in a value of the Markup type.

   For each element item where prefix replacement applies, the following
   sequence of steps is repeated until there are no more eligible
   attribute items to select in step (1):







Legg & Prager                 Experimental                     [Page 63]

RFC 4910               Robust XML Encoding Rules               July 2007


   (1) Select the attribute item with the least [normalized value] from
       amongst the attribute items of the [namespace attributes] that
       have a [local name] that is not a canonical namespace prefix
       (i.e., select from the namespace declaration attribute items that
       have not already been processed).  A [normalized value] is less
       than another [normalized value] if the former appears before the
       latter in an ordering of the values determined by comparing the
       ISO 10646 code points [UCS] of their characters, from first to
       last.  A shorter string of characters is ordered before a longer
       string of characters that is identical up to the length of the
       shorter string.

          Aside: Note that when a namespace declaration (other than for
          the default namespace) is represented as an attribute item in
          the [namespace attributes], the attribute's [prefix] is
          "xmlns", its [local name] is the namespace prefix, and its
          [normalized value] is the namespace name.

   (2) A canonical namespace prefix is unused if it is not currently the
       [prefix] of any namespace item in the [in-scope namespaces] of
       the element item.  Replace the [local name] of the selected
       attribute item with the unused canonical namespace prefix that
       has the non-negative number string with the least integer value
       (e.g., n2 is less than n10).

   (3) The selected attribute item has a corresponding namespace item in
       the [in-scope namespaces] of the element.  Replace the [prefix]
       of this corresponding namespace item with the canonical namespace
       prefix determined in step (2).

   (4) The element item and its [attributes] property, and descendent
       element items and their [attributes] properties, may depend on
       the selected attribute item to determine the binding between
       their [prefix] and [namespace name].  Replace the [prefix] of any
       such dependent element items and attribute items with the
       canonical namespace prefix determined in step (2).

       Note that a namespace prefix can be redeclared (reused).
       Replacement of the prefix does not apply to an element item
       wherein the prefix is redeclared, or to the descendants of such
       an element item.

   (5) The character data translations for values of the QName ASN.1
       type may depend on the selected attribute item to determine the
       binding between their namespace prefix and namespace name.
       Replace the namespace prefix of any such dependent character data
       translation with the canonical namespace prefix determined in
       step (2).



Legg & Prager                 Experimental                     [Page 64]

RFC 4910               Robust XML Encoding Rules               July 2007


       Note that a character data translation can appear in the
       [normalized value] of an attribute item, or as a sequence of
       character items in the [children] of an element item.

6.12.  Serialization



   The final RXER encoding is produced by serializing the Infoset
   translation as an XML document.  An implementation MUST serialize the
   Infoset translation as an XML document in such a way that the Infoset
   of the resulting XML document matches the Infoset translation, after
   ignoring the following properties:

   (1) all properties of the document item except the
       [document element],

   (2) the [base URI] of any item,

   (3) the [element content whitespace] of character items,

   (4) the [notation] of processing instruction items,

   (5) the [in-scope namespaces] of element items.

      Aside: The [in-scope namespaces] of a parent element item are only
      selectively inherited by its child element items in the Infoset
      translations of ASN.1 values.  This means that the Infoset
      reconstructed by parsing the XML document serialization of the
      original Infoset will generally have more namespace items in its
      [in-scope namespaces], but these extra namespace items will not be
      significant.

      Aside: A consequence of case (1) is that comments and PIs before
      and after the document element are permitted.

   In general, there is more than one possible serialization for any
   given Infoset translation.  Section 6.12.1 highlights some important
   considerations in producing a correct serialization and discusses
   some of the serialization options.

   Section 6.12.2 applies to CRXER encodings and limits the
   serialization options so that each distinct Infoset has only one
   possible serialization.

6.12.1.  Non-Canonical Serialization



   This section discusses aspects of Infoset serialization for
   non-canonical RXER encodings, but is not an exhaustive list of the
   options for non-canonical serialization.



Legg & Prager                 Experimental                     [Page 65]

RFC 4910               Robust XML Encoding Rules               July 2007


   If one or more character items have a [character code] in the range
   U+0001 to U+0008, U+000B to U+000C, or U+000E to U+001F, or one or
   more characters in any attribute's [normalized value] are in the
   range U+0001 to U+0008, U+000B to U+000C, or U+000E to U+001F, then
   the Infoset translation MUST be serialized as an XML version 1.1
   document; otherwise, the Infoset translation is serialized as either
   an XML version 1.0 or version 1.1 document.

   A non-canonical RXER encoding may use any of the allowed character
   encoding schemes for XML.  RXER encoders and decoders MUST support
   the UTF-8 character encoding.

   An element item may be serialized as an empty-element tag if it has
   no items in its [children].

   Attributes of an element can appear in any order since the
   [attributes] and [namespace attributes] of an element item are
   unordered.

   Ampersand ('&', U+0026) and open angle bracket ('<', U+003C)
   characters in the [normalized value] of an attribute item must be
   escaped appropriately [XML10][XML11] (with a character reference or a
   predefined entity reference).  Double quote (U+0022) and single quote
   (U+0027) characters in an attribute item's [normalized value] may
   also need to be escaped.  Character items with the [character code]
   U+0026 (ampersand, '&') or U+003C (open angle bracket, '<') must be
   escaped appropriately (with a character reference, a predefined
   entity reference or a CDATA section).

   Line break normalization by XML processors allows some freedom in how
   a character item for a line feed character (U+000A) is serialized:

   (1) If XML version 1.0 is selected, then a character item with the
       [character code] U+000A (line feed) is serialized as either a
       line feed character (U+000A), a carriage return character
       (U+000D) followed by a line feed character (U+000A), or just a
       carriage return character (U+000D) provided the next item is not
       a character item that is serialized as a line feed character
       (U+000A).

   (2) If XML version 1.1 is selected, then a character item with the
       [character code] U+000A (line feed) is serialized as either a
       line feed character (U+000A), a next line character (U+0085), a
       line separator character (U+2028), a carriage return character
       (U+000D) followed by a line feed character (U+000A), a carriage
       return character (U+000D) followed by a next line character





Legg & Prager                 Experimental                     [Page 66]

RFC 4910               Robust XML Encoding Rules               July 2007


       (U+0085), or just a carriage return character (U+000D) provided
       the next item is not a character item that is serialized as a
       line feed (U+000A) or next line (U+0085) character.

         Aside: All these sequences will be normalized to a line feed
         character (U+000A) during decoding.

   A character item with the [character code] U+000D (carriage return),
   U+0085 (next line), or U+2028 (line separator) must be serialized as
   a character reference to protect the character from line break
   normalization during decoding.

   The attribute value normalization performed by XML processors allows
   some freedom in how a space character (U+0020) is serialized:

   (1) If XML version 1.0 is selected, then a space character (U+0020)
       in an attribute item's [normalized value] is serialized as either
       a space character (U+0020), a tab character (U+0009), a carriage
       return character (U+000D), a line feed character (U+000A), a
       carriage return character (U+000D) followed by a line feed
       character (U+000A), or just a carriage return character (U+000D)
       provided the next character in the [normalized value] is not
       serialized as a line feed character (U+000A).

   (2) If XML version 1.1 is selected, then a space character (U+0020)
       in an attribute item's [normalized value] is serialized as either
       a space character (U+0020), a tab character (U+0009), a carriage
       return character (U+000D), a line feed character (U+000A), a next
       line character (U+0085), a line separator character (U+2028), a
       carriage return character (U+000D) followed by a line feed
       character (U+000A), a carriage return character (U+000D) followed
       by a next line character (U+0085), or just a carriage return
       character (U+000D) provided the next character in the
       [normalized value] is not serialized as a line feed (U+000A) or
       next line (U+0085) character.

          Aside: All these sequences will be normalized to a space
          character (U+0020) during decoding, through a combination of
          line break normalization and attribute value normalization.

   Each tab (U+0009), line feed (U+000A), or carriage return (U+000D)
   character in an attribute item's [normalized value] must be
   serialized as a character reference to protect the character from
   attribute value normalization during decoding.  In addition, if XML
   version 1.1 is selected, then each next line (U+0085) or line
   separator (U+2028) character must be serialized as a character
   reference.




Legg & Prager                 Experimental                     [Page 67]

RFC 4910               Robust XML Encoding Rules               July 2007


   Parsed entity references may be used (unless the environment in which
   the RXER encoding is used disallows entity references).  If entity
   references to other than the predefined entities are used, then the
   XML document containing the RXER encoding must necessarily contain a
   document type declaration, and the internal or external subset of the
   document type definition must contain entity declarations for those
   entities.

6.12.2.  Canonical Serialization



   This section discusses Infoset serialization for CRXER encodings.
   The serialization of an Infoset for a CRXER encoding is restricted so
   that each distinct Infoset has only one possible serialization as an
   XML document.

      Aside: These restrictions have been chosen so as to be consistent
      with Canonical XML [CXML], where possible.

   The document SHALL be encoded in UTF-8 without a leading Byte Order
   Mark [UCS].

   The XMLDecl of the document SHALL be <?xml version="1.1"?>.

   A document type declaration (doctypedecl) SHALL NOT be used.

      Aside: This has the effect of excluding entity references, except
      those for the predefined entities (e.g., &).

   A single line feed character (U+000A) MUST be inserted immediately
   before the document element.

   No other white space characters are permitted before or after the
   document element.

   There SHALL NOT be any PIs or comments before or after the document
   element.

   An element item MUST NOT be serialized as an empty-element tag.

      Aside: If an element item has no items in its [children], then it
      is serialized as a start-tag followed by an end-tag.

   There SHALL NOT be any white space characters immediately before the
   closing '>' of an element's start-tag and end-tag.  The white space
   preceding each attribute SHALL be exactly one space character
   (U+0020).  There SHALL NOT be any white space characters immediately
   before or after the equals sign (U+003D) in an attribute.




Legg & Prager                 Experimental                     [Page 68]

RFC 4910               Robust XML Encoding Rules               July 2007


   The delimiter for attribute values SHALL be the double quote
   character (U+0022).

   Namespace declaration attributes MUST appear before any other
   attributes of an element.  A namespace declaration for the default
   namespace, if present, MUST appear as the first attribute.  The
   remaining namespace declaration attributes MUST appear in
   lexicographic order based on [local name].

      Aside: In particular, this means that xmlns:n10 comes before
      xmlns:n2.

   The attributes that are not namespace declarations MUST be
   lexicographically ordered on [namespace name] as the primary key and
   [local name] as the secondary key.

   CDATA sections SHALL NOT be used.

   Each ampersand character ('&', U+0026) in an attribute item's
   [normalized value] MUST be serialized as the entity reference &.
   Each open angle bracket character ('<', U+003C) in an attribute
   item's [normalized value] MUST be serialized as the entity reference
   <.  Each double quote character (U+0022) in an attribute item's
   [normalized value] MUST be serialized as the entity reference ".
   Each character in the range U+0001 to U+001F or U+007F to U+009F in
   an attribute item's [normalized value] MUST be serialized as a
   character reference.  No other character in a [normalized value] is
   permitted to be serialized as an entity reference or character
   reference.

   Each character item with the [character code] U+0026 (the ampersand
   character) MUST be serialized as the entity reference &.  Each
   character item with the [character code] U+003C (the open angle
   bracket character) MUST be serialized as the entity reference <.
   Each character item with the [character code] U+003E (the closing
   angle bracket character) MUST be serialized as the entity reference
   >.  Each character item with a [character code] in the range
   U+0001 to U+0008, U+000B to U+001F, or U+007F to U+009F MUST be
   serialized as a character reference.  No other character item is
   permitted to be serialized as an entity reference or character
   reference.

   Character references, where they are permitted, SHALL use uppercase
   hexadecimal with no leading zeroes.  For example, the carriage return
   character is represented as .

   A space character (U+0020) in an attribute item's [normalized value]
   MUST be serialized as a single U+0020 character.



Legg & Prager                 Experimental                     [Page 69]

RFC 4910               Robust XML Encoding Rules               July 2007


   A character item with the [character code] U+000A MUST be serialized
   as a single U+000A character.

   The white space separating the [target] and [content] in the
   serialization of a processing instruction item SHALL be exactly one
   space character (U+0020).

      Aside: A processing instruction or comment can only appear in a
      CRXER encoding if it is embedded in a Markup value.

6.12.3.  Unicode Normalization in XML Version 1.1



   XML Version 1.1 recommends, but does not absolutely require, that
   text be normalized according to Unicode Normalization Form C
   [UNICODE].  ASN.1 has no similar requirement on abstract values of
   string types, and ASN.1 canonical encoding rules depend on the code
   points of characters being preserved.

   To accommodate both requirements, applications SHOULD normalize
   abstract values of ASN.1 character string types according to Unicode
   Normalization Form C at the time the values are created, but MUST NOT
   normalize a previously decoded abstract value of an ASN.1 character
   string type prior to re-encoding it.  An application may, of course,
   normalize a decoded abstract value for other purposes, such as
   display to a user.

6.13.  Syntax-Based Canonicalization



   ASN.1 encoding rules are designed to preserve abstract values, but
   not to preserve every detail of each transfer syntax that is used.
   In the case of RXER, this means that the Infoset representation of an
   abstract value is not necessarily preserved when the abstract value
   is decoded and re-encoded (regardless of the encoding rules used).
   However, syntax-based canonicalization for XML documents (e.g.,
   Canonical XML [CXML]) depends on the Infoset of an XML document being
   preserved.  The Infoset representation of an XML document containing
   the RXER encoding of an ASN.1 abstract value potentially changes if
   that value is decoded and re-encoded, disrupting the Canonical XML
   representation.  Extra normalization is required if RXER is to be
   usefully deployed in environments where syntax-based canonicalization
   is used.

   Prior to applying syntax-based canonicalization to an XML document,
   any element items in the Infoset representation of the document that
   correspond to the value of an ASN.1 top-level NamedType or have
   content and attributes that correspond to an ASN.1 abstract value
   MUST be replaced by the translation of the value according to CRXER.




Legg & Prager                 Experimental                     [Page 70]

RFC 4910               Robust XML Encoding Rules               July 2007


   If an application uses Canonical XML but has no knowledge of RXER,
   then it will not know to normalize RXER encodings.  If RXER is
   deployed into an environment containing such applications, then the
   Infoset translation for CRXER SHOULD be used for all RXER encodings.

7.  Transfer Syntax Identifiers



7.1.  RXER Transfer Syntax



   The following OBJECT IDENTIFIER has been assigned by xmled.org to
   identify the Robust XML Encoding Rules, under an arc assigned to
   xmled.org by the Internet Assigned Numbers Authority (IANA):

      { iso(1) identified-organization(3) dod(6)
        internet(1) private(4) enterprise(1)
        xmled(21472) asnx(1) encoding(1) rxer(0) }

   This OBJECT IDENTIFIER would be used, for example, to describe the
   transfer syntax for an RXER encoded data-value in an EMBEDDED PDV
   value.

7.2.  CRXER Transfer Syntax



   The following OBJECT IDENTIFIER has been assigned by xmled.org to
   identify the Canonical Robust XML Encoding Rules, under an arc
   assigned to xmled.org by the IANA:

      { iso(1) identified-organization(3) dod(6)
        internet(1) private(4) enterprise(1)
        xmled(21472) asnx(1) encoding(1) crxer(1) }

   This OBJECT IDENTIFIER would be used, for example, to describe the
   transfer syntax for a CRXER encoded data-value in an EMBEDDED PDV
   value.

8.  Relationship to XER



   The Robust XML Encoding Rules (RXER) and the XML Encoding Rules (XER)
   [X.693] are separate, distinctly different and incompatible ASN.1
   encoding rules for producing XML markup from ASN.1 abstract values.
   RXER is therefore unrelated to the XML value notation of X.680
   [X.680].

   This section describes some of the major differences between RXER and
   XER.






Legg & Prager                 Experimental                     [Page 71]

RFC 4910               Robust XML Encoding Rules               July 2007


   There are essentially two varieties of XER:  BASIC-XER (with a
   canonical form called CANONICAL-XER) and EXTENDED-XER.  The
   significant difference between the two varieties is that XER encoding
   instructions are used by EXTENDED-XER, but are ignored by BASIC-XER
   (and therefore by CANONICAL-XER).  There isn't a canonical variant of
   EXTENDED-XER.  Characteristics that are common to BASIC-XER and
   EXTENDED-XER will simply be noted as being characteristics of XER.

   Elements and attributes are the fundamental discrete structures of an
   XML document.  Not surprisingly, schema languages for XML typically
   have the means to describe, name, and reference global (i.e.,
   top-level) elements and attributes.  Global type definitions are seen
   more as a convenience for defining the contents of elements and
   attributes.  Traditional ASN.1 has the means to define global types
   (and other global constructs that support the definition of types)
   but nothing akin to a global element or attribute definition.  The
   fundamental difference between RXER and XER is in how this omission
   is addressed.

   With XER, type definitions are also regarded as being element
   definitions by default, or as attribute definitions in the presence
   of an XER ATTRIBUTE encoding instruction.  In some circumstances an
   anonymous Type is required to define an element, which leads to
   element names like <BOOLEAN> and <SEQUENCE>.  NamedType notation also
   defines local elements, and there are some curious cases in
   EXTENDED-XER where NamedType notation can define a global type.  So
   under XER, types can be defined by either Type or NamedType notation,
   and elements and attributes can also be defined by either Type or
   NamedType notation.

   With RXER, types are only defined by Type notation and elements and
   attributes are only defined by NamedType notation.  Global element
   and attribute definitions are made possible by top-level NamedType
   notation in an RXER encoding control section.

   RXER, with its clean separation of Type notation for types and
   NamedType notation for elements and attributes, is a better basis
   than XER for translating an ASN.1 specification into an XML
   representation (i.e., ASN.X [ASN.X]) or a compatible XML Schema,
   where type, element, and attribute definitions are also distinctly
   separate constructs.

   There is usually a requirement on applications specified in ASN.1 to
   maintain backward compatibility with the encodings generated by
   previous versions.  The encodings in question are typically BER.
   Even with the backward-compatibility constraint there is still
   considerable leeway for specification writers to rewrite the earlier
   specification.  For example, they could rename types, factor out an



Legg & Prager                 Experimental                     [Page 72]

RFC 4910               Robust XML Encoding Rules               July 2007


   in-line type definition as a defined type (or the reverse), or
   replace a type definition with an equivalent parameterized reference.
   These changes produce no change to BER, DER, CER [X.690], Packed
   Encoding Rules (PER) [X.691], or Generic String Encoding Rules (GSER)
   [GSER] encodings (so specification writers have felt free to make
   such changes to improve their specification), but can change the
   names of elements in the XER encoding because XER uses types as
   element definitions.  The RXER encoding is immune to this problem,
   thus RXER encodings are more stable than XER encodings over
   successive revisions of an ASN.1 specification (which explains the
   first 'R' in RXER).  This has an obvious benefit for
   interoperability.

   RXER has special provisions for encoding values of the QName and
   Markup types.  QName is used to hold qualified names and Markup can
   be used to hold arbitrary untyped markup.  XER doesn't recognize any
   special types like these, but it is possible to get the same effects
   as RXER's QName and Markup types by using XER encoding instructions.
   Since CANONICAL-XER ignores encoding instructions, this means that
   under XER an application can either support qualified names and
   untyped markup, or support canonical XML encodings, but not both.  In
   contrast, CRXER has canonicalization rules for qualified names and
   for Markup.  Furthermore, EXTENDED-XER does not address the issues of
   normalization of untyped data for other ASN.1 canonical encoding
   rules (e.g., for DER; see Section 4.1.2) or normalization of XML
   encodings for syntax-based canonicalization (e.g., for Canonical XML;
   see Section 6.13).

   Both EXTENDED-XER and RXER use encoding instructions to define
   attributes, union types, and list types, among other things.  Since
   CANONICAL-XER ignores encoding instructions, this means that under
   XER an application must choose between making use of attributes,
   union types, list types, etc., or supporting canonical XML encodings.
   In contrast, the canonicalization rules for CRXER encompass all the
   encoding instructions for RXER.

9.  Security Considerations



   RXER does not necessarily enable the exact BER octet encoding of
   values of the TeletexString, VideotexString, GraphicString, or
   GeneralString types to be reconstructed, so a transformation from DER
   to RXER and back to DER may not reproduce the original DER encoding.
   This is a result of inadequate normalization of values of these
   string types in DER.  A character in a TeletexString value (for
   example) that corresponds to a specific ISO 10646 character can be
   encoded for BER in a variety of ways that are indistinguishable in an





Legg & Prager                 Experimental                     [Page 73]

RFC 4910               Robust XML Encoding Rules               July 2007


   RXER re-encoding of the TeletexString value.  DER does not mandate
   one of these possible character encodings in preference to all
   others.

   Because of the above, RXER MUST NOT be used to re-encode, whether for
   storage or transmission, ASN.1 abstract values whose original DER or
   CER encoding must be recoverable, and whose type definitions involve
   the TeletexString, VideotexString, GraphicString, or GeneralString
   type.  Such recovery is needed for the verification of digital
   signatures.  In such cases, protocols ought to use DER or a DER-
   reversible encoding.  In other cases where ASN.1 canonical encoding
   rules are used, values of the Markup type must be normalized as
   described in Section 4.1.2.

   A transformation from CRXER to BER and back to CRXER does reproduce
   the original CRXER encoding, therefore it is safe to use BER, DER, or
   CER to re-encode ASN.1 abstract values whose original CRXER encoding
   must be recoverable.

   Digital signatures may also be calculated on the Canonical XML
   representation of an XML document.  If RXER encodings appear in such
   documents, then applications must normalize the encodings as
   described in Section 6.13.

   The null character (U+0000) cannot be represented in XML and hence
   cannot be transmitted in an RXER encoding.  Null characters in
   abstract values of ASN.1 string types will be dropped if the values
   are RXER encoded; therefore, RXER MUST NOT be used by applications
   that attach significance to the null character.

   When interpreting security-sensitive fields, and in particular fields
   used to grant or deny access, implementations MUST ensure that any
   comparisons are done on the underlying abstract value, regardless of
   the particular encoding used.  Comparisons of Markup values MUST
   operate as though the values have been normalized as specified in
   Section 4.1.2.

10.  Acknowledgements



   The technology described in this document is the product of a
   research project begun jointly by Adacel Technologies Limited and
   Deakin University, and subsequently refined and completed by eB2Bcom.









Legg & Prager                 Experimental                     [Page 74]

RFC 4910               Robust XML Encoding Rules               July 2007


11.  IANA Considerations



   The IANA has registered a new XML namespace in accordance with RFC
   3688 [XMLREG].

   URI:  urn:ietf:params:xml:ns:asnx

   Registrant Contact:  Steven Legg <steven.legg@eb2bcom.com>

   XML:  None

12.  References



12.1.  Normative References



   [BCP14]    Bradner, S., "Key words for use in RFCs to Indicate
              Requirement Levels", BCP 14, RFC 2119, March 1997.

   [UTF-8]    Yergeau, F., "UTF-8, a transformation format of ISO
              10646", RFC 3629, November 2003.

   [XMLREG]   Mealling, M., "The IETF XML Registry", RFC 3688, January
              2004.

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

   [RXEREI]   Legg, S., "Encoding Instructions for the Robust XML
              Encoding Rules (RXER)", RFC 4911, July 2007.

   [ASN.X]    Legg, S., "Abstract Syntax Notation X (ASN.X)", RFC 4912,
              July 2007.

   [X.680]    ITU-T Recommendation X.680 (07/02) | ISO/IEC 8824-1,
              Information technology - Abstract Syntax Notation One
              (ASN.1):  Specification of basic notation.

   [X.680-1]  ITU-T Recommendation X.680 (2002) Amendment 1 (10/03) |
              ISO/IEC 8824-1:2002/Amd 1:2004, Support for EXTENDED-XER.

   [X.681]    ITU-T Recommendation X.681 (07/02) | ISO/IEC 8824-2,
              Information technology - Abstract Syntax Notation One
              (ASN.1):  Information object specification.

   [X.682]    ITU-T Recommendation X.682 (07/02) | ISO/IEC 8824-3,
              Information technology - Abstract Syntax Notation One
              (ASN.1):  Constraint specification.



Legg & Prager                 Experimental                     [Page 75]

RFC 4910               Robust XML Encoding Rules               July 2007


   [X.683]    ITU-T Recommendation X.683 (07/02) | ISO/IEC 8824-4,
              Information technology - Abstract Syntax Notation One
              (ASN.1):  Parameterization of ASN.1 specifications.

   [X.690]    ITU-T Recommendation X.690 (07/02) | ISO/IEC 8825-1,
              Information technology - ASN.1 encoding rules:
              Specification of Basic Encoding Rules (BER), Canonical
              Encoding Rules (CER) and Distinguished Encoding Rules
              (DER).

   [UCS]      ISO/IEC 10646-1:2000, Information technology - Universal
              Multiple-Octet Coded Character Set (UCS) - Part 1:
              Architecture and Basic Multilingual Plane.

   [UNICODE]  The Unicode Consortium, "The Unicode Standard, Version
              4.0", Boston, MA, Addison-Wesley Developers Press, 2003.
              ISBN 0-321-18578-1.

   [XML10]    Bray, T., Paoli, J., Sperberg-McQueen, C., Maler, E. and
              F. Yergeau, "Extensible Markup Language (XML) 1.0 (Fourth
              Edition)", W3C Recommendation,
              http://www.w3.org/TR/2006/REC-xml-20060816, August 2006.

   [XML11]    Bray, T., Paoli, J., Sperberg-McQueen, C., Maler, E.,
              Yergeau, F., and J. Cowan, "Extensible Markup Language
              (XML) 1.1 (Second Edition)", W3C Recommendation,
              http://www.w3.org/TR/2006/REC-xml11-20060816, August 2006.

   [XMLNS10]  Bray, T., Hollander, D., Layman, A., and R. Tobin,
              "Namespaces in XML 1.0 (Second Edition)", W3C
              Recommendation,
              http://www.w3.org/TR/2006/REC-xml-names-20060816, August
              2006.

   [XMLNS11]  Bray, T., Hollander, D., Layman, A. and R. Tobin,
              "Namespaces in XML 1.1 (Second Edition)", W3C
              Recommendation,
              http://www.w3.org/TR/2006/REC-xml-names11-20060816, August
              2006.

   [INFOSET]  Cowan, J. and R. Tobin, "XML Information Set (Second
              Edition)", W3C Recommendation,
              http://www.w3.org/TR/2004/REC-xml-infoset-20040204,
              February 2004.







Legg & Prager                 Experimental                     [Page 76]

RFC 4910               Robust XML Encoding Rules               July 2007


   [XSD1]     Thompson, H., Beech, D., Maloney, M. and N. Mendelsohn,
              "XML Schema Part 1: Structures Second Edition", W3C
              Recommendation,
              http://www.w3.org/TR/2004/REC-xmlschema-1-20041028/,
              October 2004.

12.2.  Informative References



   [GSER]     Legg, S., "Generic String Encoding Rules (GSER) for ASN.1
              Types", RFC 3641, October 2003.

   [X.691]    ITU-T Recommendation X.691 (07/02) | ISO/IEC 8825-4:2002,
              Information technology - ASN.1 encoding rules:
              Specification of Packed Encoding Rules (PER).

   [X.693]    ITU-T Recommendation X.693 (12/01) | ISO/IEC 8825-4:2002,
              Information technology - ASN.1 encoding rules: XML
              encoding rules (XER).

   [XSD2]     Biron, P. and A. Malhotra, "XML Schema Part 2: Datatypes
              Second Edition", W3C Recommendation,
              http://www.w3.org/TR/2004/REC-xmlschema-2-20041028/,
              October 2004.

   [CXML]     Boyer, J., "Canonical XML Version 1.0", W3C
              Recommendation,
              http://www.w3.org/TR/2001/REC-xml-c14n-20010315, March
              2001.























Legg & Prager                 Experimental                     [Page 77]

RFC 4910               Robust XML Encoding Rules               July 2007


Appendix A.  Additional Basic Definitions Module



   This appendix is normative.

   AdditionalBasicDefinitions
       { iso(1) identified-organization(3) dod(6)
         internet(1) private(4) enterprise(1)
         xmled(21472) asnx(1) module(0) basic(0) }

   -- Copyright (C) The IETF Trust (2007).  This version of
   -- this ASN.1 module is part of RFC 4910; see the RFC itself
   -- for full legal notices.
   --
   -- Regarding this ASN.1 module or any portion of it, the authors
   -- make no guarantees and are not responsible for any damage
   -- resulting from its use.  The authors grant irrevocable permission
   -- to anyone to use, modify, and distribute it in any way that does
   -- not diminish the rights of anyone else to use, modify, and
   -- distribute it, provided that redistributed derivative works do
   -- not contain misleading author or version information.
   -- Derivative works need not be licensed under similar terms.

   DEFINITIONS
   RXER INSTRUCTIONS
   AUTOMATIC TAGS
   EXTENSIBILITY IMPLIED ::= BEGIN

   Markup ::= CHOICE {
       text    SEQUENCE {
           prolog      UTF8String (SIZE(1..MAX)) OPTIONAL,
           prefix      NCName OPTIONAL,
           attributes  UTF8String (SIZE(1..MAX)) OPTIONAL,
           content     UTF8String (SIZE(1..MAX)) OPTIONAL
       }
   }

   AnyURI ::= UTF8String (CONSTRAINED BY
                  { -- conforms to the format of a URI -- })

   NCName ::= UTF8String (CONSTRAINED BY
                  { -- conforms to the NCName production of
                    -- Namespaces in XML 1.0 -- })

   Name ::= UTF8String (CONSTRAINED BY
                  { -- conforms to the Name production of XML -- })






Legg & Prager                 Experimental                     [Page 78]

RFC 4910               Robust XML Encoding Rules               July 2007


   QName ::= SEQUENCE {
       namespace-name  AnyURI OPTIONAL,
       local-name      NCName
   }

   ENCODING-CONTROL RXER

       TARGET-NAMESPACE "urn:ietf:params:xml:ns:asnx" PREFIX "asnx"

       COMPONENT context [ATTRIBUTE] [LIST] SEQUENCE OF prefix NCName

   END

Authors' Addresses



   Dr. Steven Legg
   eB2Bcom
   Suite 3, Woodhouse Corporate Centre
   935 Station Street
   Box Hill North, Victoria 3129
   AUSTRALIA

   Phone: +61 3 9896 7830
   Fax:   +61 3 9896 7801
   EMail: steven.legg@eb2bcom.com


   Dr. Daniel Prager

   EMail: dap@austhink.com





















Legg & Prager                 Experimental                     [Page 79]

RFC 4910               Robust XML Encoding Rules               July 2007


Full Copyright Statement



   Copyright (C) The IETF Trust (2007).

   This document is subject to the rights, licenses and restrictions
   contained in BCP 78, and except as set forth therein, the authors
   retain all their rights.

   This document and the information contained herein are provided on an
   "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
   OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND
   THE INTERNET ENGINEERING TASK FORCE DISCLAIM 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.

Intellectual Property



   The IETF takes no position regarding the validity or scope of any
   Intellectual Property Rights or other rights that might be claimed to
   pertain to the implementation or use of the technology described in
   this document or the extent to which any license under such rights
   might or might not be available; nor does it represent that it has
   made any independent effort to identify any such rights.  Information
   on the procedures with respect to rights in RFC documents can be
   found in BCP 78 and BCP 79.

   Copies of IPR disclosures made to the IETF Secretariat and any
   assurances of licenses to be made available, or the result of an
   attempt made to obtain a general license or permission for the use of
   such proprietary rights by implementers or users of this
   specification can be obtained from the IETF on-line IPR repository at
   http://www.ietf.org/ipr.

   The IETF invites any interested party to bring to its attention any
   copyrights, patents or patent applications, or other proprietary
   rights that may cover technology that may be required to implement
   this standard.  Please address the information to the IETF at
   ietf-ipr@ietf.org.

Acknowledgement



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







Legg & Prager                 Experimental                     [Page 80]