Network Working Group C. Newman Document: draft-ietf-lemonade-rfc2192bis-02.txt Sun Microsystems Expires: November 2006 A. Melnikov Intended category: Standards Track Isode Ltd. S. H. Maes Oracle Corporation May 2006 IMAP URL Scheme Status of this Memo By submitting this Internet-Draft, each author represents that any applicable patent or other IPR claims of which he or she is aware have been or will be disclosed, and any of which he or she becomes aware will be disclosed, in accordance with Section 6 of BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress". The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html. A revised version of this draft document will be submitted to the RFC editor as a Proposed Standard for the Internet Community. Discussion and suggestions for improvement are requested, and should be sent to the IMAPEXT Mailing list <ietf-imapext@imc.org>. Distribution of this draft is unlimited. Copyright Notice Copyright (C) The Internet Society (2006). Abstract Newman & Co Expires: November 2006 FORMFEED[Page 1]
INTERNET DRAFT IMAP URL Scheme May 2006 IMAP [IMAP4] is a rich protocol for accessing remote message stores. It provides an ideal mechanism for accessing public mail- ing list archives as well as private and shared message stores. This document defines a URL scheme for referencing objects on an IMAP server. 1. Conventions used in this document The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119. 2. IMAP scheme The IMAP URL scheme is used to designate IMAP servers, mailboxes, messages, MIME bodies [MIME], and search programs on Internet hosts accessible using the IMAP protocol. The IMAP URL follows the common Internet scheme syntax as defined in [URI-GEN] <<except that clear text passwords are not permit- ted>>. If :<port> is omitted, the port defaults to 143. An IMAP URL takes one of the following forms: imap://<iserver>/ imap://<iserver>/<enc-list-mailbox>;TYPE=<list-type> imap://<iserver>/<enc-mailbox>[uidvalidity][?<enc-search>] imap://<iserver>/<enc-mailbox>[uidvalidity]<iuid> [isection][ipartial] The first form is used to refer to an IMAP server, the second form refers to a list of mailboxes, the third form refers to the con- tents of a mailbox or a set of messages resulting from a search, and the final form refers to a specific message or message part, and possibly a byte range in that part. Note that the syntax here is informal. The authoritative formal syntax for IMAP URLs is defined in section 11. The partial specifier semantics conforms to [IMAP4] partial specifiers. 3. IMAP userinfo component Newman & Co Expires: November 2006 FORMFEED[Page 2]
INTERNET DRAFT IMAP URL Scheme May 2006 3.1. IMAP mailbox naming scope The "enc-user" part of the "iuserinfo" component, if present, denotes mailbox naming scope. If it is absent, the IMAP URL can only reference mailboxes with globally unique names, i.e. mailboxes with names that don't change depending on the user the client authenticated as to the IMAP server. <<AM: Note that not all IMAP implementations support globally unique names>> For example, a personal mailbox described by the following URL <imap://michael@example.org/INBOX> is most likely be different from a personal mailbox described by <imap://bester@example.org/INBOX>, even though both URLs use the same mailbox name. <<Chris: whenever possible the shared folder's name for a mailbox should be used to avoid mailbox naming scope issues. AM: There is no standard way of learning a globally unique name for a mailbox. However by looking at the NAMESPACE command result it is possible to see if a mailbox name is globally unique or not.>> 3.2. IMAP User Name and Authentication Mechanism The userinfo component [URI-GEN] of an IMAP URI may contains an IMAP user Name (a.k.a. authorization identity [SASL], "enc-user") and/or an authentication mechanism. (Note that the "enc-user" also defines a mailbox naming scope as described in section 3.1). They are used in the "LOGIN" or "AUTHENTICATE" commands after making the connection to the IMAP server. If no user name or authentication mechanism is supplied, the client must authenticate as anonymous to the server. If the server adver- tises AUTH=ANONYMOUS IMAP capability, the client MUST use the AUTHENTICATE command with ANONYMOUS [ANONYMOUS] SASL mechanism. If SASL ANONYMOUS is not available, the user name "anonymous" is used with the "LOGIN" command and the password is supplied as the Inter- net e-mail address of the end user accessing the resource. The lat- ter option is given in order to provide for interoperability with deployed servers. If the URL doesn't supply a user name, the program interpreting the IMAP URL SHOULD request one from the user (if it is an interactive program) or configuration. Note that as described in RFC 3501, the LOGIN command MUST NOT be used when the IMAP server advertises the LOGINDISABLED capability. An authentication mechanism can be expressed by adding ";AUTH=<enc- Newman & Co Expires: November 2006 FORMFEED[Page 3]
INTERNET DRAFT IMAP URL Scheme May 2006 auth-type>" to the end of the user name. When such an <enc-auth- type> is indicated, the client SHOULD request appropriate creden- tials from that mechanism and use the "AUTHENTICATE" command instead of the "LOGIN" command. If no user name is specified, one SHOULD be obtained from the mechanism or requested from the user as appropriate. The string ";AUTH=*" indicates that the client SHOULD select an appropriate authentication mechanism. It MAY use any mechanism listed in the response to the CAPABILITY command (or CAPABILITY response code) or use an out of band security service resulting in a PREAUTH connection. If no user name is specified and no appro- priate authentication mechanisms are available, the client SHOULD fall back to anonymous login as described above. This allows a URL which grants read-write access to authorized users, and read-only anonymous access to other users. If a user name is included with no authentication mechanism, then ";AUTH=*" is assumed. Since URLs can easily come from untrusted sources, care must be taken when resolving a URL which requires or requests any sort of authentication. If authentication credentials are supplied to the wrong server, it may compromise the security of the user's account. The program resolving the URL should make sure it meets at least one of the following criteria in this case: (1) The URL comes from a trusted source, such as a referral server which the client has validated and trusts according to site policy. Note that user entry of the URL may or may not count as a trusted source, depending on the experience level of the user and site pol- icy. (2) Explicit local site policy permits the client to connect to the server in the URL. For example, if the client knows the site domain name, site policy may dictate that any hostname ending in that domain is trusted. (3) The user confirms that connecting to that domain name with the specified credentials and/or mechanism is permitted. (4) A mechanism is used which validates the server before passing potentially compromising client credentials. (5) An authentication mechanism is used which will not reveal information to the server which could be used to compromise future connections. URLs which do not include a user name must be treated with extra care, since they are more likely to compromise the user's primary account. A URL containing ";AUTH=*" must also be treated with extra care since it might fall back on a weaker security mechanism. Newman & Co Expires: November 2006 FORMFEED[Page 4]
INTERNET DRAFT IMAP URL Scheme May 2006 Finally, clients are discouraged from using a plain text password as a fallback with ";AUTH=*" unless the connection has strong encryption. A program interpreting IMAP URLs MAY cache open connections to an IMAP server for later re-use. If a URL contains a user name, only connections authenticated as that user may be re-used. If a URL does not contain a user name or authentication mechanism, then only an anonymous connection may be re-used. If a URL contains an authentication mechanism without a user name, then any non-anony- mous connection may be re-used. Note that if unsafe or reserved characters such as " " or ";" are present in the user name or authentication mechanism, they MUST be encoded as described in [URI-GEN]. 3.3. Limitations of enc-user As per sections 3.1 and 3.2 the IMAP URI enc-user has two purposes: 1) It provides context for user-specific mailbox paths such as "INBOX" (section 3.1). 2) It specifies that resolution of the URL requires logging in as that user and limits use of that URL to only that user (section 3.2). An obvious limitation of using the same field for both purposes is that the URL can only be resolved by the mailbox owner. URLAUTH overrides the second purpose of the enc-user in the IMAP URI and by default permits the URI to be resolved by any user per- mitted by the access identifier. URLAUTH is described in section 7.1. 4. IMAP server An IMAP URL referring to an IMAP server has the following form: imap://<iserver>/ A program interpreting this URL would issue the standard set of commands it uses to present a view of the contents of an IMAP server. This is likely to be semanticly equivalent to one of the following URLs: imap://<iserver>/;TYPE=LIST imap://<iserver>/;TYPE=LSUB Newman & Co Expires: November 2006 FORMFEED[Page 5]
INTERNET DRAFT IMAP URL Scheme May 2006 See section 5 for details on how such URLs should be processed. The program interpreting this URL SHOULD use the LSUB form if it supports mailbox subscriptions. 5. Lists of mailboxes An IMAP URL referring to a list of mailboxes has the following form: imap://<iserver>/<enc-list-mailbox>;TYPE=<list-type> The <list-type> may be either "LIST" or "LSUB", and is case insen- sitive. The field ";TYPE=<list-type>" MUST be included. The <enc-list-mailbox> is any argument suitable for the list-mail- box field of the IMAP [IMAP4] LIST or LSUB commands. The field <enc-list-mailbox> may be omitted, in which case the program inter- preting the IMAP URL may use "*" or "%" as the <enc-list-mailbox>. The program SHOULD use "%" if it supports a hierarchical view, oth- erwise it SHOULD use "*". Note that if unsafe or reserved characters such as " " or "%" are present in <enc-list-mailbox> they MUST be encoded as described in [URI-GEN]. If the character "/" is present in enc-list-mailbox, it SHOULD NOT be encoded. If the <enc-list-mailbox> field is omitted or is defined to be "*" or "%" the following procedure for listing mailboxes is RECOM- MENDED: The program SHOULD first use the NAMESPACE [NAMESPACE] command (if supported by the server) to discover available namespaces, for example: C: A001 NAMESPACE S: * NAMESPACE (("" "/")) NIL (("Public Folders/" "/")) S: A001 OK NAMESPACE command completed After that it should issue a LIST/LSUB command (as described by <list-type>) for each namespace, suppressing redundant list com- mands if the server lists mailboxes for multiple namespaces as a result of a single list command. For example (continuing the exam- ple above), the client first issues <A002 LIST "" *>. If the server returns information for mailboxes under the shared namespace prefix "Public Folders/" the client should omit the subsequent <A003 LIST "Public Folders/" *> command. Newman & Co Expires: November 2006 FORMFEED[Page 6]
INTERNET DRAFT IMAP URL Scheme May 2006 If the NAMESPACE command is not supported by the IMAP server, the client SHOULD issue <tag <list-type> "" <enc-list-mailbox>>. <<Add suggestion to use "LIST %" level by level? Reference \HasChildren and \NoInferiors.>> 6. Lists of messages An IMAP URL referring to a list of messages has the following form: imap://<iserver>/<enc-mailbox>[uidvalidity][?<enc-search>] The <enc-mailbox> field is used as the argument to the IMAP4 "SELECT" command. Note that if unsafe or reserved characters such as " ", ";", or "?" are present in <enc-mailbox> they MUST be encoded as described in [URI-GEN]. If the character "/" is present in enc-mailbox, it SHOULD NOT be encoded. The [uidvalidity] field is optional. If it is present, it MUST be the argument to the IMAP4 UIDVALIDITY status response at the time the URL was created. This SHOULD be used by the program interpret- ing the IMAP URL to determine if the URL is stale. The [?<enc-search>] field is optional. If it is not present, the contents of the mailbox SHOULD be presented by the program inter- preting the URL. If it is present, it SHOULD be used as the argu- ments following an IMAP4 SEARCH command with unsafe characters such as " " (which are likely to be present in the <enc-search>) encoded as described in [URI-GEN]. Note that quoted strings and non-syn- chronizing literals [LITERAL+] are allowed in the <enc-search> con- tent, however synchronizing literals are not allowed, as their presence would effectively mean that the agent interpreting IMAP URLs need to parse an <enc-search> content, find all synchrnonizing literals and perform proper command continuation request handling (see sections 4.3 and 7 of [IMAP4]). 7. A specific message or message part An IMAP URL referring to a specific message or message part has the following form: imap://<iserver>/<enc-mailbox>[uidvalidity]<iuid> [isection][ipartial] The <enc-mailbox> and [uidvalidity] are as defined above. Newman & Co Expires: November 2006 FORMFEED[Page 7]
INTERNET DRAFT IMAP URL Scheme May 2006 If [uidvalidity] is present in this form, it SHOULD be used by the program interpreting the URL to determine if the URL is stale. The <iuid> refers to an IMAP4 message UID, and SHOULD be used as the <set> argument to the IMAP4 "UID FETCH" command. The [isection] field is optional. If not present, the URL refers to the entire Internet message as returned by the IMAP command "UID FETCH <uid> BODY.PEEK[]". If present, the URL refers to the object returned by a "UID FETCH <uid> BODY.PEEK[<section>]" command. The type of the object may be determined with a "UID FETCH <uid> BODYS- TRUCTURE" command and locating the appropriate part in the result- ing BODYSTRUCTURE. Note that unsafe characters in [isection] MUST be encoded as described in [URI-GEN]. The [ipartial] field is optional. If present, it effectively appends "<<partial-range>>" to the end of the UID FETCH BODY.PEEK[<section>] command constructed as described in the previ- ous paragraph. In other words it allows the client to request a byte range of the message/message part. 7.1 URLAUTH authorized URL <<URLAUTH are only supported by IMAP server advertising URLAUTH IMAP capability [URLAUTH].>> 7.1.1. Concepts 7.1.1.1. URLAUTH The URLAUTH is a component, appended at the end of a URL, which conveys authorization to access the data addressed by that URL. It contains an authorized access identifier, an authorization mecha- nism name, and an authorization token. The authorization token is generated from the URL, the authorized access identifer, authoriza- tion mechanism name, and a mailbox access key. 7.1.1.2. Mailbox Access Key The mailbox access key is a random string with at least 128 bits of Newman & Co Expires: November 2006 FORMFEED[Page 8]
INTERNET DRAFT IMAP URL Scheme May 2006 entropy. It is generated by software (not by the human user), and MUST be unpredictable. Each user has a table of mailboxes and an associated mailbox access key for each mailbox. Consequently, the mailbox access key is per- user and per-mailbox. In other words, two users sharing the same mailbox each have a different mailbox access key for that mailbox, and each mailbox accessed by a single user also has a different mailbox access key. 7.1.1.3. Authorized Access Identifier The authorized access identifier restricts use of the URLAUTH authorized URL to certain users authorized on the server, as described in section 7.1.2. 7.1.1.4. Authorization Mechanism The authorization mechanism is the algorithm by which the URLAUTH is generated and subsequently verified, using the mailbox access key. 7.1.1.5. Authorization Token The authorization token is a deterministic string of at least 128 bits which an entity with knowledge of the secret mailbox access key and URL authorization mechanism can use to verify the URL. 7.1.2. URLAUTH extensions to IMAP URL A specific message or message part IMAP URL can optionally contain ";EXPIRE=<datetime>" and ";URLAUTH=<access>:<mech>:<token>". The URLAUTH is comprised of ";URLAUTH=<access>:<mech>:<token>", and MUST be at the end of the URL. When ";EXPIRE=<datetime>" is used, this indicates the latest date and time that the URL is valid. After that date and time, the URL has expired and server implementations MUST reject the URL. If ";EXPIRE=<datetime>" is not used, the URL has no expiration, <<but Newman & Co Expires: November 2006 FORMFEED[Page 9]
INTERNET DRAFT IMAP URL Scheme May 2006 still can be revoked as discussed below>>. The URLAUTH takes the form ";URLAUTH=<access>:<mech>:<token>". It is composed of three parts. The <access> portion provides the authorized access identifiers which may constrain the operations and users that are permitted to use this URL. The <mech> portion provides the authorization mechanism used by the IMAP server to generate the authorization token that follows. The <token> portion provides authorization token. <<GENURLAUTH [URLAUTH].>> The "submit+" access identifier prefix, followed by a userid, indi- cates that only a userid authorized as a message submission entity on behalf of the specified userid is permitted to use this URL. The IMAP server does not validate the specified userid but does validate that the IMAP session has an authorization identity that is authorized as a message submission entity. The authorized mes- sage submission entity MUST validate the userid prior to contacting the IMAP server. The "user+" access identifier prefix, followed by a userid, indi- cates that use of this URL is limited to IMAP sessions which are logged in as the specified userid (that is, have authorization identity as that userid). Note: if a SASL mechanism which provides both authorization and authentication identifiers is used to authenticate to the IMAP server, the "user+" access identifier MUST match the authoriza- tion identifier. If the SASL mechanism can't transport the authorization identifier, the "user+" access identifier MUST match the authorization identifier derived from the authentica- tion identifier (see [SASL]). The "authuser" access identifier indicates that use of this URL is limited to IMAP sessions which are logged in as an authorized user (that is, have authorization identity as an authorized user) of that IMAP server. Use of this URL is prohibited to anonymous IMAP sessions. The "anonymous" access identifier indicates that use of this URL is not restricted by session authorization identity; that is, any IMAP session in authenticated or selected state (as defined in [IMAP4]), including anonymous sessions, may issue a URLFETCH [URLAUTH] using this URL. The authorization token is represented as an ASCII-encoded hexadec- imal string, which is used to authorize the URL. The length and the calculation of the authorization token depends upon the mecha- nism used; but, in all cases, the authorization token is at least Newman & Co Expires: November 2006 FORMFEED[Page 10]
INTERNET DRAFT IMAP URL Scheme May 2006 128 bits (and therefore at least 32 hexadecimal digits). 8. Relative IMAP URLs Relative IMAP URLs are permitted and are resolved according to the rules defined in [URI-GEN] with one exception. In IMAP URLs, parameters are treated as part of the normal path with respect to relative URL resolution. This is believed to be the behavior of the installed base and is likely to be documented in a future revi- sion of the relative URL specification. The following observations are also important: The <iauth> grammar element is considered part of the user name for purposes of resolving relative IMAP URLs. This means that unless a new login/server specification is included in the relative URL, the authentication mechanism is inherited from a base IMAP URL. URLs always use "/" as the hierarchy delimiter for the purpose of resolving paths in relative URLs. IMAP4 permits the use of any hierarchy delimiter in mailbox names. For this reason, relative mailbox paths will only work if the mailbox uses "/" as the hierar- chy delimiter. Relative URLs may be used on mailboxes which use other delimiters, but in that case, the entire mailbox name MUST be specified in the relative URL or inherited as a whole from the base URL. The base URL for a list of mailboxes or messages which was referred to by an IMAP URL is always the referring IMAP URL itself. The base URL for a message or message part which was referred to by an IMAP URL may be more complicated to determine. The program inter- preting the relative URL will have to check the headers of the MIME entity and any enclosing MIME entities in order to locate the "Con- tent-Location" header. This header is used to determine the base URL as defined in section 5 of [MHTML]. For example, if the refer- ring IMAP URL contains a "/;SECTION=1.2" parameter, then the MIME headers for section 1.2, for section 1, and for the enclosing mes- sage itself SHOULD be checked in that order for "Content-Location" headers. 9. Multinational Considerations IMAP4 [IMAP4] section 5.1.3 includes a convention for encoding non- US-ASCII characters in IMAP mailbox names. Because this convention is private to IMAP, it is necessary to convert IMAP's encoding to one that can be more easily interpreted by a URL display program. Newman & Co Expires: November 2006 FORMFEED[Page 11]
INTERNET DRAFT IMAP URL Scheme May 2006 For this reason, IMAP's modified UTF-7 encoding for mailboxes MUST be converted to UTF-8 [UTF-8]. Since 8-bit characters are not per- mitted in URLs, the UTF-8 characters are encoded as required by the URL specification [URI-GEN], section 2.1. Sample code is included in Appendix A to demonstrate this conversion. IMAP usernames are UTF-8 strings and MUST be encoded as required by the URL specification [URI-GEN], section 2.1. Also note that IMAP SEARCH criteria can contain non US-ASCII char- acters. 8-bit octets in those strings MUST be encoded as required by the URL specification [URI-GEN], section 2.1. 10. Examples The following examples demonstrate how an IMAP4 client program might translate various IMAP4 URLs into a series of IMAP4 commands. Commands sent from the client to the server are prefixed with "C:", and responses sent from the server to the client are prefixed with "S:". The URL: <imap://minbari.example.org/gray-council;UIDVALIDITY=385759045/; UID=20/;PARTIAL=0.1024> Results in the following client commands: <connect to minbari.example.org, port 143> C: A001 LOGIN ANONYMOUS sheridan@babylon5.example.org C: A002 SELECT gray-council <client verifies the UIDVALIDITY matches> C: A003 UID FETCH 20 BODY.PEEK[]<0.1024> The URL: <imap://michael@minbari.example.org/users.*;type=list> Results in the following client commands: <client requests password from user> <connect to minbari.example.org imap server, activate strong encryption> C: A001 LOGIN MICHAEL zipper C: A002 LIST "" users.* Newman & Co Expires: November 2006 FORMFEED[Page 12]
INTERNET DRAFT IMAP URL Scheme May 2006 The URL: <imap://psicorp.example.org/~peter/%E6%97%A5%E6%9C%AC%E8%AA%9E/ %E5%8F%B0%E5%8C%97> Results in the following client commands: <connect to psicorp.example.org, port 143> C: A001 LOGIN ANONYMOUS bester@psycop.psicorp.example.org C: A002 SELECT ~peter/&ZeVnLIqe-/&U,BTFw- <commands the client uses for viewing the contents of a mailbox> The URL: <imap://;AUTH=GSSAPI@minbari.example.org/gray-council/;uid=20/ ;section=1.2> Results in the following client commands: <connect to minbari.example.org, port 143> C: A001 AUTHENTICATE GSSAPI <authentication exchange> C: A002 SELECT gray-council C: A003 UID FETCH 20 BODY.PEEK[1.2] If the following relative URL is located in that body part: <;section=1.4> This could result in the following client commands: C: A004 UID FETCH 20 (BODY.PEEK[1.2.MIME] BODY.PEEK[1.MIME] BODY.PEEK[HEADER.FIELDS (Content-Base Content-Location)]) <Client looks for Content-Base or Content-Location headers in result. If no such headers, then it does the following> C: A005 UID FETCH 20 BODY.PEEK[1.4] Newman & Co Expires: November 2006 FORMFEED[Page 13]
INTERNET DRAFT IMAP URL Scheme May 2006 The URL: <imap://;AUTH=*@minbari.example.org/gray%20council? SUBJECT%20shadows> Could result in the following: <connect to minbari.example.org, port 143> C: A001 CAPABILITY S: * CAPABILITY IMAP4rev1 AUTH=DIGEST-MD5 S: A001 OK C: A002 AUTHENTICATE DIGEST-MD5 <authentication exchange> S: A002 OK user lennier authenticated C: A003 SELECT "gray council" ... C: A004 SEARCH SUBJECT shadows S: * SEARCH 8 10 13 14 15 16 S: A004 OK SEARCH completed C: A005 FETCH 8,10,13:16 ALL ... NOTE: In this final example, the client has implementation depen- dent choices. The authentication mechanism could be anything, including PREAUTH. And the final FETCH command could fetch more or less information about the messages, depending on what it wishes to display to the user. The URL: <imap://john;AUTH=*@minbari.example.org/babylon5/personel? charset%20UTF-8%20SUBJECT%20%7B14+%7D%0D%0A%D0%98%D0%B2% D0%B0%D0%BD%D0%BE%D0%B2%D0%B0> shows that 8-bit data can be sent using non-synchronizing literals [LITERAL+]. This could result in the following: <connect to minbari.example.org, port 143> C: A001 CAPABILITY S: * CAPABILITY IMAP4rev1 LITERAL+ AUTH=DIGEST-MD5 S: A001 OK C: A002 AUTHENTICATE DIGEST-MD5 <authentication exchange> S: A002 OK user john authenticated C: A003 SELECT babylon5/personel ... C: A004 SEARCH CHARSET UTF-8 SUBJECT {14+} C: XXXXXXXXXXXXXX S: * SEARCH 7 10 12 Newman & Co Expires: November 2006 FORMFEED[Page 14]
INTERNET DRAFT IMAP URL Scheme May 2006 S: A004 OK SEARCH completed C: A005 FETCH 7,10,12 ALL ... Where XXXXXXXXXXXXXX is 14 bytes of UTF-8 encoded data as specified in the URL above. 11. Security Considerations Security considerations discussed in the IMAP specification [IMAP4] and the URI specification [URI-GEN] are relevant. Security consid- erations related to authenticated URLs are discussed in section 3 of this document. Many email clients store the plain text password for later use after logging into an IMAP server. Such clients MUST NOT use a stored password in response to an IMAP URL without explicit permis- sion from the user to supply that password to the specified host name. 12. ABNF for IMAP URL scheme Formal syntax is defined using ABNF [ABNF], extending the ABNF rules in section 9 of [IMAP4]. Elements not defined here can be found in the [ABNF], [IMAP4], [IMAPABNF] or [URI-GEN]. Strings are not case sensitive and free insertion of linear-white-space is not permitted. sub-delims-sh = "!" / "$" / "'" / "(" / ")" / "*" / "+" / "," ;; <<same as sub-delims, but without ";", "&" ;; and "=">> uchar = unreserved / sub-delims-sh / pct-encoded achar = uchar / "&" / "=" ;; <<same as unreserved / sub-delims / ;; pct-encoded, but ";" is disallowed>> bchar = achar / ":" / "@" / "/" enc-auth-type = 1*achar ; encoded version of [IMAP4] "auth-type" Newman & Co Expires: November 2006 FORMFEED[Page 15]
INTERNET DRAFT IMAP URL Scheme May 2006 enc-list-mailbox = 1*bchar ; encoded version of [IMAP4] "list-mailbox" enc-mailbox = 1*bchar ; encoded version of [IMAP4] "mailbox" enc-search = 1*bchar ; encoded version of [IMAPABNF] ; "search-program". Note that IMAP4 ; literals may not be used in ; a "search-program", i.e. only ; quoted or non-synchronizing ; literals (if the server supports ; LITERAL+ [LITERAL+]) are allowed. enc-section = 1*bchar ; encoded version of [IMAP4] "section-spec" enc-user = 1*achar ; encoded version of [IMAP4] authorization identity ; or "userid". imapurl = "imap://" iserver "/" [ icommand ] <<The following 3 constructs are only to be used in the IMAP URLAUTH extension:>> authimapurl = "imap://" iserver "/" imessagepart ; <<"imapurl" with "[icommand] == imessagepart">> authimapurlfull = authimapurl iurlauth ; <<"imapurl" with "[icommand] == ; imessagepart iurlauth">> authimapurlrump = authimapurl iurlauth-rump <<What is the best way of having 2 formats - the full IMAP URL as used in clients and the rump URL which is to be URLAUTHed to produce a full IMAP URL?>> enc-urlauth = 32*HEXDIG Newman & Co Expires: November 2006 FORMFEED[Page 16]
INTERNET DRAFT IMAP URL Scheme May 2006 iurlauth = iurlauth-rump iua-verifier iua-verifier = ":" uauth-mechanism ":" enc-urlauth iurlauth-rump = [expire] ";URLAUTH=" access access = ("submit+" enc-user) / ("user+" enc-user) / "authuser" / "anonymous" expire = ";EXPIRE=" date-time ; date-time defined in [DATETIME] uauth-mechanism = "INTERNAL" / 1*(alpha / digit / "-" / ".") ; case-insensitive ; new mechanisms MUST be registered with IANA <<Should "alpha" and "digit" be in uppercase?>> <<was: just "mechanism">> iauth = ";AUTH=" ( "*" / enc-auth-type ) icommand = imailboxlist / imessagelist / imessagepart [iurlauth] imailboxlist = [enc-list-mailbox] ";TYPE=" list-type imailbox-ref = enc-mailbox [uidvalidity] ; <<new production in this draft>> imessagelist = imailbox-ref [ "?" enc-search ] ; "enc-search" is [URI-GEN] "query". imessagepart = imailbox-ref iuid [isection] [ipartial] ipartial = "/;PARTIAL=" partial-range isection = "/;SECTION=" enc-section iserver = [iuserinfo "@"] host [ ":" port ] ; This is the same as "authority" defined ; in [URI-GEN]. See [URI-GEN] for "host" ; and "port" definitions. Newman & Co Expires: November 2006 FORMFEED[Page 17]
INTERNET DRAFT IMAP URL Scheme May 2006 iuid = "/;UID=" nz-number ; See [IMAP4] for "nz-number" definition iuserinfo = enc-user [iauth] / [enc-user] iauth ; conforms to the generic syntax of ; "userinfo" as defined in [URI-GEN]. list-type = "LIST" / "LSUB" partial-range = number ["." nz-number] ; partial fetch uidvalidity = ";UIDVALIDITY=" nz-number ; See [IMAP4] for "nz-number" definition 13. IANA Considerations IANA is requested to update "imap" definition in the "Uniform Resource Identifier scheme registry" to point to this document. The registration template (as per <<draft-hansen-2717bis-2718bis- uri-guidelines-XX.txt>>) is specified in section 13.1 of this docu- ment. 13.1. IANA Registration of imap: URI Scheme This section provides the information required to register the imap: URI scheme. URI scheme name: imap Status: permanent URI scheme syntax: See section 12 of this RFC. URI scheme semantics: The imap: URI scheme is used to designate IMAP servers, mail- boxes, messages, MIME bodies [MIME], and search programs on Inter- net hosts accessible using the IMAP protocol. <<Discuss configura- tion storage versa access here?>> There is no MIME type associated with this URI. Newman & Co Expires: November 2006 FORMFEED[Page 18]
INTERNET DRAFT IMAP URL Scheme May 2006 Encoding considerations: See Section 9 of <<[this RFC]>>. Applications/protocols that use this URI scheme name: The imap: URI is intended to be used by applications that might need access to IMAP mailstore. Such applications may include (but not limited to) IMAP-capable web browsers; IMAP clients that wish to access a mailbox, message, or edit a message on the server using [CATENATE]; [SUBMIT] clients and servers that are requested to assemble a complete message on submission using [BURL]. Interoperability considerations: A widely deployed IMAP client Mozilla/Thubderbird use a differ- ent imap: scheme internally. <<Is this relevant?>> Security considerations: See Security Considerations (Section 11) of <<[this RFC]>>. Contact: Alexey Melnikov <alexey.melnikov@isode.com> Author/Change controller: IESG References: [This RFC] and [IMAP4]. 14. References 14.1. Normative References [KEYWORDS] Bradner, "Key words for use in RFCs to Indicate Require- ment Levels", RFC 2119, Harvard University, March 1997. [IMAP4] Crispin, M., "Internet Message Access Protocol - Version 4rev1", RFC 3501, University of Washington, March 2003. Newman & Co Expires: November 2006 FORMFEED[Page 19]
INTERNET DRAFT IMAP URL Scheme May 2006 [IMAPABNF] Melnikov, A., and C. Daboo, "Collected extensions to IMAP4 ABNF", work in progress, draft-melnikov-imap-ext-abnf-XX.txt. [MHTML] Palme, J., Hopmann, A. and N. Shelness, "MIME Encapsulation of Aggregate Documents, such as HTML (MHTML)", RFC 2557, March 1999. [ABNF] Crocker, Overell, "Augmented BNF for Syntax Specifications: ABNF", RFC 4234, October 2005. [MIME] Freed, N., Borenstein, N., "Multipurpose Internet Mail Extensions", RFC 2045, Innosoft, First Virtual, November 1996. [URI-GEN] Berners-Lee, T., Fielding, R. and L. Masinter, "Uniform Resource Identifier (URI): Generic Syntax", RFC 3986, January 2005. [UTF-8] Yergeau, F., "UTF-8, a transformation format of ISO 10646", STD 63, RFC 3629, November 2003. [NAMESPACE] Gahrns, M. and C. Newman, "IMAP4 Namespace", RFC 2342, May 1998. [LITERAL+] Myers, J., "IMAP4 non-synchronizing literals", RFC 2088, January 1997. [ANONYMOUS] K. Zeilenga (Ed.), "The Anonymous SASL Mechanism", work in progress, draft-ietf-sasl-anon-XX.txt (Obsoletes RFC 2245). [DATETIME] Klyne, G., and Newman, C., "Date and Time on the Inter- net: Timestamps", RFC 3339, July 2002. 14.2. Informative References [SUBMIT] Gellens, R. and Klensin, J., "Message Submission for Mail", draft-gellens-submit-bis-02.txt. [BURL] Newman, C. "Message Composition", work in progress, draft- ietf-lemonade-burl-XX.txt. [CATENATE] Resnick, P. "Internet Message Access Protocol (IMAP) CATENATE Extension", work in progress, draft-ietf-lemonade-cate- nate-XX.txt. Newman & Co Expires: November 2006 FORMFEED[Page 20]
INTERNET DRAFT IMAP URL Scheme May 2006 [SASL] <<Myers, J., "Simple Authentication and Security Layer (SASL)", RFC 2222, October 1997>>. 15. Author's Address Chris Newman Sun Microsystems 3401 Centrelake Dr., Suite 410 Ontario, CA 91761 EMail: chris.newman@sun.com Alexey Melnikov (Editor) Isode Limited 5 Castle Business Village 36 Station Road Hampton, Middlesex TW12 2BX, UK Email: Alexey.Melnikov@isode.com URI: http://www.melnikov.ca/ Stephane H. Maes (Editor) Oracle Corporation 500 Oracle Parkway M/S 4op634 Redwood Shores, CA 94065 USA Phone: +1-650-607-6296 Email: stephane.maes@oracle.com Appendix A. Sample code Here is sample C source code to convert between URL paths and IMAP mail- box names, taking into account mapping between IMAP's modified UTF-7 [IMAP4] and hex-encoded UTF-8 which is more appropriate for URLs. This code has not been rigorously tested nor does it necessarily behave rea- sonably with invalid input, but it should serve as a useful example. This code just converts the mailbox portion of the URL and does not deal with parameters, query or server components of the URL. <<Does the code comply with the new URI rules?>> #include <stdio.h> #include <string.h> /* hexadecimal lookup table */ Newman & Co Expires: November 2006 FORMFEED[Page 21]
INTERNET DRAFT IMAP URL Scheme May 2006 static char hex[] = "0123456789ABCDEF"; /* URL unsafe printable characters */ static char urlunsafe[] = " \"#%&+:;<=>?@[\\]^`{|}"; /* UTF7 modified base64 alphabet */ static char base64chars[] = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+,"; #define UNDEFINED 64 /* UTF16 definitions */ #define UTF16MASK 0x03FFUL #define UTF16SHIFT 10 #define UTF16BASE 0x10000UL #define UTF16HIGHSTART 0xD800UL #define UTF16HIGHEND 0xDBFFUL #define UTF16LOSTART 0xDC00UL #define UTF16LOEND 0xDFFFUL /* Convert an IMAP mailbox to a URL path * dst needs to have roughly 4 times the storage space of src * Hex encoding can triple the size of the input * UTF-7 can be slightly denser than UTF-8 * (worst case: 8 octets UTF-7 becomes 9 octets UTF-8) */ void MailboxToURL(char *dst, char *src) { unsigned char c, i, bitcount; unsigned long ucs4, utf16, bitbuf; unsigned char base64[256], utf8[6]; /* initialize modified base64 decoding table */ memset(base64, UNDEFINED, sizeof (base64)); for (i = 0; i < sizeof (base64chars); ++i) { base64[base64chars[i]] = i; } /* loop until end of string */ while (*src != '\0') { c = *src++; /* deal with literal characters and &- */ if (c != '&' || *src == '-') { if (c < ' ' || c > '~' || strchr(urlunsafe, c) != NULL) { /* hex encode if necessary */ dst[0] = '%'; dst[1] = hex[c >> 4]; dst[2] = hex[c & 0x0f]; dst += 3; Newman & Co Expires: November 2006 FORMFEED[Page 22]
INTERNET DRAFT IMAP URL Scheme May 2006 } else { /* encode literally */ *dst++ = c; } /* skip over the '-' if this is an &- sequence */ if (c == '&') ++src; } else { /* convert modified UTF-7 -> UTF-16 -> UCS-4 -> UTF-8 -> HEX */ bitbuf = 0; bitcount = 0; ucs4 = 0; while ((c = base64[(unsigned char) *src]) != UNDEFINED) { ++src; bitbuf = (bitbuf << 6) | c; bitcount += 6; /* enough bits for a UTF-16 character? */ if (bitcount >= 16) { bitcount -= 16; utf16 = (bitcount ? bitbuf >> bitcount : bitbuf) & 0xffff; /* convert UTF16 to UCS4 */ if (utf16 >= UTF16HIGHSTART && utf16 <= UTF16HIGHEND) { ucs4 = (utf16 - UTF16HIGHSTART) << UTF16SHIFT; continue; } else if (utf16 >= UTF16LOSTART && utf16 <= UTF16LOEND) { ucs4 += utf16 - UTF16LOSTART + UTF16BASE; } else { ucs4 = utf16; } /* convert UTF-16 range of UCS4 to UTF-8 */ if (ucs4 <= 0x7fUL) { utf8[0] = ucs4; i = 1; } else if (ucs4 <= 0x7ffUL) { utf8[0] = 0xc0 | (ucs4 >> 6); utf8[1] = 0x80 | (ucs4 & 0x3f); i = 2; } else if (ucs4 <= 0xffffUL) { utf8[0] = 0xe0 | (ucs4 >> 12); utf8[1] = 0x80 | ((ucs4 >> 6) & 0x3f); utf8[2] = 0x80 | (ucs4 & 0x3f); i = 3; } else { utf8[0] = 0xf0 | (ucs4 >> 18); utf8[1] = 0x80 | ((ucs4 >> 12) & 0x3f); utf8[2] = 0x80 | ((ucs4 >> 6) & 0x3f); Newman & Co Expires: November 2006 FORMFEED[Page 23]
INTERNET DRAFT IMAP URL Scheme May 2006 utf8[3] = 0x80 | (ucs4 & 0x3f); i = 4; } /* convert utf8 to hex */ for (c = 0; c < i; ++c) { dst[0] = '%'; dst[1] = hex[utf8[c] >> 4]; dst[2] = hex[utf8[c] & 0x0f]; dst += 3; } } } /* skip over trailing '-' in modified UTF-7 encoding */ if (*src == '-') ++src; } } /* terminate destination string */ *dst = '\0'; } /* Convert hex coded UTF-8 URL path to modified UTF-7 IMAP mailbox * dst should be about twice the length of src to deal with non-hex * coded URLs */ void URLtoMailbox(char *dst, char *src) { unsigned int utf8pos, utf8total, i, c, utf7mode, bitstogo, utf16flag; unsigned long ucs4, bitbuf; unsigned char hextab[256]; /* initialize hex lookup table */ memset(hextab, 0, sizeof (hextab)); for (i = 0; i < sizeof (hex); ++i) { hextab[hex[i]] = i; if (isupper(hex[i])) hextab[tolower(hex[i])] = i; } utf7mode = 0; utf8total = 0; bitstogo = 0; while ((c = *src) != '\0') { ++src; /* undo hex-encoding */ if (c == '%' && src[0] != '\0' && src[1] != '\0') { c = (hextab[src[0]] << 4) | hextab[src[1]]; src += 2; } /* normal character? */ Newman & Co Expires: November 2006 FORMFEED[Page 24]
INTERNET DRAFT IMAP URL Scheme May 2006 if (c >= ' ' && c <= '~') { /* switch out of UTF-7 mode */ if (utf7mode) { if (bitstogo) { *dst++ = base64chars[(bitbuf << (6 - bitstogo)) & 0x3F]; } *dst++ = '-'; utf7mode = 0; } *dst++ = c; /* encode '&' as '&-' */ if (c == '&') { *dst++ = '-'; } continue; } /* switch to UTF-7 mode */ if (!utf7mode) { *dst++ = '&'; utf7mode = 1; } /* Encode US-ASCII characters as themselves */ if (c < 0x80) { ucs4 = c; utf8total = 1; } else if (utf8total) { /* save UTF8 bits into UCS4 */ ucs4 = (ucs4 << 6) | (c & 0x3FUL); if (++utf8pos < utf8total) { continue; } } else { utf8pos = 1; if (c < 0xE0) { utf8total = 2; ucs4 = c & 0x1F; } else if (c < 0xF0) { utf8total = 3; ucs4 = c & 0x0F; } else { /* NOTE: can't convert UTF8 sequences longer than 4 */ utf8total = 4; ucs4 = c & 0x03; } continue; } /* loop to split ucs4 into two utf16 chars if necessary */ utf8total = 0; Newman & Co Expires: November 2006 FORMFEED[Page 25]
INTERNET DRAFT IMAP URL Scheme May 2006 do { if (ucs4 >= UTF16BASE) { ucs4 -= UTF16BASE; bitbuf = (bitbuf << 16) | ((ucs4 >> UTF16SHIFT) + UTF16HIGHSTART); ucs4 = (ucs4 & UTF16MASK) + UTF16LOSTART; utf16flag = 1; } else { bitbuf = (bitbuf << 16) | ucs4; utf16flag = 0; } bitstogo += 16; /* spew out base64 */ while (bitstogo >= 6) { bitstogo -= 6; *dst++ = base64chars[(bitstogo ? (bitbuf >> bitstogo) : bitbuf) & 0x3F]; } } while (utf16flag); } /* if in UTF-7 mode, finish in ASCII */ if (utf7mode) { if (bitstogo) { *dst++ = base64chars[(bitbuf << (6 - bitstogo)) & 0x3F]; } *dst++ = '-'; } /* tie off string */ *dst = '\0'; } Appendix B. List of changes since RFC 2192 <<Needs checking and updating>> Updated boilerplate, list of editor's, etc. Updated references. Updated ABNF not to use _, to use SP instead of SPACE. Updated example domains to use example.org. Fixed ABNF error in "imessagelist" non-terminal. Updated ABNF, due to changes in RFC 3501, IMAPABNF and RFC 3986. Renamed "iuserauth" non-terminal to "iuserinfo". Clarified that the userinfo component describes both authorization identity and mailbox naming scope. Allow for non-synchronizing literals in "enc-search". Added "ipartial" specifier that denotes a partial fetch. Moved URLAUTH text from RFC <<XXXX>> to this document. Newman & Co Expires: November 2006 FORMFEED[Page 26]
INTERNET DRAFT IMAP URL Scheme May 2006 Intellectual Property The IETF takes no position regarding the validity or scope of any Intellectual Property Rights or other rights that might be claimed to pertain to the implementation or use of the technology described in this document or the extent to which any license under such rights might or might not be available; nor does it represent that it has made any independent effort to identify any such rights. Information on the procedures with respect to rights in RFC documents can be found in BCP 78 and BCP 79. Copies of IPR disclosures made to the IETF Secretariat and any assurances of licenses to be made available, or the result of an attempt made to obtain a general license or permission for the use of such proprietary rights by implementers or users of this specification can be obtained from the IETF on-line IPR repository at http://www.ietf.org/ipr. The IETF invites any interested party to bring to its attention any copyrights, patents or patent applications, or other proprietary rights that may cover technology that may be required to implement this standard. Please address the information to the IETF at ietf-ipr@ietf.org. Full Copyright Statement Copyright (C) The Internet Society (2006). This document is subject to the rights, licenses and restrictions contained in BCP 78, and except as set forth therein, the authors retain all their rights. This document and the information contained herein are provided on an "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Acknowledgement Funding for the RFC Editor function is currently provided by the Internet Society. Newman & Co Expires: November 2006 FORMFEED[Page 27]
