idnits 2.17.1 draft-leiba-imap-implement-guide-04.txt: ** The Abstract section seems to be numbered Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** Cannot find the required boilerplate sections (Copyright, IPR, etc.) in this document. Expected boilerplate is as follows today (2024-04-25) according to https://trustee.ietf.org/license-info : IETF Trust Legal Provisions of 28-dec-2009, Section 6.a: This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. IETF Trust Legal Provisions of 28-dec-2009, Section 6.b(i), paragraph 2: Copyright (c) 2024 IETF Trust and the persons identified as the document authors. All rights reserved. IETF Trust Legal Provisions of 28-dec-2009, Section 6.b(i), paragraph 3: This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. 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 Internet-Drafts being working documents. ** The document seems to lack a 1id_guidelines paragraph about 6 months document validity. ** The document seems to lack a 1id_guidelines paragraph about the list of current Internet-Drafts. ** The document seems to lack a 1id_guidelines paragraph about the list of Shadow Directories. ** The document is more than 15 pages and seems to lack a Table of Contents. == No 'Intended status' indicated for this document; assuming Proposed Standard 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 document seems to lack separate sections for Informative/Normative References. All references will be assumed normative when checking for downward references. ** The abstract seems to contain references ([RFC-2060]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. 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 (April 1998) is 9507 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: 'UIDVALIDITY 824708485' is mentioned on line 256, but not defined == Missing Reference: 'UNSEEN 9921' is mentioned on line 257, but not defined == Missing Reference: 'UNSEEN' is mentioned on line 262, but not defined == Missing Reference: 'TRYCREATE' is mentioned on line 451, but not defined == Missing Reference: 'READ-ONLY' is mentioned on line 453, but not defined == Missing Reference: 'UIDVALIDITY 12345' is mentioned on line 594, but not defined -- Looks like a reference, but probably isn't: '1' on line 650 ** Obsolete normative reference: RFC 2060 (Obsoleted by RFC 3501) ** Downref: Normative reference to an Informational RFC: RFC 2180 -- Possible downref: Non-RFC (?) normative reference: ref. 'NAMESPACE' Summary: 14 errors (**), 0 flaws (~~), 7 warnings (==), 4 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group B. Leiba 3 Internet Draft IBM T.J. Watson Research Center 4 Document: draft-leiba-imap-implement-guide-04.txt November 1997 5 Expires April 1998 7 IMAP4 Implementation Recommendations 9 Status of this Document 11 This document provides information for the Internet community. This 12 document does not specify an Internet standard of any kind. 13 Distribution of this document is unlimited. 15 This document is an Internet Draft. Internet Drafts are working 16 documents of the Internet Engineering Task Force (IETF), its Areas, 17 and its Working Groups. Note that other groups may also distribute 18 working documents as Internet Drafts. 20 Internet Drafts are draft documents valid for a maximum of six 21 months. Internet Drafts may be updated, replaced, or obsoleted by 22 other documents at any time. It is not appropriate to use Internet 23 Drafts as reference material or to cite them other than as a "working 24 draft" or "work in progress". 26 To learn the current status of any Internet-Draft, please check the 27 1id-abstracts.txt listing contained in the Internet-Drafts Shadow 28 Directories on ds.internic.net, nic.nordu.net, ftp.isi.edu, or 29 munnari.oz.au. 31 A revised version of this draft document will be submitted to the RFC 32 editor. Discussion and suggestions for improvement are requested. 33 This document will expire by the end of April 1998. 35 1. Abstract 37 The IMAP4 specification [RFC-2060] describes a rich protocol for use 38 in building clients and servers for storage, retrieval, and 39 manipulation of electronic mail. Because the protocol is so rich and 40 has so many implementation choices, there are often trade-offs that 41 must be made and issues that must be considered when designing such 42 clients and servers. This document attempts to outline these issues 43 and to make recommendations in order to make the end products as 44 interoperable as possible. 46 Internet DRAFT Implementation Recommendations November 1997 48 2. Conventions used in this document 50 In examples, "C:" indicates lines sent by a client that is connected 51 to a server. "S:" indicates lines sent by the server to the client. 53 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 54 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 55 document are to be interpreted as described in [RFC-2119]. 57 3. Interoperability Issues and Recommendations 59 3.1. Accessibility 61 This section describes the issues related to access to servers and 62 server resources. Concerns here include data sharing and maintenance 63 of client/server connections. 65 3.1.1. Multiple Accesses of the Same Mailbox 67 One strong point of IMAP4 is that, unlike POP3, it allows for 68 multiple simultaneous access to a single mailbox. A user can, thus, 69 read mail from a client at home while the client in the office is 70 still connected; or the help desk staff can all work out of the same 71 inbox, all seeing the same pool of questions. An important point 72 about this capability, though is that NO SERVER IS GUARANTEED TO 73 SUPPORT THIS. If you are selecting an IMAP server and this facility 74 is important to you, be sure that the server you choose to install, 75 in the configuration you choose to use, supports it. 77 If you are designing a client, you MUST NOT assume that you can 78 access the same mailbox more than once at a time. That means 79 1. you must handle gracefully the failure of a SELECT command if 80 the server refuses the second SELECT, 81 2. you must handle reasonably the severing of your connection (see 82 "Severed Connections", below) if the server chooses to allow the 83 second SELECT by forcing the first off, 84 3. you must avoid making multiple connections to the same mailbox 85 in your own client (for load balancing or other such reasons), 86 and 87 4. you must avoid using the STATUS command on a mailbox that you 88 have selected (with some server implementations the STATUS 89 command has the same problems with multiple access as do the 90 SELECT and EXAMINE commands). 92 A further note about STATUS: The STATUS command is sometimes used to 93 check a non-selected mailbox for new mail. This mechanism MUST NOT 94 be used to check for new mail in the selected mailbox; section 5.2 of 96 Internet DRAFT Implementation Recommendations November 1997 98 [RFC-2060] specifically forbids this in its last paragraph. Further, 99 since STATUS takes a mailbox name it is an independent operation, not 100 operating on the selected mailbox. Because of this, the information 101 it returns is not necessarily in synchronization with the selected 102 mailbox state. 104 3.1.2. Severed Connections 106 The client/server connection may be severed for one of three reasons: 107 the client severs the connection, the server severs the connection, 108 or the connection is severed by outside forces beyond the control of 109 the client and the server (a telephone line drops, for example). 110 Clients and servers must both deal with these situations. 112 When the client wants to sever a connection, it's usually because it 113 has finished the work it needed to do on that connection. The client 114 SHOULD send a LOGOUT command, wait for the tagged response, and then 115 close the socket. But note that, while this is what's intended in 116 the protocol design, there isn't universal agreement here. Some 117 contend that sending the LOGOUT and waiting for the two responses 118 (untagged BYE and tagged OK) is wasteful and unnecessary, and that 119 the client can simply close the socket. The server should interpret 120 the closed socket as a log out by the client. The counterargument is 121 that it's useful from the standpoint of cleanup, problem 122 determination, and the like, to have an explicit client log out, 123 because otherwise there is no way for the server to tell the 124 difference between "closed socket because of log out" and "closed 125 socket because communication was disrupted". If there is a 126 client/server interaction problem, a client which routinely 127 terminates a session by breaking the connection without a LOGOUT will 128 make it much more difficult to determine the problem. 130 Because of this disagreement, server designers must be aware that 131 some clients might close the socket without sending a LOGOUT. In any 132 case, whether or not a LOGOUT was sent, the server SHOULD NOT 133 implicitly expunge any messages from the selected mailbox. If a 134 client wants the server to do so, it MUST send a CLOSE or EXPUNGE 135 command explicitly. 137 When the server wants to sever a connection it's usually due to an 138 inactivity timeout or is because a situation has arisen that has 139 changed the state of the mail store in a way that the server can not 140 communicate to the client. The server SHOULD send an untagged BYE 141 response to the client and then close the socket. Sending an 142 untagged BYE response before severing allows the server to send a 143 human-readable explanation of the problem to the client, which the 144 client may then log, display to the user, or both (see section 7.1.5 145 of [RFC-2060]). 147 Internet DRAFT Implementation Recommendations November 1997 149 3.2. Scaling 151 IMAP4 has many features that allow for scalability, as mail stores 152 become larger and more numerous. Large numbers of users, mailboxes, 153 and messages, and very large messages require thought to handle 154 efficiently. This document will not address the administrative 155 issues involved in large numbers of users, but we will look at the 156 other items. 158 3.2.1. Flood Control 160 There are three situations when a client can make a request that will 161 result in a very large response - too large for the client reasonably 162 to deal with: there are a great many mailboxes available, there are a 163 great many messages in the selected mailbox, or there is a very large 164 message part. The danger here is that the end user will be stuck 165 waiting while the server sends (and the client processes) an enormous 166 response. In all of these cases there are things a client can do to 167 reduce that danger. 169 There is also the case where a client can flood a server, by sending 170 an arbitratily long command. We'll discuss that issue, too, in this 171 section. 173 3.2.1.1. Listing Mailboxes 175 Some servers present Usenet newsgroups to IMAP users. Newsgroups, 176 and other such hierarchical mailbox structures, can be very numerous 177 but may have only a few entries at the top level of hierarchy. Also, 178 some servers are built against mail stores that can, unbeknownst to 179 the server, have circular hierarchies - that is, it's possible for 180 "a/b/c/d" to resolve to the same file structure as "a", which would 181 then mean that "a/b/c/d/b" is the same as "a/b", and the hierarchy 182 will never end. The LIST response in this case will be unlimited. 184 Clients that will have trouble with this are those that use 185 C: 001 LIST "" * 186 to determine the mailbox list. Because of this, clients SHOULD NOT 187 use an unqualified "*" that way in the LIST command. A safer 188 approach is to list each level of hierarchy individually, allowing 189 the user to traverse the tree one limb at a time, thus: 191 C: 001 LIST "" % 192 S: * LIST () "/" Banana 193 S: * LIST ...etc... 194 S: 001 OK done 196 Internet DRAFT Implementation Recommendations November 1997 198 and then 199 C: 002 LIST "" Banana/% 200 S: * LIST () "/" Banana/Apple 201 S: * LIST ...etc... 202 S: 002 OK done 204 Using this technique the client's user interface can give the user 205 full flexibility without choking on the voluminous reply to "LIST *". 206 Of course, it is still possible that the reply to 207 C: 005 LIST "" alt.fan.celebrity.% 208 may be thousands of entries long, and there is, unfortunately, 209 nothing the client can do to protect itself from that. This has not 210 yet been a notable problem. 212 Servers that may export circular hierarchies (any server that 213 directly presents a UNIX file system, for instance) SHOULD limit the 214 hierarchy depth to prevent unlimited LIST responses. A suggested 215 depth limit is 20 hierarchy levels. 217 3.2.1.2. Fetching the List of Messages 219 When a client selects a mailbox, it is given a count, in the untagged 220 EXISTS response, of the messages in the mailbox. This number can be 221 very large. In such a case it might be unwise to use 222 C: 004 FETCH 1:* ALL 223 to populate the user's view of the mailbox. One good method to avoid 224 problems with this is to batch the requests, thus: 226 C: 004 FETCH 1:50 ALL 227 S: * 1 FETCH ...etc... 228 S: 004 OK done 229 C: 005 FETCH 51:100 ALL 230 S: * 51 FETCH ...etc... 231 S: 005 OK done 232 C: 006 FETCH 101:150 ALL 233 ...etc... 235 Using this method, another command, such as "FETCH 6 BODY[1]" can be 236 inserted as necessary, and the client will not have its access to the 237 server blocked by a storm of FETCH replies. (Such a method could be 238 reversed to fetch the LAST 50 messages first, then the 50 prior to 239 that, and so on.) 241 As a smart extension of this, a well designed client, prepared for 242 very large mailboxes, will not automatically fetch data for all 243 messages AT ALL. Rather, the client will populate the user's view 244 only as the user sees it, possibly pre-fetching selected information, 245 and only fetching other information as the user scrolls to it. For 247 Internet DRAFT Implementation Recommendations November 1997 249 example, to select only those messages beginning with the first 250 unseen one: 252 C: 003 SELECT INBOX 253 S: * 10000 EXISTS 254 S: * 80 RECENT 255 S: * FLAGS (\Answered \Flagged \Deleted \Draft \Seen) 256 S: * OK [UIDVALIDITY 824708485] UID validity status 257 S: * OK [UNSEEN 9921] First unseen message 258 S: 003 OK [READ-WRITE] SELECT completed 259 C: 004 FETCH 9921:* ALL 260 ... etc... 262 If the server does not return an OK [UNSEEN] response, the client may 263 use SEARCH UNSEEN to obtain that value. 265 This mechanism is good as a default presentation method, but only 266 works well if the default message order is acceptable. A client may 267 want to present various sort orders to the user (by subject, by date 268 sent, by sender, and so on) and in that case (lacking a SORT 269 extension on the server side) the client WILL have to retrieve all 270 message descriptors. A client that provides this service SHOULD NOT 271 do it by default and SHOULD inform the user of the costs of choosing 272 this option for large mailboxes. 274 3.2.1.3. Fetching a Large Body Part 276 The issue here is similar to the one for a list of messages. In the 277 BODYSTRUCTURE response the client knows the size, in bytes, of the 278 body part it plans to fetch. Suppose this is a 70 MB video clip. 279 The client can use partial fetches to retrieve the body part in 280 pieces, avoiding the problem of an uninterruptible 70 MB literal 281 coming back from the server: 283 C: 022 FETCH 3 BODY[1]<0.20000> 284 S: * 3 FETCH (FLAGS(\Seen) BODY[1]<0> {20000} 285 S: ...data...) 286 S: 022 OK done 287 C: 023 FETCH 3 BODY[1]<20001.20000> 288 S: * 3 FETCH (BODY[1]<20001> {20000} 289 S: ...data...) 290 S: 023 OK done 291 C: 024 FETCH 3 BODY[1]<40001.20000> 292 ...etc... 294 Internet DRAFT Implementation Recommendations November 1997 296 3.2.1.4. BODYSTRUCTURE vs. Entire Messages 298 Because FETCH BODYSTRUCTURE is necessary in order to determine the 299 number of body parts, and, thus, whether a message has "attachments", 300 clients often use FETCH FULL as their normal method of populating the 301 user's view of a mailbox. The benefit is that the client can display 302 a paperclip icon or some such indication along with the normal 303 message summary. However, this comes at a significant cost with some 304 server configurations. The parsing needed to generate the FETCH 305 BODYSTRUCTURE response may be time-consuming compared with that 306 needed for FETCH ENVELOPE. The client developer should consider this 307 issue when deciding whether the ability to add a paperclip icon is 308 worth the tradeoff in performance, especially with large mailboxes. 310 Some clients, rather than using FETCH BODYSTRUCTURE, use FETCH BODY[] 311 (or the equivalent FETCH RFC822) to retrieve the entire message. 312 They then do the MIME parsing in the client. This may give the 313 client slightly more flexibility in some areas (access, for instance, 314 to header fields that aren't returned in the BODYSTRUCTURE and 315 ENVELOPE responses), but it can cause severe performance problems by 316 forcing the transfer of all body parts when the user might only want 317 to see some of them - a user logged on by modem and reading a small 318 text message with a large ZIP file attached may prefer to read the 319 text only and save the ZIP file for later. Therefore, a client 320 SHOULD NOT normally retrieve entire messages and SHOULD retrieve 321 message body parts selectively. 323 3.2.1.5. Long Command Lines 325 A client can wind up building a very long command line in an effort 326 to try to be efficient about requesting information from a server. 327 This can typically happen when a client builds a message set from 328 selected messages and doesn't recognise that contiguous blocks of 329 messages may be group in a range. Suppose a user selects all 10,000 330 messages in a large mailbox and then unselects message 287. The 331 client could build that message set as "1:286,288:10000", but a 332 client that doesn't handle that might try to enumerate each message 333 individually and build "1,2,3,4, [and so on] ,9999,10000". Adding 334 that to the fetch command results in a command line that's almost 335 49,000 octets long, and, clearly, one can construct a command line 336 that's even longer. 338 A client SHOULD limit the length of the command lines it generates to 339 approximately 1000 octets (including all quoted strings but not 340 including literals). If the client is unable to group things into 341 ranges so that the command line is within that length, it SHOULD 342 split the request into multiple commands. The client SHOULD use 343 literals instead of long quoted strings, in order to keep the command 345 Internet DRAFT Implementation Recommendations November 1997 347 length down. 349 For its part, a server SHOULD allow for a command line of at least 350 8000 octets. This provides plenty of leeway for accepting reasonable 351 length commands from clients. The server SHOULD send a BAD response 352 to a command that does not end within the server's maximum accepted 353 command length. 355 3.2.2. Subscriptions 357 The client isn't the only entity that can get flooded: the end user, 358 too, may need some flood control. The IMAP4 protocol provides such 359 control in the form of subscriptions. Most servers support the 360 SUBSCRIBE, UNSUBSCRIBE, and LSUB commands, and many users choose to 361 narrow down a large list of available mailboxes by subscribing to the 362 ones that they usually want to see. Clients, with this in mind, 363 SHOULD give the user a way to see only subscribed mailboxes. A 364 client that never uses the LSUB command takes a significant usability 365 feature away from the user. Of course, the client would not want to 366 hide the LIST command completely; the user needs to be able to go 367 both ways. 369 3.2.3. Searching 371 IMAP SEARCH commands can become particularly troublesome (that is, 372 slow) on mailboxes containing a large number of messages. So let's 373 put a few things in perspective in that regard. 375 The flag searches SHOULD be fast. The flag searches (ALL, [UN]SEEN, 376 [UN]ANSWERED, [UN]DELETED, [UN]DRAFT, [UN]FLAGGED, NEW, OLD, RECENT) 377 are known to be used by clients for the client's own use (for 378 instance, some clients use "SEARCH UNSEEN" to find unseen mail and 379 "SEARCH DELETED" to warn the user before expunging messages). 381 Other searches, particularly the text searches (HEADER, TEXT, BODY) 382 are initiated by the user, rather than by the client itself, and 383 somewhat slower performance can be tolerated, since the user is aware 384 that the search is being done (and is probably aware that it might be 385 time-consuming). 387 The client MAY allow other commands to be sent to the server while a 388 SEARCH is in progress, but at the time of this writing there is 389 little or no server support for parallel processing of multiple 390 commands in the same session (and see "Multiple Accesses of the Same 391 Mailbox" above for a description of the dangers of trying to work 392 around this by doing your SEARCH in another session). 394 Internet DRAFT Implementation Recommendations November 1997 396 Another word about text searches: some servers, built on database 397 back-ends with indexed search capabilities, may return search results 398 that do not match the IMAP spec's "case-insensitive substring" 399 requirements. While these servers are in violation of the protocol, 400 there is little harm in the violation as long as the search results 401 are used only to response to a user's request. Still, developers of 402 such servers should be aware that they ARE violating the protocol, 403 should think carefully about that behaviour, and MUST be certain that 404 their servers respond accurately to the flag searches for the reasons 405 outlined above. 407 In addition, servers SHOULD support CHARSET UTF-8 in searches. 409 3.3 Avoiding Invalid Requests 411 IMAP4 provides ways for a server to tell a client in advance what is 412 and isn't permitted in some circumstances. Clients SHOULD use these 413 features to avoid sending requests that a well designed client would 414 know to be invalid. This section explains this in more detail. 416 3.3.1. The CAPABILITY Command 418 All IMAP4 clients SHOULD use the CAPABILITY command to determine what 419 version of IMAP and what optional features a server supports. The 420 client SHOULD NOT send IMAP4rev1 commands and arguments to a server 421 that does not advertize IMAP4rev1 in its CAPABILITY response. 422 Similarly, the client SHOULD NOT send IMAP4 commands that no longer 423 exist in IMAP4rev1 to a server that does not advertize IMAP4 in its 424 CAPABILITY response. An IMAP4rev1 server is NOT required to support 425 obsolete IMAP4 or IMAP2bis commands (though some do; do not let this 426 fact lull you into thinking that it's valid to send such commands to 427 an IMAP4rev1 server). 429 A client SHOULD NOT send commands to probe for the existance of 430 certain extensions. All standard and standards-track extensions 431 include CAPABILITY tokens indicating their presense. All private and 432 experimental extensions SHOULD do the same, and clients that take 433 advantage of them SHOULD use the CAPABILITY response to determine 434 whether they may be used or not. 436 3.3.2. Don't Do What the Server Says You Can't 438 In many cases, the server, in response to a command, will tell the 439 client something about what can and can't be done with a particular 440 mailbox. The client SHOULD pay attention to this information and 441 SHOULD NOT try to do things that it's been told it can't do. 443 Internet DRAFT Implementation Recommendations November 1997 445 Examples: 446 * Do not try to SELECT a mailbox that has the \Noselect flag set. 447 * Do not try to CREATE a sub-mailbox in a mailbox that has the 448 \Noinferiors flag set. 449 * Do not respond to a failing COPY or APPEND command by trying to 450 CREATE the target mailbox if the server does not respond with a 451 [TRYCREATE] response code. 452 * Do not try to expunge a mailbox that has been selected with the 453 [READ-ONLY] response code. 455 3.4. Miscellaneous Protocol Considerations 457 We describe here a number of important protocol-related issues, the 458 misunderstanding of which has caused significant interoperability 459 problems in IMAP4 implementations. One general item is that every 460 implementer should be certain to take note of and to understand 461 section 2.2.2 and the preamble to section 7 of the IMAP4rev1 spec 462 [RFC-2060]. 464 3.4.1. Well Formed Protocol 466 We cannot stress enough the importance of adhering strictly to the 467 protocol grammar. The specification of the protocol is quite rigid; 468 do not assume that you can insert blank space for "readability" if 469 none is called for. Keep in mind that there are parsers out there 470 that will crash if there are protocol errors. There are clients that 471 will report every parser burp to the user. And in any case, 472 information that cannot be parsed is information that is lost. Be 473 careful in your protocol generation. And see "A Word About Testing", 474 below. 476 In particular, note that the string in the INTERNALDATE response is 477 NOT an RFC-822 date string - that is, it is not in the same format as 478 the first string in the ENVELOPE response. Since most clients will, 479 in fact, accept an RFC-822 date string in the INTERNALDATE response, 480 it's easy to miss this in your interoperability testing. But it will 481 cause a problem with some client, so be sure to generate the correct 482 string for this field. 484 3.4.2. Special Characters 486 Certain characters, currently the double-quote and the backslash, may 487 not be sent as-is inside a quoted string. These characters MUST be 488 preceded by the escape character if they are in a quoted string, or 489 else the string must be sent as a literal. Both clients and servers 490 MUST handle this, both on output (they must send these characters 492 Internet DRAFT Implementation Recommendations November 1997 494 properly) and on input (they must be able to receive escaped 495 characters in quoted strings). Example: 497 C: 001 LIST "" % 498 S: * LIST () "" INBOX 499 S: * LIST () "\\" TEST 500 S: * LIST () "\\" {12} 501 S: "My" mailbox 502 S: 001 OK done 503 C: 002 LIST "" "\"My\" mailbox\\%" 504 S: * LIST () "\\" {17} 505 S: "My" mailbox\Junk 506 S: 002 OK done 508 Note that in the example the server sent the hierarchy delimiter as 509 an escaped character in the quoted string and sent the mailbox name 510 containing imbedded double-quotes as a literal. The client used only 511 quoted strings, escaping both the backslash and the double-quote 512 characters. 514 The CR and LF characters may be sent ONLY in literals; they are not 515 allowed, even if escaped, inside quoted strings. 517 And while we're talking about special characters: the IMAP spec, in 518 the section titled "Mailbox International Naming Convention", 519 describes how to encode mailbox names in modified UTF-7. 520 Implementations MUST adhere to this in order to be interoperable in 521 the international market, and servers SHOULD validate mailbox names 522 sent by client and reject names that do not conform. 524 3.4.3. UIDs and UIDVALIDITY 526 Servers that support existing back-end mail stores often have no good 527 place to save UIDs for messages. Often the existing mail store will 528 not have the concept of UIDs in the sense that IMAP has: strictly 529 increasing, never re-issued, 32-bit integers. Some servers solve 530 this by storing the UIDs in a place that's accessible to end users, 531 allowing for the possibility that the users will delete them. Others 532 solve it by re-assigning UIDs every time a mailbox is selected. 534 The server SHOULD maintain UIDs permanently for all messages if it 535 can. If that's not possible, the server MUST change the UIDVALIDITY 536 value for the mailbox whenever any of the UIDs may have become 537 invalid. Clients MUST recognize that the UIDVALIDITY has changed and 538 MUST respond to that condition by throwing away any information that 539 they have saved about UIDs in that mailbox. There have been many 540 problems in this area when clients have failed to do this; in the 541 worst case it will result in loss of mail when a client deletes the 543 Internet DRAFT Implementation Recommendations November 1997 545 wrong piece of mail by using a stale UID. 547 It seems to be a common myth that "the UIDVALIDITY and the UID, taken 548 together, form a 64-bit identifier that uniquely identifies a message 549 on a server". This is absolutely NOT TRUE. There is no assurance 550 that the UIDVALIDITY values of two mailboxes be different, so the 551 UIDVALIDITY in no way identifies a mailbox. The ONLY purpose of 552 UIDVALIDITY is, as its name indicates, to give the client a way to 553 check the validity of the UIDs it has cached. While it is a valid 554 implementation choice to put these values together to make a 64-bit 555 identifier for the message, the important concept here is that UIDs 556 are not unique between mailboxes; they are only unique WITHIN a given 557 mailbox. 559 Some server implementations have toyed with making UIDs unique across 560 the entire server. This is inadvisable, in that it limits the life 561 of UIDs unnecessarily. The UID is a 32-bit number and will run out 562 in reasonably finite time if it's global across the server. If you 563 assign UIDs sequentially in one mailbox, you will not have to start 564 re-using them until you have had, at one time or another, 2**32 565 different messages in that mailbox. In the global case, you will 566 have to reuse them once you have had, at one time or another, 2**32 567 different messages in the entire mail store. Suppose your server has 568 around 8000 users registered (2**13). That gives an average of 2**19 569 UIDs per user. Suppose each user gets 32 messages (2**5) per day. 570 That gives you 2**14 days (16000+ days = about 45 years) before you 571 run out. That may seem like enough, but multiply the usage just a 572 little (a lot of spam, a lot of mailing list subscriptions, more 573 users) and you limit yourself too much. 575 What's worse is that if you have to wrap the UIDs, and, thus, you 576 have to change UIDVALIDITY and invalidate the UIDs in the mailbox, 577 you have to do it for EVERY mailbox in the system, since they all 578 share the same UID pool. If you assign UIDs per mailbox and you have 579 a problem, you only have to kill the UIDs for that one mailbox. 581 Under extreme circumstances (and this is extreme, indeed), the server 582 may have to invalidate UIDs while a mailbox is in use by a client - 583 that is, the UIDs that the client knows about in its active mailbox 584 are no longer valid. In that case, the server MUST immediately 585 change the UIDVALIDITY and MUST communicate this to the client. The 586 server MAY do this by sending an unsolicited UIDVALIDITY message, in 587 the same form as in response to the SELECT command. Clients MUST be 588 prepared to handle such a message and the possibly coincident failure 589 of the command in process. For example: 591 Internet DRAFT Implementation Recommendations November 1997 593 C: 032 UID STORE 382 +Flags.silent \Deleted 594 S: * OK [UIDVALIDITY 12345] New UIDVALIDITY value! 595 S: 032 NO UID command rejeced because UIDVALIDITY changed! 596 C: ...invalidates local information and re-fetches... 597 C: 033 FETCH 1:* UID 598 ...etc... 600 At the time of the writing of this document, the only server known to 601 do this does so only under the following condition: the client 602 selects INBOX, but there is not yet a physical INBOX file created. 603 Nonetheless, the SELECT succeeds, exporting an empty INBOX with a 604 temporary UIDVALIDITY of 1. While the INBOX remains selected, mail 605 is delivered to the user, which creates the real INBOX file and 606 assigns a permanent UIDVALIDITY (that is likely not to be 1). The 607 server reports the change of UIDVALIDITY, but as there were no 608 messages before, so no UIDs have actually changed, all the client 609 must do is accept the change in UIDVALIDITY. 611 Alternatively, a server may force the client to re-select the 612 mailbox, at which time it will obtain a new UIDVALIDITY value. To do 613 this, the server closes this client session (see "Severed 614 Connections" above) and the client then reconnects and gets back in 615 synch. Clients MUST be prepared for either of these behaviours. 617 We do not know of, nor do we anticipate the future existance of, a 618 server that changes UIDVALIDITY while there are existing messages, 619 but clients MUST be prepared to handle this eventuality. 621 3.4.4. FETCH Responses 623 When a client asks for certain information in a FETCH command, the 624 server MAY return the requested information in any order, not 625 necessarily in the order that it was requested. Further, the server 626 MAY return the information in separate FETCH responses and MAY also 627 return information that was not explicitly requested (to reflect to 628 the client changes in the state of the subject message). Some 629 examples: 631 C: 001 FETCH 1 UID FLAGS INTERNALDATE 632 S: * 5 FETCH (FLAGS (\Deleted)) 633 S: * 1 FETCH (FLAGS (\Seen) INTERNALDATE "..." UID 345) 634 S: 001 OK done 635 (In this case, the responses are in a different order. Also, the 636 server returned a flag update for message 5, which wasn't part of the 637 client's request.) 639 Internet DRAFT Implementation Recommendations November 1997 641 C: 002 FETCH 2 UID FLAGS INTERNALDATE 642 S: * 2 FETCH (INTERNALDATE "...") 643 S: * 2 FETCH (UID 399) 644 S: * 2 FETCH (FLAGS ()) 645 S: 002 OK done 646 (In this case, the responses are in a different order and were 647 returned in separate responses.) 649 C: 003 FETCH 2 BODY[1] 650 S: * 2 FETCH (FLAGS (\Seen) BODY[1] {14} 651 S: Hello world! 652 S: ) 653 S: 003 OK done 654 (In this case, the FLAGS response was added by the server, since 655 fetching the body part caused the server to set the \Seen flag.) 657 Because of this characteristic a client MUST be ready to receive any 658 FETCH response at any time and should use that information to update 659 its local information about the message to which the FETCH response 660 refers. A client MUST NOT assume that any FETCH responses will come 661 in any particular order, or even that any will come at all. If after 662 receiving the tagged response for a FETCH command the client finds 663 that it did not get all of the information requested, the client 664 SHOULD send a NOOP command to the server to ensure that the server 665 has an opportunity to send any pending EXPUNGE responses to the 666 client (see [RFC-2180]). 668 3.4.5. RFC822.SIZE 670 Some back-end mail stores keep the mail in a canonical form, rather 671 than retaining the original MIME format of the messages. This means 672 that the server must reassemble the message to produce a MIME stream 673 when a client does a fetch such as RFC822 or BODY[], requesting the 674 entire message. It also may mean that the server has no convenient 675 way to know the RFC822.SIZE of the message. Often, such a server 676 will actually have to build the MIME stream to compute the size, only 677 to throw the stream away and report the size to the client. 679 When this is the case, some servers have chosen to estimate the size, 680 rather than to compute it precisely. Such an estimate allows the 681 client to display an approximate size to the user and to use the 682 estimate in flood control considerations (q.v.), but requires that 683 the client not use the size for things such as allocation of buffers, 684 because those buffers might then be too small to hold the actual MIME 685 stream. Instead, a client SHOULD use the size that's returned in the 686 literal when you fetch the data. 688 The protocol requires that the RFC822.SIZE value returned by the 690 Internet DRAFT Implementation Recommendations November 1997 692 server be EXACT. Estimating the size is a protocol violation, and 693 server designers must be aware that, despite the performance savings 694 they might realize in using an estimate, this practice will cause 695 some clients to fail in various ways. If possible, the server SHOULD 696 compute the RFC822.SIZE for a particular message once, and then save 697 it for later retrieval. If that's not possible, the server MUST 698 compute the value exactly every time. Incorrect estimates do cause 699 severe interoperability problems with some clients. 701 3.4.6. Expunged Messages 703 If the server allows multiple connections to the same mailbox, it is 704 often possible for messages to be expunged in one client unbeknownst 705 to another client. Since the server is not allowed to tell the 706 client about these expunged messages in response to a FETCH command, 707 the server may have to deal with the issue of how to return 708 information about an expunged message. There was extensive 709 discussion about this issue, and the results of that discussion are 710 summarized in [RFC-2180]. See that reference for a detailed 711 explanation and for recommendations. 713 3.4.7. The Namespace Issue 715 Namespaces are a very muddy area in IMAP4 implementation right now 716 (see [NAMESPACE] for a proposal to clear the water a bit). Until the 717 issue is resolved, the important thing for client developers to 718 understand is that some servers provide access through IMAP to more 719 than just the user's personal mailboxes, and, in fact, the user's 720 personal mailboxes may be "hidden" somewhere in the user's default 721 hierarchy. The client, therefore, SHOULD provide a setting wherein 722 the user can specify a prefix to be used when accessing mailboxes. 723 If the user's mailboxes are all in "~/mail/", for instance, then the 724 user can put that string in the prefix. The client would then put 725 the prefix in front of any name pattern in the LIST and LSUB 726 commands: 727 C: 001 LIST "" ~/mail/% 728 (See also "Reference Names in the LIST Command" below.) 730 3.4.8. Creating Special-Use Mailboxes 732 It may seem at first that this is part of the namespace issue; it is 733 not, and is only indirectly related to it. A number of clients like 734 to create special-use mailboxes with particular names. Most 735 commonly, clients with a "trash folder" model of message deletion 736 want to create a mailbox with the name "Trash" or "Deleted". Some 737 clients want to create a "Drafts" mailbox, an "Outbox" mailbox, or a 739 Internet DRAFT Implementation Recommendations November 1997 741 "Sent Mail" mailbox. And so on. There are two major 742 interoperability problems with this practice: 743 1. different clients may use different names for mailboxes with 744 similar functions (such as "Trash" and "Deleted"), or may manage the 745 same mailboxes in different ways, causing problems if a user switches 746 between clients and 747 2. there is no guarantee that the server will allow the creation of 748 the desired mailbox. 750 The client developer is, therefore, well advised to consider 751 carefully the creation of any special-use mailboxes on the server, 752 and, further, the client MUST NOT require such mailbox creation - 753 that is, if you do decide to do this, you MUST handle gracefully the 754 failure of the CREATE command and behave reasonably when your 755 special-use mailboxes do not exist and can not be created. 757 In addition, the client developer SHOULD provide a convenient way for 758 the user to select the names for any special-use mailboxes, allowing 759 the user to make these names the same in all clients s/he uses and to 760 put them where s/he wants them. 762 3.4.9. Reference Names in the LIST Command 764 Many implementers of both clients and servers are confused by the 765 "reference name" on the LIST command. The reference name is intended 766 to be used in much the way a "cd" (change directory) command is used 767 on Unix, PC DOS, Windows, and OS/2 systems. That is, the mailbox 768 name is interpreted in much the same way as a file of that name would 769 be found if one had done a "cd" command into the directory specified 770 by the reference name. For example, in Unix we have the following: 772 > cd /u/jones/junk 773 > vi banana [file is "/u/jones/junk/banana"] 774 > vi stuff/banana [file is "/u/jones/junk/stuff/banana"] 775 > vi /etc/hosts [file is "/etc/hosts"] 777 The interoperability problems with this, in practice, are several. 778 First, while some IMAP servers are built on Unix or PC file systems, 779 many others are not, and the file system semantics do not make sense 780 in those configurations. Second, while some IMAP servers expose the 781 underlying file system to the clients, others allow access only to 782 the user's personal mailboxes, or to some other limited set of files, 783 making such file-system-like semantics less meaningful. Third, 784 because the IMAP spec leaves the interpretation of the reference name 785 as "implementation-dependent", the various server implementations 786 handle it in vastly differing ways, and fourth, many implementers 787 simply do not understand it and misuse it, do not use it, or ignore 788 it as a result. 790 Internet DRAFT Implementation Recommendations November 1997 792 The following statement gets somewhat into the religious issues that 793 we've tried to avoid scrupulously here; so be it: because of the 794 confusion around the reference name, its use by a client is a 795 dangerous thing, prone to result in interoperability problems. There 796 are servers that interpret it as originally intended; there are 797 servers that ignore it completely; there are servers that simply 798 prepend it to the mailbox name (with or without inserting a hierarchy 799 delimiter in between). Because a client can't know which of these 800 four behaviours to expect, a client SHOULD NOT use a reference name 801 itself, expecting a particular server behavior. However, a client 802 SHOULD permit a USER, by configuration, to use a reference name. 804 There is in no way universal agreement about the use or non-use of 805 the reference name. The last words here are, "Be aware." 807 3.4.10. ALERT Response Codes 809 The protocol spec is very clear on the matter of what to do with 810 ALERT response codes, and yet there are many clients that violate it 811 so it needs to be said anyway: "The human-readable text contains a 812 special alert that MUST be presented to the user in a fashion that 813 calls the user's attention to the message." Enough said. Do it. 815 3.4.11. Deleting Mailboxes 817 The protocol does not guarantee that a client may delete a mailbox 818 that is not empty, though on some servers it is permissible and is, 819 in fact, much faster than the alternative or deleting all the 820 messages from the client. If the client chooses to try to take 821 advantage of this possibility it MUST be prepared to use the other 822 method in the even that the more convenient one fails. Further, a 823 client SHOULD NOT try to delete the mailbox that it has selected, but 824 should first close that mailbox; some servers do not permit the 825 deletion of the selected mailbox. 827 That said, a server SHOULD permit the deletion of a non-empty 828 mailbox; there's little reason to pass this work on to the client. 829 Moreover, forbidding this prevents the deletion of a mailbox that for 830 some reason can not be opened or expunged, leading to possible 831 denial-of-service problems. 833 Internet DRAFT Implementation Recommendations November 1997 835 Example: 836 [User tells the client to delete mailbox BANANA, which is 837 currently selected...] 838 C: 008 CLOSE 839 S: 008 OK done 840 C: 009 DELETE BANANA 841 S: 009 NO Delete failed; mailbox is not empty. 842 C: 010 SELECT BANANA 843 S: * ... untagged SELECT responses 844 S: 010 OK done 845 C: 011 STORE 1:* +FLAGS.SILENT \DELETED 846 S: 011 OK done 847 C: 012 CLOSE 848 S: 012 OK done 849 C: 013 DELETE BANANA 850 S: 013 OK done 852 3.5. A Word About Testing 854 Since the whole point of IMAP is interoperability, and since 855 interoperability can not be tested in a vacuum, the final 856 recommendation of this treatise is, "Test against EVERYTHING." Test 857 your client against every server you can get an account on. Test 858 your server with every client you can get your hands on. Many 859 clients make limited test versions available on the Web for the 860 downloading. Many server owners will give serious client developers 861 guest accounts for testing. Contact them and ask. NEVER assume that 862 because your client works with one or two servers, or because your 863 server does fine with one or two clients, you will interoperate well 864 in general. 866 In particular, in addition to everything else, be sure to test 867 against the reference implementations: the PINE client, the 868 University of Washington server, and the Cyrus server. 870 See the following URLs on the web for more information here: 871 IMAP Products and Sources: http://www.imap.org/products.html 872 IMC MailConnect: http://www.imc.org/imc-mailconnect 874 4. Security Considerations 876 This document describes behaviour of clients and servers that use the 877 IMAP4 protocol, and as such, has the same security considerations as 878 described in [RFC-2060]. 880 Internet DRAFT Implementation Recommendations November 1997 882 5. References 884 [RFC-2060], Crispin, M., "Internet Message Access Protocol - Version 885 4rev1", RFC 2060, University of Washington, December 1996. 887 [RFC-2119], Bradner, S., "Key words for use in RFCs to Indicate 888 Requirement Levels", RFC 2119, Harvard University, March 1997. 890 [RFC-2180], Gahrns, M., "IMAP4 Multi-Accessed Mailbox Practice", RFC 891 2180, Microsoft, July 1997. 893 [NAMESPACE], Gahrns, M. & Newman, C., "IMAP4 Namespace", draft 894 document , June 1997. 896 6. Acknowledgments 898 To be completed... 900 7. Author's Address 902 Barry Leiba 903 IBM T.J. Watson Research Center 904 30 Saw Mill River Road 905 Hawthorne, NY 10532 907 Phone: 1-914-784-7941 908 Email: leiba@watson.ibm.com