| Rfc | 1176 |
|---|
| Title | Interactive Mail Access Protocol: Version 2 |
|---|
| Author | M.R. Crispin |
|---|
| Date | August
1990 |
|---|
| Format: | TXT, HTML |
|---|
| Obsoletes | RFC1064 |
|---|
| Status: | EXPERIMENTAL |
|---|
|
Network Working Group M. Crispin
Request for Comments: 1176 Washington
Obsoletes: RFC 1064 August 1990
INTERACTIVE MAIL ACCESS PROTOCOL - VERSION 2
Status of this Memo
This RFC suggests a method for personal computers and workstations to
dynamically access mail from a mailbox server ("repository"). It
obosoletes RFC 1064. This RFC specifies an Experimental Protocol for
the Internet community. Discussion and suggestions for improvement
are requested. Please refer to the current edition of the "IAB
Official Protocol Standards" for the standardization state and status
of this protocol. Distribution of this memo is unlimited.
Introduction
The intent of the Interactive Mail Access Protocol, Version 2 (IMAP2)
is to allow a workstation, personal computer, or similar small
machine to access electronic mail from a mailbox server. Since the
distinction between personal computers and workstations is blurring
over time, it is desirable to have a single solution that addresses
the need in a general fashion. IMAP2 is the "glue" of a distributed
electronic mail system consisting of a family of client and server
implementations on a wide variety of platforms, from small single-
tasking personal computing engines to complex multi-user timesharing
systems.
Although different in many ways from the Post Office Protocols (POP2
and POP3, hereafter referred to collectively as "POP") described in
RFC 937 and RFC 1081, IMAP2 may be thought of as a functional
superset of these. RFC 937 was used as a model for this RFC. There
was a cognizant reason for this; POP deals with a similar problem,
albeit with a less comprehensive solution, and it was desirable to
offer a basis for comparison.
Like POP, IMAP2 specifies a means of accessing stored mail and not of
posting mail; this function is handled by a mail transfer protocol
such as SMTP (RFC 821).
This protocol assumes a reliable data stream such as provided by TCP
or any similar protocol. When TCP is used, the IMAP2 server listens
on port 143.
System Model and Philosophy
Electronic mail is a primary means of communication for the widely
spread Internet community. The advent of distributed personal
computers and workstations has forced a significant rethinking of the
mechanisms employed to manage electronic mail. With mainframes, each
user tends to receive and process mail at the computer he uses most
of the time, his "primary host". The first inclination of many users
when an independent workstation is placed in front of them is to
begin receiving mail at the workstation, and many vendors have
implemented facilities to do this. However, this approach has
several disadvantages:
(1) Personal computers and many workstations have a software
design that gives full control of all aspects of the system to the
user at the console. As a result, background tasks such as
receiving mail may not run for long periods of time; either
because the user is asking to use all the machine's resources, or
because the user has (perhaps accidentally) manipulated the
environment in such a way that it prevents mail reception. In
many personal computers, the operating system is single-tasking
and this is the only mode of operation. Any of these conditions
could lead to repeated failed delivery attempts by outside agents.
(2) The hardware failure of a single machine can keep its user
"off the air" for a considerable time, since repair of individual
units may be delayed. Given the growing number of personal
computers and workstations spread throughout office environments,
quick repair of such systems is not assured. On the other hand, a
central mainframe is generally repaired soon after failure.
(3) Personal computers and workstations are often not backed up
with as much diligence as a central mainframe, if at all.
(4) It is more difficult to keep track of mailing addresses when
each person is associated with a distinct machine. Consider the
difficulty in keeping track of many postal addresses or phone
numbers, particularly if there was no single address or phone
number for an organization through which you could reach any
person in that organization. Traditionally, electronic mail on
the ARPANET involved remembering a name and one of several "hosts"
(machines) whose name reflected the organization in which the
individual worked. This was suitable at a time when most
organizations had only one central host. It is less satisfactory
today unless the concept of a host is changed to refer to an
organizational entity and not a particular machine.
(5) It is difficult to keep a multitude of heterogeneous machines
working properly with complex mailing protocols, making it
difficult to move forward as progress is made in electronic
communication and as new standards emerge. Each system has to
worry about receiving incoming mail, routing and delivering
outgoing mail, formatting, storing, and providing for the
stability of mailboxes over a variety of possible filing and
mailing protocols.
Consequently, while a personal computer or workstation may be viewed
as an Internet host in the sense that it implements TCP/IP, it should
not be viewed as the entity that contains the user's mailbox.
Instead, a mail server machine ("server", sometimes called a
"repository") should hold the mailbox, and the personal computer or
workstation (hereafter referred to as a "client") should access the
mailbox via mail transactions.
Because the mail server machine is isolated from direct user
manipulation, it should achieve high software reliability easily,
and, as a shared resource, it should also achieve high hardware
reliability, perhaps through redundancy. The mail server may be
accessed from arbitrary locations, allowing users to read mail across
campus, town, or country using commonly available clients.
Furthermore, the same user may access his mailbox from different
clients at different times, and multiple users may access the same
mailbox simultaneously.
The mail server acts an an interface among users, data storage, and
other mailers. A mail access protocol retrieves messages, accesss
and changes properties of messages, and otherwise manages mailboxes.
This differs from some approaches (e.g., Unix mail via NFS) in that
the mail access protocol is used for all message manipulations,
isolating the user and the client from all knowledge of how the data
storage is used. This means that the mail server can use the data
storage in whatever way is most efficient to organize the mail in
that particular environment, without having to worry about storage
representation compatibility across different machines.
A mail access protocol further differs in that it transmits
information only on demand. A well-designed mail access protocol
requires considerably less network traffic than Unix mail via NFS,
particularly when the mail file is large. The result is that a mail
access protocol can scale well to situations of large mailboxes or
networks with high latency or low speed.
In defining a mail access protocol, it is important to keep in mind
that the client and server form a macrosystem, in which it should be
possible to exploit the strong points of both while compensating for
each other's weaknesses. Furthermore, it is desirable to allow for a
growth path beyond the hoary text-only RFC 822 protocol, specifically
in the area of attachments and multi-media mail, to ease the eventual
transition to ISO solutions.
Unlike POP, IMAP2 has extensive features for remote searching and
parsing of messages on the server. A free text search (optionally
with other searching) can be made in the entire mailbox by the server
and the results made available to the client without the client
having to transfer the entire mailbox and searching itself. Since
remote parsing of a message into a structured (and standard format)
"envelope" is available, a client can display envelope information
and implement commands such as REPLY without having any understanding
of how to parse RFC 822, etc. headers. The effect of this is
twofold: it further improves the ability to scale well in instances
where network traffic must be reduced, and it reduces the complexity
of the client program.
Additionally, IMAP2 offers several facilities for managing individual
message state and the mailbox as a whole beyond the simple "delete
message" functionality of POP. Another benefit of IMAP2 is the use
of tagged responses to reduce the possibility of synchronization
errors and the concept of state on the client (a "local cache") that
the server may update without explicit request by the client. These
concepts and how they are used are explained under "Implementation
Discussion" below.
In spite of this functional richness, IMAP2 is a small protocol.
Although servers should implement the full set of IMAP2 functions, a
simple client can be written that uses IMAP2 in much the way as a POP
client.
A related protocol to POP and IMAP2 is the DMSP protocol of PCMAIL
(RFC 1056). IMAP2 differs from DMSP more fundamentally, reflecting a
differing architecture from PCMAIL. PCMAIL is either an online
("interactive mode"), or offline ("batch mode") system with long-term
shared state. Some POP based systems are also offline; in such
systems, since there is no long-term shared state POP is little more
than a download mechanism of the "mail file" to the client. IMAP2-
based software is primarily an online system in which real-time and
simultaneous mail access were considered important.
In PCMAIL, there is a long-term client/server relationship in which
some mailbox state is preserved on the client. There is a
registration of clients used by a particular user, and the client
keeps a set of "descriptors" for each message that summarize the
message. The server and client synchronize their states when the
DMSP connection starts up, and, if a client has not accessed the
server for a while, the client does a complete reset (reload) of its
state from the server.
In IMAP2-based software, the client/server relationship lasts only
for the duration of the TCP connection. All mailbox state is
maintained on the server. There is no registration of clients. The
function of a descriptor is handled by a structured representation of
the message "envelope" as noted above. There is no client/server
synchronization since the client does not remember state between
IMAP2 connections. This is not a problem since in general the client
never needs the entire state of the mailbox in a single session,
therefore there isn't much overhead in fetching the state information
that is needed as it is needed.
There are also some functional differences between IMAP2 and DMSP.
DMSP has functions for sending messages, printing messages, listing
mailboxes, and changing passwords; these are done outside IMAP2.
DMSP has 16 binary flags of which 8 are defined by the system. IMAP2
has flag names; there are currently 5 defined system flag names and a
facility for some number (30 in the current implementations) of user
flag names. IMAP2 has a sophisticated message search facility in the
server to identify interesting messages based on dates, addresses,
flag status, or textual contents without compelling the client to
fetch this data for every message.
It was felt that maintaining state on the client is advantageous only
in those cases where the client is only used by a single user, or if
there is some means on the client to restrict access to another
user's data. It can be a serious disadvantage in an environment in
which multiple users routinely use the same client, the same user
routinely uses different clients, and where there are no access
restrictions on the client. It was also observed that most user mail
access is to a small set of "interesting" messages, which were either
new mail or mail based on some user-selected criteria. Consequently,
IMAP2 was designed to easily identify those "interesting" messages so
that the client could fetch the state of those messages and not those
that were not "interesting".
The Protocol
The IMAP2 protocol consists of a sequence of client commands and
server responses, with server data interspersed between the
responses. Unlike most Internet protocols, commands and responses
are tagged. That is, a command begins with a unique identifier
(typically a short alphanumeric sequence such as a Lisp "gensym"
function would generate e.g., A0001, A0002, etc.), called a tag. The
response to this command is given the same tag from the server.
Additionally, the server may send an arbitrary amount of "unsolicited
data", which is identified by the special reserved tag of "*". There
is another special reserved tag, "+", discussed below.
The server must be listening for a connection. When a connection is
opened the server sends an unsolicited OK response as a greeting
message and then waits for commands.
The client opens a connection and waits for the greeting. The client
must not send any commands until it has received the greeting from
the server.
Once the greeting has been received, the client may begin sending
commands and is not under any obligation to wait for a server
response to this command before sending another command, within the
constraints of TCP flow control. When commands are received the
server acts on them and responds with command responses, often
interspersed with data. The effect of a command can not be
considered complete until a command response with a tag matching the
command is received from the server.
Although all known IMAP2 servers at the time of this writing process
commands to completion before processing the next command, it is not
required that a server do so. However, many commands can affect the
results of other commands, creating processing-order dependencies
(or, for SEARCH and FIND, ambiguities about which data is associated
with which command). All implementations that operate in a non-
lockstep fashion must recognize such dependencies and defer or
synchronize execution as necessary. In general, such multi-
processing is limited to consecutive FETCH commands.
Generally, the first command from the client is a LOGIN command with
user name and password arguments to establish identity and access
authorization, unless this has already been accomplished through
other means, e.g. Kerberos. Until identity and access authorization
have been established, no operations other than LOGIN or LOGOUT are
permitted.
Once identity and authorization have been established, the client
must send a SELECT command to access the desired mailbox; no mailbox
is selected by default. SELECT's argument is implementation-
dependent; however the word "INBOX" must be implemented to mean the
primary or default mailbox for this user, independent of any other
server semantics. On a successful SELECT, the server will send a
list of valid flags, number of messages, and number of messages
arrived since last access for this mailbox as unsolicited data,
followed by an OK response. The client may terminate access to this
mailbox and access a different one with another SELECT command.
The client reads mailbox information with FETCH commands. The actual
data is transmitted via the unsolicited data mechanism (that is,
FETCH should be viewed as instructing the server to include the
desired data along with any other data it wishes to transmit to the
client). There are three major categories of data that may be
fetched.
The first category is data that is associated with a message as an
entity in the mailbox. There are now three such items of data: the
"internal date", the "RFC 822 size", and the "flags". The internal
date is the date and time that the message was placed in the mailbox.
The RFC 822 size is subject to deletion in the future; it is the size
in bytes of the message, expressed as an RFC 822 text string.
Current clients only use it as part of a status display line. The
flags are a list of status flags associated with the message (see
below). All the first category data can be fetched by using the
macro-fetch word "FAST"; that is, "FAST" expands to "(FLAGS
INTERNALDATE RFC822.SIZE)".
The second category is that data that describes the composition and
delivery information of a message; that is, information such as the
message sender, recipient lists, message-ID, subject, etc. This is
the information that is stored in the message header in RFC 822
format message and is traditionally called the "envelope". [Note:
this should not be confused with the SMTP (RFC 821) envelope, which
is strictly limited to delivery information.] IMAP2 defines a
structured and unambiguous representation for the envelope that is
particularly suited for Lisp-based parsers. A client can use the
envelope for operations such as replying and not worry about RFC 822
at all. Envelopes are discussed in more detail below. The first two
categories of data can be fetched together by using the macro-fetch
word "ALL"; that is, "ALL" expands to "(FLAGS INTERNALDATE
RFC822.SIZE ENVELOPE)".
The third category is that data that is intended for direct human
viewing. The present RFC 822 based IMAP2 defines three such items:
RFC822.HEADER, RFC822.TEXT, and RFC822 (the latter being the two
former appended together in a single text string). RFC822.HEADER is
the "raw", unprocessed RFC 822 format header of the message.
Fetching "RFC822" is equivalent to fetching the RFC 822
representation of the message as stored on the mailbox without any
filtering or processing.
An intelligent client will "FETCH ALL" for some (or all) of the
messages in the mailbox for use as a presentation menu, and when the
user wishes to read a particular message will "FETCH RFC822.TEXT" to
get the message body. A more primitive client could, of course,
simply "FETCH RFC822" a`la POP-type functionality.
The client can alter certain data (currently only the flags) by a
STORE command. As an example, a message is deleted from a mailbox by
a STORE command that includes the \DELETED flag as a flag being set.
Other client operations include copying a message to another mailbox
(COPY command), permanently removing deleted messages (EXPUNGE
command), checking for new messages (CHECK command), and searching
for messages that match certain criteria (SEARCH command).
The client terminates the session with the LOGOUT command. The
server returns a "BYE" followed by an "OK".
A Typical Scenario
Client Server
------ ------
{Wait for Connection}
{Open Connection} -->
<-- * OK IMAP2 Server Ready
{Wait for command}
A001 LOGIN Fred Secret -->
<-- A001 OK User Fred logged in
{Wait for command}
A002 SELECT INBOX -->
<-- * FLAGS (Meeting Notice \Answered
\Flagged \Deleted \Seen)
<-- * 19 EXISTS
<-- * 2 RECENT
<-- A0002 OK Select complete
{Wait for command}
A003 FETCH 1:19 ALL -->
<-- * 1 Fetch (......)
...
<-- * 18 Fetch (......)
<-- * 19 Fetch (......)
<-- A003 OK Fetch complete
{Wait for command}
A004 FETCH 8 RFC822.TEXT -->
<-- * 8 Fetch (RFC822.TEXT {893}
...893 characters of text...
<-- )
<-- A004 OK Fetch complete
{Wait for command}
A005 STORE 8 +Flags \Deleted -->
<-- * 8 Store (Flags (\Deleted
\Seen))
<-- A005 OK Store complete
{Wait for command}
A006 EXPUNGE -->
<-- * 19 EXISTS
<-- * 8 EXPUNGE
<-- * 18 EXISTS
<-- A006 Expunge complete
{Wait for command}
A007 LOGOUT -->
<-- * BYE IMAP2 server quitting
<-- A007 OK Logout complete
{Close Connection} --><-- {Close connection}
{Go back to start}
Conventions
The following terms are used in a meta-sense in the syntax
specification below:
An ASCII-STRING is a sequence of arbitrary ASCII characters.
An ATOM is a sequence of ASCII characters delimited by SP or CRLF.
A CHARACTER is any ASCII character except """", "{", CR, LF, "%",
or "\".
A CRLF is an ASCII carriage-return character followed immediately
by an ASCII linefeed character.
A NUMBER is a sequence of the ASCII characters that represent
decimal numerals ("0" through "9"), delimited by SP, CRLF, ",", or
":".
A SP is the ASCII space character.
A TEXT_LINE is a human-readable sequence of ASCII characters up to
but not including a terminating CRLF.
A common field in the IMAP2 protocol is a STRING, which may be an
ATOM, QUOTED-STRING (a sequence of CHARACTERs inside double-quotes),
or a LITERAL. A literal consists of an open brace ("{"), a number, a
close brace ("}"), a CRLF, and then an ASCII-STRING of n characters,
where n is the value of the number inside the brace. In general, a
string should be represented as an ATOM or QUOTED-STRING if at all
possible. The semantics for QUOTED-STRING or LITERAL are checked
before those for ATOM; therefore an ATOM used in a STRING may only
contain CHARACTERs. Literals are most often sent from the server to
the client; in the rare case of a client to server literal there is a
special consideration (see the "+ text" response below).
Another important field is the SEQUENCE, which identifies a set of
messages by consecutive numbers from 1 to n where n is the number of
messages in the mailbox. A sequence may consist of a single number,
a pair of numbers delimited by colon (equivalent to all numbers
between those two numbers), or a list of single numbers or number
pairs. For example, the sequence 2,4:7,9,12:15 is equivalent to
2,4,5,6,7,9,12,13,14,15 and identifies all those messages.
Definitions of Commands and Responses
Summary of Commands and Responses
Commands || Responses
-------- || -------
tag NOOP || tag OK text
tag LOGIN user password || tag NO text
tag LOGOUT || tag BAD text
tag SELECT mailbox || * number message_data
tag BBOARD bulletin_board || * FLAGS flag_list
tag FIND MAILBOXES pattern || * SEARCH sequence
tag FIND BBOARDS pattern || * BBOARD string
tag CHECK || * MAILBOX string
tag EXPUNGE || * BYE text
tag COPY sequence mailbox || * OK text
tag FETCH sequence data || * NO text
tag STORE sequence data value || * BAD text
tag SEARCH search_program || + text
Commands
tag NOOP
The NOOP command returns an OK to the client. By itself, it does
nothing, but certain things may happen as side effects. For
example, server implementations that implicitly check the mailbox
for new mail may do so as a result of this command. The primary
use of this command is to for the client to see if the server is
still alive (and notify the server that the client is still alive,
for those servers that have inactivity autologout timers).
tag LOGIN user password
The LOGIN command identifies the user to the server and carries
the password authenticating this user. This information is used
by the server to control access to the mailboxes.
EXAMPLE: A001 LOGIN SMITH SESAME
logs in as user SMITH with password SESAME.
tag LOGOUT
The LOGOUT command informs the server that the client is done with
the session. The server should send an unsolicited BYE response
before the (tagged) OK response, and then close the network
connection.
tag SELECT mailbox
The SELECT command selects a particular mailbox. The server must
check that the user is permitted read access to this mailbox.
Before returning an OK to the client, the server must send the
following unsolicited data to the client:
FLAGS mailbox's defined flags
<n> EXISTS the number of messages in the mailbox
<n> RECENT the number of new messages in the mailbox
in order to define the initial state of the mailbox at the client.
Multiple SELECT commands are permitted in a session, in which case
the previous mailbox is automatically deselected when a new SELECT
is made.
The default mailbox for the SELECT command is INBOX, which is a
special name reserved to mean "the primary mailbox for this user
on this server". The format of other mailbox names is operating
system dependent (as of this writing, it reflects the filename
path of the mailbox file on the current servers).
It is customary, although not required, for the text of an OK
response to the SELECT command to begin with either "[READ-ONLY]"
or "[READ-WRITE]" to show the mailbox's access status.
EXAMPLE: A002 SELECT INBOX
selects the default mailbox.
tag BBOARD bulletin_board
The BBOARD command is equivalent to SELECT, and returns the same
output. However, it differs from SELECT in that its argument is a
shared mailbox (bulletin board) name instead of an ordinary
mailbox. The format of a bulletin name is implementation
specific, although it is strongly encouraged to use something that
resembles a name in a generic sense and not a file or mailbox name
on the particular system. There is no requirement that a bulletin
board name be a mailbox name or a file name (in particular, Unix
netnews has a completely different namespace from mailbox or file
names).
Support for BBOARD is optional.
tag FIND MAILBOXES pattern
The FIND MAILBOXES command accepts as an argument a pattern
(including wildcards) that specifies some set of mailbox names
that are usable by the SELECT command. The format of mailboxes is
implementation dependent. The special mailbox name INBOX is not
included in the output.
Two wildcard characters are defined; "*" specifies any number
(including zero) characters may match at this position and "%"
specifies a single character may match at this position. For
example, FOO*BAR will match FOOBAR, FOOD.ON.THE.BAR and FOO.BAR,
whereas FOO%BAR will match only FOO.BAR. "*" will match all
mailboxes.
The FIND MAILBOXES command will return some set of unsolicited
MAILBOX replies that have as their value a single mailbox name.
EXAMPLE: A002 FIND MAILBOXES *
* MAILBOX FOOBAR
* MAILBOX GENERAL
A002 FIND completed
Although the use of explicit file or path names for mailboxes is
discouraged by this standard, it may be unavoidable. It is
important that the value returned in the MAILBOX unsolicited reply
be usable in the SELECT command without remembering any path
specification that may have been used in the FIND MAILBOXES
pattern.
Support for FIND MAILBOXES is optional. If a client's attempt
returns BAD as a response then the client can make no assumptions
about what mailboxes exist on the server other than INBOX.
tag FIND BBOARDS pattern
The FIND BBOARDS command accepts as an argument a pattern that
specifies some set of bulletin board names that are usable by the
BBOARD command. Wildcards are permitted as in FIND MAILBOXES.
The FIND BBOARDS command will return some set of unsolicited
BBOARD replies that have as their value a single bulletin board
name.
EXAMPLE: A002 FIND BBOARDS *
* BBOARD FOOBAR
* BBOARD GENERAL
A002 FIND completed
Support for FIND BBOARDS is optional. If a client's attempt
returns BAD as a response then the client can make no assumptions
about what bulletin boards exist on the server, or that they exist
at all.
tag CHECK
The CHECK command forces a check for new messages and a rescan of
the mailbox for internal change for those implementations that
allow multiple simultaneous read/write access to the same mailbox.
It is recommend that periodic implicit checks for new mail be done
by servers as well. The server should send unsolicited EXISTS and
RECENT responses with the current status before returning an OK to
the client.
tag EXPUNGE
The EXPUNGE command permanently removes all messages with the
\DELETED flag set in its flags from the mailbox. Before returning
an OK to the client, for each message that is removed, an
unsolicited EXPUNGE response is sent. The message number for each
successive message in the mailbox is immediately decremented by 1;
this means that if the last 5 messages in a 9-message mail file
are expunged you will receive 5 unsolicited EXPUNGE responses for
message 5. To ensure mailbox integrity and server/client
synchronization, it is recommended that the server do an implicit
check before commencing the expunge and again when the expunge is
completed. Furthermore, if the server allows multiple
simultaneous access to the same mail file the server must lock the
mail file for exclusive access while an expunge is taking place.
EXPUNGE is not allowed if the user does not have write access to
this mailbox.
tag COPY sequence mailbox
The COPY command copies the specified message(s) to the specified
destination mailbox. If the destination mailbox does not exist,
the server should create it. Before returning an OK to the
client, the server should return an unsolicited <n> COPY response
for each message copied. A copy should set the \SEEN flag for all
messages that were successfully copied (provided, of course, that
the user has write access to this mailbox).
EXAMPLE: A003 COPY 2:4 MEETING
copies messages 2, 3, and 4 to mailbox "MEETING".
COPY is not allowed if the user does not have write access to the
destination mailbox.
tag FETCH sequence data
The FETCH command retrieves data associated with a message in the
mailbox. The data items to be fetched may be either a single atom
or an S-expression list. The currently defined data items that
can be fetched are:
ALL Macro equivalent to:
(FLAGS INTERNALDATE RFC822.SIZE ENVELOPE)
ENVELOPE The envelope of the message. The envelope is
computed by the server by parsing the RFC 822
header into the component parts, defaulting
various fields as necessary.
FAST Macro equivalent to:
(FLAGS INTERNALDATE RFC822.SIZE)
FLAGS The flags that are set for this message.
This may include the following system flags:
\RECENT Message arrived since the
previous time this mailbox
was read
\SEEN Message has been read
\ANSWERED Message has been answered
\FLAGGED Message is "flagged" for
urgent/special attention
\DELETED Message is "deleted" for
removal by later EXPUNGE
INTERNALDATE The date and time the message was written to
the mailbox.
RFC822 The message in RFC 822 format. The \SEEN
flag is implicitly set; if this causes the
flags to change they should be included as
part of the fetch results. This is the
concatenation of RFC822.HEADER and RFC822.TEXT.
RFC822.HEADER The "raw" RFC 822 format header of the message
as stored on the server.
RFC822.SIZE The number of characters in the message as
expressed in RFC 822 format.
RFC822.TEXT The text body of the message, omitting the
RFC 822 header. The \SEEN flag is implicitly
set as with RFC822 above.
EXAMPLES:
A003 FETCH 2:4 ALL
fetches the flags, internal date, RFC 822 size, and envelope
for messages 2, 3, and 4.
A004 FETCH 3 RFC822
fetches the RFC 822 representation for message 3.
A005 FETCH 4 (FLAGS RFC822.HEADER)
fetches the flags and RFC 822 format header for message 4.
Note: An attempt to FETCH already-transmitted data may have no
result. See the Implementation Discussion below.
tag STORE sequence data value
The STORE command alters data associated with a message in the
mailbox. The currently defined data items that can be stored are:
FLAGS Replace the flags for the message with the
argument (in flag list format).
+FLAGS Add the flags in the argument to the
message's flag list.
-FLAGS Remove the flags in the argument from the
message's flag list.
STORE is not allowed if the user does not have write access to
this mailbox.
EXAMPLE: A003 STORE 2:4 +FLAGS (\DELETED)
marks messages 2, 3, and 4 for deletion.
tag SEARCH search_criteria
The SEARCH command searches the mailbox for messages that match
the given set of criteria. The unsolicited SEARCH <1#number>
response from the server is a list of messages that express the
intersection (AND function) of all the messages which match that
criteria. For example,
A003 SEARCH DELETED FROM "SMITH" SINCE 1-OCT-87
returns the message numbers for all deleted messages from Smith
that were placed in the mail file since October 1, 1987.
In all search criteria which use strings, a message matches the
criteria if the string is a case-independent substring of that
field. The currently defined criteria are:
ALL All messages in the mailbox; the default
initial criterion for ANDing.
ANSWERED Messages with the \ANSWERED flag set.
BCC string Messages which contain the specified string
in the envelope's BCC field.
BEFORE date Messages whose internal date is earlier than
the specified date.
BODY string Messages which contain the specified string
in the body of the message.
CC string Messages which contain the specified string
in the envelope's CC field.
DELETED Messages with the \DELETED flag set.
FLAGGED Messages with the \FLAGGED flag set.
FROM string Messages which contain the specified string
in the envelope's FROM field.
KEYWORD flag Messages with the specified flag set.
NEW Messages which have the \RECENT flag set but
not the \SEEN flag. This is functionally
equivalent to "RECENT UNSEEN".
OLD Messages which do not have the \RECENT flag
set.
ON date Messages whose internal date is the same as
the specified date.
RECENT Messages which have the \RECENT flag set.
SEEN Messages which have the \SEEN flag set.
SINCE date Messages whose internal date is later than
the specified date.
SUBJECT string Messages which contain the specified string
in the envelope's SUBJECT field.
TEXT string Messages which contain the specified string.
TO string Messages which contain the specified string in
the envelope's TO field.
UNANSWERED Messages which do not have the \ANSWERED flag
set.
UNDELETED Messages which do not have the \DELETED flag
set.
UNFLAGGED Messages which do not have the \FLAGGED flag
set.
UNKEYWORD flag Messages which do not have the specified flag
set.
UNSEEN Messages which do not have the \SEEN flag set.
Responses
tag OK text
This response identifies successful completion of the command with
that tag. The text is a line of human-readable text that may be
useful in a protocol telemetry log for debugging purposes.
tag NO text
This response identifies unsuccessful completion of the command
with that tag. The text is a line of human-readable text that
probably should be displayed to the user in an error report by the
client.
tag BAD text
This response identifies faulty protocol received from the client;
The text is a line of human-readable text that should be recorded
in any telemetry as part of a bug report to the maintainer of the
client.
* number message_data
This response occurs as a result of several different commands.
The message_data is one of the following:
EXISTS The specified number of messages exists in the mailbox.
RECENT The specified number of messages have arrived since the
previous time this mailbox was read.
EXPUNGE The specified message number has been permanently
removed from the mailbox, and the next message in the
mailbox (if any) becomes that message number.
STORE data
Obsolete and functionally equivalent to FETCH.
FETCH data
This is the principle means by which data about a
message is returned to the client. The data is in a
Lisp-like S-expression property list form. The current
properties are:
ENVELOPE An S-expression format list that describes the
envelope of a message. The envelope is computed
by the server by parsing the RFC 822 header into
the component parts, defaulting various fields
as necessary.
The fields of the envelope are in the following
order: date, subject, from, sender, reply-to, to,
cc, bcc, in-reply-to, and message-id. The date,
subject, in-reply-to, and message-id fields are
strings. The from, sender, reply-to, to, cc,
and bcc fields are lists of addresses.
An address is an S-expression format list that
describes an electronic mail address. The fields
of an address are in the following order:
personal name, source-route (a.k.a. the
at-domain-list in SMTP), mailbox name, and
host name.
Any field of an envelope or address that is
not applicable is presented as the atom NIL.
Note that the server must default the reply-to
and sender fields from the from field; a client is
not expected to know to do this.
FLAGS An S-expression format list of flags that are set
for this message. This may include the following
system flags:
\RECENT Message arrived since the
previous time this mailbox
was read
\SEEN Message has been read
\ANSWERED Message has been answered
\FLAGGED Message is "flagged" for
urgent/special attention
\DELETED Message is "deleted" for
removal by later EXPUNGE
INTERNALDATE A string containing the date and time the
message was written to the mailbox.
RFC822 A string expressing the message in RFC 822
format.
RFC822.HEADER A string expressing the RFC 822 format
header of the message
RFC822.SIZE A number indicating the number of
characters in the message as expressed
in RFC 822 format.
RFC822.TEXT A string expressing the text body of the
message, omitting the RFC 822 header.
* FLAGS flag_list
This response occurs as a result of a SELECT command. The flag
list are the list of flags (at a minimum, the system-defined
flags) that are applicable for this mailbox. Flags other than the
system flags are a function of the server implementation.
* SEARCH number(s)
This response occurs as a result of a SEARCH command. The
number(s) refer to those messages that match the search criteria.
Each number is delimited by a space, e.g., "SEARCH 2 3 6".
* BBOARD string
This response occurs as a result of a FIND BBOARDS command. The
string is a bulletin board name that matches the pattern in the
command.
* MAILBOX string
This response occurs as a result of a FIND MAILBOXES command. The
string is a mailbox name that matches the pattern in the command.
* BYE text
This response identifies that the server is about to close the
connection. The text is a line of human-readable text that should
be displayed to the user in a status report by the client. This
may be sent as part of a normal logout sequence, or as a panic
shutdown announcement by the server. It is also used by some
servers as an announcement of an inactivity autologout.
* OK text
This response identifies normal operation on the server. No
special action by the client is called for, however, the text
should be displayed to the user in some fashion. This is
currently only used by servers at startup as a greeting message to
show they are ready to accept the first command.
* NO text
This response identifies a warning from the server that does not
affect the overall results of any particular request. The text is
a line of human-readable text that should be presented to the user
as a warning of improper operation.
* BAD text
This response identifies a serious error at the server; it may
also indicate faulty protocol from the client in which a tag could
not be parsed. The text is a line of human-readable text that
should be presented to the user as a serious or possibly fatal
error. It should also be recorded in any telemetry as part of a
bug report to the maintainer of the client and server.
+ text
This response identifies that the server is ready to accept the
text of a literal from the client. Normally, a command from the
client is a single text line. If the server detects an error in
the command, it can simply discard the remainder of the line. It
cannot do this for commands that contain literals, since a literal
can be an arbitrarily long amount of text, and the server may not
even be expecting a literal. This mechanism is provided so the
client knows not to send a literal until the server expects it,
preserving client/server synchronization.
In practice, this condition is rarely encountered. In the current
protocol, the only client command likely to contain a literal is
the LOGIN command. Consider a server that validates the user
before checking the password. If the password contains "funny"
characters and hence is sent as a literal, then if the user is
invalid an error would occur before the password is parsed.
No such synchronization protection is provided for literals sent
from the server to the client, for performance reasons. Any
synchronization problems in this direction would be caused by a
bug in the client or server.
Sample IMAP2 session
The following is a transcript of an IMAP2 session. Server output is
identified by "S:" and client output by "U:". In cases where lines
are too long to fit within the boundaries of this document, the line
is continued on the next line.
S: * OK SUMEX-AIM.Stanford.EDU Interim Mail Access Protocol II Service
6.1(349) at Thu, 9 Jun 88 14:58:30 PDT
U: a001 login crispin secret
S: a002 OK User CRISPIN logged in at Thu, 9 Jun 88 14:58:42 PDT, job 76
U: a002 select inbox
S: * FLAGS (Bugs SF Party Skating Meeting Flames Request AI Question
Note \XXXX \YYYY \Answered \Flagged \Deleted \Seen)
S: * 16 EXISTS
S: * 0 RECENT
S: a002 OK Select complete
U: a003 fetch 16 all
S: * 16 Fetch (Flags (\Seen) InternalDate " 9-Jun-88 12:55:44 PDT"
RFC822.Size 637 Envelope ("Sat, 4 Jun 88 13:27:11 PDT"
"INFO-MAC Mail Message" (("Larry Fagan" NIL "FAGAN"
"SUMEX-AIM.Stanford.EDU")) (("Larry Fagan" NIL "FAGAN"
"SUMEX-AIM.Stanford.EDU")) (("Larry Fagan" NIL "FAGAN"
"SUMEX-AIM.Stanford.EDU")) ((NIL NIL "rindflEISCH"
"SUMEX-AIM.Stanford.EDU")) NIL NIL NIL
"<12403828905.13.FAGAN@SUMEX-AIM.Stanford.EDU>"))
S: a003 OK Fetch completed
U: a004 fetch 16 rfc822
S: * 16 Fetch (RFC822 {637}
S: Mail-From: RINDFLEISCH created at 9-Jun-88 12:55:43
S: Mail-From: FAGAN created at 4-Jun-88 13:27:12
S: Date: Sat, 4 Jun 88 13:27:11 PDT
S: From: Larry Fagan <FAGAN@SUMEX-AIM.Stanford.EDU>
S: To: rindflEISCH@SUMEX-AIM.Stanford.EDU
S: Subject: INFO-MAC Mail Message
S: Message-ID: <12403828905.13.FAGAN@SUMEX-AIM.Stanford.EDU>
S: ReSent-Date: Thu, 9 Jun 88 12:55:43 PDT
S: ReSent-From: TC Rindfleisch <Rindfleisch@SUMEX-AIM.Stanford.EDU>
S: ReSent-To: Yeager@SUMEX-AIM.Stanford.EDU,
Crispin@SUMEX-AIM.Stanford.EDU
S: ReSent-Message-ID:
<12405133897.80.RINDFLEISCH@SUMEX-AIM.Stanford.EDU>
S:
S: The file is <info-mac>usenetv4-55.arc ...
S: Larry
S: -------
S: )
S: a004 OK Fetch completed
U: a005 logout
S: * BYE DEC-20 IMAP II server terminating connection
S: a005 OK SUMEX-AIM.Stanford.EDU Interim Mail Access Protocol
Service logout
Implementation Discussion
There are several advantages to the scheme of tags and unsolicited
responses. First, the infamous synchronization problems of SMTP and
similar protocols do not happen with tagged commands; a command is
not considered satisfied until a response with the same tag is seen.
Tagging allows an arbitrary amount of other responses ("unsolicited"
data) to be sent by the server with no possibility of the client
losing synchronization. Compare this with the problems that FTP or
SMTP clients have with continuation, partial completion, and
commentary reply codes.
Another advantage is that a non-lockstep client implementation is
possible. The client could send a command, and entrust the handling
of the server responses to a different process that would signal the
client when the tagged response comes in. Under certain
circumstances, the client may have more than one command outstanding.
It was observed that synchronization problems can occur with literals
if the literal is not recognized as such. Fortunately, the cases in
which this can happen are rare; a mechanism (the special "+" tag
response) was introduced to handle those few cases. The proper way
to address this problem is probably to move towards a record-oriented
architecture instead of the text stream model provided by TCP.
An IMAP2 client must maintain a local cache of data from the mailbox.
This cache is an incomplete model of the mailbox, and at startup is
empty. A listener processes all unsolicited data, and updates the
cache based on this data. If a tagged response arrives, the listener
unblocks the process that sent the tagged request.
Unsolicited data needs some discussion. Unlike most protocols, in
which the server merely does the client's bidding, an IMAP2 server
has a semi-autonomous role. By sending "unsolicited data", the
server is in effect sending a command to the client -- to update or
extend the client's cache with new information from the server. In
other words, a "fetch" command is merely a request to the server to
ensure that the client's cache has the most up-to-date version of the
requested information. A server acknowledgement to the "fetch" is a
statement that all the requested data has been sent.
Although no current server does this, a server is not obliged by the
protocol to send data that it has already sent and is unchanged. An
exception to this is the actual message text fetching operations
(RFC822, RFC822.HEADER, and RFC822.TEXT), owing to the possibly
excessive resource consumption of maintaining this data in a cache.
It can not be assumed that a FETCH will transmit any data; only that
an OK to the FETCH means that the client's cache has the most up-to-
date information.
When a mailbox is selected, the initial unsolicited data from the
server arrives. The first piece of data is the number of messages.
By sending a new EXISTS unsolicited data message the server causes
the client to resize its cache (this is how newly arrived mail is
handled). If the client attempts to access information from the
cache, it will encounter empty spots that will trigger "fetch"
requests. The request would be sent, some unsolicited data including
the answer to the fetch will flow back, and then the "fetch" response
will unblock the client.
People familiar with demand-paged virtual memory operating system
design will recognize this model as being similar to page-fault
handling on a demand-paged system.
Formal Syntax
The following syntax specification uses the augmented Backus-Naur
Form (BNF) notation as specified in RFC 822 with one exception; the
delimiter used with the "#" construct is a single space (SP) and not
a comma.
address ::= "(" addr_name SP addr_adl SP addr_mailbox SP
addr_host ")"
addr_adl ::= nil / string
addr_host ::= nil / string
addr_mailbox ::= nil / string
addr_name ::= nil / string
bboard ::= "BBOARD" SP string
check ::= "CHECK"
copy ::= "COPY" SP sequence SP mailbox
data ::= ("FLAGS" SP flag_list / "SEARCH" SP 1#number /
"BYE" SP text_line / "OK" SP text_line /
"NO" SP text_line / "BAD" SP text_line)
date ::= string in form "dd-mmm-yy hh:mm:ss-zzz"
envelope ::= "(" env_date SP env_subject SP env_from SP
env_sender SP env_reply-to SP env_to SP
env_cc SP env_bcc SP env_in-reply-to SP
env_message-id ")"
env_bcc ::= nil / "(" 1*address ")"
env_cc ::= nil / "(" 1*address ")"
env_date ::= string
env_from ::= nil / "(" 1*address ")"
env_in-reply-to ::= nil / string
env_message-id ::= nil / string
env_reply-to ::= nil / "(" 1*address ")"
env_sender ::= nil / "(" 1*address ")"
env_subject ::= nil / string
env_to ::= nil / "(" 1*address ")"
expunge ::= "EXPUNGE"
fetch ::= "FETCH" SP sequence SP ("ALL" / "FAST" /
fetch_att / "(" 1#fetch_att ")")
fetch_att ::= "ENVELOPE" / "FLAGS" / "INTERNALDATE" /
"RFC822" / "RFC822.HEADER" / "RFC822.SIZE" /
"RFC822.TEXT"
find ::= "FIND" SP find_option SP string
find_option ::= "MAILBOXES" / "BBOARDS"
flag_list ::= ATOM / "(" 1#ATOM ")"
literal ::= "{" NUMBER "}" CRLF ASCII-STRING
login ::= "LOGIN" SP userid SP password
logout ::= "LOGOUT"
mailbox ::= "INBOX" / string
msg_copy ::= "COPY"
msg_data ::= (msg_exists / msg_recent / msg_expunge /
msg_fetch / msg_copy)
msg_exists ::= "EXISTS"
msg_expunge ::= "EXPUNGE"
msg_fetch ::= ("FETCH" / "STORE") SP "(" 1#("ENVELOPE" SP
envelope / "FLAGS" SP "(" 1#(recent_flag
flag_list) ")" / "INTERNALDATE" SP date /
"RFC822" SP string / "RFC822.HEADER" SP string /
"RFC822.SIZE" SP NUMBER / "RFC822.TEXT" SP
string) ")"
msg_recent ::= "RECENT"
msg_num ::= NUMBER
nil ::= "NIL"
noop ::= "NOOP"
password ::= string
recent_flag ::= "\RECENT"
ready ::= "+" SP text_line
request ::= tag SP (noop / login / logout / select / check /
expunge / copy / fetch / store / search / find /
bboard) CRLF
response ::= tag SP ("OK" / "NO" / "BAD") SP text_line CRLF
search ::= "SEARCH" SP 1#("ALL" / "ANSWERED" /
"BCC" SP string / "BEFORE" SP string /
"BODY" SP string / "CC" SP string / "DELETED" /
"FLAGGED" / "KEYWORD" SP atom / "NEW" / "OLD" /
"ON" SP string / "RECENT" / "SEEN" /
"SINCE" SP string / "TEXT" SP string /
"TO" SP string / "UNANSWERED" / "UNDELETED" /
"UNFLAGGED" / "UNKEYWORD" / "UNSEEN")
select ::= "SELECT" SP mailbox
sequence ::= NUMBER / (NUMBER "," sequence) / (NUMBER ":"
sequence)
store ::= "STORE" SP sequence SP store_att
store_att ::= ("+FLAGS" SP flag_list / "-FLAGS" SP flag_list /
"FLAGS" SP flag_list)
string ::= atom / """" 1*character """" / literal
system_flags ::= "\ANSWERED" SP "\FLAGGED" SP "\DELETED" SP
"\SEEN"
tag ::= atom
unsolicited ::= "*" SP (msg_num SP msg_data / data) CRLF
userid ::= string
Implementation Status
This information is current as of this writing.
The University of Washington has developed an electronic mail client
library called the "C-Client". It provides complete IMAP2, SMTP, and
local mailbox (both /usr/spool/mail and mail.txt formats) services in
a well-defined way to a user interface main program. Using the C-
Client, the University of Washington has created an operational
client for BSD Unix and two operational clients (one basic, one
advanced) for the NeXT.
Stanford University/SUMEX has developed operational IMAP2 clients for
Xerox Lisp machines, Texas Instruments Explorers, and the Apple
Macintosh. The core of the Macintosh client is an early version of
the C-Client. SUMEX has also developed IMAP2 servers for TOPS-20 and
BSD Unix.
All of the above software is in production use, with enthusiastic
local user communities. Active development continues on the
Macintosh and C-Client based clients and the BSD Unix server. This
software is freely available from the University of Washington and
SUMEX.
IMAP2 software exists for other platforms; for example Nippon
Telephone and Telegraph (NTT) has developed an operational IMAP2
client for the NTT ELIS. Several organizations are working on a PC
client.
IMAP2 can be used to access mailboxes at very remote sites, where
echo delays and frequent outages make TELNET and running a local mail
reader intolerable. For example, from a desktop workstation on the
University of Washington local network the author routinely uses
IMAP2 to read and manage mailboxes on various University of
Washington local servers, at two systems at Stanford University, at a
Milnet site, and at a site in Tokyo, Japan.
This specification does not make any formal definition of size
restrictions, but the DEC-20 server has the following limitations:
. length of a mailbox: 7,077,888 characters
. maximum number of messages: 18,432 messages
. length of a command line: 10,000 characters
. length of the local host name: 64 characters
. length of a "short" argument: 39 characters
. length of a "long" argument: 491,520 characters
. maximum amount of data output in a single fetch:
655,360 characters
To date, nobody has run up against any of these limitations, many of
which are substantially larger than most current user mail reading
programs.
Acknowledgements
Bill Yeager and Rich Acuff both contributed invaluable suggestions in
the evolution of IMAP2 from the original IMAP. James Rice pointed
out several ambiguities in the previous IMAP2 specification and
otherwise would not allow me to leave bad enough along. Laurence
Lundblade reviewed a draft of this version and made several helpful
suggestions.
Many dedicated individuals have worked on IMAP2 software, including:
Mark Crispin, Frank Gilmurray, Christopher Lane, Hiroshi Okuno,
Christopher Schmidt, and Bill Yeager.
Any mistakes, flaws, or sins of omission in this IMAP2 protocol
specification are, however, strictly my own; and the mention of any
name above does not imply an endorsement.
Security Considerations
Security issues are not discussed in this memo.
Author's Address
Mark R. Crispin
Panda Programming
6158 Lariat Loop NE
Bainbridge Island, WA 98110-2020
Phone: (206) 842-2385
EMail: mrc@Tomobiki-Cho.CAC.Washington.EDU