ASCII Escaping of Unicode Characters

Versions: 00 01 02 03 04 05 06 07 rfc5137                               
Network Working Group                                         J. Klensin
Internet-Draft                                          January 29, 2007
Expires: August 2, 2007

                  ASCII Escaping of Unicode Characters

Status of this Memo

   By submitting this Internet-Draft, each author represents that any
   applicable patent or other IPR claims of which he or she is aware
   have been or will be disclosed, and any of which he or she becomes
   aware will be disclosed, in accordance with Section 6 of BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF), its areas, and its working groups.  Note that
   other groups may also distribute working documents as Internet-

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress."

   The list of current Internet-Drafts can be accessed at

   The list of Internet-Draft Shadow Directories can be accessed at

   This Internet-Draft will expire on August 2, 2007.

Copyright Notice

   Copyright (C) The IETF Trust (2007).


   There are a number of circumstances in which an escape mechanism is
   needed in conjunction with a protocol to encode characters that
   cannot be represented or transmitted directly.  With ASCII coding the
   traditional escape has been either the decimal or hexadecimal offset
   of the character, written in a variety of different ways.  The move
   to Unicode, where characters occupy two or more octets and may be
   coded in several different forms, has further complicated the
   question of escapes.  This document discusses some options now in use
   and makes a proposal for general use in new IETF protocols and

Klensin                  Expires August 2, 2007                 [Page 1]

Internet-Draft               Unicode Escapes                January 2007

   protocols that are now being internationalized.

Warning: Interim Draft

   This version of the specification is an interim draft, intended to
   correct (or at least note) obvious errors and reflect some of the
   discussion on the mailing list in order to help focus the discussion
   on remaining critical issues.  It is not complete, nor does it claim
   to accurately reflect all of the discussions so far.

Table of Contents

   1.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  3
     1.1.  Context and Background . . . . . . . . . . . . . . . . . .  3
     1.2.  Terminology  . . . . . . . . . . . . . . . . . . . . . . .  4
     1.3.  Discussion List  . . . . . . . . . . . . . . . . . . . . .  4
   2.  Encodings that Represent Unicode Code Points . . . . . . . . .  4
     2.1.  Unicode Table Position versus UTF-8 Octets . . . . . . . .  4
   3.  Referring to Unicode Characters  . . . . . . . . . . . . . . .  5
   4.  Syntax for Code Point Escapes  . . . . . . . . . . . . . . . .  5
   5.  Presentation Variants for Unicode Code Points  . . . . . . . .  6
     5.1.  The C Programming Language: Backslash-U  . . . . . . . . .  6
     5.2.  HTML and XML . . . . . . . . . . . . . . . . . . . . . . .  7
     5.3.  Perl: A Hexadecimal String . . . . . . . . . . . . . . . .  7
     5.4.  Java: Escaped UTF-16 . . . . . . . . . . . . . . . . . . .  7
   6.  Security Considerations  . . . . . . . . . . . . . . . . . . .  7
   7.  Acknowledgments  . . . . . . . . . . . . . . . . . . . . . . .  8
   8.  Change log . . . . . . . . . . . . . . . . . . . . . . . . . .  8
     8.1.  Changes in -01 . . . . . . . . . . . . . . . . . . . . . .  8
   9.  References . . . . . . . . . . . . . . . . . . . . . . . . . .  8
     9.1.  Normative References . . . . . . . . . . . . . . . . . . .  8
     9.2.  Informative References . . . . . . . . . . . . . . . . . .  9
   Author's Address . . . . . . . . . . . . . . . . . . . . . . . . .  9
   Intellectual Property and Copyright Statements . . . . . . . . . . 10

Klensin                  Expires August 2, 2007                 [Page 2]

Internet-Draft               Unicode Escapes                January 2007

1.  Introduction

1.1.  Context and Background

   There are a number of circumstances in which an escape mechanism is
   needed in conjunction with a protocol to encode characters that
   cannot be represented or transmitted directly.  With ASCII [ASCII]
   coding the traditional escape has been either the decimal or
   hexadecimal offset of the character, written in a variety of
   different ways.  For example, in different contexts, we have seen
   %dNN or %NN for the decimal form, %NN, %xNN, X'nn', and %X'NN' for
   the hexadecimal form. "%NN" has become popular in recent years to
   represent a hexadecimal value without further qualification, perhaps
   as a consequence of its use in URLs and their prevalence.  There are
   even some applications around in which octal forms are used and,
   while they do not generalize well, the MIME Quoted-Printable and
   Encoded-word forms can be thought of as yet another set of escapes.
   So, even for the fairly simple cases of ASCII and standard built by
   extending ASCII, such as the ISO 8859 family, we have been living
   with several different escaping forms, each the result of some

   When one moves to Unicode [Unicode] [ISO10646], where characters
   occupy two or more octets and may be coded in several different
   forms, the question of escapes becomes even more complicated.  In
   particular, we have seen fairly extensive use of both hexadecimal
   representations of the UTF-8 encoding [RFC3629] of a character and
   variations on the U+NNNN[N[N]] notation commonly used in conjunction
   with the Unicode Standard.  This document proposes that new
   protocols, and protocols being internationalized, SHOULD use some
   contextually-appropriate variation on the latter unless other
   considerations outweigh those described here.

   This recommendation is not applicable to protocols that already
   accept native UTF-8 or some other encoding of Unicode.  In general,
   when protocols are internationalized, it is preferable to accept
   those forms rather than using escapes.  This recommendation applies
   to cases, including transition arrangements, in which that is not

   In addition to the protocol contexts addressed in this specification,
   escapes to represent Unicode characters also appear in presentations
   to users, i.e., in user interfaces (UI).  The formats specified in,
   and the reasoning of, this document may be applicable in UI contexts
   as well, but this is not a proposal to standardize UI or presentation

Klensin                  Expires August 2, 2007                 [Page 3]

Internet-Draft               Unicode Escapes                January 2007

1.2.  Terminology

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   document are to be interpreted as described in [RFC2119].

1.3.  Discussion List

   Discussion of this document should be addressed to the mailing list.

2.  Encodings that Represent Unicode Code Points

   There are many different ways to designate, encode, or call out a
   Unicode character.  Given adequate decoding facilities, all of these
   other than the formal character name are equivalent.  However, when
   information about characters is to be processed by people,
   information about the Unicode code point is preferable to a further
   encoding of the encoded form of the character.  It is also desirable
   to use hexadecimal references to code points because the Unicode
   Standard is organized on a hexadecimal basis.

   These issues are discussed in the following subsections.

2.1.  Unicode Table Position versus UTF-8 Octets

   There are two major families of ways to represent Unicode characters.
   One uses the code point position in the table in some representation
   (see the next section), the other encodes the octets of the UTF-8
   encoding.  Some other options are possible, but they have been rare
   in practice.  This specification recommends that, in the absence of
   compelling reasons to do otherwise, the Unicode code point forms
   SHOULD be used rather than the UTF-8 ones.  There are several reasons
   for this, including:
   o  One reason for the success of many IETF protocols is that they use
      human-interpretable text forms to communicate, rather than
      encodings that generally require computer programs (or hand
      simulation of algorithms) to decode.  This suggests that the
      presentation form should reference the Unicode tables for
      characters and to do so as simply as possible.
   o  The nature of UTF-8 implies that a decimal or hexadecimal numeral
      representation of UTF-8 requires conversion to the UTF-8 form,
      then conversion from the UTF-8 form to a Unicode character
      position form in order to look the character up in a table.  That
      may be appropriate in some cases where the goal is really to
      represent the UTF-8 form but, in general, it just obscures desired
      information and makes errors more likely and debugging harder.

Klensin                  Expires August 2, 2007                 [Page 4]

Internet-Draft               Unicode Escapes                January 2007

   o  Except for characters in the ASCII subset of Unicode (U+0000
      through U+007F), the character code position form is generally
      more compact than forms based on coding UTF-8 octets, sometimes
      much more compact.

   The same considerations that apply to encoding of UTF-8 octets also
   apply to more compact ACE encodings such as the "bootstring" encoding
   [RFC3492] with or without its "Punycode" profile.

3.  Referring to Unicode Characters

   Regardless of what decisions are made about escapes for Unicode
   characters in protocol or similar contexts, references to Unicode
   characters in text SHOULD use the U+NNN[N[N]] syntax for code point
   references specified in the Unicode Standard, where the NNN... string
   consists of hexadecimal numbers.

4.  Syntax for Code Point Escapes

   There are many options for code point escapes, some of which are
   summarized below.  All are equivalent in content and semantics -- the
   differences lie in syntax.  The best choice of syntax for a
   particular protocol or other application depends on that application:
   one form may simply "fit" better in a given context than others.  It
   is clear, however, that hexadecimal values are preferable to other
   alternatives: Systems based on decimal or octal offsets SHOULD NOT be

   Since this specification does not recommend one specific syntax,
   protocols specifications that use escapes MUST define the syntax they
   are using, including any necessary escapes to permit the escape
   sequence to be used literally.

   The application designer selecting a format should consider at least
   the following factors:
   o  If similar or related protocols already use one form, it may be
      best to select that form for consistency and predictability.
   o  A Unicode code point can fall in the range from U+0000 to
      U+10FFFF.  Different escape systems may use four, five, six, or
      eight hexadecimal digits.  To avoid clever syntax tricks and the
      consequent risk of confusion and errors, forms that use explicit
      string terminators are generally preferred over other
      alternatives.  In many contexts, symmetric paired delimiters are
      easier to recognize and understand than visually-unrelated ones.

Klensin                  Expires August 2, 2007                 [Page 5]

Internet-Draft               Unicode Escapes                January 2007

   o  Forms that require decoding surrogate pairs share most of the
      problems that appear with encoding of UTF-8 octets and SHOULD NOT,
      in general, be used.

5.  Presentation Variants for Unicode Code Points

   There are a number of different ways to represent a Unicode code
   point position.  No one of them appears to be "best" for all
   contexts.  In addition, when an escape is needed for the escape
   mechanism itself, the optimal one of those might differ from one
   context to another.

   Some forms that are in popular use and that might reasonably be
   considered for use in a given protocol, are described below and
   identified with a current-use context when feasible.

5.1.  The C Programming Language: Backslash-U

   The forms
   \UNNNNNNNN (for any Unicode character) and
   \uNNNN (for Unicode characters in plane 0)
   are utilized in the C Programming Language [ISO-C] when an ASCII
   escape for embedded Unicode characters is needed.

   Specifically, in ABNF [RFC4234], [[anchor10: Note in Draft: The ABNF
   that follows is _not_ valid because ABNF literal strings are not
   case-sensitive.  Once more substantive issues are resolved, this
   syntax will need to be corrected, either to escape the "u" and "U"
   (at least) or to note an exception from the standard ABNF rules.  If
   the charaters are escaped, a note will be necessary that the escapes
   are references to ASCII (or Unicode) character abstractions, not a
   limitation to the use of those particular octets.]]
   EmbeddedUnicodeChar =  BMP-form / Full-form
   Hex-quad =  4*4HexDigit
   BMP-form =  "\u" Hex-quad
   Full-form =  "\U" 2*2Hex-quad
   HexDigit =  "0" / "1" / "2"/ "3"/ "4"/ "5"/ "6"/ "7"/ "8"/ "9"/ "A"/
      "B" / "C"/ "D"/ "E"/ "F"

   There are disadvantages of this form which may be significant.
   First, the use of a case variation (between "u" for the four digit
   form and "U" for the eight digit form) may not seem natural in
   environments in which upper and lower case characters are generally
   considered equivalent and might be confusing to people who are not
   very familiar with Latin-based alphabets (although those people might
   have even more trouble reading relevant English text and
   explanations).  Second, the very fact that there are several

Klensin                  Expires August 2, 2007                 [Page 6]

Internet-Draft               Unicode Escapes                January 2007

   different conventions that start in \u or \U may become a source of
   confusion as people make incorrect assumptions about what they are
   looking at.  The similarity between this convention and the
   surrogate-using Java one (see Section 5.4) are particularly
   unfortunate examples of this.

5.2.  HTML and XML

   HTML and XML use the form &#xNNNN;.  Like the Perl form, this form
   has a clear terminator, reducing ambiguity.  However, it is generally
   considered ugly and awkward outside of its native HTML, XML, and
   similar contexts.

5.3.  Perl: A Hexadecimal String

   Perl uses the form \x(NNN...).  The advantage of this form is that
   there are explicit delimiters, resolving the issue of having
   variable-length strings or using the case-change mechanism of the
   proposed form to distinguish between Plane 0 and more general forms.
   Some other programming languages would tend to favor X'NNN...' forms
   for hexadecimal strings and perhaps U'NNNN...' for Unicode-specific
   strings, but those forms do not seem to be in use around the IETF.

5.4.  Java: Escaped UTF-16

   Java uses the form \uNNNN, but can represent characters outside Plane
   0 (i.e., above U+FFFF) only by the use of surrogate pairs.  Decoding
   (or de-mapping) surrogates raises some of the same issues as the use
   of UTF-8 octets discussed above.  For characters in Plane 0, the Java
   form is identical to the recommended Plane 0-only form recommended

6.  Security Considerations

   This document proposes a specific mechanism for encoding Unicode
   characters when other considerations do not apply.  Since the
   encoding is unambiguous and normalization issues are not involved, it
   should not introduce any security issues that are not present as a
   result of simple use of non-ASCII characters, no matter how they are
   encoded.  The mechanism suggested should slightly lower the risks of
   confusing users with encoded characters by making the identity of the
   characters being used somewhat more obvious than some of the

Klensin                  Expires August 2, 2007                 [Page 7]

Internet-Draft               Unicode Escapes                January 2007

7.  Acknowledgments

   This document was produced in response to a series of discussions
   within the IETF Applications Area and as part of work on email
   internationalization and internationalized domain name updates.  It
   is a synthesis of a large number of discussions, the comments of the
   participants in which are gratefully acknowledged.  The help of Mark
   Davis in constructing a list of alternative presentations and
   selecting among them was especially important.

   Stephane Bortzmeyer, Frank Ellermann, Clive D.W. Feather, Bill
   McQuillan, Simon Josefsson, and Julian Reschke provided careful
   reading and some corrections and suggestions on the initial draft.
   Taken together, their suggestions motivated the significant revision
   of this document and its recommendations between version -00 and
   version -01.

8.  Change log

   [[anchor14: RFC Editor: Please remove this section before

8.1.  Changes in -01

   o  Corrected ABNF syntax for Hex-quad and Full-form.

9.  References

9.1.  Normative References

              International Organization for Standardization,
              "Information Technology - Universal Multiple- Octet Coded
              Character Set (UCS)"", ISO/IEC 10646:2003, December 2003.

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

   [RFC3629]  Yergeau, F., "UTF-8, a transformation format of ISO
              10646", STD 63, RFC 3629, November 2003.

   [RFC4234]  Crocker, D., Ed. and P. Overell, "Augmented BNF for Syntax
              Specifications: ABNF", RFC 4234, October 2005.

   [Unicode]  The Unicode Consortium, "The Unicode Standard, Version
              5.0", 2006.

Klensin                  Expires August 2, 2007                 [Page 8]

Internet-Draft               Unicode Escapes                January 2007

              (Addison-Wesley, 2006.  ISBN 0-321-48091-0).

9.2.  Informative References

   [ASCII]    American National Standards Institute (formerly United
              States of America Standards Institute), "USA Code for
              Information Interchange", ANSI X3.4-1968, 1968.

              ANSI X3.4-1968 has been replaced by newer versions with
              slight modifications, but the 1968 version remains
              definitive for the Internet.

   [ISO-C]    International Organization for Standardization,
              "Information technology --  Programming languages -- C",
              ISO/IEC 9899:1999, 1999.

   [RFC3492]  Costello, A., "Punycode: A Bootstring encoding of Unicode
              for Internationalized Domain Names in Applications
              (IDNA)", RFC 3492, March 2003.

Author's Address

   John C Klensin
   1770 Massachusetts Ave, #322
   Cambridge, MA  02140

   Phone: +1 617 245 1457

Klensin                  Expires August 2, 2007                 [Page 9]

Internet-Draft               Unicode Escapes                January 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

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

   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


   Funding for the RFC Editor function is provided by the IETF
   Administrative Support Activity (IASA).

Klensin                  Expires August 2, 2007                [Page 10]