source: draft-ietf-iri-3987bis/draft-ietf-iri-3987bis.xml @ 28

Last change on this file since 28 was 28, checked in by duerst@…, 9 years ago

Fixing Ticket #29 (include tag ranges in iprivate production):

  • Created text that points out the difference from RFC 3987, as requested by Larry. The text reads:

Added the tag range (U+E0000-E0FFF) to the iprivate production. Some IRIs generated with the new syntax may fail to pass very strict checks relying on the old syntax.

  • To keep this text (the "Change Log" section is to be removed by the RFC editor), created a new section just above the "Change Log" section, entitled "Main Changes Since RFC 3987".
  • Moved Larry's OLD/NEW text about the processing model to this section (It may need some more tweaking.) from the Change Log section.
  • Added a dummy subsection to the "Major Changes" section to indicate that there are other major changes that need to go in this section.
  • Property svn:executable set to *
File size: 127.2 KB
Line 
1<?xml version="1.0"?>
2<!DOCTYPE rfc SYSTEM "rfc2629.dtd" [
3<!ENTITY rfc1738 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.1738.xml">
4<!ENTITY rfc2045 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2045.xml">
5<!ENTITY rfc2119 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2119.xml">
6<!ENTITY rfc2130 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2130.xml">
7<!ENTITY rfc2141 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2141.xml">
8<!ENTITY rfc2192 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2192.xml">
9<!ENTITY rfc2277 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2277.xml">
10<!ENTITY rfc2368 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2368.xml">
11<!ENTITY rfc2384 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2384.xml">
12<!ENTITY rfc2396 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2396.xml">
13<!ENTITY rfc2397 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2397.xml">
14<!ENTITY rfc2616 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2616.xml">
15<!ENTITY rfc2640 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.2640.xml">
16<!ENTITY rfc3490 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.3490.xml">
17<!ENTITY rfc3491 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.3491.xml">
18<!ENTITY rfc3629 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.3629.xml">
19<!ENTITY rfc3986 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.3986.xml">
20<!ENTITY rfc3986 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.3987.xml">
21<!ENTITY rfc5890 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.5890.xml">
22<!ENTITY rfc5891 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.RFC.5891.xml">
23]>
24<?rfc strict='yes'?>
25
26<?xml-stylesheet type='text/css' href='rfc2629.css' ?>
27<?xml-stylesheet type='text/xsl' href='rfc2629.xslt' ?>
28<?rfc symrefs='yes'?>
29<?rfc sortrefs='yes'?>
30<?rfc iprnotified="no" ?>
31<?rfc toc='yes'?>
32<?rfc compact='yes'?>
33<?rfc subcompact='no'?>
34<rfc ipr="pre5378Trust200902" docName="draft-ietf-iri-3987bis-03" category="std" xml:lang="en" obsoletes="3987">
35<front>
36<title abbrev="IRIs">Internationalized Resource Identifiers (IRIs)</title>
37
38  <author initials="M.J." surname="Duerst" fullname='Martin Duerst'>
39    <!-- (Note: Please write "Duerst" with u-umlaut wherever
40      possible, for example as "D&#252;rst" in XML and HTML) -->
41  <organization abbrev="Aoyama Gakuin University">Aoyama Gakuin University</organization>
42  <address>
43  <postal>
44  <street>5-10-1 Fuchinobe</street>
45  <city>Sagamihara</city>
46  <region>Kanagawa</region>
47  <code>229-8558</code>
48  <country>Japan</country>
49  </postal>
50  <phone>+81 42 759 6329</phone>
51  <facsimile>+81 42 759 6495</facsimile>
52  <email>duerst@it.aoyama.ac.jp</email>
53  <uri>http://www.sw.it.aoyama.ac.jp/D%C3%BCrst/<!-- (Note: This is the percent-encoded form of an IRI)--></uri>
54  </address>
55</author>
56
57<author initials="M.L." surname="Suignard" fullname="Michel Suignard">
58   <organization>Unicode Consortium</organization>
59   <address>
60   <postal>
61   <street></street>
62   <street>P.O. Box 391476</street>
63   <city>Mountain View</city>
64   <region>CA</region>
65   <code>94039-1476</code>
66   <country>U.S.A.</country>
67   </postal>
68   <phone>+1-650-693-3921</phone>
69   <email>michel@unicode.org</email>
70   <uri>http://www.suignard.com</uri>
71   </address>
72</author>
73<author initials="L." surname="Masinter" fullname="Larry Masinter">
74   <organization>Adobe</organization>
75   <address>
76   <postal>
77   <street>345 Park Ave</street>
78   <city>San Jose</city>
79   <region>CA</region>
80   <code>95110</code>
81   <country>U.S.A.</country>
82   </postal>
83   <phone>+1-408-536-3024</phone>
84   <email>masinter@adobe.com</email>
85   <uri>http://larry.masinter.net</uri>
86   </address>
87</author>
88
89<date year="2010" month="October" day="25"/>
90<area>Applications</area>
91<workgroup>Internationalized Resource Identifiers (iri)</workgroup>
92<keyword>IRI</keyword>
93<keyword>Internationalized Resource Identifier</keyword>
94<keyword>UTF-8</keyword>
95<keyword>URI</keyword>
96<keyword>URL</keyword>
97<keyword>IDN</keyword>
98<keyword>LEIRI</keyword>
99
100<abstract>
101<t>This document defines the Internationalized Resource Identifier
102(IRI) protocol element, as an extension of the Uniform Resource
103Identifier (URI).  An IRI is a sequence of characters from the
104Universal Character Set (Unicode/ISO 10646). Grammar and processing
105rules are given for IRIs and related syntactic forms.</t>
106
107<t>In addition, this document provides named additional rule sets
108for processing otherwise invalid IRIs, in a way that supports
109other specifications that wish to mandate common behavior for
110'error' handling. In particular, rules used in some XML languages
111(LEIRI) and web applications are given.</t>
112
113<t>Defining IRI as new protocol element (rather than updating or
114extending the definition of URI) allows independent orderly
115transitions: other protocols and languages that use URIs must
116explicitly choose to allow IRIs.</t>
117
118<t>Guidelines are provided for the use and deployment of IRIs and
119related protocol elements when revising protocols, formats, and
120software components that currently deal only with URIs.</t>
121
122</abstract>
123  <note title='RFC Editor: Please remove the next paragraph before publication.'>
124    <t>This document is intended to update RFC 3987 and move towards IETF
125    Draft Standard.  For discussion and comments on this
126    draft, please join the IETF IRI WG by subscribing to the mailing
127    list public-iri@w3.org. For a list of open issues, please see
128    the issue tracker of the WG at http://trac.tools.ietf.org/wg/iri/trac/report/1.</t>
129</note>
130</front>
131<middle>
132
133<section title="Introduction">
134
135<section title="Overview and Motivation" anchor="overview">
136
137<t>A Uniform Resource Identifier (URI) is defined in <xref
138target="RFC3986"/> as a sequence of characters chosen from a limited
139subset of the repertoire of US-ASCII <xref target="ASCII"/>
140characters.</t>
141
142<t>The characters in URIs are frequently used for representing words
143of natural languages.  This usage has many advantages: Such URIs are
144easier to memorize, easier to interpret, easier to transcribe, easier
145to create, and easier to guess. For most languages other than English,
146however, the natural script uses characters other than A - Z. For many
147people, handling Latin characters is as difficult as handling the
148characters of other scripts is for those who use only the Latin
149alphabet. Many languages with non-Latin scripts are transcribed with
150Latin letters. These transcriptions are now often used in URIs, but
151they introduce additional difficulties.</t>
152
153<t>The infrastructure for the appropriate handling of characters from
154additional scripts is now widely deployed in operating system and
155application software. Software that can handle a wide variety of
156scripts and languages at the same time is increasingly common. Also,
157an increasing number of protocols and formats can carry a wide range of
158characters.</t>
159
160<t>URIs are used both as a protocol element (for transmission and
161processing by software) and also a presentation element (for display
162and handling by people who read, interpret, coin, or guess them). The
163transition between these roles is more difficult and complex when
164dealing with the larger set of characters than allowed for URIs in
165<xref target="RFC3986"/>. </t>
166
167<t>This document defines the protocol element called Internationalized
168Resource Identifier (IRI), which allow applications of URIs to be
169extended to use resource identifiers that have a much wider repertoire
170of characters. It also provides corresponding "internationalized"
171versions of other constructs from <xref target="RFC3986"/>, such as
172URI references. The syntax of IRIs is defined in <xref
173target="syntax"/>.
174</t>
175
176<t>Using characters outside of A - Z in IRIs adds a number of
177difficulties. <xref target="Bidi"/> discusses the special case of
178bidirectional IRIs using characters from scripts written
179right-to-left.  <xref target="equivalence"/> discusses various forms
180of equivalence between IRIs. <xref target="IRIuse"/> discusses the use
181of IRIs in different situations.  <xref target="guidelines"/> gives
182additional informative guidelines.  <xref target="security"/>
183discusses IRI-specific security considerations.</t>
184</section> <!-- overview -->
185
186<section title="Applicability" anchor="Applicability">
187
188<t>IRIs are designed to allow protocols and software that deal with
189URIs to be updated to handle IRIs. A "URI scheme" (as defined by <xref
190target="RFC3986"/> and registered through the IANA process defined in
191<xref target="RFC4395bis"/> also serves as an "IRI scheme". Processing of
192IRIs is accomplished by extending the URI syntax while retaining (and
193not expanding) the set of "reserved" characters, such that the syntax
194for any URI scheme may be uniformly extended to allow non-ASCII
195characters. In addition, following parsing of an IRI, it is possible
196to construct a corresponding URI by first encoding characters outside
197of the allowed URI range and then reassembling the components.
198</t>
199
200<t>Practical use of IRIs forms in place of URIs forms depends on the
201following conditions being met:</t>
202
203<t><list style="hanging">
204   
205<t hangText="a.">A protocol or format element MUST be explicitly designated to be
206  able to carry IRIs. The intent is to avoid introducing IRIs into
207  contexts that are not defined to accept them.  For example, XML
208  schema <xref target="XMLSchema"/> has an explicit type "anyURI" that
209  includes IRIs and IRI references. Therefore, IRIs and IRI references
210  can be in attributes and elements of type "anyURI".  On the other
211  hand, in the <xref target="RFC2616"/> definition of HTTP/1.1, the
212  Request URI is defined as a URI, which means that direct use of IRIs
213  is not allowed in HTTP requests.</t>
214
215<t hangText="b.">The protocol or format carrying the IRIs MUST have a
216  mechanism to represent the wide range of characters used in IRIs,
217  either natively or by some protocol- or format-specific escaping
218  mechanism (for example, numeric character references in <xref
219  target="XML1"/>).</t>
220
221<t hangText="c.">The URI scheme definition, if it explicitly allows a
222  percent sign ("%") in any syntactic component, SHOULD define the
223  interpretation of sequences of percent-encoded octets (using "%XX"
224  hex octets) as octet from sequences of UTF-8 encoded strings; this
225  is recommended in the guidelines for registering new schemes, <xref
226  target="RFC4395bis"/>.  For example, this is the practice for IMAP URLs
227  <xref target="RFC2192"/>, POP URLs <xref target="RFC2384"/> and the
228  URN syntax <xref target="RFC2141"/>). Note that use of
229  percent-encoding may also be restricted in some situations, for
230  example, URI schemes that disallow percent-encoding might still be
231  used with a fragment identifier which is percent-encoded (e.g.,
232  <xref target="XPointer"/>). See <xref target="UTF8use"/> for further
233  discussion.</t>
234</list></t>
235
236</section> <!-- applicability -->
237
238<section title="Definitions" anchor="sec-Definitions">
239 
240<t>The following definitions are used in this document; they follow the
241terms in <xref target="RFC2130"/>, <xref target="RFC2277"/>, and
242<xref target="ISO10646"/>.</t>
243<t><list style="hanging">
244   
245<t hangText="character:">A member of a set of elements used for the
246    organization, control, or representation of data. For example,
247    "LATIN CAPITAL LETTER A" names a character.</t>
248   
249<t hangText="octet:">An ordered sequence of eight bits considered as a
250    unit.</t>
251   
252<t hangText="character repertoire:">A set of characters (set in the
253    mathematical sense).</t>
254   
255<t hangText="sequence of characters:">A sequence of characters (one
256    after another).</t>
257   
258<t hangText="sequence of octets:">A sequence of octets (one after
259    another).</t>
260   
261<t hangText="character encoding:">A method of representing a sequence
262    of characters as a sequence of octets (maybe with variants). Also,
263    a method of (unambiguously) converting a sequence of octets into a
264    sequence of characters.</t>
265   
266<t hangText="charset:">The name of a parameter or attribute used to
267    identify a character encoding.</t>
268   
269<t hangText="UCS:">Universal Character Set. The coded character set
270    defined by ISO/IEC 10646 <xref target="ISO10646"/> and the Unicode
271    Standard <xref target="UNIV4"/>.</t>
272   
273<t hangText="IRI reference:">Denotes the common usage of an
274    Internationalized Resource Identifier. An IRI reference may be
275    absolute or relative.  However, the "IRI" that results from such a
276    reference only includes absolute IRIs; any relative IRI references
277    are resolved to their absolute form.  Note that in <xref
278    target="RFC2396"/> URIs did not include fragment identifiers, but
279    in <xref target="RFC3986"/> fragment identifiers are part of
280    URIs.</t>
281   
282<t hangText="URL:">The term "URL" was originally used <xref
283   target="RFC1738"/> for roughly what is now called a "URI".  Books,
284   software and documentation often refers to URIs and IRIs using the
285   "URL" term. Some usages restrict "URL" to those URIs which are not
286   URNs. Because of the ambiguity of the term using the term "URL" is
287   NOT RECOMMENDED in formal documents.</t>
288
289<t hangText="LEIRI (Legacy Extended IRI) processing:">  This term was used in
290   various XML specifications to refer
291   to strings that, although not valid IRIs, were acceptable input to
292   the processing rules in <xref target="LEIRIspec" />.</t>
293
294<t hangText="(Web Address, Hypertext Reference, HREF):"> These terms have been
295   added in this document for convenience, to allow other
296   specifications to refer to those strings that, although not valid
297   IRIs, are acceptable input to the processing rules in <xref
298   target="webaddress"/>. This usage corresponds to the parsing rules
299   of some popular web browsing applications.
300   ISSUE: Need to find a good name/abbreviation for these.</t>
301   
302<t hangText="running text:">Human text (paragraphs, sentences,
303   phrases) with syntax according to orthographic conventions of a
304   natural language, as opposed to syntax defined for ease of
305   processing by machines (e.g., markup, programming languages).</t>
306   
307<t hangText="protocol element:">Any portion of a message that affects
308    processing of that message by the protocol in question.</t>
309   
310<t hangText="presentation element:">A presentation form corresponding
311    to a protocol element; for example, using a wider range of
312    characters.</t>
313   
314<t hangText="create (a URI or IRI):">With respect to URIs and IRIs,
315     the term is used for the initial creation. This may be the
316     initial creation of a resource with a certain identifier, or the
317     initial exposition of a resource under a particular
318     identifier.</t>
319   
320<t hangText="generate (a URI or IRI):">With respect to URIs and IRIs,
321     the term is used when the identifier is generated by derivation
322     from other information.</t>
323
324<t hangText="parsed URI component:">When a URI processor parses a URI
325   (following the generic syntax or a scheme-specific syntax, the result
326   is a set of parsed URI components, each of which has a type
327   (corresponding to the syntactic definition) and a sequence of URI
328   characters.  </t>
329
330<t hangText="parsed IRI component:">When an IRI processor parses
331   an IRI directly, following the general syntax or a scheme-specific
332   syntax, the result is a set of parsed IRI components, each of
333   which has a type (corresponding to the syntactice definition)
334   and a sequence of IRI characters. (This definition is analogous
335   to "parsed URI component".)</t>
336
337<t hangText="IRI scheme:">A URI scheme may also be known as
338   an "IRI scheme" if the scheme's syntax has been extended to
339   allow non-US-ASCII characters according to the rules in this
340   document.</t>
341
342</list></t>
343</section> <!-- definitions -->
344<section title="Notation" anchor="sec-Notation">
345     
346<t>RFCs and Internet Drafts currently do not allow any characters
347outside the US-ASCII repertoire. Therefore, this document uses various
348special notations to denote such characters in examples.</t>
349     
350<t>In text, characters outside US-ASCII are sometimes referenced by
351using a prefix of 'U+', followed by four to six hexadecimal
352digits.</t>
353
354<t>To represent characters outside US-ASCII in examples, this document
355uses two notations: 'XML Notation' and 'Bidi Notation'.</t>
356
357<t>XML Notation uses a leading '&amp;#x', a trailing ';', and the
358hexadecimal number of the character in the UCS in between. For
359example, &amp;#x44F; stands for CYRILLIC CAPITAL LETTER YA. In this
360notation, an actual '&amp;' is denoted by '&amp;amp;'.</t>
361
362<t>Bidi Notation is used for bidirectional examples: Lower case
363letters stand for Latin letters or other letters that are written left
364to right, whereas upper case letters represent Arabic or Hebrew
365letters that are written right to left.</t>
366
367<t>To denote actual octets in examples (as opposed to percent-encoded
368octets), the two hex digits denoting the octet are enclosed in "&lt;"
369and "&gt;".  For example, the octet often denoted as 0xc9 is denoted
370here as &lt;c9&gt;.</t>
371
372<t> In this document, the key words "MUST", "MUST NOT", "REQUIRED",
373"SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY",
374and "OPTIONAL" are to be interpreted as described in <xref
375target="RFC2119"/>.</t>
376
377</section> <!-- notation -->
378</section> <!-- introduction -->
379
380<section title="IRI Syntax" anchor="syntax">
381<t>This section defines the syntax of Internationalized Resource
382Identifiers (IRIs).</t>
383
384<t>As with URIs, an IRI is defined as a sequence of characters, not as
385a sequence of octets. This definition accommodates the fact that IRIs
386may be written on paper or read over the radio as well as stored or
387transmitted digitally.  The same IRI might be represented as different
388sequences of octets in different protocols or documents if these
389protocols or documents use different character encodings (and/or
390transfer encodings).  Using the same character encoding as the
391containing protocol or document ensures that the characters in the IRI
392can be handled (e.g., searched, converted, displayed) in the same way
393as the rest of the protocol or document.</t>
394
395<section title="Summary of IRI Syntax" anchor="summary">
396
397<t>IRIs are defined by extending the URI syntax in <xref
398target="RFC3986"/>, but extending the class of unreserved characters
399by adding the characters of the UCS (Universal Character Set, <xref
400target="ISO10646"/>) beyond U+007F, subject to the limitations given
401in the syntax rules below and in <xref target="limitations"/>.</t>
402
403<t>The syntax and use of components and reserved characters is the
404same as that in <xref target="RFC3986"/>. Each "URI scheme" thus also
405functions as an "IRI scheme", in that scheme-specific parsing rules
406for URIs of a scheme are be extended to allow parsing of IRIs using
407the same parsing rules.</t>
408
409<t>All the operations defined in <xref target="RFC3986"/>, such as the
410resolution of relative references, can be applied to IRIs by
411IRI-processing software in exactly the same way as they are for URIs
412by URI-processing software.</t>
413
414<t>Characters outside the US-ASCII repertoire MUST NOT be reserved and
415therefore MUST NOT be used for syntactical purposes, such as to
416delimit components in newly defined schemes. For example, U+00A2, CENT
417SIGN, is not allowed as a delimiter in IRIs, because it is in the
418'iunreserved' category. This is similar to the fact that it is not
419possible to use '-' as a delimiter in URIs, because it is in the
420'unreserved' category.</t>
421
422</section> <!-- summary -->
423<section title="ABNF for IRI References and IRIs" anchor="abnf">
424
425<t>An ABNF definition for IRI references (which are the most general
426concept and the start of the grammar) and IRIs is given here. The
427syntax of this ABNF is described in <xref target="STD68"/>. Character
428numbers are taken from the UCS, without implying any actual binary
429encoding. Terminals in the ABNF are characters, not octets.</t>
430
431<t>The following grammar closely follows the URI grammar in <xref
432target="RFC3986"/>, except that the range of unreserved characters is
433expanded to include UCS characters, with the restriction that private
434UCS characters can occur only in query parts. The grammar is split
435into two parts: Rules that differ from <xref target="RFC3986"/>
436because of the above-mentioned expansion, and rules that are the same
437as those in <xref target="RFC3986"/>. For rules that are different
438than those in <xref target="RFC3986"/>, the names of the non-terminals
439have been changed as follows. If the non-terminal contains 'URI', this
440has been changed to 'IRI'. Otherwise, an 'i' has been prefixed.</t>
441
442<!--
443for line length measuring in artwork (max 72 chars, three chars at start):
444      1         2         3         4         5         6         7
445456789012345678901234567890123456789012345678901234567890123456789012
446-->
447<figure>
448<preamble>The following rules are different from those in <xref target="RFC3986"/>:</preamble>
449<artwork>
450IRI            = scheme ":" ihier-part [ "?" iquery ]
451                 [ "#" ifragment ]
452
453ihier-part     = "//" iauthority ipath-abempty
454               / ipath-absolute
455               / ipath-rootless
456               / ipath-empty
457
458IRI-reference  = IRI / irelative-ref
459
460absolute-IRI   = scheme ":" ihier-part [ "?" iquery ]
461
462irelative-ref  = irelative-part [ "?" iquery ] [ "#" ifragment ]
463
464irelative-part = "//" iauthority ipath-abempty
465               / ipath-absolute
466               / ipath-noscheme
467               / ipath-empty
468
469iauthority     = [ iuserinfo "@" ] ihost [ ":" port ]
470iuserinfo      = *( iunreserved / pct-form / sub-delims / ":" )
471ihost          = IP-literal / IPv4address / ireg-name
472
473pct-form       = pct-encoded
474
475ireg-name      = *( iunreserved / sub-delims )
476
477ipath          = ipath-abempty   ; begins with "/" or is empty
478               / ipath-absolute  ; begins with "/" but not "//"
479               / ipath-noscheme  ; begins with a non-colon segment
480               / ipath-rootless  ; begins with a segment
481               / ipath-empty     ; zero characters
482
483ipath-abempty  = *( path-sep isegment )
484ipath-absolute = path-sep [ isegment-nz *( path-sep isegment ) ]
485ipath-noscheme = isegment-nz-nc *( path-sep isegment )
486ipath-rootless = isegment-nz *( path-sep isegment )
487ipath-empty    = 0&lt;ipchar&gt;
488path-sep       = "/"
489
490isegment       = *ipchar
491isegment-nz    = 1*ipchar
492isegment-nz-nc = 1*( iunreserved / pct-form / sub-delims
493                     / "@" )
494               ; non-zero-length segment without any colon ":"                     
495
496ipchar         = iunreserved / pct-form / sub-delims / ":"
497               / "@"
498 
499iquery         = *( ipchar / iprivate / "/" / "?" )
500
501ifragment      = *( ipchar / "/" / "?" / "#" )
502
503iunreserved    = ALPHA / DIGIT / "-" / "." / "_" / "~" / ucschar
504
505ucschar        = %xA0-D7FF / %xF900-FDCF / %xFDF0-FFEF
506               / %x10000-1FFFD / %x20000-2FFFD / %x30000-3FFFD
507               / %x40000-4FFFD / %x50000-5FFFD / %x60000-6FFFD
508               / %x70000-7FFFD / %x80000-8FFFD / %x90000-9FFFD
509               / %xA0000-AFFFD / %xB0000-BFFFD / %xC0000-CFFFD
510               / %xD0000-DFFFD / %xE1000-EFFFD
511
512iprivate       = %xE000-F8FF / %xE0000-E0FFF / %xF0000-FFFFD
513               / %x100000-10FFFD
514</artwork>
515</figure>
516
517<t>Some productions are ambiguous. The "first-match-wins" (a.k.a. "greedy")
518algorithm applies. For details, see <xref target="RFC3986"/>.</t>
519
520<figure>
521<preamble>The following rules are the same as those in <xref target="RFC3986"/>:</preamble>
522<artwork>
523scheme         = ALPHA *( ALPHA / DIGIT / "+" / "-" / "." )
524 
525port           = *DIGIT
526 
527IP-literal     = "[" ( IPv6address / IPvFuture  ) "]"
528 
529IPvFuture      = "v" 1*HEXDIG "." 1*( unreserved / sub-delims / ":" )
530 
531IPv6address    =                            6( h16 ":" ) ls32
532               /                       "::" 5( h16 ":" ) ls32
533               / [               h16 ] "::" 4( h16 ":" ) ls32
534               / [ *1( h16 ":" ) h16 ] "::" 3( h16 ":" ) ls32
535               / [ *2( h16 ":" ) h16 ] "::" 2( h16 ":" ) ls32
536               / [ *3( h16 ":" ) h16 ] "::"    h16 ":"   ls32
537               / [ *4( h16 ":" ) h16 ] "::"              ls32
538               / [ *5( h16 ":" ) h16 ] "::"              h16
539               / [ *6( h16 ":" ) h16 ] "::"
540               
541h16            = 1*4HEXDIG
542ls32           = ( h16 ":" h16 ) / IPv4address
543
544IPv4address    = dec-octet "." dec-octet "." dec-octet "." dec-octet
545
546dec-octet      = DIGIT                 ; 0-9
547               / %x31-39 DIGIT         ; 10-99
548               / "1" 2DIGIT            ; 100-199
549               / "2" %x30-34 DIGIT     ; 200-249
550               / "25" %x30-35          ; 250-255
551           
552pct-encoded    = "%" HEXDIG HEXDIG
553
554unreserved     = ALPHA / DIGIT / "-" / "." / "_" / "~"
555reserved       = gen-delims / sub-delims
556gen-delims     = ":" / "/" / "?" / "#" / "[" / "]" / "@"
557sub-delims     = "!" / "$" / "&amp;" / "'" / "(" / ")"
558               / "*" / "+" / "," / ";" / "="
559</artwork></figure>
560
561<t>This syntax does not support IPv6 scoped addressing zone identifiers.</t>
562
563</section> <!-- abnf -->
564
565</section> <!-- syntax -->
566
567<section title="Processing IRIs and related protocol elements" anchor="processing">
568
569<t>IRIs are meant to replace URIs in identifying resources within new
570versions of protocols, formats, and software components that use a
571UCS-based character repertoire.  Protocols and components may use and
572process IRIs directly. However, there are still numerous systems and
573protocols which only accept URIs or components of parsed URIs; that is,
574they only accept sequences of characters within the subset of US-ASCII
575characters allowed in URIs. </t>
576
577<t>This section defines specific processing steps for IRI consumers
578which establish the relationship between the string given and the
579interpreted derivatives. These
580processing steps apply to both IRIs and IRI references (i.e., absolute
581or relative forms); for IRIs, some steps are scheme specific. </t>
582
583<section title="Converting to UCS" anchor="ucsconv"> 
584 
585<t>Input that is already in a Unicode form (i.e., a sequence of Unicode
586 characters or an octet-stream representing a Unicode-based character
587 encoding such as UTF-8 or UTF-16) should be left as is and not
588 normalized (see (see <xref target="normalization"/>).</t>
589
590  <t>An IRI or IRI reference is a sequence of characters from the UCS.
591    For IRIs that are not already in a Unicode form
592    (as when written on paper, read aloud, or represented in a text stream
593    using a legacy character encoding), convert the IRI to Unicode.
594    Note that some character encodings or transcriptions can be converted
595    to or represented by more than one sequence of Unicode characters.
596    Ideally the resulting IRI would use a normalized form,
597    such as Unicode Normalization Form C <xref target="UTR15"/>
598    (see <xref target='ladder'/> Normalization and Comparison),
599    since that ensures a stable, consistent representation
600    that is most likely to produce the intended results.
601    Implementers and users are cautioned that, while denormalized character sequences are valid,
602    they might be difficult for other users or processes to reproduce
603    and might lead to unexpected results.
604  </t>
605
606<t> In other cases (written on paper, read aloud, or otherwise
607 represented independent of any character encoding) represent the IRI
608 as a sequence of characters from the UCS normalized according to
609 Unicode Normalization Form C (NFC, <xref target="UTR15"/>).</t>
610</section> <!-- ucsconv -->
611
612<section title="Parse the IRI into IRI components">
613
614<t>Parse the IRI, either as a relative reference (no scheme)
615or using scheme specific processing (according to the scheme
616given); the result resulting in a set of parsed IRI components.
617(NOTE: FIX BEFORE RELEASE: INTENT IS THAT ALL IRI SCHEMES
618THAT USE GENERIC SYNTAX AND ALLOW NON-ASCII AUTHORITY CAN
619ONLY USE AUTHORITY FOR NAMES THAT FOLLOW PUNICODE.)
620 </t>
621
622<t>NOTE: The result of parsing into components will correspond result
623in a correspondence of subtrings of the IRI according to the part
624matched.  For example, in <xref target="HTML5"/>, the protocol
625components of interest are SCHEME (scheme), HOST (ireg-name), PORT
626(port), the PATH (ipath after the initial "/"), QUERY (iquery),
627FRAGMENT (ifragment), and AUTHORITY (iauthority).
628</t>
629
630<t>Subsequent processing rules are sometimes used to define other
631syntactic components. For example, <xref target="HTML5"/> defines APIs
632for IRI processing; in these APIs:
633
634<list style="hanging">
635<t hangText="HOSTSPECIFIC"> the substring that follows
636the substring matched by the iauthority production, or the whole
637string if the iauthority production wasn't matched.</t>
638<t hangText="HOSTPORT"> if there is a scheme component and a port
639component and the port given by the port component is different than
640the default port defined for the protocol given by the scheme
641component, then HOSTPORT is the substring that starts with the
642substring matched by the host production and ends with the substring
643matched by the port production, and includes the colon in between the
644two. Otherwise, it is the same as the host component.
645</t>
646</list>
647</t>
648</section> <!-- parse -->
649
650<section title="General percent-encoding of IRI components" anchor="compmapping">
651   
652<t>For most IRI components, it is possible to map the IRI component
653to an equivalent URI component by percent-encoding those characters
654not allowed in URIs. Previous processing steps will have removed
655some characters, and the interpretation of reserved characters will
656have already been done (with the syntactic reserved characters outside
657of the IRI component). This mapping is defined for all sequences
658of Unicode characters, whether or not they are valid for the component
659in question. </t>
660   
661<t>For each character which is not allowed in a valid URI (NOTE: WHAT
662IS THE RIGHT REFERENCE HERE), apply the following steps. </t>
663
664<t><list style="hanging">
665
666<t hangText="Convert to UTF-8">Convert the character to a sequence of
667  one or more octets using UTF-8 <xref target="RFC3629"/>.</t>
668
669<t hangText="Percent encode">Convert each octet of this sequence to %HH,
670   where HH is the hexadecimal notation of the octet value. The
671   hexadecimal notation SHOULD use uppercase letters. (This is the
672   general URI percent-encoding mechanism in Section 2.1 of <xref
673   target="RFC3986"/>.)</t>
674   
675</list></t>
676
677<t>Note that the mapping is an identity transformation for parsed URI
678components of valid URIs, and is idempotent: applying the mapping a
679second time will not change anything.</t>
680</section> <!-- general conversion -->
681
682<section title="Mapping ireg-name" anchor="dnsmapping">
683
684<t>Schemes that allow non-ASCII based characters
685in the reg-name (ireg-name) position MUST convert the ireg-name
686component of an IRI as follows:</t>
687
688<t>Replace the ireg-name part of the IRI by the part converted using
689the ToASCII operation specified in Section 4.1 of <xref
690target="RFC3490"/> on each dot-separated label, and by using U+002E
691(FULL STOP) as a label separator, with the flag UseSTD3ASCIIRules set
692to FALSE, and with the flag AllowUnassigned set to FALSE.
693The ToASCII operation may
694fail, but this would mean that the IRI cannot be resolved.
695In such cases, if the domain name conversion fails, then the
696entire IRI conversion fails. Processors that have no mechanism for
697signalling a failure MAY instead substitute an otherwise
698invalid host name, although such processing SHOULD be avoided.
699 </t>
700
701<t>For example, the IRI
702<vspace/>"http://r&amp;#xE9;sum&amp;#xE9;.example.org"<vspace/> MAY be
703converted to <vspace/>"http://xn--rsum-bad.example.org"<vspace/>;
704conversion to percent-encoded form, e.g.,
705 <vspace/>"http://r%C3%A9sum%C3%A9.example.org", MUST NOT be performed. </t>
706
707<t><list style="hanging"> 
708
709<t hangText="Note:">Domain Names may appear in parts of an IRI other
710than the ireg-name part.  It is the responsibility of scheme-specific
711implementations (if the Internationalized Domain Name is part of the
712scheme syntax) or of server-side implementations (if the
713Internationalized Domain Name is part of 'iquery') to apply the
714necessary conversions at the appropriate point. Example: Trying to
715validate the Web page at<vspace/>
716http://r&amp;#xE9;sum&amp;#xE9;.example.org would lead to an IRI of
717<vspace/>http://validator.w3.org/check?uri=http%3A%2F%2Fr&amp;#xE9;sum&amp;#xE9;.<vspace/>example.org,
718which would convert to a URI
719of<vspace/>http://validator.w3.org/check?uri=http%3A%2F%2Fr%C3%A9sum%C3%A9.<vspace/>example.org.
720The server-side implementation is responsible for making the
721necessary conversions to be able to retrieve the Web page.</t>
722
723<t hangText="Note:">In this process, characters allowed in URI
724references and existing percent-encoded sequences are not encoded further.
725(This mapping is similar to, but different from, the encoding applied
726when arbitrary content is included in some part of a URI.)
727
728For example, an IRI of
729<vspace/>"http://www.example.org/red%09ros&amp;#xE9;#red"
730(in XML notation) is converted to
731<vspace/>"http://www.example.org/red%09ros%C3%A9#red", not to
732something like
733<vspace/>"http%3A%2F%2Fwww.example.org%2Fred%2509ros%C3%A9%23red".
734((DESIGN QUESTION: What about e.g. http://r%C3%A9sum%C3%A9.example.org in an IRI? Will that get converted to punycode, or not?))
735
736</t>
737
738</list></t>
739</section> <!-- dnsmapping -->
740
741<section title="Mapping query components" anchor="querymapping">
742
743<t>((NOTE: SEE ISSUES LIST))
744
745For compatibility with existing deployed HTTP infrastructure,
746the following special case applies for schemes "http" and "https"
747and IRIs whose origin has a document charset other than one which
748is UCS-based (e.g., UTF-8 or UTF-16). In such a case, the "query"
749component of an IRI is mapped into a URI by using the document
750charset rather than UTF-8 as the binary representation before
751pct-encoding. This mapping is not applied for any other scheme
752or component.</t>
753
754</section> <!-- querymapping -->
755
756<section title="Mapping IRIs to URIs" anchor="mapping">
757
758<t>The canonical mapping from a IRI to URI is defined by applying the
759mapping above (from IRI to URI components) and then reassembling a URI
760from the parsed URI components using the original punctuation that
761delimited the IRI components. </t>
762
763</section> <!-- mapping -->
764
765<section title="Converting URIs to IRIs" anchor="URItoIRI">
766
767<t>In some situations, for presentation and further processing,
768it is desirable to convert a URI into an equivalent IRI in which
769natural characters are represented directly rather than
770percent encoded. Of course, every URI is already an IRI in
771its own right without any conversion, and in general there
772This section gives one such procedure for this conversion.
773</t>
774
775<t>
776The conversion described in this section, if given a valid URI, will
777result in an IRI that maps back to the URI used as an input for the
778conversion (except for potential case differences in percent-encoding
779and for potential percent-encoded unreserved characters).
780
781However, the IRI resulting from this conversion may differ
782from the original IRI (if there ever was one).</t> 
783
784<t>URI-to-IRI conversion removes percent-encodings, but not all
785percent-encodings can be eliminated. There are several reasons for
786this:</t>
787
788<t><list style="hanging">
789
790<t hangText="1.">Some percent-encodings are necessary to distinguish
791    percent-encoded and unencoded uses of reserved characters.</t>
792
793<t hangText="2.">Some percent-encodings cannot be interpreted as sequences
794    of UTF-8 octets.<vspace blankLines="1"/>
795    (Note: The octet patterns of UTF-8 are highly regular.
796    Therefore, there is a very high probability, but no guarantee,
797    that percent-encodings that can be interpreted as sequences of UTF-8
798    octets actually originated from UTF-8. For a detailed discussion,
799    see <xref target="Duerst97"/>.)</t>
800
801<t hangText="3.">The conversion may result in a character that is not
802    appropriate in an IRI. See <xref target="abnf"/>, <xref target="visual"/>,
803      and <xref target="limitations"/> for further details.</t>
804
805<t hangText="4.">IRI to URI conversion has different rules for
806    dealing with domain names and query parameters.</t>
807
808</list></t>
809
810<t>Conversion from a URI to an IRI MAY be done by using the following
811steps:
812
813<list style="hanging">
814<t hangText="1.">Represent the URI as a sequence of octets in
815       US-ASCII.</t>
816
817<t hangText="2.">Convert all percent-encodings ("%" followed by two
818      hexadecimal digits) to the corresponding octets, except those
819      corresponding to "%", characters in "reserved", and characters
820      in US-ASCII not allowed in URIs.</t> 
821
822<t hangText="3.">Re-percent-encode any octet produced in step 2 that
823      is not part of a strictly legal UTF-8 octet sequence.</t>
824
825
826<t hangText="4.">Re-percent-encode all octets produced in step 3 that
827      in UTF-8 represent characters that are not appropriate according
828      to <xref target="abnf"/>, <xref target="visual"/>, and <xref
829      target="limitations"/>.</t> 
830
831<t hangText="5.">Interpret the resulting octet sequence as a sequence
832      of characters encoded in UTF-8.</t>
833
834<t hangText="6.">URIs known to contain domain names in the reg-name
835      component SHOULD convert punycode-encoded domain name labels to
836      the corresponding characters using the ToUnicode procedure. </t>
837</list></t>
838
839<t>This procedure will convert as many percent-encoded characters as
840possible to characters in an IRI. Because there are some choices when
841step 4 is applied (see <xref target="limitations"/>), results may
842vary.</t>
843
844<t>Conversions from URIs to IRIs MUST NOT use any character
845encoding other than UTF-8 in steps 3 and 4, even if it might be
846possible to guess from the context that another character encoding
847than UTF-8 was used in the URI.  For example, the URI
848"http://www.example.org/r%E9sum%E9.html" might with some guessing be
849interpreted to contain two e-acute characters encoded as
850iso-8859-1. It must not be converted to an IRI containing these
851e-acute characters. Otherwise, in the future the IRI will be mapped to
852"http://www.example.org/r%C3%A9sum%C3%A9.html", which is a different
853URI from "http://www.example.org/r%E9sum%E9.html".</t>
854
855<section title="Examples">
856
857<t>This section shows various examples of converting URIs to IRIs.
858Each example shows the result after each of the steps 1 through 6 is
859applied. XML Notation is used for the final result.  Octets are
860denoted by "&lt;" followed by two hexadecimal digits followed by
861"&gt;".</t>
862
863<t>The following example contains the sequence "%C3%BC", which is a
864strictly legal UTF-8 sequence, and which is converted into the actual
865character U+00FC, LATIN SMALL LETTER U WITH DIAERESIS (also known as
866u-umlaut).
867
868<list style="hanging">
869<t hangText="1.">http://www.example.org/D%C3%BCrst</t>
870<t hangText="2.">http://www.example.org/D&lt;c3&gt;&lt;bc&gt;rst</t>
871<t hangText="3.">http://www.example.org/D&lt;c3&gt;&lt;bc&gt;rst</t>
872<t hangText="4.">http://www.example.org/D&lt;c3&gt;&lt;bc&gt;rst</t>
873<t hangText="5.">http://www.example.org/D&amp;#xFC;rst</t>
874<t hangText="6.">http://www.example.org/D&amp;#xFC;rst</t>
875</list>
876</t>
877
878<t>The following example contains the sequence "%FC", which might
879represent U+00FC, LATIN SMALL LETTER U WITH DIAERESIS, in
880the<vspace/>iso-8859-1 character encoding.  (It might represent other
881characters in other character encodings. For example, the octet
882&lt;fc&gt; in iso-8859-5 represents U+045C, CYRILLIC SMALL LETTER
883KJE.)  Because &lt;fc&gt; is not part of a strictly legal UTF-8
884sequence, it is re-percent-encoded in step 3.
885
886
887<list style="hanging">
888<t hangText="1.">http://www.example.org/D%FCrst</t>
889<t hangText="2.">http://www.example.org/D&lt;fc&gt;rst</t>
890<t hangText="3.">http://www.example.org/D%FCrst</t>
891<t hangText="4.">http://www.example.org/D%FCrst</t>
892<t hangText="5.">http://www.example.org/D%FCrst</t>
893<t hangText="6.">http://www.example.org/D%FCrst</t>
894</list>
895</t>
896
897<t>The following example contains "%e2%80%ae", which is the percent-encoded<vspace/>UTF-8
898character encoding of U+202E, RIGHT-TO-LEFT OVERRIDE. <xref target="visual"/>
899forbids the direct use of this character in an IRI. Therefore, the
900corresponding octets are re-percent-encoded in step 4. This example shows
901that the case (upper- or lowercase) of letters used in percent-encodings may not be preserved.
902The example also contains a punycode-encoded domain name label (xn--99zt52a),
903which is not converted.
904
905<list style="hanging">
906<t hangText="1.">http://xn--99zt52a.example.org/%e2%80%ae</t>
907<t hangText="2.">http://xn--99zt52a.example.org/&lt;e2&gt;&lt;80&gt;&lt;ae&gt;</t>
908<t hangText="3.">http://xn--99zt52a.example.org/&lt;e2&gt;&lt;80&gt;&lt;ae&gt;</t>
909<t hangText="4.">http://xn--99zt52a.example.org/%E2%80%AE</t>
910<t hangText="5.">http://xn--99zt52a.example.org/%E2%80%AE</t>
911<t hangText="6.">http://&amp;#x7D0D;&amp;#x8C46;.example.org/%E2%80%AE</t>
912</list></t>
913
914<t>Note that the label "xn--99zt52a" is converted to U+7D0D U+8C46
915(Japanese Natto). ((EDITOR NOTE: There is some inconsistency in this note.))</t>
916
917</section> <!-- examples -->
918</section> <!-- URItoIRI -->
919</section> <!-- processing -->
920<section title="Bidirectional IRIs for Right-to-Left Languages" anchor="Bidi">
921
922<t>Some UCS characters, such as those used in the Arabic and Hebrew
923scripts, have an inherent right-to-left (rtl) writing direction. IRIs
924containing these characters (called bidirectional IRIs or Bidi IRIs)
925require additional attention because of the non-trivial relation
926between logical representation (used for digital representation and
927for reading/spelling) and visual representation (used for
928display/printing).</t>
929
930<t>Because of the complex interaction between the logical representation,
931the visual representation, and the syntax of a Bidi IRI, a balance is
932needed between various requirements.
933The main requirements are<list style="hanging">
934<t hangText="1.">user-predictable conversion between visual and
935    logical representation;</t>
936<t hangText="2.">the ability to include a wide range of characters
937    in various parts of the IRI; and</t>
938<t hangText="3.">minor or no changes or restrictions for
939      implementations.</t>
940</list></t>
941
942<section title="Logical Storage and Visual Presentation" anchor="visual">
943
944<t>When stored or transmitted in digital representation, bidirectional
945IRIs MUST be in full logical order and MUST conform to the IRI syntax
946rules (which includes the rules relevant to their scheme). This
947ensures that bidirectional IRIs can be processed in the same way as
948other IRIs.</t> <t>Bidirectional IRIs MUST be rendered by using the
949Unicode Bidirectional Algorithm <xref target="UNIV4"/>, <xref
950target="UNI9"/>.  Bidirectional IRIs MUST be rendered in the same way
951as they would be if they were in a left-to-right embedding; i.e., as
952if they were preceded by U+202A, LEFT-TO-RIGHT EMBEDDING (LRE), and
953followed by U+202C, POP DIRECTIONAL FORMATTING (PDF).  Setting the
954embedding direction can also be done in a higher-level protocol (e.g.,
955the dir='ltr' attribute in HTML).</t> 
956
957<t>There is no requirement to use the above embedding if the display
958is still the same without the embedding. For example, a bidirectional
959IRI in a text with left-to-right base directionality (such as used for
960English or Cyrillic) that is preceded and followed by whitespace and
961strong left-to-right characters does not need an embedding.  Also, a
962bidirectional relative IRI reference that only contains strong
963right-to-left characters and weak characters and that starts and ends
964with a strong right-to-left character and appears in a text with
965right-to-left base directionality (such as used for Arabic or Hebrew)
966and is preceded and followed by whitespace and strong characters does
967not need an embedding.</t>
968
969<t>In some other cases, using U+200E, LEFT-TO-RIGHT MARK (LRM), may be
970sufficient to force the correct display behavior.  However, the
971details of the Unicode Bidirectional algorithm are not always easy to
972understand. Implementers are strongly advised to err on the side of
973caution and to use embedding in all cases where they are not
974completely sure that the display behavior is unaffected without the
975embedding.</t>
976
977<t>The Unicode Bidirectional Algorithm (<xref target="UNI9"/>, section
9784.3) permits higher-level protocols to influence bidirectional
979rendering. Such changes by higher-level protocols MUST NOT be used if
980they change the rendering of IRIs.</t> 
981
982<t>The bidirectional formatting characters that may be used before or
983after the IRI to ensure correct display are not themselves part of the
984IRI.  IRIs MUST NOT contain bidirectional formatting characters (LRM,
985RLM, LRE, RLE, LRO, RLO, and PDF). They affect the visual rendering of
986the IRI but do not appear themselves. It would therefore not be
987possible to input an IRI with such characters correctly.</t>
988
989</section> <!-- visual -->
990<section title="Bidi IRI Structure" anchor="bidi-structure">
991
992<t>The Unicode Bidirectional Algorithm is designed mainly for running
993text.  To make sure that it does not affect the rendering of
994bidirectional IRIs too much, some restrictions on bidirectional IRIs
995are necessary. These restrictions are given in terms of delimiters
996(structural characters, mostly punctuation such as "@", ".", ":",
997and<vspace/>"/") and components (usually consisting mostly of letters
998and digits).</t>
999
1000<t>The following syntax rules from <xref target="abnf"/> correspond to
1001components for the purpose of Bidi behavior: iuserinfo, ireg-name,
1002isegment, isegment-nz, isegment-nz-nc, ireg-name, iquery, and
1003ifragment.</t>
1004
1005<t>Specifications that define the syntax of any of the above
1006components MAY divide them further and define smaller parts to be
1007components according to this document. As an example, the restrictions
1008of <xref target="RFC3490"/> on bidirectional domain names correspond
1009to treating each label of a domain name as a component for schemes
1010with ireg-name as a domain name.  Even where the components are not
1011defined formally, it may be helpful to think about some syntax in
1012terms of components and to apply the relevant restrictions.  For
1013example, for the usual name/value syntax in query parts, it is
1014convenient to treat each name and each value as a component. As
1015another example, the extensions in a resource name can be treated as
1016separate components.</t>
1017
1018<t>For each component, the following restrictions apply:</t>
1019<t>
1020<list style="hanging">
1021
1022<t hangText="1.">A component SHOULD NOT use both right-to-left and
1023  left-to-right characters.</t>
1024
1025<t hangText="2.">A component using right-to-left characters SHOULD
1026  start and end with right-to-left characters.</t>
1027
1028</list></t>
1029
1030<t>The above restrictions are given as "SHOULD"s, rather than as
1031"MUST"s.  For IRIs that are never presented visually, they are not
1032relevant.  However, for IRIs in general, they are very important to
1033ensure consistent conversion between visual presentation and logical
1034representation, in both directions.</t>
1035
1036<t><list style="hanging">
1037
1038<t hangText="Note:">In some components, the above restrictions may
1039  actually be strictly enforced.  For example, <xref
1040  target="RFC3490"></xref> requires that these restrictions apply to
1041  the labels of a host name for those schemes where ireg-name is a
1042  host name.  In some other components (for example, path components)
1043  following these restrictions may not be too difficult.  For other
1044  components, such as parts of the query part, it may be very
1045  difficult to enforce the restrictions because the values of query
1046  parameters may be arbitrary character sequences.</t>
1047
1048</list></t>
1049
1050<t>If the above restrictions cannot be satisfied otherwise, the
1051affected component can always be mapped to URI notation as described
1052in <xref target="compmapping"/>. Please note that the whole component
1053has to be mapped (see also Example 9 below).</t>
1054
1055</section> <!-- bidi-structure -->
1056
1057<section title="Input of Bidi IRIs" anchor="bidiInput">
1058
1059<t>Bidi input methods MUST generate Bidi IRIs in logical order while
1060rendering them according to <xref target="visual"/>.  During input,
1061rendering SHOULD be updated after every new character is input to
1062avoid end-user confusion.</t>
1063
1064</section> <!-- bidiInput -->
1065
1066<section title="Examples">
1067
1068<t>This section gives examples of bidirectional IRIs, in Bidi
1069Notation.  It shows legal IRIs with the relationship between logical
1070and visual representation and explains how certain phenomena in this
1071relationship may look strange to somebody not familiar with
1072bidirectional behavior, but familiar to users of Arabic and Hebrew. It
1073also shows what happens if the restrictions given in <xref
1074target="bidi-structure"/> are not followed. The examples below can be
1075seen at <xref target="BidiEx"/>, in Arabic, Hebrew, and Bidi Notation
1076variants.</t>
1077
1078<t>To read the bidi text in the examples, read the visual
1079representation from left to right until you encounter a block of rtl
1080text. Read the rtl block (including slashes and other special
1081characters) from right to left, then continue at the next unread ltr
1082character.</t>
1083
1084<t>Example 1: A single component with rtl characters is inverted:
1085<vspace/>Logical representation:
1086"http://ab.CDEFGH.ij/kl/mn/op.html"<vspace/>Visual representation:
1087"http://ab.HGFEDC.ij/kl/mn/op.html"<vspace/> Components can be read
1088one by one, and each component can be read in its natural
1089direction.</t>
1090
1091<t>Example 2: More than one consecutive component with rtl characters
1092is inverted as a whole: <vspace/>Logical representation:
1093"http://ab.CDE.FGH/ij/kl/mn/op.html"<vspace/>Visual representation:
1094"http://ab.HGF.EDC/ij/kl/mn/op.html"<vspace/> A sequence of rtl
1095components is read rtl, in the same way as a sequence of rtl words is
1096read rtl in a bidi text.</t>
1097
1098<t>Example 3: All components of an IRI (except for the scheme) are
1099rtl.  All rtl components are inverted overall: <vspace/>Logical
1100representation:
1101"http://AB.CD.EF/GH/IJ/KL?MN=OP;QR=ST#UV"<vspace/>Visual
1102representation: "http://VU#TS=RQ;PO=NM?LK/JI/HG/FE.DC.BA"<vspace/> The
1103whole IRI (except the scheme) is read rtl. Delimiters between rtl
1104components stay between the respective components; delimiters between
1105ltr and rtl components don't move.</t>
1106
1107<t>Example 4: Each of several sequences of rtl components is inverted
1108on its own: <vspace/>Logical representation:
1109"http://AB.CD.ef/gh/IJ/KL.html"<vspace/>Visual representation:
1110"http://DC.BA.ef/gh/LK/JI.html"<vspace/> Each sequence of rtl
1111components is read rtl, in the same way as each sequence of rtl words
1112in an ltr text is read rtl.</t>
1113
1114<t>Example 5: Example 2, applied to components of different kinds:
1115<vspace/>Logical representation: "http://ab.cd.EF/GH/ij/kl.html"
1116<vspace/>Visual representation:
1117"http://ab.cd.HG/FE/ij/kl.html"<vspace/> The inversion of the domain
1118name label and the path component may be unexpected, but it is
1119consistent with other bidi behavior.  For reassurance that the domain
1120component really is "ab.cd.EF", it may be helpful to read aloud the
1121visual representation following the bidi algorithm. After
1122"http://ab.cd." one reads the RTL block "E-F-slash-G-H", which
1123corresponds to the logical representation.
1124</t>
1125
1126<t>Example 6: Same as Example 5, with more rtl components:
1127<vspace/>Logical representation:
1128"http://ab.CD.EF/GH/IJ/kl.html"<vspace/>Visual representation:
1129"http://ab.JI/HG/FE.DC/kl.html"<vspace/> The inversion of the domain
1130name labels and the path components may be easier to identify because
1131the delimiters also move.</t>
1132
1133<t>Example 7: A single rtl component includes digits: <vspace/>Logical
1134representation: "http://ab.CDE123FGH.ij/kl/mn/op.html"<vspace/>Visual
1135representation: "http://ab.HGF123EDC.ij/kl/mn/op.html"<vspace/>
1136Numbers are written ltr in all cases but are treated as an additional
1137embedding inside a run of rtl characters. This is completely
1138consistent with usual bidirectional text.</t>
1139
1140<t>Example 8 (not allowed): Numbers are at the start or end of an rtl
1141component:<vspace/>Logical representation:
1142"http://ab.cd.ef/GH1/2IJ/KL.html"<vspace/>Visual representation:
1143"http://ab.cd.ef/LK/JI1/2HG.html"<vspace/> The sequence "1/2" is
1144interpreted by the bidi algorithm as a fraction, fragmenting the
1145components and leading to confusion. There are other characters that
1146are interpreted in a special way close to numbers; in particular, "+",
1147"-", "#", "$", "%", ",", ".", and ":".</t>
1148
1149<t>Example 9 (not allowed): The numbers in the previous example are
1150percent-encoded: <vspace/>Logical representation:
1151"http://ab.cd.ef/GH%31/%32IJ/KL.html",<vspace/>Visual representation:
1152"http://ab.cd.ef/LK/JI%32/%31HG.html"</t>
1153
1154<t>Example 10 (allowed but not recommended): <vspace/>Logical
1155representation: "http://ab.CDEFGH.123/kl/mn/op.html"<vspace/>Visual
1156representation: "http://ab.123.HGFEDC/kl/mn/op.html"<vspace/>
1157Components consisting of only numbers are allowed (it would be rather
1158difficult to prohibit them), but these may interact with adjacent RTL
1159components in ways that are not easy to predict.</t>
1160
1161<t>Example 11 (allowed but not recommended): <vspace/>Logical
1162representation: "http://ab.CDEFGH.123ij/kl/mn/op.html"<vspace/>Visual
1163representation: "http://ab.123.HGFEDCij/kl/mn/op.html"<vspace/>
1164Components consisting of numbers and left-to-right characters are
1165allowed, but these may interact with adjacent RTL components in ways
1166that are not easy to predict.</t>
1167</section><!-- examples -->
1168</section><!-- bidi -->
1169
1170<section title="Normalization and Comparison" anchor="equivalence">
1171
1172<t><list style="hanging"><t hangText="Note:">The structure and much of
1173  the material for this section is taken from section 6 of <xref
1174  target="RFC3986"></xref>; the differences are due to the specifics
1175  of IRIs.</t></list></t>
1176
1177<t>One of the most common operations on IRIs is simple comparison:
1178Determining whether two IRIs are equivalent, without using the IRIs to
1179access their respective resource(s). A comparison is performed
1180whenever a response cache is accessed, a browser checks its history to
1181color a link, or an XML parser processes tags within a
1182namespace. Extensive normalization prior to comparison of IRIs may be
1183used by spiders and indexing engines to prune a search space or reduce
1184duplication of request actions and response storage.</t>
1185
1186<t>IRI comparison is performed for some particular purpose. Protocols
1187or implementations that compare IRIs for different purposes will often
1188be subject to differing design trade-offs in regards to how much
1189effort should be spent in reducing aliased identifiers. This section
1190describes various methods that may be used to compare IRIs, the
1191trade-offs between them, and the types of applications that might use
1192them.</t>
1193
1194<section title="Equivalence">
1195
1196<t>Because IRIs exist to identify resources, presumably they should be
1197considered equivalent when they identify the same resource. However,
1198this definition of equivalence is not of much practical use, as there
1199is no way for an implementation to compare two resources to determine
1200if they are "the same" unless it has full knowledge or control of
1201them. For this reason, determination of equivalence or difference of
1202IRIs is based on string comparison, perhaps augmented by reference to
1203additional rules provided by URI scheme definitions.  We use the terms
1204"different" and "equivalent" to describe the possible outcomes of such
1205comparisons, but there are many application-dependent versions of
1206equivalence.</t>
1207
1208<t>Even when it is possible to determine that two IRIs are equivalent,
1209IRI comparison is not sufficient to determine whether two IRIs
1210identify different resources. For example, an owner of two different
1211domain names could decide to serve the same resource from both,
1212resulting in two different IRIs. Therefore, comparison methods are
1213designed to minimize false negatives while strictly avoiding false
1214positives.</t>
1215
1216<t>In testing for equivalence, applications should not directly
1217compare relative references; the references should be converted to
1218their respective target IRIs before comparison. When IRIs are compared
1219to select (or avoid) a network action, such as retrieval of a
1220representation, fragment components (if any) should be excluded from
1221the comparison.</t>
1222
1223<t>Applications using IRIs as identity tokens with no relationship to
1224a protocol MUST use the Simple String Comparison (see <xref
1225target="stringcomp"></xref>).  All other applications MUST select one
1226of the comparison practices from the Comparison Ladder (see <xref
1227target="ladder"></xref>.</t>
1228</section> <!-- equivalence -->
1229
1230
1231<section title="Preparation for Comparison">
1232<t>Any kind of IRI comparison REQUIRES that any additional contextual
1233processing is first performed, including undoing higher-level
1234escapings or encodings in the protocol or format that carries an
1235IRI. This preprocessing is usually done when the protocol or format is
1236parsed.</t>
1237
1238<t>Examples of contextual preprocessing steps are described in <xref
1239target="LEIRIHREF"/>. </t>
1240
1241<t>Examples of such escapings or encodings are entities and
1242numeric character references in <xref target="HTML4"></xref> and <xref
1243target="XML1"></xref>. As an example,
1244"http://example.org/ros&amp;eacute;" (in HTML),
1245"http://example.org/ros&amp;#233;" (in HTML or XML), and
1246<vspace/>"http://example.org/ros&amp;#xE9;" (in HTML or XML) are all
1247resolved into what is denoted in this document (see <xref
1248target="sec-Notation"></xref>) as "http://example.org/ros&amp;#xE9;"
1249(the "&amp;#xE9;" here standing for the actual e-acute character, to
1250compensate for the fact that this document cannot contain non-ASCII
1251characters).</t>
1252
1253<t>Similar considerations apply to encodings such as Transfer Codings
1254in HTTP (see <xref target="RFC2616"></xref>) and Content Transfer
1255Encodings in MIME (<xref target="RFC2045"></xref>), although in these
1256cases, the encoding is based not on characters but on octets, and
1257additional care is required to make sure that characters, and not just
1258arbitrary octets, are compared (see <xref
1259target="stringcomp"></xref>).</t>
1260
1261</section> <!-- preparation -->
1262
1263<section title="Comparison Ladder" anchor="ladder">
1264
1265<t>In practice, a variety of methods are used to test IRI
1266equivalence. These methods fall into a range distinguished by the
1267amount of processing required and the degree to which the probability
1268of false negatives is reduced. As noted above, false negatives cannot
1269be eliminated. In practice, their probability can be reduced, but this
1270reduction requires more processing and is not cost-effective for all
1271applications.</t>
1272
1273
1274<t>If this range of comparison practices is considered as a ladder,
1275the following discussion will climb the ladder, starting with
1276practices that are cheap but have a relatively higher chance of
1277producing false negatives, and proceeding to those that have higher
1278computational cost and lower risk of false negatives.</t>
1279
1280<section title="Simple String Comparison" anchor="stringcomp">
1281
1282<t>If two IRIs, when considered as character strings, are identical,
1283then it is safe to conclude that they are equivalent.  This type of
1284equivalence test has very low computational cost and is in wide use in
1285a variety of applications, particularly in the domain of parsing. It
1286is also used when a definitive answer to the question of IRI
1287equivalence is needed that is independent of the scheme used and that
1288can be calculated quickly and without accessing a network. An example
1289of such a case is XML Namespaces (<xref
1290target="XMLNamespace"></xref>).</t>
1291
1292
1293<t>Testing strings for equivalence requires some basic precautions.
1294This procedure is often referred to as "bit-for-bit" or
1295"byte-for-byte" comparison, which is potentially misleading. Testing
1296strings for equality is normally based on pair comparison of the
1297characters that make up the strings, starting from the first and
1298proceeding until both strings are exhausted and all characters are
1299found to be equal, until a pair of characters compares unequal, or
1300until one of the strings is exhausted before the other.</t>
1301
1302<t>This character comparison requires that each pair of characters be
1303put in comparable encoding form. For example, should one IRI be stored
1304in a byte array in UTF-8 encoding form and the second in a UTF-16
1305encoding form, bit-for-bit comparisons applied naively will produce
1306errors. It is better to speak of equality on a character-for-character
1307rather than on a byte-for-byte or bit-for-bit basis.  In practical
1308terms, character-by-character comparisons should be done codepoint by
1309codepoint after conversion to a common character encoding form.
1310
1311When comparing character by character, the comparison function MUST
1312NOT map IRIs to URIs, because such a mapping would create additional
1313spurious equivalences. It follows that an IRI SHOULD NOT be modified
1314when being transported if there is any chance that this IRI might be
1315used in a context that uses Simple String Comparison.</t>
1316
1317
1318<t>False negatives are caused by the production and use of IRI
1319aliases. Unnecessary aliases can be reduced, regardless of the
1320comparison method, by consistently providing IRI references in an
1321already normalized form (i.e., a form identical to what would be
1322produced after normalization is applied, as described below).
1323Protocols and data formats often limit some IRI comparisons to simple
1324string comparison, based on the theory that people and implementations
1325will, in their own best interest, be consistent in providing IRI
1326references, or at least be consistent enough to negate any efficiency
1327that might be obtained from further normalization.</t>
1328</section> <!-- stringcomp -->
1329
1330<section title="Syntax-Based Normalization">
1331
1332<figure><preamble>Implementations may use logic based on the
1333definitions provided by this specification to reduce the probability
1334of false negatives. This processing is moderately higher in cost than
1335character-for-character string comparison. For example, an application
1336using this approach could reasonably consider the following two IRIs
1337equivalent:</preamble>
1338
1339<artwork>
1340   example://a/b/c/%7Bfoo%7D/ros&amp;#xE9;
1341   eXAMPLE://a/./b/../b/%63/%7bfoo%7d/ros%C3%A9
1342</artwork></figure>
1343
1344<t>Web user agents, such as browsers, typically apply this type of IRI
1345normalization when determining whether a cached response is
1346available. Syntax-based normalization includes such techniques as case
1347normalization, character normalization, percent-encoding
1348normalization, and removal of dot-segments.</t>
1349
1350<section title="Case Normalization">
1351
1352<t>For all IRIs, the hexadecimal digits within a percent-encoding
1353triplet (e.g., "%3a" versus "%3A") are case-insensitive and therefore
1354should be normalized to use uppercase letters for the digits A-F.</t>
1355
1356<t>When an IRI uses components of the generic syntax, the component
1357syntax equivalence rules always apply; namely, that the scheme and
1358US-ASCII only host are case insensitive and therefore should be
1359normalized to lowercase. For example, the URI
1360"HTTP://www.EXAMPLE.com/" is equivalent to
1361"http://www.example.com/". Case equivalence for non-ASCII characters
1362in IRI components that are IDNs are discussed in <xref
1363target="schemecomp"></xref>.  The other generic syntax components are
1364assumed to be case sensitive unless specifically defined otherwise by
1365the scheme.</t>
1366
1367<t>Creating schemes that allow case-insensitive syntax components
1368containing non-ASCII characters should be avoided. Case normalization
1369of non-ASCII characters can be culturally dependent and is always a
1370complex operation. The only exception concerns non-ASCII host names
1371for which the character normalization includes a mapping step derived
1372from case folding.</t>
1373
1374</section> <!-- casenorm -->
1375
1376<section title="Character Normalization" anchor="normalization">
1377
1378<t>The Unicode Standard <xref target="UNIV4"></xref> defines various
1379equivalences between sequences of characters for various
1380purposes. Unicode Standard Annex #15 <xref target="UTR15"></xref>
1381defines various Normalization Forms for these equivalences, in
1382particular Normalization Form C (NFC, Canonical Decomposition,
1383followed by Canonical Composition) and Normalization Form KC (NFKC,
1384Compatibility Decomposition, followed by Canonical Composition).</t>
1385
1386<t> IRIs already in Unicode MUST NOT be normalized before parsing or
1387interpreting. In many non-Unicode character encodings, some text
1388cannot be represented directly. For example, the word "Vietnam" is
1389natively written "Vi&amp;#x1EC7;t Nam" (containing a LATIN SMALL
1390LETTER E WITH CIRCUMFLEX AND DOT BELOW) in NFC, but a direct
1391transcoding from the windows-1258 character encoding leads to
1392"Vi&amp;#xEA;&amp;#x323;t Nam" (containing a LATIN SMALL LETTER E WITH
1393CIRCUMFLEX followed by a COMBINING DOT BELOW). Direct transcoding of
1394other 8-bit encodings of Vietnamese may lead to other
1395representations.</t>
1396
1397<t>Equivalence of IRIs MUST rely on the assumption that IRIs are
1398appropriately pre-character-normalized rather than apply character
1399normalization when comparing two IRIs. The exceptions are conversion
1400from a non-digital form, and conversion from a non-UCS-based character
1401encoding to a UCS-based character encoding. In these cases, NFC or a
1402normalizing transcoder using NFC MUST be used for interoperability. To
1403avoid false negatives and problems with transcoding, IRIs SHOULD be
1404created by using NFC. Using NFKC may avoid even more problems; for
1405example, by choosing half-width Latin letters instead of full-width
1406ones, and full-width instead of half-width Katakana.</t>
1407
1408
1409<t>As an example,
1410"http://www.example.org/r&amp;#xE9;sum&amp;#xE9;.html" (in XML
1411Notation) is in NFC. On the other hand,
1412"http://www.example.org/re&amp;#x301;sume&amp;#x301;.html" is not in
1413NFC.</t>
1414
1415<t>The former uses precombined e-acute characters, and the latter uses
1416"e" characters followed by combining acute accents. Both usages are
1417defined as canonically equivalent in <xref target="UNIV4"></xref>.</t>
1418
1419<t><list style="hanging">
1420
1421<t hangText="Note:">
1422Because it is unknown how a particular sequence of characters is being
1423treated with respect to character normalization, it would be
1424inappropriate to allow third parties to normalize an IRI
1425arbitrarily. This does not contradict the recommendation that when a
1426resource is created, its IRI should be as character normalized as
1427possible (i.e., NFC or even NFKC). This is similar to the
1428uppercase/lowercase problems.  Some parts of a URI are case
1429insensitive (for example, the domain name). For others, it is unclear
1430whether they are case sensitive, case insensitive, or something in
1431between (e.g., case sensitive, but with a multiple choice selection if
1432the wrong case is used, instead of a direct negative result).  The
1433best recipe is that the creator use a reasonable capitalization and,
1434when transferring the URI, capitalization never be
1435changed.</t></list></t>
1436
1437<t>Various IRI schemes may allow the usage of Internationalized Domain
1438Names (IDN) <xref target="RFC3490"></xref> either in the ireg-name
1439part or elsewhere. Character Normalization also applies to IDNs, as
1440discussed in <xref target="schemecomp"></xref>.</t>
1441</section> <!-- charnorm -->
1442
1443<section title="Percent-Encoding Normalization">
1444
1445<t>The percent-encoding mechanism (Section 2.1 of <xref
1446target="RFC3986"></xref>) is a frequent source of variance among
1447otherwise identical IRIs. In addition to the case normalization issue
1448noted above, some IRI producers percent-encode octets that do not
1449require percent-encoding, resulting in IRIs that are equivalent to
1450their nonencoded counterparts. These IRIs should be normalized by
1451decoding any percent-encoded octet sequence that corresponds to an
1452unreserved character, as described in section 2.3 of <xref
1453target="RFC3986"></xref>.</t>
1454
1455<t>For actual resolution, differences in percent-encoding (except for
1456the percent-encoding of reserved characters) MUST always result in the
1457same resource.  For example, "http://example.org/~user",
1458"http://example.org/%7euser", and "http://example.org/%7Euser", must
1459resolve to the same resource.</t>
1460
1461<t>If this kind of equivalence is to be tested, the percent-encoding
1462of both IRIs to be compared has to be aligned; for example, by
1463converting both IRIs to URIs (see Section 3.1), eliminating escape
1464differences in the resulting URIs, and making sure that the case of
1465the hexadecimal characters in the percent-encoding is always the same
1466(preferably upper case). If the IRI is to be passed to another
1467application or used further in some other way, its original form MUST
1468be preserved.  The conversion described here should be performed only
1469for local comparison.</t>
1470
1471</section> <!-- pctnorm -->
1472
1473<section title="Path Segment Normalization">
1474
1475<t>The complete path segments "." and ".." are intended only for use
1476within relative references (Section 4.1 of <xref
1477target="RFC3986"></xref>) and are removed as part of the reference
1478resolution process (Section 5.2 of <xref target="RFC3986"></xref>).
1479However, some implementations may incorrectly assume that reference
1480resolution is not necessary when the reference is already an IRI, and
1481thus fail to remove dot-segments when they occur in non-relative
1482paths.  IRI normalizers should remove dot-segments by applying the
1483remove_dot_segments algorithm to the path, as described in Section
14845.2.4 of <xref target="RFC3986"></xref>.</t>
1485
1486</section> <!-- pathnorm -->
1487</section> <!-- ladder -->
1488
1489<section title="Scheme-Based Normalization" anchor="schemecomp">
1490
1491<t>The syntax and semantics of IRIs vary from scheme to scheme, as
1492described by the defining specification for each
1493scheme. Implementations may use scheme-specific rules, at further
1494processing cost, to reduce the probability of false negatives. For
1495example, because the "http" scheme makes use of an authority
1496component, has a default port of "80", and defines an empty path to be
1497equivalent to "/", the following four IRIs are equivalent:</t>
1498
1499<figure><artwork>
1500   http://example.com
1501   http://example.com/
1502   http://example.com:/
1503   http://example.com:80/</artwork></figure>
1504
1505<t>In general, an IRI that uses the generic syntax for authority with
1506an empty path should be normalized to a path of "/". Likewise, an
1507explicit ":port", for which the port is empty or the default for the
1508scheme, is equivalent to one where the port and its ":" delimiter are
1509elided and thus should be removed by scheme-based normalization. For
1510example, the second IRI above is the normal form for the "http"
1511scheme.</t>
1512
1513<t>Another case where normalization varies by scheme is in the
1514handling of an empty authority component or empty host
1515subcomponent. For many scheme specifications, an empty authority or
1516host is considered an error; for others, it is considered equivalent
1517to "localhost" or the end-user's host. When a scheme defines a default
1518for authority and an IRI reference to that default is desired, the
1519reference should be normalized to an empty authority for the sake of
1520uniformity, brevity, and internationalization. If, however, either the
1521userinfo or port subcomponents are non-empty, then the host should be
1522given explicitly even if it matches the default.</t>
1523
1524<t>Normalization should not remove delimiters when their associated
1525component is empty unless it is licensed to do so by the scheme
1526specification. For example, the IRI "http://example.com/?" cannot be
1527assumed to be equivalent to any of the examples above. Likewise, the
1528presence or absence of delimiters within a userinfo subcomponent is
1529usually significant to its interpretation.  The fragment component is
1530not subject to any scheme-based normalization; thus, two IRIs that
1531differ only by the suffix "#" are considered different regardless of
1532the scheme.</t>
1533 
1534<t>Some IRI schemes allow the usage of Internationalized Domain
1535Names (IDN) <xref target='RFC5890'></xref> either in their ireg-name
1536part or elswhere. When in use in IRIs, those names SHOULD
1537conform to the definition of U-Label in <xref
1538target='RFC5890'></xref>. An IRI containing an invalid IDN cannot
1539successfully be resolved. For legibility purposes, they
1540SHOULD NOT be converted into ASCII Compatible Encoding (ACE).</t>
1541
1542<t>Scheme-based normalization may also consider IDN
1543components and their conversions to punycode as equivalent. As an
1544example, "http://r&amp;#xE9;sum&amp;#xE9;.example.org" may be
1545considered equivalent to
1546"http://xn--rsum-bpad.example.org".</t><t>Other scheme-specific
1547normalizations are possible.</t>
1548
1549</section> <!-- schemenorm -->
1550
1551<section title="Protocol-Based Normalization">
1552
1553<t>Substantial effort to reduce the incidence of false negatives is
1554often cost-effective for web spiders. Consequently, they implement
1555even more aggressive techniques in IRI comparison. For example, if
1556they observe that an IRI such as</t>
1557
1558<figure><artwork>
1559   http://example.com/data</artwork></figure>
1560<t>redirects to an IRI differing only in the trailing slash</t>
1561<figure><artwork>
1562   http://example.com/data/</artwork></figure>
1563
1564<t>they will likely regard the two as equivalent in the future.  This
1565kind of technique is only appropriate when equivalence is clearly
1566indicated by both the result of accessing the resources and the common
1567conventions of their scheme's dereference algorithm (in this case, use
1568of redirection by HTTP origin servers to avoid problems with relative
1569references).</t>
1570
1571</section> <!-- protonorm -->
1572</section> <!-- equivalence -->
1573</section> 
1574
1575<section title="Use of IRIs" anchor="IRIuse">
1576
1577<section title="Limitations on UCS Characters Allowed in IRIs" anchor="limitations">
1578
1579<t>This section discusses limitations on characters and character
1580sequences usable for IRIs beyond those given in <xref target="abnf"/>
1581and <xref target="visual"/>. The considerations in this section are
1582relevant when IRIs are created and when URIs are converted to
1583IRIs.</t>
1584
1585<t>
1586
1587<list style="hanging"><t hangText="a.">The repertoire of characters allowed
1588    in each IRI component is limited by the definition of that component.
1589    For example, the definition of the scheme component does not allow
1590    characters beyond US-ASCII.
1591    <vspace blankLines="1"/>
1592    (Note: In accordance with URI practice, generic IRI
1593    software cannot and should not check for such limitations.)</t>
1594
1595<t hangText="b.">The UCS contains many areas of characters for which
1596    there are strong visual look-alikes. Because of the likelihood of
1597    transcription errors, these also should be avoided. This includes
1598    the full-width equivalents of Latin characters, half-width
1599    Katakana characters for Japanese, and many others. It also
1600    includes many look-alikes of "space", "delims", and "unwise",
1601    characters excluded in <xref target="RFC3491"/>.</t>
1602   
1603</list>
1604</t>
1605
1606<t>Additional information is available from <xref target="UNIXML"/>.
1607    <xref target="UNIXML"/> is written in the context of running text
1608    rather than in that of identifiers. Nevertheless, it discusses
1609    many of the categories of characters not appropriate for IRIs.</t>
1610</section> <!-- limitations -->
1611
1612<section title="Software Interfaces and Protocols">
1613
1614<t>Although an IRI is defined as a sequence of characters, software
1615interfaces for URIs typically function on sequences of octets or other
1616kinds of code units. Thus, software interfaces and protocols MUST
1617define which character encoding is used.</t>
1618
1619<t>Intermediate software interfaces between IRI-capable components and
1620URI-only components MUST map the IRIs per <xref target="mapping"/>,
1621when transferring from IRI-capable to URI-only components.
1622
1623This mapping SHOULD be applied as late as possible. It SHOULD NOT be
1624applied between components that are known to be able to handle IRIs.</t>
1625</section> <!-- software -->
1626
1627<section title="Format of URIs and IRIs in Documents and Protocols">
1628
1629<t>Document formats that transport URIs may have to be upgraded to allow
1630the transport of IRIs. In cases where the document as a whole
1631has a native character encoding, IRIs MUST also be encoded in this
1632character encoding and converted accordingly by a parser or interpreter.
1633
1634IRI characters not expressible in the native character encoding SHOULD
1635be escaped by using the escaping conventions of the document format if
1636such conventions are available. Alternatively, they MAY be
1637percent-encoded according to <xref target="mapping"/>. For example, in
1638HTML or XML, numeric character references SHOULD be used. If a
1639document as a whole has a native character encoding and that character
1640encoding is not UTF-8, then IRIs MUST NOT be placed into the document
1641in the UTF-8 character encoding.</t>
1642
1643<t>((UPDATE THIS NOTE)) Note: Some formats already accommodate IRIs,
1644although they use different terminology. HTML 4.0 <xref
1645target="HTML4"/> defines the conversion from IRIs to URIs as
1646error-avoiding behavior. XML 1.0 <xref target="XML1"/>, XLink <xref
1647target="XLink"/>, XML Schema <xref target="XMLSchema"/>, and
1648specifications based upon them allow IRIs. Also, it is expected that
1649all relevant new W3C formats and protocols will be required to handle
1650IRIs <xref target="CharMod"/>.</t>
1651
1652</section> <!-- format -->
1653
1654<section title="Use of UTF-8 for Encoding Original Characters" anchor="UTF8use">
1655
1656<t>This section discusses details and gives examples for point c) in
1657<xref target="Applicability"/>. To be able to use IRIs, the URI
1658corresponding to the IRI in question has to encode original characters
1659into octets by using UTF-8.  This can be specified for all URIs of a
1660URI scheme or can apply to individual URIs for schemes that do not
1661specify how to encode original characters.  It can apply to the whole
1662URI, or only to some part. For background information on encoding
1663characters into URIs, see also Section 2.5 of <xref
1664target="RFC3986"/>.</t>
1665
1666<t>For new URI schemes, using UTF-8 is recommended in <xref
1667target="RFC4395bis"/>.  Examples where UTF-8 is already used are the URN
1668syntax <xref target="RFC2141"/>, IMAP URLs <xref target="RFC2192"/>,
1669and POP URLs <xref target="RFC2384"/>.  On the other hand, because the
1670HTTP URI scheme does not specify how to encode original characters,
1671only some HTTP URLs can have corresponding but different IRIs.</t>
1672
1673<t>For example, for a document with a URI
1674of<vspace/>"http://www.example.org/r%C3%A9sum%C3%A9.html", it is
1675possible to construct a corresponding IRI (in XML notation, see <xref
1676target="sec-Notation"/>):
1677"http://www.example.org/r&amp;#xE9;sum&amp;#xE9;.html" ("&amp;#xE9;"
1678stands for the e-acute character, and "%C3%A9" is the UTF-8 encoded
1679and percent-encoded representation of that character). On the other
1680hand, for a document with a URI of
1681"http://www.example.org/r%E9sum%E9.html", the percent-encoding octets
1682cannot be converted to actual characters in an IRI, as the
1683percent-encoding is not based on UTF-8.</t>
1684
1685<t>For most URI schemes, there is no need to upgrade their scheme
1686definition in order for them to work with IRIs.  The main case where
1687upgrading makes sense is when a scheme definition, or a particular
1688component of a scheme, is strictly limited to the use of US-ASCII
1689characters with no provision to include non-ASCII characters/octets
1690via percent-encoding, or if a scheme definition currently uses highly
1691scheme-specific provisions for the encoding of non-ASCII characters.
1692An example of this is the mailto: scheme <xref target="RFC2368"/>.</t>
1693
1694<t>This specification updates the IANA registry of URI schemes to note
1695their applicability to IRIs, see <xref target="iana"/>.  All IRIs use
1696URI schemes, and all URIs with URI schemes can be used as IRIs, even
1697though in some cases only by using URIs directly as IRIs, without any
1698conversion.</t>
1699
1700<t>Scheme definitions can impose restrictions on the syntax of
1701scheme-specific URIs; i.e., URIs that are admissible under the generic
1702URI syntax <xref target="RFC3986"/> may not be admissible due to
1703narrower syntactic constraints imposed by a URI scheme
1704specification. URI scheme definitions cannot broaden the syntactic
1705restrictions of the generic URI syntax; otherwise, it would be
1706possible to generate URIs that satisfied the scheme-specific syntactic
1707constraints without satisfying the syntactic constraints of the
1708generic URI syntax. However, additional syntactic constraints imposed
1709by URI scheme specifications are applicable to IRI, as the
1710corresponding URI resulting from the mapping defined in <xref
1711target="mapping"/> MUST be a valid URI under the syntactic
1712restrictions of generic URI syntax and any narrower restrictions
1713imposed by the corresponding URI scheme specification.</t>
1714
1715<t>The requirement for the use of UTF-8 generally applies to all parts
1716of a URI.  However, it is possible that the capability of IRIs to
1717represent a wide range of characters directly is used just in some
1718parts of the IRI (or IRI reference). The other parts of the IRI may
1719only contain US-ASCII characters, or they may not be based on
1720UTF-8. They may be based on another character encoding, or they may
1721directly encode raw binary data (see also <xref
1722target="RFC2397"/>). </t>
1723
1724<t>For example, it is possible to have a URI reference
1725of<vspace/>"http://www.example.org/r%E9sum%E9.xml#r%C3%A9sum%C3%A9",
1726where the document name is encoded in iso-8859-1 based on server
1727settings, but where the fragment identifier is encoded in UTF-8 according
1728to <xref target="XPointer"/>. The IRI corresponding to the above
1729URI would be (in XML notation)<vspace/>"http://www.example.org/r%E9sum%E9.xml#r&amp;#xE9;sum&amp;#xE9;".</t>
1730
1731<t>Similar considerations apply to query parts. The functionality
1732of IRIs (namely, to be able to include non-ASCII characters) can
1733only be used if the query part is encoded in UTF-8.</t>
1734
1735</section> <!-- utf8 -->
1736
1737<section title="Relative IRI References">
1738<t>Processing of relative IRI references against a base is handled
1739straightforwardly; the algorithms of <xref target="RFC3986"/> can
1740be applied directly, treating the characters additionally allowed
1741in IRI references in the same way that unreserved characters are in URI
1742references.</t>
1743
1744</section> <!-- relative -->
1745</section> <!-- IRIuse -->
1746
1747<section title="Liberal handling of otherwise invalid IRIs" anchor="LEIRIHREF">
1748
1749<t>(EDITOR NOTE: This Section may move to an appendix.)
1750 
1751Some technical specifications and widely-deployed software have
1752allowed additional variations and extensions of IRIs to be used in
1753syntactic components. This section describes two widely-used
1754preprocessing agreements. Other technical specifications may wish to
1755reference a syntactic component which is "a valid IRI or a string that
1756will map to a valid IRI after this preprocessing algorithm". These two
1757variants are known as <xref target="LEIRI">Legacy Extended IRI or
1758LEIRI</xref>, and <xref target="HTML5">Web Address</xref>).
1759</t>
1760
1761<t>Future technical specifications SHOULD NOT allow conforming
1762producers to produce, or conforming content to contain, such forms,
1763as they are not interoperable with other IRI consuming software.</t>
1764
1765<section title="LEIRI processing"  anchor="LEIRIspec">
1766  <t>This section defines Legacy Extended IRIs (LEIRIs).
1767    The syntax of Legacy Extended IRIs is the same as that for IRIs,
1768    except that the ucschar production is replaced by the leiri-ucschar production:</t>
1769<figure>
1770
1771<artwork>
1772  leiri-ucschar  = " " / "&lt;" / "&gt;" / '"' / "{" / "}" / "|"
1773                   / "\" / "^" / "`" / %x0-1F / %x7F-D7FF
1774                   / %xE000-FFFD / %x10000-10FFFF
1775</artwork>
1776
1777<postamble>
1778  Among other extensions, processors based on this specification also
1779  did not enforce the restriction on bidirectional formatting
1780  characters in <xref target="visual"></xref>, and the iprivate
1781  production becomes redundant.</postamble>
1782</figure>
1783
1784<t>To convert a string allowed as a LEIRI to an IRI, each character
1785allowed in leiri-ucschar but not in ucschar must be percent-encoded
1786using <xref target="compmapping"/>.</t>
1787</section> <!-- leiriproc -->
1788
1789<section title="Web Address processing" anchor="webaddress">
1790
1791<t>Many popular web browsers have taken the approach of being quite
1792liberal in what is accepted as a "URL" or its relative
1793forms. This section describes their behavior in terms of a preprocessor
1794which maps strings into the IRI space for subsequent parsing and
1795interpretation as an IRI.</t>
1796
1797<t>In some situations, it might be appropriate to describe the syntax
1798that a liberal consumer implementation might accept as a "Web
1799Address" or "Hypertext Reference" or "HREF". However,
1800technical specifications SHOULD restrict the syntactic form allowed by compliant producers
1801to the IRI or IRI reference syntax defined in this document
1802even if they want to mandate this processing.</t>
1803
1804<t>
1805Summary:
1806<list style="symbols">
1807   <t>Leading and trailing whitespace is removed.</t>
1808   <t>Some additional characters are removed.</t>
1809   <t>Some additional characters are allowed and escaped (as with LEIRI).</t>
1810   <t>If interpreting an IRI as a URI, the pct-encoding of the query
1811   component of the parsed URI component depends on operational
1812   context.</t>
1813</list>
1814</t>
1815
1816<t>Each string provided may have an associated charset (called
1817the HREF-charset here); this defaults to UTF-8.
1818For web browsers interpreting HTML, the document
1819charset of a string is determined:
1820
1821<list style="hanging">
1822<t hangText="If the string came from a script (e.g. as an argument to
1823 a method)">The HRef-charset is the script's charset.</t>
1824
1825<t hangText="If the string came from a DOM node (e.g. from an
1826  element)">The node has a Document, and the HRef-charset is the
1827  Document's character encoding.</t>
1828
1829<t hangText="If the string had a HRef-charset defined when the string was
1830created or defined">The HRef-charset is as defined.</t>
1831
1832</list></t>
1833
1834<t>If the resulting HRef-charset is a unicode based character encoding
1835(e.g., UTF-16), then use UTF-8 instead.</t>
1836
1837
1838<figure>
1839<preamble>The syntax for Web Addresses is obtained by replacing the 'ucschar',
1840  pct-form, and path-sep rules with the href-ucschar, href-pct-form, and href-path-sep
1841  rules below. In addition, some characters are stripped.</preamble>
1842
1843<artwork type='abnf'>
1844  href-ucschar  = " " / "&lt;" / "&gt;" / DQUOTE / "{" / "}" / "|"
1845                   / "\" / "^" / "`" / %x0-1F / %x7F-D7FF
1846                   / %xE000-FFFD / %x10000-10FFFF
1847  href-pct-form = pct-encoded / "%"
1848  href-path-sep = "/" / "\"
1849  href-strip    = &lt;to be done&gt;
1850</artwork>
1851
1852<postamble>
1853(NOTE: NEED TO FIX THESE SETS TO MATCH HTML5; NOT SURE ABOUT NEXT SENTENCE)
1854browsers did not enforce the restriction on bidirectional formatting
1855  characters in <xref target="visual"></xref>, and the iprivate
1856  production becomes redundant.</postamble>
1857</figure>
1858
1859<t>'Web Address processing' requires the following additional
1860preprocessing steps:
1861
1862<list style="numbers">
1863
1864<t>Leading and trailing instances of space (U+0020),
1865CR (U+000A), LF (U+000D), and TAB (U+0009) characters are removed.</t>
1866
1867<t>strip all characters in href-strip.</t>
1868  <t>Percent-encode all characters in href-ucschar not in ucschar.</t>
1869  <t>Replace occurrences of "%" not followed by two hexadecimal digits by "%25".</t>
1870  <t>Convert backslashes ('\') matching href-path-sep to forward slashes ('/').</t>
1871</list></t>
1872</section> <!-- webaddress -->
1873
1874<section title="Characters not allowed in IRIs" anchor="notAllowed">
1875
1876<t>This section provides a list of the groups of characters and code
1877points that are allowed by LEIRI or HREF but are not allowed in IRIs or are
1878allowed in IRIs only in the query part. For each group of characters,
1879advice on the usage of these characters is also given, concentrating
1880on the reasons for why they are excluded from IRI use.</t>
1881
1882<t>
1883
1884<list><t>Space (U+0020): Some formats and applications use space as a
1885delimiter, e.g. for items in a list. Appendix C of <xref
1886target="RFC3986"></xref> also mentions that white space may have to be
1887added when displaying or printing long URIs; the same applies to long
1888IRIs. This means that spaces can disappear, or can make the what is
1889intended as a single IRI or IRI reference to be treated as two or more
1890separate IRIs.</t>
1891
1892<t>Delimiters "&lt;" (U+003C), "&gt;" (U+003E), and '"' (U+0022):
1893Appendix C of <xref target="RFC3986"></xref> suggests the use of
1894double-quotes ("http://example.com/") and angle brackets
1895(&lt;http://example.com/&gt;) as delimiters for URIs in plain
1896text. These conventions are often used, and also apply to IRIs.  Using
1897these characters in strings intended to be IRIs would result in the
1898IRIs being cut off at the wrong place.</t>
1899
1900<t>Unwise characters "\" (U+005C), "^" (U+005E), "`"
1901(U+0060), "{" (U+007B), "|" (U+007C), and "}" (U+007D): These
1902characters originally have been excluded from URIs because the
1903respective codepoints are assigned to different graphic characters in
1904some 7-bit or 8-bit encoding. Despite the move to Unicode, some of
1905these characters are still occasionally displayed differently on some
1906systems, e.g. U+005C may appear as a Japanese Yen symbol on some
1907systems. Also, the fact that these characters are not used in URIs or
1908IRIs has encouraged their use outside URIs or IRIs in contexts that
1909may include URIs or IRIs. If a string with such a character were used
1910as an IRI in such a context, it would likely be interpreted
1911piecemeal.</t>
1912
1913<t>The controls (C0 controls, DEL, and C1 controls, #x0 - #x1F #x7F -
1914#x9F): There is generally no way to transmit these characters reliably
1915as text outside of a charset encoding.  Even when in encoded form,
1916many software components silently filter out some of these characters,
1917or may stop processing alltogether when encountering some of
1918them. These characters may affect text display in subtle, unnoticable
1919ways or in drastic, global, and irreversible ways depending on the
1920hardware and software involved. The use of some of these characters
1921would allow malicious users to manipulate the display of an IRI and
1922its context in many situations.</t>
1923
1924<t>Bidi formatting characters (U+200E, U+200F, U+202A-202E): These
1925characters affect the display ordering of characters. If IRIs were
1926allowed to contain these characters and the resulting visual display
1927transcribed. they could not be converted back to electronic form
1928(logical order) unambiguously. These characters, if allowed in IRIs,
1929might allow malicious users to manipulate the display of IRI and its
1930context.</t>
1931
1932<t>Specials (U+FFF0-FFFD): These code points provide functionality
1933beyond that useful in an IRI, for example byte order identification,
1934annotation, and replacements for unknown characters and objects. Their
1935use and interpretation in an IRI would serve no purpose and might lead
1936to confusing display variations.</t>
1937
1938<t>Private use code points (U+E000-F8FF, U+F0000-FFFFD,
1939U+100000-10FFFD): Display and interpretation of these code points is
1940by definition undefined without private agreement. Therefore, these
1941code points are not suited for use on the Internet. They are not
1942interoperable and may have unpredictable effects.</t>
1943
1944<t>Tags (U+E0000-E0FFF): These characters provide a way to language
1945tag in Unicode plain text. They are not appropriate for IRIs because
1946language information in identifiers cannot reliably be input,
1947transmitted (e.g. on a visual medium such as paper), or
1948recognized.</t>
1949
1950<t>Non-characters (U+FDD0-FDEF, U+1FFFE-1FFFF, U+2FFFE-2FFFF,
1951U+3FFFE-3FFFF, U+4FFFE-4FFFF, U+5FFFE-5FFFF, U+6FFFE-6FFFF,
1952U+7FFFE-7FFFF, U+8FFFE-8FFFF, U+9FFFE-9FFFF, U+AFFFE-AFFFF,
1953U+BFFFE-BFFFF, U+CFFFE-CFFFF, U+DFFFE-DFFFF, U+EFFFE-EFFFF,
1954U+FFFFE-FFFFF, U+10FFFE-10FFFF): These code points are defined as
1955non-characters. Applications may use some of them internally, but are
1956not prepared to interchange them.</t>
1957
1958</list></t>
1959
1960<t>LEIRI preprocessing disallowed some code points and
1961code units:
1962
1963<list><t>Surrogate code units (D800-DFFF): These do not represent
1964Unicode codepoints.</t></list></t>
1965</section> <!-- notallowed -->
1966</section> <!-- lieirihref -->
1967 
1968<section title="URI/IRI Processing Guidelines (Informative)" anchor="guidelines">
1969
1970<t>This informative section provides guidelines for supporting IRIs in
1971the same software components and operations that currently process
1972URIs: Software interfaces that handle URIs, software that allows users
1973to enter URIs, software that creates or generates URIs, software that
1974displays URIs, formats and protocols that transport URIs, and software
1975that interprets URIs. These may all require modification before
1976functioning properly with IRIs. The considerations in this section
1977also apply to URI references and IRI references.</t>
1978
1979<section title="URI/IRI Software Interfaces">
1980<t>Software interfaces that handle URIs, such as URI-handling APIs and
1981protocols transferring URIs, need interfaces and protocol elements
1982that are designed to carry IRIs.</t>
1983
1984<t>In case the current handling in an API or protocol is based on
1985US-ASCII, UTF-8 is recommended as the character encoding for IRIs, as
1986it is compatible with US-ASCII, is in accordance with the
1987recommendations of <xref target="RFC2277"/>, and makes converting to
1988URIs easy. In any case, the API or protocol definition must clearly
1989define the character encoding to be used.</t>
1990
1991<t>The transfer from URI-only to IRI-capable components requires no
1992mapping, although the conversion described in <xref
1993target="URItoIRI"/> above may be performed. It is preferable not to
1994perform this inverse conversion unless it is certain this can be done
1995correctly.</t>
1996</section>
1997
1998<section title="URI/IRI Entry">
1999
2000<t>Some components allow users to enter URIs into the system
2001by typing or dictation, for example. This software must be updated to allow
2002for IRI entry.</t>
2003
2004<t>A person viewing a visual representation of an IRI (as a sequence
2005of glyphs, in some order, in some visual display) or hearing an IRI
2006will use an entry method for characters in the user's language to
2007input the IRI. Depending on the script and the input method used, this
2008may be a more or less complicated process.</t>
2009
2010<t>The process of IRI entry must ensure, as much as possible, that the
2011restrictions defined in <xref target="abnf"/> are met. This may be
2012done by choosing appropriate input methods or variants/settings
2013thereof, by appropriately converting the characters being input, by
2014eliminating characters that cannot be converted, and/or by issuing a
2015warning or error message to the user.</t>
2016
2017<t>As an example of variant settings, input method editors for East
2018Asian Languages usually allow the input of Latin letters and related
2019characters in full-width or half-width versions. For IRI input, the
2020input method editor should be set so that it produces half-width Latin
2021letters and punctuation and full-width Katakana.</t>
2022
2023<t>An input field primarily or solely used for the input of URIs/IRIs
2024might allow the user to view an IRI as it is mapped to a URI.  Places
2025where the input of IRIs is frequent may provide the possibility for
2026viewing an IRI as mapped to a URI. This will help users when some of
2027the software they use does not yet accept IRIs.</t>
2028
2029<t>An IRI input component interfacing to components that handle URIs,
2030but not IRIs, must map the IRI to a URI before passing it to these
2031components.</t>
2032
2033<t>For the input of IRIs with right-to-left characters, please see
2034<xref target="bidiInput"></xref>.</t>
2035</section>
2036
2037<section title="URI/IRI Transfer between Applications">
2038
2039<t>Many applications (for example, mail user agents) try to detect
2040URIs appearing in plain text. For this, they use some heuristics based
2041on URI syntax. They then allow the user to click on such URIs and
2042retrieve the corresponding resource in an appropriate (usually
2043scheme-dependent) application.</t>
2044
2045<t>Such applications would need to be upgraded, in order to use the
2046IRI syntax as a base for heuristics. In particular, a non-ASCII
2047character should not be taken as the indication of the end of an IRI.
2048Such applications also would need to make sure that they correctly
2049convert the detected IRI from the character encoding of the document
2050or application where the IRI appears, to the character encoding used
2051by the system-wide IRI invocation mechanism, or to a URI (according to
2052<xref target="mapping"/>) if the system-wide invocation mechanism only
2053accepts URIs.</t>
2054
2055<t>The clipboard is another frequently used way to transfer URIs and
2056IRIs from one application to another. On most platforms, the clipboard
2057is able to store and transfer text in many languages and scripts.
2058Correctly used, the clipboard transfers characters, not octets, which
2059will do the right thing with IRIs.</t>
2060</section>
2061
2062<section title="URI/IRI Generation">
2063
2064<t>Systems that offer resources through the Internet, where those
2065resources have logical names, sometimes automatically generate URIs
2066for the resources they offer. For example, some HTTP servers can
2067generate a directory listing for a file directory and then respond to
2068the generated URIs with the files.</t>
2069
2070<t>Many legacy character encodings are in use in various file systems.
2071Many currently deployed systems do not transform the local character
2072representation of the underlying system before generating URIs.</t>
2073
2074<t>For maximum interoperability, systems that generate resource
2075identifiers should make the appropriate transformations. For example,
2076if a file system contains a file named
2077"r&amp;#xE9;sum&amp;#xE9;.html", a server should expose this as
2078"r%C3%A9sum%C3%A9.html" in a URI, which allows use of
2079"r&amp;#xE9;sum&amp;#xE9;.html" in an IRI, even if locally the file
2080name is kept in a character encoding other than UTF-8.
2081</t>
2082
2083<t>This recommendation particularly applies to HTTP servers. For FTP
2084servers, similar considerations apply; see <xref target="RFC2640"/>.</t>
2085</section>
2086
2087<section title="URI/IRI Selection" anchor="selection">
2088<t>In some cases, resource owners and publishers have control over the
2089IRIs used to identify their resources. This control is mostly
2090executed by controlling the resource names, such as file names,
2091directly.</t>
2092
2093<t>In these cases, it is recommended to avoid choosing IRIs that are
2094easily confused. For example, for US-ASCII, the lower-case ell ("l") is
2095easily confused with the digit one ("1"), and the upper-case oh ("O") is
2096easily confused with the digit zero ("0"). Publishers should avoid
2097confusing users with "br0ken" or "1ame" identifiers.</t>
2098
2099<t>Outside the US-ASCII repertoire, there are many more opportunities for
2100confusion; a complete set of guidelines is too lengthy to include
2101here. As long as names are limited to characters from a single script,
2102native writers of a given script or language will know best when
2103ambiguities can appear, and how they can be avoided. What may look
2104ambiguous to a stranger may be completely obvious to the average
2105native user. On the other hand, in some cases, the UCS contains
2106variants for compatibility reasons; for example, for typographic purposes.
2107These should be avoided wherever possible. Although there may be exceptions,
2108newly created resource names should generally be in NFKC
2109<xref target="UTR15"></xref> (which means that they are also in NFC).</t>
2110
2111<t>As an example, the UCS contains the "fi" ligature at U+FB01
2112for compatibility reasons.
2113Wherever possible, IRIs should use the two letters "f" and "i" rather
2114than the "fi" ligature. An example where the latter may be used is
2115in the query part of an IRI for an explicit search for a word written
2116containing the "fi" ligature.</t>
2117
2118<t>In certain cases, there is a chance that characters from different
2119scripts look the same. The best known example is the similarity of the
2120Latin "A", the Greek "Alpha", and the Cyrillic "A". To avoid such
2121cases, IRIs should only be created where all the characters in a
2122single component are used together in a given language. This usually
2123means that all of these characters will be from the same script, but
2124there are languages that mix characters from different scripts (such
2125as Japanese).  This is similar to the heuristics used to distinguish
2126between letters and numbers in the examples above. Also, for Latin,
2127Greek, and Cyrillic, using lowercase letters results in fewer
2128ambiguities than using uppercase letters would.</t>
2129</section>
2130
2131<section title="Display of URIs/IRIs" anchor="display">
2132<t>
2133In situations where the rendering software is not expected to display
2134non-ASCII parts of the IRI correctly using the available layout and font
2135resources, these parts should be percent-encoded before being displayed.</t>
2136
2137<t>For display of Bidi IRIs, please see <xref target="visual"/>.</t>
2138</section>
2139
2140<section title="Interpretation of URIs and IRIs">
2141<t>Software that interprets IRIs as the names of local resources should
2142accept IRIs in multiple forms and convert and match them with the
2143appropriate local resource names.</t>
2144
2145<t>First, multiple representations include both IRIs in the native
2146character encoding of the protocol and also their URI counterparts.</t>
2147
2148<t>Second, it may include URIs constructed based on character
2149encodings other than UTF-8. These URIs may be produced by user agents that do
2150not conform to this specification and that use legacy character encodings to
2151convert non-ASCII characters to URIs. Whether this is necessary, and what
2152character encodings to cover, depends on a number of factors, such as
2153the legacy character encodings used locally and the distribution of
2154various versions of user agents. For example, software for Japanese
2155may accept URIs in Shift_JIS and/or EUC-JP in addition to UTF-8.</t>
2156
2157<t>Third, it may include additional mappings to be more user-friendly
2158and robust against transmission errors. These would be similar to how
2159some servers currently treat URIs as case insensitive or perform
2160additional matching to account for spelling errors. For characters
2161beyond the US-ASCII repertoire, this may, for example, include
2162ignoring the accents on received IRIs or resource names. Please note
2163that such mappings, including case mappings, are language
2164dependent.</t>
2165
2166<t>It can be difficult to identify a resource unambiguously if too
2167many mappings are taken into consideration. However, percent-encoded
2168and not percent-encoded parts of IRIs can always be clearly distinguished.
2169Also, the regularity of UTF-8 (see <xref target="Duerst97"/>) makes the
2170potential for collisions lower than it may seem at first.</t>
2171</section>
2172
2173<section title="Upgrading Strategy">
2174<t>Where this recommendation places further constraints on software
2175for which many instances are already deployed, it is important to
2176introduce upgrades carefully and to be aware of the various
2177interdependencies.</t>
2178
2179<t>If IRIs cannot be interpreted correctly, they should not be created,
2180generated, or transported. This suggests that upgrading URI interpreting
2181software to accept IRIs should have highest priority.</t>
2182
2183<t>On the other hand, a single IRI is interpreted only by a single or
2184very few interpreters that are known in advance, although it may be
2185entered and transported very widely.</t>
2186
2187<t>Therefore, IRIs benefit most from a broad upgrade of software to be
2188able to enter and transport IRIs. However, before an
2189individual IRI is published, care should be taken to upgrade the corresponding
2190interpreting software in order to cover the forms expected to be
2191received by various versions of entry and transport software.</t>
2192
2193<t>The upgrade of generating software to generate IRIs instead of using a
2194local character encoding should happen only after the service is upgraded
2195to accept IRIs. Similarly, IRIs should only be generated when the service
2196accepts IRIs and the intervening infrastructure and protocol is known
2197to transport them safely.</t>
2198
2199<t>Software converting from URIs to IRIs for display should be upgraded
2200only after upgraded entry software has been widely deployed to the
2201population that will see the displayed result.</t>
2202
2203
2204<t>Where there is a free choice of character encodings, it is often
2205possible to reduce the effort and dependencies for upgrading to IRIs
2206by using UTF-8 rather than another encoding. For example, when a new
2207file-based Web server is set up, using UTF-8 as the character encoding
2208for file names will make the transition to IRIs easier. Likewise, when
2209a new Web form is set up using UTF-8 as the character encoding of the
2210form page, the returned query URIs will use UTF-8 as the character
2211encoding (unless the user, for whatever reason, changes the character
2212encoding) and will therefore be compatible with IRIs.</t>
2213
2214
2215<t>These recommendations, when taken together, will allow for the
2216extension from URIs to IRIs in order to handle characters other than
2217US-ASCII while minimizing interoperability problems. For
2218considerations regarding the upgrade of URI scheme definitions, see
2219<xref target="UTF8use"/>.</t>
2220
2221</section>
2222</section> <!-- guidelines -->
2223
2224<section title="IANA Considerations" anchor="iana">
2225
2226<t>RFC Editor and IANA note: Please Replace RFC XXXX with the
2227number of this document when it issues as an RFC. </t>
2228
2229<t>IANA maintains a registry of "URI schemes". A "URI scheme" also
2230serves an "IRI scheme". </t>
2231
2232<t>To clarify that the URI scheme registration process also applies to
2233IRIs, change the description of the "URI schemes" registry
2234header to say "[RFC4395] defines an IANA-maintained registry of URI
2235Schemes. These registries include the Permanent and Provisional URI
2236Schemes.  RFC XXXX updates this registry to designate that schemes may
2237also indicate their usability as IRI schemes.</t>
2238
2239<t> Update "per RFC 4395" to "per RFC 4395 and RFC XXXX".
2240</t>
2241
2242</section> <!-- IANA -->
2243   
2244<section title="Security Considerations" anchor="security">
2245<t>The security considerations discussed in <xref target="RFC3986"/>
2246also apply to IRIs. In addition, the following issues require
2247particular care for IRIs.</t>
2248<t>Incorrect encoding or decoding can lead to security problems.
2249In particular, some UTF-8 decoders do not check against overlong
2250byte sequences. As an example, a "/" is encoded with the byte 0x2F
2251both in UTF-8 and in US-ASCII, but some UTF-8 decoders also wrongly
2252interpret the sequence 0xC0 0xAF as a "/". A sequence such as "%C0%AF.."
2253may pass some security tests and then be interpreted
2254as "/.." in a path if UTF-8 decoders are fault-tolerant, if conversion
2255and checking are not done in the right order, and/or if reserved
2256characters and unreserved characters are not clearly distinguished.</t>
2257
2258<t>There are various ways in which "spoofing" can occur with IRIs.
2259"Spoofing" means that somebody may add a resource name that looks the
2260same or similar to the user, but that points to a different resource.
2261The added resource may pretend to be the real resource by looking
2262very similar but may contain all kinds of changes that may be
2263difficult to spot and that can cause all kinds of problems.
2264Most spoofing possibilities for IRIs are extensions of those for URIs.</t>
2265
2266<t>Spoofing can occur for various reasons. First, a user's normalization expectations or actual normalization
2267when entering an IRI or  transcoding an IRI from a legacy character
2268encoding do not match the normalization used on the
2269server side. Conceptually, this is no different from the problems
2270surrounding the use of case-insensitive web servers. For example,
2271a popular web page with a mixed-case name ("http://big.example.com/PopularPage.html")
2272might be "spoofed" by someone who is able to create "http://big.example.com/popularpage.html".
2273However, the use of unnormalized character sequences, and of additional
2274mappings for user convenience, may increase the chance for spoofing.
2275Protocols and servers that allow the creation of resources with
2276names that are not normalized are particularly vulnerable to such
2277attacks. This is an inherent
2278security problem of the relevant protocol, server, or resource
2279and is not specific to IRIs, but it is mentioned here for completeness.</t>
2280
2281<t>Spoofing can occur in various IRI components, such as the
2282domain name part or a path part. For considerations specific
2283to the domain name part, see <xref target="RFC3491"/>.
2284For the path part, administrators of sites that allow independent
2285users to create resources in the same sub area may have to be careful
2286to check for spoofing.</t>
2287
2288<t>Spoofing can occur because in the UCS many characters look very similar. Details are discussed in <xref target="selection"/>.
2289Again, this is very similar to spoofing possibilities on US-ASCII,
2290e.g., using "br0ken" or "1ame" URIs.</t>
2291
2292<t>Spoofing can occur when URIs with percent-encodings based on various
2293character encodings are accepted to deal with older user agents. In some
2294cases, particularly for Latin-based resource names, this is usually easy to
2295detect because UTF-8-encoded names, when interpreted and viewed as
2296legacy character encodings, produce mostly garbage.</t><t>When
2297concurrently used character encodings have a similar structure but there
2298are no characters that have exactly the same encoding, detection is more
2299difficult.</t>
2300
2301<t>Spoofing can occur with bidirectional IRIs, if the restrictions
2302in <xref target="bidi-structure"/> are not followed. The same visual
2303representation may be interpreted as different logical representations,
2304and vice versa. It is also very important that a correct Unicode bidirectional
2305implementation be used.</t><t>The use of Legacy Extended IRIs introduces additional security issues.</t>
2306</section><!-- security -->
2307
2308<section title="Acknowledgements">
2309<t>For contributions to this update, we would like to thank Ian Hickson, Michael Sperberg-McQueen, Dan Connolly, Norman Walsh, Richard Tobin, Henry S. Thomson, and the XML Core Working Group of the W3C.</t>
2310
2311<t>The discussion on the issue addressed here started a long time
2312ago. There was a thread in the HTML working
2313group in August 1995 (under the topic of "Globalizing URIs") and in the
2314www-international mailing list in July 1996 (under the topic of
2315"Internationalization and URLs"), and there were ad-hoc meetings at the Unicode
2316conferences in September 1995 and September 1997.</t>
2317
2318<t>For contributions to the previous version of this document, RFC 3987, many thanks go to
2319Francois Yergeau, Matitiahu Allouche,
2320Roy Fielding, Tim Berners-Lee, Mark Davis,
2321M.T. Carrasco Benitez, James Clark, Tim Bray, Chris Wendt, Yaron Goland,
2322Andrea Vine, Misha Wolf, Leslie Daigle, Ted Hardie, Bill Fenner, Margaret Wasserman,
2323Russ Housley, Makoto MURATA, Steven Atkin,
2324Ryan Stansifer, Tex Texin, Graham Klyne, Bjoern Hoehrmann, Chris Lilley, Ian Jacobs,
2325Adam Costello, Dan Oscarson, Elliotte Rusty Harold, Mike J. Brown,
2326Roy Badami, Jonathan Rosenne, Asmus Freytag, Simon Josefsson, Carlos Viegas Damasio,
2327Chris Haynes, Walter Underwood, and many others.</t>
2328<t>A definition of HyperText Reference was initially produced by Ian Hixson,
2329and further edited by Dan Connolly and C. M. Spergerg-McQueen.</t>
2330<t>Thanks to the Internationalization Working
2331Group (I18N WG) of the World Wide Web Consortium (W3C),
2332and the members of the W3C
2333I18N Working Group and Interest Group for their contributions and their
2334work on <xref target="CharMod"/>. Thanks also go
2335to the members of many other W3C Working Groups for adopting IRIs, and to
2336the members of the Montreal IAB Workshop on Internationalization and
2337Localization for their review.</t>
2338</section>
2339
2340<section title="Main Changes Since RFC 3987">
2341  <t>This section describes the main changes since <xref target="RFC3987"></xref>.</t>
2342  <section title="Major restructuring of IRI processing model" anchor="forkChanges">
2343    <t>Major restructuring of IRI processing model to make scheme-specific translation necessary to handle IDNA requirements and for consistency with web implementations. </t>
2344    <t>Starting with IRI, you want one of:
2345      <list style="hanging">
2346        <t hangText="a"> IRI components (IRI parsed into UTF8 pieces)</t>
2347        <t hangText="b"> URI components (URI parsed into ASCII pieces, encoded correctly) </t>
2348        <t hangText="c"> whole URI  (for passing on to some other system that wants whole URIs) </t>
2349      </list></t>
2350   
2351    <section title="OLD WAY">
2352      <t><list style="numbers">
2353       
2354        <t>Pct-encoding on the whole thing to a URI.
2355          (c1) If you want a (maybe broken) whole URI, you might
2356          stop here.</t>
2357       
2358        <t>Parsing the URI into URI components.
2359          (b1) If you want (maybe broken) URI components, stop here.</t>
2360       
2361        <t> Decode the components (undoing the pct-encoding).
2362          (a) if you want IRI components, stop here.</t>
2363       
2364        <t> reencode:  Either using a different encoding some components
2365          (for domain names, and query components in web pages, which
2366          depends on the component, scheme and context), and otherwise
2367          using pct-encoding.
2368          (b2) if you want (good) URI components, stop here.</t>
2369       
2370        <t> reassemble the reencoded components.
2371          (c2) if you want a (*good*) whole URI stop here.</t>
2372      </list>
2373       
2374      </t>
2375     
2376    </section>
2377   
2378    <section title="NEW WAY">
2379      <t>
2380        <list style="numbers">
2381         
2382          <t> Parse the IRI into IRI components using the generic syntax.
2383            (a) if you want IRI components, stop here.</t>
2384         
2385          <t> Encode each components, using pct-encoding, IDN encoding, or
2386            special query part encoding depending on the component
2387            scheme or context. (b) If you want URI components, stop here.</t>
2388          <t> reassemble the a whole URI from URI components.
2389            (c) if you want a whole URI stop here.</t>
2390        </list></t>
2391    </section>
2392    <section title="Extension of Syntax">
2393      <t>Added the tag range (U+E0000-E0FFF) to the iprivate production.
2394        Some IRIs generated with the new syntax may fail to pass very strict checks
2395        relying on the old syntax. But characters in this range should be extremely infrequent
2396        anyway.</t>
2397    </section>
2398    <section title="More to be added"><t>TODO: There are more main changes that need to be
2399      documented in this section.</t></section>
2400</section>
2401
2402<section title="Change Log">
2403
2404<t>Note to RFC Editor: Please completely remove this section before publication.</t>
2405
2406<section title='Changes after draft-ietf-iri-3987bis-01'>
2407    <t>Changes from draft-ietf-iri-3987bis-01 onwards are available as changesets
2408      in the IETF tools subversion repository at
2409      http://trac.tools.ietf.org/wg/iri/trac/log/draft-ietf-iri-3987bis/draft-ietf-iri-3987bis.xml.</t>
2410</section>
2411 
2412<section title='Changes from draft-duerst-iri-bis-07 to draft-ietf-iri-3987bis-00'>
2413     <t>Changed draft name, date, last paragraph of abstract, and titles in change log, and added this section
2414     in moving from draft-duerst-iri-bis-07 (personal submission) to draft-ietf-iri-3987bis-00 (WG document).</t>
2415</section>
2416
2417<section title="Changes from -06 to -07 of draft-duerst-iri-bis">
2418  <t>Major restructuring of the processing model, see <xref target="forkChanges"></xref>.</t>
2419</section>
2420</section>
2421
2422<section title='Changes from -00 to -01'><t><list style="symbols">
2423  <t>Removed 'mailto:' before mail addresses of authors.</t>
2424  <t>Added "&lt;to be done&gt;" as right side of 'href-strip' rule. Fixed '|' to '/' for
2425    alternatives.</t>
2426</list></t>
2427</section>
2428
2429<section title="Changes from -05 to -06 of draft-duerst-iri-bis-00"><t><list style="symbols">
2430<t>Add HyperText Reference, change abstract, acks and references for it</t>
2431<t>Add Masinter back as another editor.</t>
2432<t>Masinter integrates HRef material from HTML5 spec.</t>
2433<t>Rewrite introduction sections to modernize.</t>
2434</list></t>
2435</section>
2436
2437<section title="Changes from -04 to -05 of draft-duerst-iri-bis"><t><list style="symbols"><t>Updated references.</t><t>Changed IPR text to pre5378Trust200902.</t></list></t>
2438</section>
2439
2440<section title="Changes from -03 to -04 of draft-duerst-iri-bis"><t><list style="symbols"><t>Added explicit abbreviation for LEIRIs.</t><t>Mentioned LEIRI references.</t><t>Completed text in LEIRI section about tag characters and about specials.</t></list></t>
2441</section>
2442
2443<section title="Changes from -02 to -03 of draft-duerst-iri-bis"><t><list style="symbols"><t>Updated some references.</t><t>Updated Michel Suginard's coordinates.</t></list></t>
2444</section>
2445
2446<section title="Changes from -01 to -02 of draft-duerst-iri-bis"><t><list style="symbols"><t>Added tag range to iprivate (issue private-include-tags-115).</t><t>Added Specials (U+FFF0-FFFD) to Legacy Extended IRIs.</t></list></t>
2447</section>
2448<section title="Changes from -00 to -01 of draft-duerst-iri-bis"><t><list style="symbols"><t>Changed from "IRIs with Spaces/Controls" to "Legacy Extended IRI" based on input from the W3C XML Core WG. Moved the relevant subsections to the back and promoted them to a section.</t><t>Added some text re. Legacy Extended IRIs to the security section.</t><t>Added a IANA Consideration Section.</t><t>Added this Change Log Section.</t><t>Added a section about "IRIs with Spaces/Controls" (converting from a Note in RFC 3987).</t></list></t>
2449</section>
2450<section title="Changes from RFC 3987 to -00 of draft-duerst-iri-bis"><t><list><t>Fixed errata (see http://www.rfc-editor.org/cgi-bin/errataSearch.pl?rfc=3987).</t></list></t>
2451</section>
2452</section>
2453</middle>
2454
2455<back>
2456<references title="Normative References">
2457
2458<reference anchor="ASCII">
2459<front>
2460<title>Coded Character Set -- 7-bit American Standard Code for Information
2461Interchange</title>
2462<author>
2463<organization>American National Standards Institute</organization>
2464</author>
2465<date year="1986"/>
2466</front>
2467<seriesInfo name="ANSI" value="X3.4"/>
2468</reference>
2469
2470<reference anchor="ISO10646">
2471<front>
2472<title>ISO/IEC 10646:2003: Information Technology -
2473Universal Multiple-Octet Coded Character Set (UCS)</title>
2474<author>
2475<organization>International Organization for Standardization</organization>
2476</author>
2477<date month="December" year="2003"/>
2478</front>
2479<seriesInfo name="ISO" value="Standard 10646"/>
2480</reference>
2481
2482&rfc2119;
2483&rfc3490;
2484&rfc3491;
2485&rfc3629;
2486&rfc3986;
2487
2488<reference anchor="STD68">
2489<front>
2490<title abbrev="ABNF">Augmented BNF for Syntax Specifications: ABNF</title>
2491<author initials="D." surname="Crocker" fullname="Dave Crocker"><organization/></author>
2492<author initials="P." surname="Overell" fullname="Paul Overell"><organization/></author>
2493<date month="January" year="2008"/></front>
2494<seriesInfo name="STD" value="68"/><seriesInfo name="RFC" value="5234"/>
2495</reference>
2496 
2497&rfc5890;
2498&rfc5891;
2499
2500<reference anchor="UNIV4">
2501<front>
2502<title>The Unicode Standard, Version 5.1.0, defined by: The Unicode Standard,
2503Version 5.0 (Boston, MA, Addison-Wesley, 2007. ISBN 0-321-48091-0),
2504as amended by Unicode 4.1.0 (http://www.unicode.org/versions/Unicode5.1.0/)</title>
2505<author><organization>The Unicode Consortium</organization></author>
2506<date year="2008" month="April"/>
2507</front>
2508</reference>
2509
2510<reference anchor="UNI9" target="http://www.unicode.org/reports/tr9/tr9-13.html">
2511<front>
2512<title>The Bidirectional Algorithm</title>
2513<author initials="M." surname="Davis" fullname="Mark Davis"><organization/></author>
2514<date year="2004" month="March"/>
2515</front>
2516<seriesInfo name="Unicode Standard Annex" value="#9"/>
2517</reference>
2518
2519<reference anchor="UTR15" target="http://www.unicode.org/unicode/reports/tr15/tr15-23.html">
2520<front>
2521<title>Unicode Normalization Forms</title>
2522<author initials="M." surname="Davis" fullname="Mark Davis"><organization/></author>
2523<author initials="M.J." surname="Duerst" fullname="Martin Duerst"><organization/></author>
2524<date year="2008" month="March"/>
2525</front>
2526<seriesInfo name="Unicode Standard Annex" value="#15"/>
2527</reference>
2528
2529</references>
2530
2531<references title="Informative References">
2532
2533<reference anchor="BidiEx" target="http://www.w3.org/International/iri-edit/BidiExamples">
2534<front>
2535<title>Examples of bidirectional IRIs</title>
2536<author><organization/></author>
2537<date year="" month=""/>
2538</front>
2539</reference>
2540
2541<reference anchor="CharMod" target="http://www.w3.org/TR/charmod-resid">
2542<front>
2543<title>Character Model for the World Wide Web: Resource Identifiers</title>
2544<author initials="M." surname="Duerst" fullname="Martin Duerst"><organization/></author>
2545<author initials="F." surname="Yergeau" fullname="Francois Yergeau"><organization/></author>
2546<author initials="R." surname="Ishida" fullname="Richard Ishida"><organization/></author>
2547<author initials="M." surname="Wolf" fullname="Misha Wolf"><organization/></author>
2548<author initials="T." surname="Texin" fullname="Tex Texin"><organization/></author>
2549<date year="2004" month="November" day="25"/>
2550</front>
2551<seriesInfo name="World Wide Web Consortium" value="Candidate Recommendation"/>
2552</reference>
2553
2554<reference anchor="Duerst97" target="http://www.ifi.unizh.ch/mml/mduerst/papers/PDF/IUC11-UTF-8.pdf">
2555<front>
2556<title>The Properties and Promises of UTF-8</title>
2557<author initials="M.J." surname="Duerst" fullname="Martin Duerst"><organization/></author>
2558<date year="1997" month="September"/>
2559</front>
2560<seriesInfo name="Proc. 11th International Unicode Conference, San Jose" value=""/>
2561</reference>
2562
2563<reference anchor="Gettys" target="http://www.w3.org/DesignIssues/ModelConsequences">
2564<front>
2565<title>URI Model Consequences</title>
2566<author initials="J." surname="Gettys" fullname="Jim Gettys"><organization/></author>
2567<date month="" year=""/>
2568</front>
2569</reference>
2570
2571<reference anchor="HTML4" target="http://www.w3.org/TR/html401/appendix/notes.html#h-B.2">
2572<front>
2573<title>HTML 4.01 Specification</title>
2574<author initials="D." surname="Raggett" fullname="Dave Raggett"><organization/></author>
2575<author initials="A." surname="Le Hors" fullname="Arnaud Le Hors"><organization/></author>
2576<author initials="I." surname="Jacobs" fullname="Ian Jacobs"><organization/></author>
2577<date year="1999" month="December" day="24"/>
2578</front>
2579<seriesInfo name="World Wide Web Consortium" value="Recommendation"/>
2580</reference>
2581
2582<reference anchor="LEIRI" target="http://www.w3.org/TR/leiri/">
2583<front>
2584<title>Legacy extended IRIs for XML resource identification</title>
2585<author initials="H." surname="Thompson" fullname="Henry Thompson"><organization/></author>
2586<author initials="R." surname="Tobin"    fullname="Richard Tobin"><organization/></author>
2587<author initials="N." surname="Walsh" fullname="Norman Walsh"><organization/></author>
2588  <date year="2008" month="November" day="3"/>
2589
2590</front>
2591<seriesInfo name="World Wide Web Consortium" value="Note"/>
2592</reference>
2593
2594
2595&rfc2045;
2596&rfc2130;
2597&rfc2141;
2598&rfc2192;
2599&rfc2277;
2600&rfc2368;
2601&rfc2384;
2602&rfc2396;
2603&rfc2397;
2604&rfc2616;
2605&rfc1738;
2606&rfc2640;
2607&rfc3987;
2608<reference anchor='RFC4395bis'>
2609  <front>
2610    <title>Guidelines and Registration Procedures for New URI/IRI Schemes</title>
2611    <author initials='T.' surname='Hansen' fullname="Tony Hansen"><organization/></author>
2612    <author initials='T.' surname='Hardie' fullname="Ted Hardie"><organization/></author>
2613    <author initials='L.' surname='Masinter' fullname="Larry Masinter"><organization/></author>
2614    <date year="2010" month='September' day="30"/>
2615    <workgroup>IRI</workgroup>
2616  </front>
2617  <seriesInfo name="Internet-Draft" value="draft-hansen-iri-4395bis-irireg-00"/>
2618</reference>
2619 
2620 
2621<reference anchor="UNIXML" target="http://www.w3.org/TR/unicode-xml/">
2622<front>
2623<title>Unicode in XML and other Markup Languages</title>
2624<author initials="M.J." surname="Duerst" fullname="Martin Duerst"><organization/></author>
2625<author initials="A." surname="Freytag" fullname="Asmus Freytag"><organization/></author>
2626<date year="2003" month="June" day="18"/>
2627</front>
2628<seriesInfo name="Unicode Technical Report" value="#20"/>
2629<seriesInfo name="World Wide Web Consortium" value="Note"/>
2630</reference>
2631 
2632<reference anchor="UTR36" target="http://unicode.org/reports/tr36/">
2633<front>
2634<title>Unicode Security Considerations</title>
2635<author initials="M." surname="Davis" fullname="Mark Davis"><organization/></author>
2636<author initials="M." surname="Suignard" fullname="Michel Suignard"><organization/></author>
2637<date year="2010" month="August" day="4"/>
2638</front>
2639<seriesInfo name="Unicode Technical Report" value="#36"/>
2640</reference>
2641
2642<reference anchor="XLink" target="http://www.w3.org/TR/xlink/#link-locators">
2643<front>
2644<title>XML Linking Language (XLink) Version 1.0</title>
2645<author initials="S." surname="DeRose" fullname="Steve DeRose"><organization/></author>
2646<author initials="E." surname="Maler" fullname="Eve Maler"><organization/></author>
2647<author initials="D." surname="Orchard" fullname="David Orchard"><organization/></author>
2648<date year="2001" month="June" day="27"/>
2649</front>
2650<seriesInfo name="World Wide Web Consortium" value="Recommendation"/>
2651</reference>
2652
2653<reference anchor="XML1" target="http://www.w3.org/TR/REC-xml">
2654  <front>
2655    <title>Extensible Markup Language (XML) 1.0 (Forth Edition)</title>
2656    <author initials="T." surname="Bray" fullname="Tim Bray"><organization/></author>
2657    <author initials="J." surname="Paoli" fullname="Jean Paoli"><organization/></author>
2658    <author initials="C.M." surname="Sperberg-McQueen" fullname="C. M. Sperberg-McQueen">
2659      <organization/></author>
2660    <author initials="E." surname="Maler" fullname="Eve Maler"><organization/></author>
2661    <author initials="F." surname="Yergeau" fullname="Francois Yergeau"><organization/></author>
2662    <date day="16" month="August" year="2006"/>
2663  </front>
2664  <seriesInfo name="World Wide Web Consortium" value="Recommendation"/>
2665</reference>
2666
2667<reference anchor="XMLNamespace" target="http://www.w3.org/TR/REC-xml-names">
2668  <front>
2669    <title>Namespaces in XML (Second Edition)</title>
2670    <author initials="T." surname="Bray" fullname="Tim Bray"><organization/></author>
2671    <author initials="D." surname="Hollander" fullname="Dave Hollander"><organization/></author>
2672    <author initials="A." surname="Layman" fullname="Andrew Layman"><organization/></author>
2673    <author initials="R." surname="Tobin" fullname="Richard Tobin"><organization></organization></author><date day="16" month="August" year="2006"/>
2674  </front>
2675  <seriesInfo name="World Wide Web Consortium" value="Recommendation"/>
2676</reference>
2677
2678<reference anchor="XMLSchema" target="http://www.w3.org/TR/xmlschema-2/#anyURI">
2679<front>
2680<title>XML Schema Part 2: Datatypes</title>
2681<author initials="P." surname="Biron" fullname="Paul Biron"><organization/></author>
2682<author initials="A." surname="Malhotra" fullname="Ashok Malhotra"><organization/></author>
2683<date year="2001" month="May" day="2"/>
2684</front>
2685<seriesInfo name="World Wide Web Consortium" value="Recommendation"/>
2686</reference>
2687
2688<reference anchor="XPointer" target="http://www.w3.org/TR/xptr-framework/#escaping">
2689<front>
2690<title>XPointer Framework</title>
2691<author initials="P." surname="Grosso" fullname="Paul Grosso"><organization/></author>
2692<author initials="E." surname="Maler" fullname="Eve Maler"><organization/></author>
2693<author initials="J." surname="Marsh" fullname="Jonathan Marsh"><organization/></author>
2694<author initials="N." surname="Walsh" fullname="Norman Walsh"><organization/></author>
2695<date year="2003" month="March" day="25"/>
2696</front>
2697<seriesInfo name="World Wide Web Consortium" value="Recommendation"/>
2698</reference>
2699
2700<reference anchor="HTML5" target="http://www.w3.org/TR/2009/WD-html5-20090423/">
2701<front>
2702<title>A vocabulary and associated APIs for HTML and XHTML</title>
2703<author initials="I." surname="Hickson" fullname="Ian Hickson"><organization>Google, Inc.</organization></author>
2704<author initials="D." surname="Hyatt" fullname="David Hyatt"><organization>Apple, Inc.</organization></author>
2705<date year="2009"  month="April" day="23"/>
2706</front>
2707<seriesInfo name="World Wide Web Consortium" value="Working Draft"/>
2708</reference>
2709
2710</references>
2711
2712<section title="Design Alternatives">
2713<t>This section briefly summarizes some design alternatives
2714considered earlier and the reasons why they were not chosen.</t>
2715<section title="New Scheme(s)">
2716<t>Introducing new schemes (for example, httpi:, ftpi:,...) or a
2717new metascheme (e.g., i:, leading to URI/IRI prefixes such as
2718i:http:, i:ftp:,...) was proposed to make IRI-to-URI conversion
2719scheme dependent or to distinguish between percent-encodings
2720resulting from IRI-to-URI conversion and percent-encodings from
2721legacy character encodings.</t>
2722
2723<t>New schemes are not needed to distinguish URIs from true IRIs (i.e.,
2724  IRIs that contain non-ASCII characters). The benefit of being able
2725  to detect the origin of percent-encodings is marginal, as UTF-8
2726  can be detected with very high reliability. Deploying new schemes is
2727  extremely hard, so not requiring new schemes for IRIs makes
2728  deployment of IRIs vastly easier. Making conversion scheme dependent
2729  is highly inadvisable and would be encouraged by separate schemes for IRIs.
2730  Using a uniform convention for conversion from IRIs to URIs makes
2731  IRI implementation orthogonal to the introduction of actual new
2732  schemes.</t>
2733</section>
2734<section title="Character Encodings Other Than UTF-8">
2735<t>At an early stage, UTF-7 was considered as an alternative to
2736UTF-8 when IRIs are converted to URIs. UTF-7 would not have needed
2737percent-encoding and  in most cases would have been shorter than
2738percent-encoded UTF-8.</t>
2739<t>Using UTF-8 avoids a double layering and overloading of the use of
2740   the "+" character. UTF-8 is fully compatible with US-ASCII and has
2741   therefore been recommended by the IETF, and is being used widely.</t>
2742 
2743  <t>UTF-7 has never been used much and is now clearly being
2744   discouraged. Requiring implementations to convert from UTF-8
2745   to UTF-7 and back would be an additional implementation burden.</t>
2746</section> <!-- notutf8 -->
2747<section title="New Encoding Convention">
2748<t>Instead of using the existing percent-encoding convention
2749of URIs, which is based on octets, the idea was to create a new
2750encoding convention; for example, to use "%u" to introduce
2751UCS code points.</t>
2752<t>Using the existing octet-based percent-encoding mechanism
2753does not need an upgrade of the URI syntax and does not
2754need corresponding server upgrades.</t>
2755</section> <!-- new encoding -->
2756<section title="Indicating Character Encodings in the URI/IRI">
2757<t>Some proposals suggested indicating the character encodings used
2758in an URI or IRI with some new syntactic convention in the URI itself,
2759similar to the "charset" parameter for e-mails and Web pages.
2760As an example, the label in square brackets in
2761"http://www.example.org/ros[iso-8859-1]&amp;#xE9;" indicated that
2762the following "&amp;#xE9;" had to be interpreted as iso-8859-1.</t>
2763<t>If UTF-8 is used exclusively, an upgrade to the URI syntax is not needed.
2764It avoids potentially multiple labels that have to be copied correctly
2765in all cases, even on the side of a bus or on a napkin, leading to
2766usability problems (and being prohibitively annoying).
2767Exclusively using UTF-8 also reduces transcoding errors and confusion.</t>
2768</section> <!-- indicating -->
2769</section>
2770</back>
2771</rfc>
Note: See TracBrowser for help on using the repository browser.