idnits 2.17.1 draft-ietf-idnabis-protocol-12.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** The document seems to lack a License Notice according IETF Trust Provisions of 28 Dec 2009, Section 6.b.i or Provisions of 12 Sep 2009 Section 6.b -- however, there's a paragraph with a matching beginning. Boilerplate error? (You're using the IETF Trust Provisions' Section 6.b License Notice from 12 Feb 2009 rather than one of the newer Notices. See https://trustee.ietf.org/license-info/.) Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- -- The draft header indicates that this document obsoletes RFC3490, but the abstract doesn't seem to mention this, which it should. -- The draft header indicates that this document updates RFC3492, but the abstract doesn't seem to mention this, which it should. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year (Using the creation date from RFC3492, updated by this document, for RFC5378 checks: 2002-01-10) -- The document seems to contain a disclaimer for pre-RFC5378 work, and may have content which was first submitted before 10 November 2008. The disclaimer is necessary when there are original authors that you have been unable to contact, or if some do not wish to grant the BCP78 rights to the IETF Trust. If you are able to get all authors (current and original) to grant those rights, you can and should remove the disclaimer; otherwise, the disclaimer is needed and you can ignore this comment. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (May 8, 2009) is 5466 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'RFC1123' is defined on line 673, but no explicit reference was found in the text == Unused Reference: 'Unicode-PropertyValueAliases' is defined on line 683, but no explicit reference was found in the text == Unused Reference: 'Unicode-RegEx' is defined on line 688, but no explicit reference was found in the text == Unused Reference: 'Unicode-Scripts' is defined on line 693, but no explicit reference was found in the text == Unused Reference: 'ASCII' is defined on line 705, but no explicit reference was found in the text == Unused Reference: 'RFC2136' is defined on line 719, but no explicit reference was found in the text == Unused Reference: 'RFC2181' is defined on line 723, but no explicit reference was found in the text == Unused Reference: 'RFC2535' is defined on line 726, but no explicit reference was found in the text -- Possible downref: Non-RFC (?) normative reference: ref. 'IDNA2008-BIDI' -- Possible downref: Non-RFC (?) normative reference: ref. 'Unicode-PropertyValueAliases' -- Possible downref: Non-RFC (?) normative reference: ref. 'Unicode-RegEx' -- Possible downref: Non-RFC (?) normative reference: ref. 'Unicode-Scripts' -- Possible downref: Non-RFC (?) normative reference: ref. 'Unicode-UAX15' -- Obsolete informational reference (is this intentional?): RFC 2535 (Obsoleted by RFC 4033, RFC 4034, RFC 4035) -- Obsolete informational reference (is this intentional?): RFC 2671 (Obsoleted by RFC 6891) -- Obsolete informational reference (is this intentional?): RFC 3490 (Obsoleted by RFC 5890, RFC 5891) -- Obsolete informational reference (is this intentional?): RFC 3491 (Obsoleted by RFC 5891) -- Obsolete informational reference (is this intentional?): RFC 4952 (Obsoleted by RFC 6530) Summary: 1 error (**), 0 flaws (~~), 9 warnings (==), 14 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group J. Klensin 3 Internet-Draft May 8, 2009 4 Obsoletes: 3490, 3491 5 (if approved) 6 Updates: 3492 (if approved) 7 Intended status: Standards Track 8 Expires: November 9, 2009 10 Internationalized Domain Names in Applications (IDNA): Protocol 11 draft-ietf-idnabis-protocol-12.txt 13 Status of this Memo 15 This Internet-Draft is submitted to IETF in full conformance with the 16 provisions of BCP 78 and BCP 79. This document may contain material 17 from IETF Documents or IETF Contributions published or made publicly 18 available before November 10, 2008. The person(s) controlling the 19 copyright in some of this material may not have granted the IETF 20 Trust the right to allow modifications of such material outside the 21 IETF Standards Process. Without obtaining an adequate license from 22 the person(s) controlling the copyright in such materials, this 23 document may not be modified outside the IETF Standards Process, and 24 derivative works of it may not be created outside the IETF Standards 25 Process, except to format it for publication as an RFC or to 26 translate it into languages other than English. 28 Internet-Drafts are working documents of the Internet Engineering 29 Task Force (IETF), its areas, and its working groups. Note that 30 other groups may also distribute working documents as Internet- 31 Drafts. 33 Internet-Drafts are draft documents valid for a maximum of six months 34 and may be updated, replaced, or obsoleted by other documents at any 35 time. It is inappropriate to use Internet-Drafts as reference 36 material or to cite them other than as "work in progress." 38 The list of current Internet-Drafts can be accessed at 39 http://www.ietf.org/ietf/1id-abstracts.txt. 41 The list of Internet-Draft Shadow Directories can be accessed at 42 http://www.ietf.org/shadow.html. 44 This Internet-Draft will expire on November 9, 2009. 46 Copyright Notice 48 Copyright (c) 2009 IETF Trust and the persons identified as the 49 document authors. All rights reserved. 51 This document is subject to BCP 78 and the IETF Trust's Legal 52 Provisions Relating to IETF Documents in effect on the date of 53 publication of this document (http://trustee.ietf.org/license-info). 54 Please review these documents carefully, as they describe your rights 55 and restrictions with respect to this document. 57 Abstract 59 This document is the revised protocol definition for 60 internationalized domain names (IDNs). The rationale for changes, 61 the relationship to the older specification, and important 62 terminology are provided in other documents. This document specifies 63 the protocol mechanism, called Internationalizing Domain Names in 64 Applications (IDNA), for registering and looking up IDNs in a way 65 that does not require changes to the DNS itself. IDNA is only meant 66 for processing domain names, not free text. 68 Table of Contents 70 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 5 71 1.1. Discussion Forum . . . . . . . . . . . . . . . . . . . . . 5 72 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 73 3. Requirements and Applicability . . . . . . . . . . . . . . . . 6 74 3.1. Requirements . . . . . . . . . . . . . . . . . . . . . . . 6 75 3.2. Applicability . . . . . . . . . . . . . . . . . . . . . . 6 76 3.2.1. DNS Resource Records . . . . . . . . . . . . . . . . . 7 77 3.2.2. Non-domain-name Data Types Stored in the DNS . . . . . 7 78 4. Registration Protocol . . . . . . . . . . . . . . . . . . . . 7 79 4.1. Input to IDNA Registration Process . . . . . . . . . . . . 8 80 4.2. Permitted Character and Label Validation . . . . . . . . . 8 81 4.2.1. Input Format . . . . . . . . . . . . . . . . . . . . . 8 82 4.2.2. Rejection of Characters that are not Permitted . . . . 8 83 4.2.3. Label Validation . . . . . . . . . . . . . . . . . . . 8 84 4.2.4. Registration Validation Summary . . . . . . . . . . . 9 85 4.3. Registry Restrictions . . . . . . . . . . . . . . . . . . 10 86 4.4. Punycode Conversion . . . . . . . . . . . . . . . . . . . 10 87 4.5. Insertion in the Zone . . . . . . . . . . . . . . . . . . 10 88 5. Domain Name Lookup Protocol . . . . . . . . . . . . . . . . . 10 89 5.1. Label String Input . . . . . . . . . . . . . . . . . . . . 11 90 5.2. Conversion to Unicode . . . . . . . . . . . . . . . . . . 11 91 5.3. Character Changes in Preprocessing or the User 92 Interface . . . . . . . . . . . . . . . . . . . . . . . . 11 93 5.4. A-label Input . . . . . . . . . . . . . . . . . . . . . . 12 94 5.5. Validation and Character List Testing . . . . . . . . . . 13 95 5.6. Punycode Conversion . . . . . . . . . . . . . . . . . . . 14 96 5.7. DNS Name Resolution . . . . . . . . . . . . . . . . . . . 14 97 6. Security Considerations . . . . . . . . . . . . . . . . . . . 14 98 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 15 99 8. Contributors . . . . . . . . . . . . . . . . . . . . . . . . . 15 100 9. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 15 101 10. References . . . . . . . . . . . . . . . . . . . . . . . . . . 16 102 10.1. Normative References . . . . . . . . . . . . . . . . . . . 16 103 10.2. Informative References . . . . . . . . . . . . . . . . . . 17 104 Appendix A. Local Mapping Alternatives . . . . . . . . . . . . . 18 105 A.1. Transitional Mapping Model . . . . . . . . . . . . . . . . 18 106 A.1.1. Fallback Lookup . . . . . . . . . . . . . . . . . . . 19 107 A.1.2. Two-step Lookup . . . . . . . . . . . . . . . . . . . 19 108 A.2. Internationalized Resource Identifier (IRI) Mapping 109 Model . . . . . . . . . . . . . . . . . . . . . . . . . . 20 110 Appendix B. Summary of Major Changes from IDNA2003 . . . . . . . 21 111 Appendix C. Change Log . . . . . . . . . . . . . . . . . . . . . 21 112 C.1. Changes between Version -00 and -01 of 113 draft-ietf-idnabis-protocol . . . . . . . . . . . . . . . 21 114 C.2. Version -02 . . . . . . . . . . . . . . . . . . . . . . . 22 115 C.3. Version -03 . . . . . . . . . . . . . . . . . . . . . . . 22 116 C.4. Version -04 . . . . . . . . . . . . . . . . . . . . . . . 22 117 C.5. Version -05 . . . . . . . . . . . . . . . . . . . . . . . 22 118 C.6. Version -06 . . . . . . . . . . . . . . . . . . . . . . . 23 119 C.7. Version -07 . . . . . . . . . . . . . . . . . . . . . . . 23 120 C.8. Version -08 . . . . . . . . . . . . . . . . . . . . . . . 23 121 C.9. Version -09 . . . . . . . . . . . . . . . . . . . . . . . 24 122 C.10. Version -10 . . . . . . . . . . . . . . . . . . . . . . . 24 123 C.11. Version -11 . . . . . . . . . . . . . . . . . . . . . . . 24 124 C.12. Version -12 . . . . . . . . . . . . . . . . . . . . . . . 25 125 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . . 25 127 1. Introduction 129 This document supplies the protocol definition for internationalized 130 domain names. Essential definitions and terminology for 131 understanding this document and a road map of the collection of 132 documents that make up IDNA2008 appear in [IDNA2008-Defs]. 133 Appendix B discusses the relationship between this specification and 134 the earlier version of IDNA (referred to here as "IDNA2003") and the 135 rationale for these changes, along with considerable explanatory 136 material and advice to zone administrators who support IDNs is 137 provided in another documents, notably [IDNA2008-Rationale]. 139 IDNA works by allowing applications to use certain ASCII string 140 labels (beginning with a special prefix) to represent non-ASCII name 141 labels. Lower-layer protocols need not be aware of this; therefore 142 IDNA does not changes any infrastructure. In particular, IDNA does 143 not depend on any changes to DNS servers, resolvers, or protocol 144 elements, because the ASCII name service provided by the existing DNS 145 can be used for IDNA. 147 IDNA applies only to DNS labels. The base DNS standards [RFC1034] 148 [RFC1035] and their various updates specify how to combine labels 149 into fully-qualified domain names and parse labels out of those 150 names. 152 This document describes two separate protocols, one for IDN 153 registration (Section 4) and one for IDN lookup (Section 5), that 154 share some terminology, reference data and operations. [[anchor2: 155 Note in draft: See the note in the introduction to.]]Section 5 157 1.1. Discussion Forum 159 [[anchor4: RFC Editor: please remove this section.]] 161 This work is being discussed in the IETF IDNABIS WG and on the 162 mailing list idna-update@alvestrand.no 164 2. Terminology 166 Terminology used in IDNA, but also in Unicode or other character set 167 standards and the DNS, appears in [IDNA2008-Defs]. Terminology that 168 is required as part of the IDNA definition, including the definitions 169 of "ACE", appears in that document as well. Readers of this document 170 are assumed to be familiar with [IDNA2008-Defs] and with the DNS- 171 specific terminology in RFC 1034 [RFC1034]. 173 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 174 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 175 document are to be interpreted as described in BCP 14, RFC 2119 176 [RFC2119]. 178 3. Requirements and Applicability 180 3.1. Requirements 182 IDNA makes the following requirements: 184 1. Whenever a domain name is put into an IDN-unaware domain name 185 slot (see Section 2 and [IDNA2008-Defs]), it MUST contain only 186 ASCII characters (i.e., must be either an A-label or an NR-LDH- 187 label), unless the DNS application is not subject to historical 188 recommendations for "hostname"-style names (see [RFC1034] and 189 Section 3.2.1). 191 2. Labels MUST be compared using equivalent forms: either both 192 A-Label forms or both U-Label forms. Because A-labels and 193 U-labels can be transformed into each other without loss of 194 information, these comparisons are equivalent. A pair of 195 A-labels MUST be compared as case-insensitive ASCII (as with all 196 comparisons of ASCII DNS labels). U-labels must be compared 197 as-is, without case-folding or other intermediate steps. Note 198 that it is not necessary to validate labels in order to compare 199 them. In many cases, validation may be important for other 200 reasons and SHOULD be performed. 202 3. Labels being registered MUST conform to the requirements of 203 Section 4. Labels being looked up and the lookup process MUST 204 conform to the requirements of Section 5. 206 3.2. Applicability 208 IDNA applies to all domain names in all domain name slots in 209 protocols except where it is explicitly excluded. It does not apply 210 to domain name slots which do not use the Letter/Digit/Hyphen (LDH) 211 syntax rules. 213 Because it uses the DNS, IDNA applies to many protocols that were 214 specified before it was designed. IDNs occupying domain name slots 215 in those older protocols MUST be in A-label form until and unless 216 those protocols and implementations of them are explicitly upgraded 217 to be aware of IDNs in Unicode. IDNs actually appearing in DNS 218 queries or responses MUST be A-labels. 220 IDNA is not defined for extended label types (see RFC 2671, Section 3 222 [RFC2671]). 224 3.2.1. DNS Resource Records 226 IDNA applies only to domain names in the NAME and RDATA fields of DNS 227 resource records whose CLASS is IN. See RFC 1034 [RFC1034] for 228 precise definitions of these terms. 230 The application of IDNA to DNS resource records depends entirely on 231 the CLASS of the record, and not on the TYPE except as noted below. 232 This will remain true, even as new types are defined, unless a new 233 type defines type-specific rules. Special naming conventions for SRV 234 records (and "underscore names" more generally) are incompatible with 235 IDNA coding. The first two labels on a SRV type record (the ones 236 required to start in "_") MUST NOT be A-labels or U-labels, because 237 conversion to an A-label would lose information (since the underscore 238 is not a letter, digit, or hyphen and is consequently DISALLOWED in 239 IDNs). Of course, those labels may be part of a domain that uses IDN 240 labels at higher levels in the tree. 242 3.2.2. Non-domain-name Data Types Stored in the DNS 244 Although IDNA enables the representation of non-ASCII characters in 245 domain names, that does not imply that IDNA enables the 246 representation of non-ASCII characters in other data types that are 247 stored in domain names, specifically in the RDATA field for types 248 that have structured RDATA format. For example, an email address 249 local part is stored in a domain name in the RNAME field as part of 250 the RDATA of an SOA record (hostmaster@example.com would be 251 represented as hostmaster.example.com). IDNA does not update the 252 existing email standards, which allow only ASCII characters in local 253 parts. Even though work is in progress to define 254 internationalization for email addresses [RFC4952], changes to the 255 email address part of the SOA RDATA would require action in, or 256 updates to, other standards, specifically those that specify the 257 format of the SOA RR. 259 4. Registration Protocol 261 This section defines the procedure for registering an IDN. The 262 procedure is implementation independent; any sequence of steps that 263 produces exactly the same result for all labels is considered a valid 264 implementation. 266 Note that, while the registration and lookup protocols (Section 5) 267 are very similar in most respects, they are different and 268 implementers should carefully follow the appropriate steps. 270 4.1. Input to IDNA Registration Process 272 Registration processes, especially processing by entities, such as 273 "registrars" who deal with registrants before the request actually 274 reaches the zone manager ("registry") are outside the scope of these 275 protocols and may differ significantly depending on local needs. By 276 the time a string enters the IDNA registration process as described 277 in this specification, it is expected to be in Unicode and MUST be in 278 Unicode Normalization Form C (NFC [Unicode-UAX15]). Entities 279 responsible for zone files ("registries") are expected to accept only 280 the exact string for which registration is requested, free of any 281 mappings or local adjustments. They SHOULD avoid any possible 282 ambiguity by accepting registrations only for A-labels, possibly 283 paired with the relevant U-labels so that they can verify the 284 correspondence. 286 4.2. Permitted Character and Label Validation 288 4.2.1. Input Format 290 The registry SHOULD permit submission of labels in A-label form and 291 is encouraged to accept both the A-label form and the U-label one. 292 If it does so, it MUST perform a conversion to a U-label, perform the 293 steps and tests described below, and verify that the A-label produced 294 by the step in Section 4.4 matches the one provided as input. In 295 addition, if a U-label was provided, that U-label and the one 296 obtained by conversion of the A-label MUST match exactly. If, for 297 some reason, these tests fail, the registration MUST be rejected. If 298 the conversion to a U-label is not performed, the registry MUST still 299 verify that the A-label is superficially valid, i.e., that it does 300 not violate any of the rules of Punycode [RFC3492] encoding such as 301 the prohibition on trailing hyphen-minus, appearance of non-basic 302 characters before the delimiter, and so on. Fake A-labels, i.e., 303 invalid strings that appear to be A-labels but are not, MUST NOT be 304 placed in DNS zones that support IDNA. 306 4.2.2. Rejection of Characters that are not Permitted 308 The candidate Unicode string MUST NOT contain characters in the 309 "DISALLOWED" and "UNASSIGNED" lists specified in [IDNA2008-Tables]. 311 4.2.3. Label Validation 313 The proposed label (in the form of a Unicode string, i.e., a string 314 that at least superficially appears to be a U-label) is then 315 examined, performing tests that require examination of more than one 316 character. Character order is considered to be the on-the-wire 317 order, not the display order. 319 4.2.3.1. Consecutive Hyphens 321 The Unicode string MUST NOT contain "--" (two consecutive hyphens) in 322 the third and fourth character positions. 324 4.2.3.2. Leading Combining Marks 326 The Unicode string MUST NOT begin with a combining mark or combining 327 character (see The Unicode Standard, Section 2.11 [Unicode] for an 328 exact definition). 330 4.2.3.3. Contextual Rules 332 The Unicode string MUST NOT contain any characters whose validity is 333 context-dependent, unless the validity is positively confirmed by a 334 contextual rule. To check this, each code-point marked as CONTEXTJ 335 and CONTEXTO in [IDNA2008-Tables] MUST have a non-null rule. If such 336 a code-point is missing a rule, it is invalid. If the rule exists 337 but the result of applying the rule is negative or inconclusive, the 338 proposed label is invalid. 340 NOTE: These contextual rules are required to support characters that 341 could be used, under some conditions, to produce misleading labels or 342 to cause unacceptable ambiguity in label matching and interpretation. 343 For example, labels containing zero-width characters may be permitted 344 in context with characters whose presentation forms are significantly 345 changed by the zero-width characters, while other labels in which 346 zero-width characters appear may be rejected. 347 [[anchor11: Note in draft: Should this note be moved to Rationale??? 348 It has no normative consequences here.]] 350 4.2.3.4. Labels Containing Characters Written Right to Left 352 If the proposed label contains any characters that are written from 353 right to left it MUST meet the "bidi" criteria [IDNA2008-BIDI]. 355 4.2.4. Registration Validation Summary 357 Strings that contain at least one non-ASCII character, have been 358 produced by the steps above, whose contents pass all of the tests in 359 Section 4.2, and are 63 or fewer characters long in ACE form (see 360 Section 4.4), are U-labels. 362 To summarize, tests are made in Section 4.2 for invalid characters, 363 invalid combinations of characters, for labels that are invalid even 364 if the characters they contain are valid individually, and for labels 365 that do not conform to the restrictions for strings containing right 366 to left characters. 368 4.3. Registry Restrictions 370 In addition to the rules and tests above, there are many reasons why 371 a registry could reject a label. Registries at all levels of the 372 DNS, not just the top level, establish policies about label 373 registrations. Policies are likely to be informed by the local 374 languages and may depend on many factors including what characters 375 are in the label (for example, a label may be rejected based on other 376 labels already registered). See [IDNA2008-Rationale] for a 377 discussion and recommendations about registry policies. 379 The string produced by the above steps is checked and processed as 380 appropriate to local registry restrictions. Application of those 381 registry restrictions may result in the rejection of some labels or 382 the application of special restrictions to others. 384 4.4. Punycode Conversion 386 The resulting U-label is converted to an A-label. The A-label, more 387 precisely defined elsewhere, is the encoding of the U-label according 388 to the Punycode algorithm [RFC3492] with the ACE prefix "xn--" added 389 at the beginning of the string. The resulting string must, of 390 course, conform to the length limits imposed by the DNS. This 391 document updates RFC 3492 only to the extent of replacing the 392 reference to the discussion of the ACE prefix. The ACE prefix is now 393 specified in this document rather than as part of RFC 3490 or 394 Nameprep [RFC3491] but is the same in both sets of documents. 396 The failure conditions identified in the Punycode encoding procedure 397 cannot occur if the input is a U-label as determined by the steps 398 above. 400 4.5. Insertion in the Zone 402 The A-label is registered in the DNS by insertion into a zone. 404 5. Domain Name Lookup Protocol 406 Lookup is different from registration and different tests are applied 407 on the client. Although some validity checks are necessary to avoid 408 serious problems with the protocol, the lookup-side tests are more 409 permissive and rely on the assumption that names that are present in 410 the DNS are valid. That assumption is, however, a weak one because 411 the presence of wild cards in the DNS might cause a string that is 412 not actually registered in the DNS to be successfully looked up. 414 The two steps in Section 5.2 and Section 5.3 are required. 416 [[anchor14: Note in Draft: Try to reorganize and renumber Section 5 417 (Lookup) so that it exactly parallels Section 4 (Registration). This 418 has not been done in drafts -10 through -12 because the task will be 419 much easier if the local mapping material is pulled from here (and 420 there is no point trying to align the section numbers twice).]] 422 5.1. Label String Input 424 The user supplies a string in the local character set, typically by 425 typing it or clicking on, or copying and pasting, a resource 426 identifier, e.g., a URI [RFC3986] or IRI [RFC3987] from which the 427 domain name is extracted. Alternately, some process not directly 428 involving the user may read the string from a file or obtain it in 429 some other way. Processing in this step and the next two are local 430 matters, to be accomplished prior to actual invocation of IDNA. 432 5.2. Conversion to Unicode 434 The string is converted from the local character set into Unicode, if 435 it is not already Unicode. A Unicode string may require 436 normalization as discussed in Section 4.1. The result MUST be a 437 Unicode string in NFC form. 439 5.3. Character Changes in Preprocessing or the User Interface 441 [[anchor15: Note in Draft -12. This entire section is likely to need 442 to be rewritten when we make final decisions about mapping.]] 444 The Unicode string MAY then be processed to prevent confounding of 445 user expectations. For instance, it might be reasonable, at this 446 step, to convert all upper case characters to lower case, if this 447 makes sense in the user's environment, but even this should be 448 approached with caution due to some edge cases: in the long term, it 449 is probably better for users to understand IDNs strictly in lower- 450 case, U-label, form. More generally, preprocessing may be useful to 451 smooth the transition from IDNA2003, especially for direct user 452 input, but with similar cautions. In general, IDNs appearing in 453 files and those transmitted across the network as part of protocols 454 are expected to be in either ASCII form (including A-labels) or to 455 contain U-labels, rather than being in forms requiring mapping or 456 other conversions. 458 Other examples of processing for localization might be applied, 459 especially to direct user input, at this point. They include 460 interpreting various characters as separating domain name components 461 from each other (label separators) because they either look like 462 periods or are used to separate sentences, mapping halfwidth or 463 fullwidth East Asian characters to the common form permitted in 464 labels, or giving special treatment to characters whose presentation 465 forms are dependent only on placement in the label. Such 466 localization changes are also outside the scope of this 467 specification. 469 Recommendations for preprocessing for global contexts (i.e., when 470 local considerations do not apply or cannot be used) and for maximum 471 interoperability with labels that might have been specified under 472 liberal readings of IDNA2003 are given in [IDNA2008-Rationale]. It 473 is important to note that the intent of these specifications is that 474 labels in application protocols, files, or links are intended to be 475 in U-label or A-label form. Preprocessing MUST NOT map a character 476 that is valid in a label as specified elsewhere in this document or 477 in [IDNA2008-Tables] into another character. Excessively liberal use 478 of preprocessing, especially to strings stored in files, poses a 479 threat to consistent and predictable behavior for the user even if 480 not to actual interoperability. 482 Because these transformations are local, it is important that domain 483 names that might be passed between systems (e.g., in IRIs) be 484 U-labels or A-labels and not forms that might be accepted locally as 485 a consequence of this step. This step is not standardized as part of 486 IDNA, and is not further specified here. 488 5.4. A-label Input 490 If the input to this procedure appears to be an A-label (i.e., it 491 starts in "xn--"), the lookup application MAY attempt to convert it 492 to a U-label and apply the tests of Section 5.5 and the conversion of 493 Section 5.6 to that form. If the label is converted to Unicode 494 (i.e., to U-label form) using the Punycode decoding algorithm, then 495 the processing specified in those two sections MUST be performed, and 496 the label MUST be rejected if the resulting label is not identical to 497 the original. See the Name Server Considerations section of 498 [IDNA2008-Rationale] for additional discussion on this topic. 500 That conversion and testing SHOULD be performed if the domain name 501 will later be presented to the user in native character form (this 502 requires that the lookup application be IDNA-aware). If those steps 503 are not performed, the lookup process SHOULD at least make tests to 504 determine that the string is actually an A-label, examining it for 505 the invalid formats specified in the Punycode decoding specification. 506 Applications that are not IDNA-aware will obviously omit that 507 testing; others MAY treat the string as opaque to avoid the 508 additional processing at the expense of providing less protection and 509 information to users. 511 5.5. Validation and Character List Testing 513 As with the registration procedure described in Section 4, the 514 Unicode string is checked to verify that all characters that appear 515 in it are valid as input to IDNA lookup processing. As discussed 516 above and in [IDNA2008-Rationale], the lookup check is more liberal 517 than the registration one. Putative labels with any of the following 518 characteristics MUST BE rejected prior to DNS lookup: 520 o Labels containing code points that are unassigned in the version 521 of Unicode being used by the application, i.e.,in the UNASSIGNED 522 category of [IDNA2008-Tables]. 524 o Labels that are not in NFC form as defined in [Unicode-UAX15]. 526 o Labels containing prohibited code points, i.e., those that are 527 assigned to the "DISALLOWED" category in the permitted character 528 table [IDNA2008-Tables]. 530 o Labels containing code points that are identified in 531 [IDNA2008-Tables] as "CONTEXTJ", i.e., requiring exceptional 532 contextual rule processing on lookup, but that do not conform to 533 that rule. Note that this implies that a rule much be defined, 534 not null: a character that requires a contextual rule but for 535 which the rule is null is treated in this step as having failed to 536 conform to the rule. 538 o Labels containing code points that are identified in 539 [IDNA2008-Tables] as "CONTEXTO", but for which no such rule 540 appears in the table of rules. Applications resolving DNS names 541 or carrying out equivalent operations are not required to test 542 contextual rules for "CONTEXTO" characters, only to verify that a 543 rule is defined (although they MAY make such tests to provide 544 better protection or give better information to the user). 546 o Labels whose first character is a combining mark (see 547 Section 4.2.3.2). 549 In addition, the application SHOULD apply the following test. The 550 test may be omitted in special circumstances, such as when the lookup 551 application knows that the conditions are enforced elsewhere, because 552 an attempt to look up and resolve such strings will almost certainly 553 lead to a DNS lookup failure except when wildcards are present in the 554 zone. However, applying the test is likely to give much better 555 information about the reason for a lookup failure -- information that 556 may be usefully passed to the user when that is feasible -- than DNS 557 resolution failure information alone. In any event, lookup 558 applications should avoid attempting to resolve labels that are 559 invalid under that test. 561 o Verification that the string is compliant with the requirements 562 for right to left characters, specified in [IDNA2008-BIDI]. 564 For all other strings, the lookup application MUST rely on the 565 presence or absence of labels in the DNS to determine the validity of 566 those labels and the validity of the characters they contain. If 567 they are registered, they are presumed to be valid; if they are not, 568 their possible validity is not relevant. While a lookup application 569 may reasonably issue warnings about strings it believes may be 570 problematic, applications that decline to process a string that 571 conforms to the rules above (i.e., does not look it up in the DNS) 572 are not in conformance with this protocol. 574 5.6. Punycode Conversion 576 The string that has now been validated for lookup is converted to ACE 577 form using the Punycode algorithm (with the ACE prefix added). With 578 the understanding that this summary is not normative (the steps above 579 are), the string is either 581 o in Unicode NFC form that contains no leading combining marks, 582 contains no DISALLOWED or UNASSIGNED code points, has rules 583 associated with any code points in CONTEXTJ or CONTEXTO, and, for 584 those in CONTEXTJ, to satisfies the conditions of the rules; or 586 o in A-label form, was supplied under circumstances in which the 587 U-label conversions and tests have not been performed (see 588 Section 5.4). 590 5.7. DNS Name Resolution 592 That resulting validated string is looked up in the DNS, using normal 593 DNS resolver procedures. That lookup can obviously either succeed 594 (returning information) or fail. 596 6. Security Considerations 598 Security Considerations for this version of IDNA, except for the 599 special issues associated with right to left scripts and characters, 600 are described in [IDNA2008-Defs]. Specific issues for labels 601 containing characters associated with scripts written right to left 602 appear in [IDNA2008-BIDI]. 604 7. IANA Considerations 606 IANA actions for this version of IDNA are specified in 607 [IDNA2008-Tables] and discussed informally in [IDNA2008-Rationale]. 608 The components of IDNA described in this document do not require any 609 IANA actions. 611 8. Contributors 613 While the listed editor held the pen, the original versions of this 614 document represent the joint work and conclusions of an ad hoc design 615 team consisting of the editor and, in alphabetic order, Harald 616 Alvestrand, Tina Dam, Patrik Faltstrom, and Cary Karp. This document 617 draws significantly on the original version of IDNA [RFC3490] both 618 conceptually and for specific text. This second-generation version 619 would not have been possible without the work that went into that 620 first version and its authors, Patrik Faltstrom, Paul Hoffman, and 621 Adam Costello. While Faltstrom was actively involved in the creation 622 of this version, Hoffman and Costello were not and should not be held 623 responsible for any errors or omissions. 625 9. Acknowledgments 627 This revision to IDNA would have been impossible without the 628 accumulated experience since RFC 3490 was published and resulting 629 comments and complaints of many people in the IETF, ICANN, and other 630 communities, too many people to list here. Nor would it have been 631 possible without RFC 3490 itself and the efforts of the Working Group 632 that defined it. Those people whose contributions are acknowledged 633 in RFC 3490, [RFC4690], and [IDNA2008-Rationale] were particularly 634 important. 636 Specific textual changes were incorporated into this document after 637 suggestions from the other contributors, Stephane Bortzmeyer, Vint 638 Cerf, Lisa Dusseault, Mark Davis, Paul Hoffman, Kent Karlsson, Erik 639 van der Poel, Marcos Sanz, Andrew Sullivan, Ken Whistler, and other 640 WG participants. Special thanks are due to Paul Hoffman for 641 permission to extract material from his Internet-Draft to form the 642 basis for Appendix B. 644 10. References 645 10.1. Normative References 647 [IDNA2008-BIDI] 648 Alvestrand, H. and C. Karp, "An updated IDNA criterion for 649 right-to-left scripts", July 2008, . 652 [IDNA2008-Defs] 653 Klensin, J., "Internationalized Domain Names for 654 Applications (IDNA): Definitions and Document Framework", 655 February 2009, . 658 [IDNA2008-Tables] 659 Faltstrom, P., "The Unicode Codepoints and IDNA", 660 July 2008, . 663 A version of this document is available in HTML format at 664 http://stupid.domain.name/idnabis/ 665 draft-ietf-idnabis-tables-02.html 667 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 668 STD 13, RFC 1034, November 1987. 670 [RFC1035] Mockapetris, P., "Domain names - implementation and 671 specification", STD 13, RFC 1035, November 1987. 673 [RFC1123] Braden, R., "Requirements for Internet Hosts - Application 674 and Support", STD 3, RFC 1123, October 1989. 676 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 677 Requirement Levels", BCP 14, RFC 2119, March 1997. 679 [RFC3492] Costello, A., "Punycode: A Bootstring encoding of Unicode 680 for Internationalized Domain Names in Applications 681 (IDNA)", RFC 3492, March 2003. 683 [Unicode-PropertyValueAliases] 684 The Unicode Consortium, "Unicode Character Database: 685 PropertyValueAliases", March 2008, . 688 [Unicode-RegEx] 689 The Unicode Consortium, "Unicode Technical Standard #18: 690 Unicode Regular Expressions", May 2005, 691 . 693 [Unicode-Scripts] 694 The Unicode Consortium, "Unicode Standard Annex #24: 695 Unicode Script Property", February 2008, 696 . 698 [Unicode-UAX15] 699 The Unicode Consortium, "Unicode Standard Annex #15: 700 Unicode Normalization Forms", 2006, 701 . 703 10.2. Informative References 705 [ASCII] American National Standards Institute (formerly United 706 States of America Standards Institute), "USA Code for 707 Information Interchange", ANSI X3.4-1968, 1968. 709 ANSI X3.4-1968 has been replaced by newer versions with 710 slight modifications, but the 1968 version remains 711 definitive for the Internet. 713 [IDNA2008-Rationale] 714 Klensin, J., Ed., "Internationalizing Domain Names for 715 Applications (IDNA): Issues, Explanation, and Rationale", 716 February 2009, . 719 [RFC2136] Vixie, P., Thomson, S., Rekhter, Y., and J. Bound, 720 "Dynamic Updates in the Domain Name System (DNS UPDATE)", 721 RFC 2136, April 1997. 723 [RFC2181] Elz, R. and R. Bush, "Clarifications to the DNS 724 Specification", RFC 2181, July 1997. 726 [RFC2535] Eastlake, D., "Domain Name System Security Extensions", 727 RFC 2535, March 1999. 729 [RFC2671] Vixie, P., "Extension Mechanisms for DNS (EDNS0)", 730 RFC 2671, August 1999. 732 [RFC3490] Faltstrom, P., Hoffman, P., and A. Costello, 733 "Internationalizing Domain Names in Applications (IDNA)", 734 RFC 3490, March 2003. 736 [RFC3491] Hoffman, P. and M. Blanchet, "Nameprep: A Stringprep 737 Profile for Internationalized Domain Names (IDN)", 738 RFC 3491, March 2003. 740 [RFC3986] Berners-Lee, T., Fielding, R., and L. Masinter, "Uniform 741 Resource Identifier (URI): Generic Syntax", STD 66, 742 RFC 3986, January 2005. 744 [RFC3987] Duerst, M. and M. Suignard, "Internationalized Resource 745 Identifiers (IRIs)", RFC 3987, January 2005. 747 [RFC4690] Klensin, J., Faltstrom, P., Karp, C., and IAB, "Review and 748 Recommendations for Internationalized Domain Names 749 (IDNs)", RFC 4690, September 2006. 751 [RFC4952] Klensin, J. and Y. Ko, "Overview and Framework for 752 Internationalized Email", RFC 4952, July 2007. 754 [Unicode] The Unicode Consortium, "The Unicode Standard, Version 755 5.0", 2007. 757 Boston, MA, USA: Addison-Wesley. ISBN 0-321-48091-0 759 Appendix A. Local Mapping Alternatives 761 The subsections of this appendix are temporary and represent 762 different sketches of possible replacements for Section 5.3. They do 763 not represent an assertion of WG consensus or any assertion about the 764 possibility of including one of them as part of the WG's work 765 program. Instead, they are supplied only for purposes of comparison, 766 discussion, and, should it be relevant, refinement. 768 The first paragraph of each subsection describes how the material 769 would be placed relative to the existing main document text. 770 Subsequent paragraphs are the actual suggestions, although in 771 incomplete sketch form. 773 A.1. Transitional Mapping Model 775 If this subsection were adopted, Section 5.3 would be deleted and 776 this one would be inserted after, or integrated with, Section 5.7. 778 This specification does not support the extensive mappings from one 779 character to another, including Unicode Case Folding and 780 Compatibility Character mapping, of IDNA2003. It also changes the 781 interpretations of a small number of characters relative to IDNA2003. 782 Most applications, especially those with which IDNs have been used 783 for some time, will need to maintain reasonable compatibility with 784 files created under IDNA2003 and user interfaces designed for it. 785 This section specifies additional steps to be taken to provide 786 maximum IDNA2003 compatibility. 788 If an application requires IDNA2003 backward compatibility, it MUST 789 execute the steps in one of the two subsections that immediately 790 follow. 792 A.1.1. Fallback Lookup 794 If the string validates and the resolution attempt in Section 5.7 795 successfully returns a result, the lookup process terminates with 796 that result. If validation succeeds but resolution fails, the 797 validated string is proceeded through the ToASCII operation specified 798 in IDNA2003 [RFC3490]. Assuming it produces a valid result, the 799 resulting string is compared to the previous validated one. If they 800 are not identical, a resolution attempt is made with the ToASCII 801 output and the result of that attempt is returned as the result of 802 the lookup operation. 804 Should IDNA2008 validation fail, the string is processed through 805 ToASCII and, assuming the result is valid, the resulting string is 806 resolved and the result of that attempt returned as the result of the 807 lookup operation. 809 If ToASCII (IDNA2003) conversion is attempted and fails, the lookup 810 operation behaves as if no name was found in the DNS. 812 Note that this procedure involves, at most, one DNS lookup 813 (resolution attempt). If IDNA2008 string validation, conversion, and 814 resolution succeed, no attempt is made to use IDNA2003 mechanisms. 815 The procedure does, however, require that lookup applications fully 816 support both IDNA2008 and IDNA2003 lookup operations so that the 817 fallback can occur. 819 A.1.2. Two-step Lookup 821 Prior to the resolution attempt in Section 5.7, ACE strings are 822 computed using both IDNA2003 (ToASCII) and IDNA2008 methods (as 823 specified here). Assuming both validate, those strings are compared. 824 If they are identical, or only one was valid, then a single DNS 825 resolution is performed and its result is the result of the lookup 826 operation. If both are valid but they are not identical, one 827 resolution attempt is made with each of the two ACE strings. 829 If neither string is valid as an IDN, then the lookup operation 830 fails. 832 When two resolutions are attempted, if one of the two is successful 833 and the other is not, the successful value is used as the result of 834 the lookup. If both are successful, the user or calling application 835 must be presented with a choice in some way. 837 This procedure will require two DNS lookups (resolution attempts) in 838 all cases except those in which the label string fails IDNA2008 839 validation, neither IDNA2003 or IDNA2008 can validate the string and 840 translate it to ACE form, or the strings obtained from the two 841 conversions are identical. As with the prior option, IDNA 842 implementations will need to support both the IDNA2003 algorithm and 843 tables and the IDNA2008 one. The question of how multiple results 844 from different interpretations of the same input string should be 845 handled by applications is a difficult one, with potential false 846 positive and security attack vector implications as well as the 847 possibility of general confusion. 849 In particular, if both interpretations of the name return values, the 850 lookup application has no practical way to tell whether the relevant 851 registry has applied "variant" or "bundling" techniques to ensure 852 that both domain names are under the same control or not. From that 853 perspective, the approach in the previous subsection assumes that has 854 been done (if the IDNA2003-interpretation label is present at all) 855 while this one assumes that such bundling is unlikely to have 856 occurred. 858 [[anchor25: Note in Draft: If this appendix is used, RFC3490 must be 859 moved from Informative to Normative.]] 861 A.2. Internationalized Resource Identifier (IRI) Mapping Model 863 This subsection is intended to be descriptive of an approach that 864 lies outside IDNA, rather than a normative component of it. If it 865 were adopted, Section 5.3 would be deleted and the material below 866 would be referenced, either as a non-normative Appendix in Protocol 867 or, more reasonably, as a section of Rationale. 869 IDNA2003 supported extensive mappings from one character to another, 870 including Unicode Case Folding and Compatibility Character mapping. 871 Those mappings are no longer supported on registration and are 872 inconsistent with the "exact match" lookups that people expect from 873 the DNS. Some mapping should still be supported, both for 874 compatibility with applications that assume IDNA2003 and to avoid 875 confounding user expectations. The specific mappings involved are 876 not part of IDNA, but are expected to be specified as part of a 877 revision to the IRI specification [RFC3987] and the conversion from 878 IRI form to URI form. That change leaves mapping unspecified and 879 prohibited for actual domain names, however, in practice, most domain 880 names, especially in the web applications that appear to have been 881 most important for IDNs between the publication of IDNA2003 and the 882 release of this specification, are not interpreted as themselves but 883 as abbreviated form of URIs or IRIs and hence subject to the 884 transformation rules of the latter. 886 Appendix B. Summary of Major Changes from IDNA2003 888 1. Update base character set from Unicode 3.2 to Unicode version- 889 agnostic. 891 2. Separate the definitions for the "registration" and "lookup" 892 activities. 894 3. Disallow symbol and punctuation characters except where special 895 exceptions are necessary. 897 4. Remove the mapping and normalization steps from the protocol and 898 have them instead done by the applications themselves, possibly 899 in a local fashion, before invoking the protocol. 901 5. Change the way that the protocol specifies which characters are 902 allowed in labels from "humans decide what the table of 903 codepoints contains" to "decision about codepoints are based on 904 Unicode properties plus a small exclusion list created by 905 humans". 907 6. Introduce the new concept of characters that can be used only in 908 specific contexts. 910 7. Allow typical words and names in languages such as Dhivehi and 911 Yiddish to be expressed. 913 8. Make bidirectional domain names (delimited strings of labels, 914 not just labels standing on their own) display in a less 915 surprising fashion whether they appear in obvious domain name 916 contexts or as part of running text in paragraphs. 918 9. Remove the dot separator from the mandatory part of the 919 protocol. 921 10. Make some currently-valid labels that are not actually IDNA 922 labels invalid. 924 Appendix C. Change Log 926 [[anchor28: RFC Editor: Please remove this appendix.]] 928 C.1. Changes between Version -00 and -01 of draft-ietf-idnabis-protocol 930 o Corrected discussion of SRV records. 932 o Several small corrections for clarity. 934 o Inserted more "open issue" placeholders. 936 C.2. Version -02 938 o Rewrote the "conversion to Unicode" text in Section 5.2 as 939 requested on-list. 941 o Added a comment (and reference) about EDNS0 to the "DNS Server 942 Conventions" section, which was also retitled. 944 o Made several editorial corrections and improvements in response to 945 various comments. 947 o Added several new discussion placeholder anchors and updated some 948 older ones. 950 C.3. Version -03 952 o Trimmed change log, removing information about pre-WG drafts. 954 o Incorporated a number of changes suggested by Marcos Sanz in his 955 note of 2008.07.17 and added several more placeholder anchors. 957 o Several minor editorial corrections and improvements. 959 o "Editor" designation temporarily removed because the automatic 960 posting machinery does not accept it. 962 C.4. Version -04 964 o Removed Contextual Rule appendices for transfer to Tables. 966 o Several changes, including removal of discussion anchors, based on 967 discussions at IETF 72 (Dublin) 969 o Rewrote the preprocessing material (Section 5.3) somewhat. 971 C.5. Version -05 973 o Updated part of the A-label input explanation (Section 5.4) per 974 note from Erik van der Poel. 976 C.6. Version -06 978 o Corrected a few typographical errors. 980 o Incorporated the material (formerly in Rationale) on the 981 relationship between IDNA2003 and IDNA2008 as an appendix and 982 pointed to the new definitions document. 984 o Text modified in several places to recognize the dangers of 985 interaction between DNS wildcards and IDNs. 987 o Text added to be explicit about the handling of edge and failure 988 cases in Punycode encoding and decoding. 990 o Revised for consistency with the new Definitions document and to 991 make the text read more smoothly. 993 C.7. Version -07 995 o Multiple small textual and editorial changes and clarifications. 997 o Requirement for normalization clarified to apply to all cases and 998 conditions for preprocessing further clarified. 1000 o Substantive change to Section 4.2.1, turning a SHOULD to a MUST 1001 (see note from Mark Davis, 19 November, 2008 18:14 -0800). 1003 C.8. Version -08 1005 o Added some references and altered text to improve clarity. 1007 o Changed the description of CONTEXTJ/CONTEXTO to conform to that in 1008 Tables. In other words, these are now treated as distinction 1009 categories (again), rather than as specially-flagged subsets of 1010 PROTOCOL VALID. 1012 o The discussion of label comparisons has been rewritten to make it 1013 more precise and to clarify that one does not need to verify that 1014 a string is a [valid] A-label or U-label in order to test it for 1015 equality with another string. The WG should verify that the 1016 current text is what is desired. 1018 o Other changes to reflect post-IETF discussions or editorial 1019 improvements. 1021 C.9. Version -09 1023 o Removed Security Considerations material to Defs document. 1025 o Removed the Name Server Considerations material to Rationale. 1026 That material is not normative and not needed to implement the 1027 protocol itself. 1029 o Adjusted terminology to match new version of Defs. 1031 o Removed all discussion of local mapping and option for it from 1032 registration protocol. Such mapping is now completely prohibited 1033 on Registration. 1035 o Removed some old placeholders and inquiries because no comments 1036 have been received. 1038 o Small editorial corrections. 1040 C.10. Version -10 1042 o Rewrote the registration input material slightly to further 1043 clarify the "no mapping on registration" principle. 1045 o Added placeholder notes about several tasks, notably reorganizing 1046 Section 4 and Section 5 so that subsection numbers are parallel. 1048 o Cleaned up an incorrect use of the terms "A-label" and "U-label" 1049 in the lookup phase that was spotted by Mark Davis. Inserted a 1050 note there about alternate ways to deal with the resulting 1051 terminology problem. 1053 o Added a temporarily appendix (above) to document alternate 1054 strategies for possible replacements for Section 5.3. 1056 C.11. Version -11 1058 o Removed dangling reference to "C-label" (editing error in prior 1059 draft). 1061 o Recast the last steps of the Lookup description to eliminate 1062 "apparent" (previously "putative") terminology. 1064 o Rewrote major portions of the temporary appendix that describes 1065 transitional mappings to improve clarity and add context. 1067 o Did some fine-tuning of terminology, notably in Section 3.2.1. 1069 C.12. Version -12 1071 o Extensive editorial improvements, mostly due to suggestions from 1072 Lisa Dusseault. 1074 o Conformance statements have been made consistent, especially in 1075 Section 4.2.1 and subsequent text, which said "SHOULD" in one 1076 place and then said "MAY" as the result of incomplete removal of 1077 registration-time mapping. Also clarified the definition of 1078 "registration processes" in Section 4.1 -- the previous text had 1079 confused several people. 1081 o A few new "question to the WG notes have been added about 1082 appropriateness or placement of text. If there are no comments on 1083 the mailing list, the editor will apply his own judgment. 1085 o Several of the usual small typos and other editorial errors have 1086 been corrected. 1088 o Section 5 has still not been reorganized to match Section 4 in 1089 structure and subsection numbering -- will be done as soon as the 1090 mapping decisions and references are final. 1092 Author's Address 1094 John C Klensin 1095 1770 Massachusetts Ave, Ste 322 1096 Cambridge, MA 02140 1097 USA 1099 Phone: +1 617 245 1457 1100 Email: john+ietf@jck.com