idnits 2.17.1 draft-ietf-imapext-thread-09.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** Looks like you're using RFC 2026 boilerplate. This must be updated to follow RFC 3978/3979, as updated by RFC 4748. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- ** Missing expiration date. The document expiration date should appear on the first and last page. ** The document seems to lack a 1id_guidelines paragraph about the list of Shadow Directories. == No 'Intended status' indicated for this document; assuming Proposed Standard == It seems as if not all pages are separated by form feeds - found 0 form feeds but 14 pages Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack an Introduction section. ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) ** The abstract seems to contain references ([RFC-2822], [ABNF], [NEWS], [RFC2822]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. ** The document seems to lack a both a reference to RFC 2119 and the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords. RFC 2119 keyword, line 81: '...onnected clients MUST use exactly this...' RFC 2119 keyword, line 166: '...eading algorithm MUST normalize any ms...' RFC 2119 keyword, line 174: '... MUST be interpreted as bei...' RFC 2119 keyword, line 253: '... MUST be kept consistent...' RFC 2119 keyword, line 495: '...plementations of THREAD MUST implement...' Miscellaneous warnings: ---------------------------------------------------------------------------- -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (March 2002) is 8078 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) -- Missing reference section? 'RFC 2822' on line 189 looks like a reference -- Missing reference section? 'ABNF' on line 548 looks like a reference -- Missing reference section? 'NEWS' on line 551 looks like a reference -- Missing reference section? 'RFC-2822' on line 555 looks like a reference Summary: 7 errors (**), 0 flaws (~~), 2 warnings (==), 6 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 IMAP Extensions Working Group M. Crispin 3 Internet Draft: IMAP THREAD K. Murchison 4 Document: internet-drafts/draft-ietf-imapext-thread-09.txt March 2002 6 INTERNET MESSAGE ACCESS PROTOCOL - THREAD EXTENSION 8 Status of this Memo 10 This document is an Internet-Draft and is in full conformance with 11 all provisions of Section 10 of RFC 2026. 13 Internet-Drafts are working documents of the Internet Engineering 14 Task Force (IETF), its areas, and its working groups. Note that 15 other groups may also distribute working documents as Internet- 16 Drafts. 18 Internet-Drafts are draft documents valid for a maximum of six months 19 and may be updated, replaced, or obsoleted by other documents at any 20 time. It is inappropriate to use Internet-Drafts as reference 21 material or to cite them other than as "work in progress." 23 The list of current Internet-Drafts can be accessed at 24 http://www.ietf.org/ietf/1id-abstracts.txt 26 To view the list Internet-Draft Shadow Directories, see 27 http://www.ietf.org/shadow.html. 29 A revised version of this draft document will be submitted to the RFC 30 editor as a Proposed Standard for the Internet Community. Discussion 31 and suggestions for improvement are requested, and should be sent to 32 ietf-imapext@IMC.ORG. This document will expire before 21 August 33 2002. Distribution of this memo is unlimited. 35 Abstract 37 This document describes the server-based threading extension to the 38 IMAP4rev1 protocol. This extension provides substantial performance 39 improvements for IMAP clients which offer threaded views. 41 A server which supports this extension indicates this with one or 42 more capability names consisting of "THREAD=" followed by a supported 43 threading algorithm name as described in this document. This 44 provides for future upwards-compatible extensions. 46 Extracted Subject Text 48 Threading uses a version of the subject which has specific subject 49 artifacts of deployed Internet mail software removed. Due to the 50 complexity of these artifacts, the formal syntax for the subject 51 extraction rules is ambiguous. The following procedure is followed 52 to determine the actual "base subject" which is used to thread: 54 (1) Convert any RFC 2047 encoded-words in the subject to 55 UTF-8. Convert all tabs and continuations to space. 56 Convert all multiple spaces to a single space. 58 (2) Remove all trailing text of the subject that matches 59 the subj-trailer ABNF, repeat until no more matches are 60 possible. 62 (3) Remove all prefix text of the subject that matches the 63 subj-leader ABNF. 65 (4) If there is prefix text of the subject that matches the 66 subj-blob ABNF, and removing that prefix leaves a non-empty 67 subj-base, then remove the prefix text. 69 (5) Repeat (3) and (4) until no matches remain. 71 Note: it is possible to defer step (2) until step (6), 72 but this requires checking for subj-trailer in step (4). 74 (6) If the resulting text begins with the subj-fwd-hdr ABNF 75 and ends with the subj-fwd-trl ABNF, remove the 76 subj-fwd-hdr and subj-fwd-trl and repeat from step (2). 78 (7) The resulting text is the "base subject" used in 79 threading. 81 All servers and disconnected clients MUST use exactly this algorithm 82 when threading. Otherwise there is potential for a user to get 83 inconsistent results based on whether they are running in connected 84 or disconnected IMAP mode. 86 Sent Date 88 As used in this document, the term "sent date" refers to the date and 89 time from the Date: header, adjusted by time zone. This differs from 90 date-related criteria in SEARCH, which use just the date and not the 91 time, nor adjusts by time zone. 93 Additional Commands 95 This command is an extension to the IMAP4rev1 base protocol. 97 The section header is intended to correspond with where it would be 98 located in the main document if it was part of the base 99 specification. 101 6.3.THREAD. THREAD Command 103 Arguments: threading algorithm 104 charset specification 105 searching criteria (one or more) 107 Data: untagged responses: THREAD 109 Result: OK - thread completed 110 NO - thread error: can't thread that charset or 111 criteria 112 BAD - command unknown or arguments invalid 114 The THREAD command is a variant of SEARCH with threading semantics 115 for the results. Thread has two arguments before the searching 116 criteria argument; a threading algorithm, and the searching 117 charset. Note that unlike SEARCH, the searching charset argument 118 is mandatory. 120 There is also a UID THREAD command which corresponds to THREAD the 121 way that UID SEARCH corresponds to SEARCH. 123 The THREAD command first searches the mailbox for messages that 124 match the given searching criteria using the charset argument for 125 the interpretation of strings in the searching criteria. It then 126 returns the matching messages in an untagged THREAD response, 127 threaded according to the specified threading algorithm. 129 Sorting is in ascending order. Earlier dates sort before later 130 dates; smaller sizes sort before larger sizes; and strings are 131 sorted according to ascending values established by their 132 collation algorithm (see under "Internationalization 133 Considerations"). 135 The defined threading algorithms are as follows: 137 ORDEREDSUBJECT 138 The ORDEREDSUBJECT threading algorithm is also referred to as 139 "poor man's threading." The searched messages are sorted by 140 subject and then by the sent date. The messages are then split 141 into separate threads, with each thread containing messages 142 with the same extracted subject text. Finally, the threads are 143 sorted by the sent date of the first message in the thread. 145 Note that each message in a thread is a child (as opposed to a 146 sibling) of the previous message. 148 REFERENCES 149 The REFERENCES threading algorithm is based on the algorithm 150 written by Jamie Zawinski which was used in "Netscape Mail and 151 News" versions 2.0 through 3.0. For details, see 152 http://www.jwz.org/doc/threading.html. 154 This algorithm threads the searched messages by grouping them 155 together in parent/child relationships based on which messages 156 are replies to others. The parent/child relationships are 157 built using two methods: reconstructing a message's ancestry 158 using the references contained within it; and checking the 159 subject of a message to see if it is a reply to (or forward of) 160 another. 162 Note: "Message ID" in the following description refers to a 163 normalized form of the msg-id in [RFC 2822]. The actual 164 text in an RFC 2822 may use quoting, resulting in multiple 165 ways of expressing the same Message ID. Implementations of 166 the REFERENCES threading algorithm MUST normalize any msg-id 167 in order to avoid false non-matches due to differences in 168 quoting. 170 For example, the msg-id 171 <"01KF8JCEOCBS0045PS"@xxx.yyy.com> 172 and the msg-id 173 <01KF8JCEOCBS0045PS@xxx.yyy.com> 174 MUST be interpreted as being the same Message ID. 176 The references used for reconstructing a message's ancestry are 177 found using the following rules: 179 If a message contains a [NEWS]-style References header line, 180 then use the Message IDs in the References header line as 181 the references. 183 If a message does not contain a References header line, or 184 the References header line does not contain any valid 185 Message IDs, then use the first (if any) valid Message ID 186 found in the In-Reply-To header line as the only reference 187 (parent) for this message. 189 Note: Although [RFC 2822] permits multiple Message IDs in 190 the In-Reply-To header, in actual practice this 191 discipline has not been followed. For example, 192 In-Reply-To headers have been observed with email 193 addresses after the Message ID, and there are no good 194 heuristics for software to determine the difference. 195 This is not a problem with the References header however. 197 If a message does not contain an In-Reply-To header line, or 198 the In-Reply-To header line does not contain a valid Message 199 ID, then the message does not have any references (NIL). 201 The REFERENCES algorithm is significantly more complex than 202 ORDEREDSUBJECT and consists of six main steps. These steps are 203 outlined in detail below. 205 (1) For each searched message: 207 (A) Using the Message IDs in the message's references, link 208 the corresponding messages (those whose Message-ID header 209 line contains the given reference Message ID) together as 210 parent/child. Make the first reference the parent of the 211 second (and the second a child of the first), the second the 212 parent of the third (and the third a child of the second), 213 etc. The following rules govern the creation of these 214 links: 216 If a message does not contain a Message-ID header line, 217 or the Message-ID header line does not contain a valid 218 Message ID, then assign a unique Message ID to this 219 message. 221 If two or more messages have the same Message ID, assign 222 a unique Message ID to each of the duplicates. 224 If no message can be found with a given Message ID, 225 create a dummy message with this ID. Use this dummy 226 message for all subsequent references to this ID. 228 If a message already has a parent, don't change the 229 existing link. This is done because the References 230 header line may have been truncated by a MUA. As a 231 result, there is no guarantee that the messages 232 corresponding to adjacent Message IDs in the References 233 header line are parent and child. 235 Do not create a parent/child link if creating that link 236 would introduce a loop. For example, before making 237 message A the parent of B, make sure that A is not a 238 descendent of B. 240 Note: Message ID comparisons are case-sensitive. 242 (B) Create a parent/child link between the last reference 243 (or NIL if there are no references) and the current message. 244 If the current message already has a parent, it is probably 245 the result of a truncated References header line, so break 246 the current parent/child link before creating the new 247 correct one. As in step 1.A, do not create the parent/child 248 link if creating that link would introduce a loop. Note 249 that if this message has no references, that it will now 250 have no parent. 252 Note: The parent/child links created in steps 1.A and 1.B 253 MUST be kept consistent with one another at ALL times. 255 (2) Gather together all of the messages that have no parents 256 and make them all children (siblings of one another) of a dummy 257 parent (the "root"). These messages constitute the first 258 (head) message of the threads created thus far. 260 (3) Prune dummy messages from the thread tree. Traverse each 261 thread under the root, and for each message: 263 If it is a dummy message with NO children, delete it. 265 If it is a dummy message with children, delete it, but 266 promote its children to the current level. In other words, 267 splice them in with the dummy's siblings. 269 Do not promote the children if doing so would make them 270 children of the root, unless there is only one child. 272 (4) Sort the messages under the root (top-level siblings only) 273 by sent date. In the case of an exact match on sent date or if 274 either of the Date: headers used in a comparison can not be 275 parsed, use the order in which the messages appear in the 276 mailbox (that is, by sequence number) to determine the order. 277 In the case of a dummy message, sort its children by sent date 278 and then use the first child for the top-level sort. 280 (5) Gather together messages under the root that have the same 281 extracted subject text. 283 (A) Create a table for associating extracted subjects with 284 messages. 286 (B) Populate the subject table with one message per 287 extracted subject. For each child of the root: 289 (i) Find the subject of this thread by extracting the 290 base subject from the current message, or its first child 291 if the current message is a dummy. 293 (ii) If the extracted subject is empty, skip this 294 message. 296 (iii) Lookup the message associated with this extracted 297 subject in the table. 299 (iv) If there is no message in the table with this 300 subject, add the current message and the extracted 301 subject to the subject table. 303 Otherwise, replace the message in the table with the 304 current message if the message in the table is not a 305 dummy AND either of the following criteria are true: 307 The current message is a dummy, OR 309 The message in the table is a reply or forward (its 310 original subject contains a subj-refwd part and/or a 311 "(fwd)" subj-trailer) and the current message is not. 313 (C) Merge threads with the same subject. For each child of 314 the root: 316 (i) Find the subject of this thread as in step 4.B.i 317 above. 319 (ii) If the extracted subject is empty, skip this 320 message. 322 (iii) Lookup the message associated with this extracted 323 subject in the table. 325 (iv) If the message in the table is the current message, 326 skip this message. 328 Otherwise, merge the current message with the one in the 329 table using the following rules: 331 If both messages are dummies, append the current 332 message's children to the children of the message in 333 the table (the children of both messages become 334 siblings), and then delete the current message. 336 If the message in the table is a dummy and the current 337 message is not, make the current message a child of 338 the message in the table (a sibling of it's children). 340 If the current message is a reply or forward and the 341 message in the table is not, make the current message 342 a child of the message in the table (a sibling of it's 343 children). 345 Otherwise, create a new dummy message and make both 346 the current message and the message in the table 347 children of the dummy. Then replace the message in 348 the table with the dummy message. 350 Note: Subject comparisons are case-insensitive, as 351 described under "Internationalization 352 Considerations." 354 (6) Traverse the messages under the root and sort each set of 355 siblings by sent date. Traverse the messages in such a way 356 that the "youngest" set of siblings are sorted first, and the 357 "oldest" set of siblings are sorted last (grandchildren are 358 sorted before children, etc). In the case of an exact match on 359 sent date or if either of the Date: headers used in a 360 comparison can not be parsed, use the order in which the 361 messages appear in the mailbox (that is, by sequence number) to 362 determine the order. In the case of a dummy message (which can 363 only occur with top-level siblings), use its first child for 364 sorting. 366 Example: C: A283 THREAD ORDEREDSUBJECT UTF-8 SINCE 5-MAR-2000 367 S: * THREAD (166)(167)(168)(169)(172)(170)(171) 368 (173)(174 175 176 178 181 180)(179)(177 183 369 182 188 184 185 186 187 189)(190)(191)(192) 370 (193)(194 195)(196 197 198)(199)(200 202)(201) 371 (203)(204)(205)(206 207)(208) 372 S: A283 OK THREAD completed 373 C: A284 THREAD ORDEREDSUBJECT US-ASCII TEXT "gewp" 374 S: * THREAD 375 S: A284 OK THREAD completed 376 C: A285 THREAD REFERENCES UTF-8 SINCE 5-MAR-2000 377 S: * THREAD (166)(167)(168)(169)(172)((170)(179)) 378 (171)(173)((174)(175)(176)(178)(181)(180)) 379 ((177)(183)(182)(188 (184)(189))(185 186)(187)) 380 (190)(191)(192)(193)((194)(195 196))(197 198) 381 (199)(200 202)(201)(203)(204)(205 206 207)(208) 382 S: A285 OK THREAD completed 384 Note: The line breaks in the first and third client 385 responses are for editorial clarity and do not appear in 386 real THREAD responses. 388 Additional Responses 390 This response is an extension to the IMAP4rev1 base protocol. 392 The section heading of this response is intended to correspond with 393 where it would be located in the main document. 395 7.2.THREAD. THREAD Response 397 Data: zero or more threads 399 The THREAD response occurs as a result of a THREAD or UID THREAD 400 command. It contains zero or more threads. A thread consists of 401 a parenthesized list of thread members. 403 Thread members consist of zero or more message numbers, delimited 404 by spaces, indicating successive parent and child. This continues 405 until the thread splits into multiple sub-threads, at which point 406 the thread nests into multiple sub-threads with the first member 407 of each subthread being siblings at this level. There is no limit 408 to the nesting of threads. 410 The messages numbers refer to those messages that match the search 411 criteria. For THREAD, these are message sequence numbers; for UID 412 THREAD, these are unique identifiers. 414 Example: S: * THREAD (2)(3 6 (4 23)(44 7 96)) 416 The first thread consists only of message 2. The second thread 417 consists of the messages 3 (parent) and 6 (child), after which it 418 splits into two subthreads; the first of which contains messages 4 419 (child of 6, sibling of 44) and 23 (child of 4), and the second of 420 which contains messages 44 (child of 6, sibling of 4), 7 (child of 421 44), and 96 (child of 7). Since some later messages are parents 422 of earlier messages, the messages were probably moved from some 423 other mailbox at different times. 425 -- 2 427 -- 3 428 \-- 6 429 |-- 4 430 | \-- 23 431 | 432 \-- 44 433 \-- 7 434 \-- 96 436 Example: S: * THREAD ((3)(5)) 438 In this example, 3 and 5 are siblings of a parent which does not 439 match the search criteria (and/or does not exist in the mailbox); 440 however they are members of the same thread. 442 Formal Syntax of THREAD commands and Responses 444 thread-data = "THREAD" [SP 1*thread-list] 446 thread-list = "(" thread-members / thread-nested ")" 448 thread-members = nz-number *(SP nz-number) [SP thread-nested] 450 thread-nested = 2*thread-list 452 thread = ["UID" SP] "THREAD" SP thread-algorithm 453 SP search-charset 1*(SP search-key) 455 thread-algorithm = "ORDEREDSUBJECT" / "REFERENCES" / atom 457 The following syntax describes subject extraction rules (2)-(6): 459 subject = *subj-leader [subj-middle] *subj-trailer 461 subj-refwd = ("re" / ("fw" ["d"])) *WSP [subj-blob] ":" 463 subj-blob = "[" *BLOBCHAR "]" *WSP 465 subj-fwd = subj-fwd-hdr subject subj-fwd-trl 467 subj-fwd-hdr = "[fwd:" 469 subj-fwd-trl = "]" 471 subj-leader = (*subj-blob subj-refwd) / WSP 473 subj-middle = *subj-blob (subj-base / subj-fwd) 474 ; last subj-blob is subj-base if subj-base would 475 ; otherwise be empty 477 subj-trailer = "(fwd)" / WSP 479 subj-base = NONWSP *([*WSP] NONWSP) 480 ; can be a subj-blob 482 BLOBCHAR = %x01-5a / %x5c / %x5e-7f 483 ; any CHAR except '[' and ']' 485 NONWSP = %x01-08 / %x0a-1f / %x21-7f 486 ; any CHAR other than WSP 488 Security Considerations 490 Security issues are not discussed in this memo. 492 Internationalization Considerations 494 By default, strings are threaded according to the "minimum sorting 495 collation algorithm". All implementations of THREAD MUST implement 496 the minimum sorting collation algorithm. 498 In the minimum sorting collation algorithm, the Basic Latin 499 alphabetics (U+0041 to U+005A uppercase, U+0061 to U+007A lowercase) 500 are sorted in a case-insensitive fashion; that is, "A" (U+0041) and 501 "a" (U+0061) are treated as exact equals. The characters U+005B to 502 U+0060 are sorted after the Basic Latin alphabetics; for example, 503 U+005E is sorted after U+005A and U+007A. All other characters are 504 sorted according to their octet values, as expressed in UTF-8. No 505 attempt is made to treat composed characters specially, or to do 506 case-insensitive comparisons of composed characters. 508 Note: this means, among other things, that the composed 509 characters in the Latin-1 Supplement are not compared in 510 what would be considered an ISO 8859-1 "case-insensitive" 511 fashion. Case comparison rules for characters with 512 diacriticals differ between languages; the minimum sorting 513 collation does not attempt to deal with this at all. This 514 is reserved for other sorting collations, which may be 515 language-specific. 517 ;;; *** ITEM FOR DISCUSSION *** 518 ;;; THERE IS SOME CONCERN THAT THIS MINIMUM COLLATION IS TOO MINIMAL, 519 ;;; AND THAT THE "GENERIC UNICODE SORTING COLLATION" DISCUSSED BELOW 520 ;;; NEEDS TO BE THE MINIMUM. ONE SUGGESTION IS UNICODE TECHNICAL 521 ;;; STANDARD 10 (TR-10). IF THIS IS THE MINIMUM, THAT REQUIRES THAT 522 ;;; ALL IMPLEMENTATIONS OF SORT AND THREAD BE UNICODE-SAVVY AT LEAST 523 ;;; TO THE POINT OF IMPLEMENTATION TR-10. IS THIS REALISTIC? DOES 524 ;;; THIS RAISE EXCESSIVE IMPLEMENTATION BARRIERS? 525 Other sorting collations, and the ability to change the sorting 526 collation, will be defined in a separate document dealing with IMAP 527 internationalization. 529 It is anticipated that there will be a generic Unicode sorting 530 collation, which will provide generic case-insensitivity for 531 alphabetic scripts, specification of composed character handling, and 532 language-specific sorting collations. A server which implements 533 non-default sorting collations will modify its sorting behavior 534 according to the selected sorting collation. 536 Non-English translations of "Re" or "Fw"/"Fwd" are not specified for 537 removal in the extracted subject text process. By specifying that 538 only the English forms of the prefixes are used, it becomes a simple 539 display time task to localize the prefix language for the user. If, 540 on the other hand, prefixes in multiple languages are permitted, the 541 result is a geometrically complex, and ultimately unimplementable, 542 task. In order to improve the ability to support non-English display 543 in Internet mail clients, only the English form of these prefixes 544 should be transmitted in Internet mail messages. 546 A. References 548 [ABNF] Crocker, D., and Overell, P. "Augmented BNF for Syntax 549 Specifications: ABNF", RFC 2234, November 1997. 551 [NEWS] Horton, M., and Adams, R., "Standard for interchange of USENET 552 messages", RFC-1036, AT&T Bell Laboratories and Center for Seismic 553 Studies, December, 1987. 555 [RFC-2822] Resnick, P. "Internet Message Format", RFC 2822, April 556 2001. 558 Author's Address 560 Mark R. Crispin 561 Networks and Distributed Computing 562 University of Washington 563 4545 15th Avenue NE 564 Seattle, WA 98105-4527 566 Phone: (206) 543-5762 568 EMail: MRC@CAC.Washington.EDU 570 Kenneth Murchison 571 Oceana Matrix Ltd. 572 21 Princeton Place 573 Orchard Park, NY 14127 575 Phone: (716) 662-8973 x26 577 EMail: ken@oceana.com