SIEVE Email Filtering: Notifications
draft-ietf-sieve-notify-09

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.

This Internet-Draft will expire on April 7, 2008.

Abstract

Users go to great lengths to be notified as quickly as possible that they have received new mail. Most of these methods involve polling to check for new messages periodically. A push method handled by the final delivery agent gives users quicker notifications and saves server resources. This document does not specify the notification method but it is expected that using existing instant messaging infrastructure such as XMPP, or SMS messages will be popular. This draft describes an extension to the Sieve mail filtering language that allows users to give specific rules for how and when notifications should be sent.

Changes since draft-ietf-sieve-notify-08

• Added missing IANA registry for notification methods.

Changes since draft-ietf-sieve-notify-07

• Added a new "set" modifier for URL percent-encoding.
• Clarified that notification methods must address notification loops.
• Added an implementation consideration for implementations that use URIs internally.

Changes since draft-ietf-sieve-notify-06

• Remove extract_text. The WG consensus was to move it to another document, such as Sieve MIME loops.
• Deleted markers for open issues from the document.
• Clarified that a notification mechanism can treat some URI parameters as an error.
• Added notify_method_capability test and example.
• Minor corrections to the IANA registration as a result of other changes.

Changes since draft-ietf-sieve-notify-05

• Fixed XMPP URI in one example.
• Addressed Michael's issue with how timestamp are described.
• Renamed "valid_notif_method" to "valid_notify_method".
• Added text about truncation of a textual part when it is stored in a variable using extract_text.
• Changed tagged :method argument to positional argument.
• Added text about notification throttling, identifying notification source and restricting values of the :from parameter.
• Added a requirement on documents describing notification methods to list which URI parameters must be ignored.

Changes since draft-ietf-sieve-notify-04

• Made notification method required.
• Defined "mailto" as a mandatory-to-implement method.
• Added normative reference to mailto.
• Clarified that :importance may be treated as a transport indicator.
• Clarified that :importance value can be included in the default :message, if one is not specified.
• Made the default :message implementation specific.
• Renamed the capability name from "notify" to "enotify"
• Updated IANA registration.
• Moved text about ManageSieve capability to the ManageSieve document itself.
• Removed reference to IANA registry for options.
• Some miscellaneous text cleanup and clarification.

Changes since draft-ietf-sieve-notify-03

• Added a warning that "notify" must not be used as a crappy form of "redirect".
• Added a warning about using "notify" to forward confidential information in order to bypass organization's policy.
• Fixed syntax of the :options argument - it is a string list, each string containing "<attribute>=<value>"
• Renamed :priority to :importance
• Cleaned up section about requirements on methods.

Changes since draft-ietf-sieve-notify-02

• Added :from tagged argument.
• Added Extract_text action, which allows to extract content of the first text/* part.
• Added back the ":options" parameter to the notify action.
• Added new section talking about requirements on notification method specs.
• Added more examples.

Changes since draft-ietf-sieve-notify-00

• Updated references, etc.
• Added IANA considerations section.
• Removed denotify action.
• Updated examples to use the variables extension.
• Replaced notification method with URI.
• Removed text suggesting that this extension can be used to track all Sieve actions taken.
• Changed priority to be a string.
• Added text about URI verification.
• Clarified that a notification method is allowed to perform adaptation of notification context (e.g. truncation, charset conversion, etc.). These adaptations must be documented in a document describing the notification method.
• Clarified that notify is compatible with all existing actions.
• Removed the :id parameter to the notify action.
• Added valid_notif_method test that allows to test if an notification method (URI) is supported.
• Added a new capability response to ManageSieve that allows to report supported notification types.

Table of Contents

1.  Introduction
1.1.  Conventions used in this document

2.  Capability Identifier

3.  Notify Action
3.1.  Notify Action Syntax and Semantics
3.2.  Notify parameter "method"
3.3.  Notify tag ":from"
3.4.  Notify tag ":importance"
3.5.  Notify tag ":options"
3.6.  Notify tag ":message"
3.7.  Examples
3.8.  Requirements on notification methods specifications

4.  Test valid_notify_method

5.  Test notify_method_capability

6.  Modifier encodeurl to the 'set' action

7.  Interactions with Other Sieve Actions

8.  Security Considerations

9.  IANA Considerations
9.1.  Registration of Sieve extension
9.2.  New registry for Sieve notification mechanisms

10.  Acknowledgements

11.  References
11.1.  Normative References
11.2.  Informative References

§  Authors' Addresses
§  Intellectual Property and Copyright Statements

1.  Introduction

This is an extension to the Sieve language defined by [Sieve] (Guenther, P. and T. Showalter, “Sieve: An Email Filtering Language,” August 2006.) for providing instant notifications. It defines the new action "notify".

This document does not specify the notification methods. Examples of possible notification methods are email and XMPP. To allow a mechanism for portability of scripts that use notifications, implementation of the [MailTo] (Leiba, B. and M. Haardt, “Sieve Notification Mechanism: mailto,” October 2006.) method is mandatory. Other available methods shall depend upon the implementation and configuration of the system.

1.1.  Conventions used in this document

Conventions for notations are as in [Sieve] (Guenther, P. and T. Showalter, “Sieve: An Email Filtering Language,” August 2006.) section 1.1, including the use of [ABNF] (Crocker, D., Ed. and P. Overell, “Augmented BNF for Syntax Specifications: ABNF,” October 2005.).

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 [Kwds] (Bradner, S., “Key words for use in RFCs to Indicate Requirement Levels,” March 1997.).

2.  Capability Identifier

The capability string associated with the extension defined in this document is "enotify".

3.  Notify Action

3.1.  Notify Action Syntax and Semantics

Usage:

notify [":from" string] [":importance" <"1" / "2" / "3">]
[":options" string-list]
[":message" string]
<method: string>

The Notify action specifies that a notification should be sent to a user. The format of the notification is implementation-defined and is also affected by the notification method used (see Section 3.2 (Notify parameter "method")). However, all content specified in the :message parameter SHOULD be included.

3.2.  Notify parameter "method"

The method positional parameter identifies the notification method that will be used; it is a URI [URI] (Berners-Lee, T., Fielding, R., and L. Masinter, “Uniform Resource Identifier (URI): Generic Syntax,” January 2005.). For example, the notification method can be an SMS URI [SMS‑URI] (Wilde, E. and A. Vaha-Sipila, “URI scheme for GSM Short Message Service,” August 2005.) containing a phone number, or an XMPP [XMPP] (Saint-Andre, Ed., P., “Extensible Messaging and Presence Protocol (XMPP): Core,” October 2004.) URI containing an XMPP identifier [XMPP‑URI] (Saint-Andre, P., “Internationalized Resource Identifiers (IRIs) and Uniform Resource Identifiers (URIs) for the Extensible Messaging and Presence Protocol (XMPP),” September 2005.).

The supported URI values will be site-specific, but support for the [MailTo] (Leiba, B. and M. Haardt, “Sieve Notification Mechanism: mailto,” October 2006.) method is REQUIRED in order to insure interoperability. If a URI schema is specified that the implementation does not support, the notification MUST cause an error condition. Sieve scripts can check the supported methods using the "valid_notify_method" test to be sure that they only use supported ones, to avoid such error conditions.

If the method parameter contains a supported URI schema, then the URI MUST be checked for syntactic validity. An invalid URI syntax or an unsupported URI extension MUST cause an error. An implementation MAY enforce other semantic restrictions on URIs -- for example an SMS URI can only contain phone numbers in a particular geographical region -- and will treat violations of such semantic restrictions as errors.

3.3.  Notify tag ":from"

A ":from" parameter may be used to specify an author of the notification. The syntax of this parameter's value is method-specific. Implementations SHOULD check the syntax according to the notification method specification and generate an error when a syntactically invalid ":from" parameter is specified. In order to minimize/prevent forgery of the author value, implementations SHOULD impose restrictions on what values can specified in a ":from" parameter; it is suggested that values which fail such a validity check simply be ignored rather than causing the notify action to fail.

3.4.  Notify tag ":importance"

The :importance tag specifies the importance of the delivery of the notification. The :importance tag is followed by a numeric value represented as a string: "1" (high importance), "2" (normal importance), and "3" (low importance). If no importance is given, the default value "2" SHOULD be assumed. A notification method can treat the importance value as a transport indicator. For example, it might deliver notifications of high importance quicker than notifications of normal or low importance. Some notification methods allow users to specify their state of activity (for example "busy" or "away from keyboard"). If the notification method provides this information it SHOULD be used to selectively send notifications. If, for example, the user marks herself as "busy", a notification method can require that a notification with importance of "3" is not to be sent, however the user should be notified of a notification with higher importance.

If the notification method allows users to filter messages based upon certain parameters in the message, users SHOULD be able to filter based upon importance. If the notification method does not support importance, then this parameter MUST be ignored. An implementation MAY include the importance value in the default message Section 3.6 (Notify tag ":message"), if one is not provided.

3.5.  Notify tag ":options"

The :options tag is used to send additional parameters to the notification method. Interpretation of the parameters is method-specific. This document doesn't specify any such additional parameter.

Each string in the options string list has the following syntax:
"<optionname>=<value>"

3.6.  Notify tag ":message"

The :message tag specifies the message data to be included in the notification. The entirety of the string SHOULD be sent but implementations MAY shorten the message for technical or aesthetic reasons. If the message parameter is absent, a default implementation-specific message is used. Unless specified otherwise by a particular notification mechanism, an implementation default containing at least the value of the "From" header field and the value of the "Subject" header field is RECOMMENDED.

In order to construct more complex messages the notify extension can be used together with the Sieve variables extension [Variables] (Homme, K., “Sieve Extension: Variables,” December 2005.), as shown in the examples below.

3.7.  Examples

Example 1:
    require ["enotify", "fileinto", "variables"];

    if header :contains "from" "boss@example.org" {
        notify :importance "1"
            :message "This is probably very important"
            "mailto:alm@example.com";
        # Don't send any further notifications
        stop;
    }

    if header :contains "to" "sievemailinglist@example.org" {
        # :matches is used to get the value of the Subject header
        if header :matches "Subject" "*" {
            set "subject" "${1}";
        }

        # :matches is used to get the value of the From header
        if header :matches "From" "*" {
            set "from" "${1}";
        }

        notify :importance "3"
            :message "[SIEVE] ${from}: ${subject}"
            "mailto:alm@example.com";
        fileinto "INBOX.sieve";
    }

Example 2:
    require ["enotify", "fileinto", "variables", "envelope"];

    if header :matches "from" "*@*.example.org" {
        # :matches is used to get the MAIL FROM address
        if envelope :all :matches "from" "*" {
            set "env_from" " [really: ${1}]";
        }

        # :matches is used to get the value of the Subject header
        if header :matches "Subject" "*" {
            set "subject" "${1}";
        }

        # :matches is used to get the address from the From header
        if address :matches :all "from" "*" {
            set "from_addr" "${1}";
        }

        notify :message "${from_addr}${env_from}: ${subject}"
            "mailto:alm@example.com";
    }

Example 3:
    require ["enotify", "variables"];

    set "notif_method"
    "xmpp:tim@example.com?message;subject=SIEVE;body=You%20got%20mail";

    if header :contains "subject" "Your dog" {
        set "notif_method" "sms:+14085551212";
    }

    if header :contains "to" "sievemailinglist@example.org" {
        set "notif_method" "";
    }

    if not string :is "${notif_method}" "" {
        notify "${notif_method}";
    }

    if header :contains "from" "boss@example.org" {
        # :matches is used to get the value of the Subject header
        if header :matches "Subject" "*" {
            set "subject" "${1}";
        }

        # don't need high importance notification for
        # a 'for your information'
        if not header :contains "subject" "FYI:" {
            notify :importance "1" :message "BOSS: ${subject}"
                   "sms:+14085551212";
        }
    }

3.8.  Requirements on notification methods specifications

This section describes requirements for documents that define specific Sieve notification methods.

A notification method MAY allow modification of the final notification text -- for example, truncating it if it exceeds a length limit, or modifying characters that can not be represented in the target character set. Characters in the notification text which can't be represented by the notification method SHOULD be replaced with a symbol indicating an unknown character. Allowed modifications MUST be documented in the document describing the notification method.

A notification method MAY ignore parameters specified in the Notify action.

A notification method MAY recommend the default message value to be used if the :message argument is not specified.

Notifications SHOULD include timestamps, if the notification method allows for their transmission outside of the textual message. Implementation methods which can only transmit timestamps in the textual message MAY include them in the textual message.

A notification SHOULD include means to identify/track its origin, in order to allow a recipient to stop notifications or find out how to contact the sender. This requirement is to help tracking a misconfigured or abusive origin of notifications.

Methods SHOULD NOT include any other extraneous information not specified in parameters to the notify action.

Methods MUST specify which URI parameters (if any) must be ignored, which ones must be used in the resulting notification and which ones must cause an error.

Methods MUST specify what values are returned by the notify_method_capability test Section 5 (Test notify_method_capability).

If there are errors sending the notification, the Sieve interpreter SHOULD ignore the notification and not retry indefinitely. The Sieve interpreter MAY throttle notifications; if it does, a request to send a notification can be silently ignored. Documents describing notification methods SHOULD describe how retries, throttling, duplicate suppression (if any), etc. are to be handled by implementations.

4.  Test valid_notify_method

Usage:

valid_notify_method <notification-uris: string-list>

The "valid_notify_method" test is true if the notification methods listed in the notification-uris argument are supported and they are syntactically (including URI parameters) and semantically (including implementation-specific semantic restrictions) valid. This test MUST perform exactly the same validation as would be performed on the "method" parameter to the "notify" action.

The test is true only if ALL of the listed notification methods are supported and valid.

Example 4 (partial):
          if not valid_notify_method ["mailto:",
                  "http://gw.example.net/notify?test"] {
              stop;
          }

5.  Test notify_method_capability

Usage:

notify_method_capability [COMPARATOR] [MATCH-TYPE] <notification-uri: string>
<notification-capability: string>
<key-list: string-list>

The "notify_method_capability" test retrieves the notification capability specified by the notification-capability string that is specific to the notification-uri and matches it to the values specified in the key-list. The test succeeds if a match occurs. The type of match defaults to ":is" and the default comparator is "i;ascii-casemap".

The notification-capability is case insensitive.

The notify_method_capability test MUST fail unconditionally if the specified notification-uri is syntactically invalid (as determined by the valid_notify_method test Section 4 (Test valid_notify_method)) or specifies an unsupported notification method. However this MUST NOT cause an error.

The notify_method_capability test MUST fail unconditionally if the specified notification-capability item does not exist. A script MUST NOT fail with an error if the item does not exist. This allows scripts to be written that handle nonexistent items gracefully.

This document defines a single notification-capability value "online", which is described below. Additional notification-capability values may be defined by a Standard Track or Experimental RFC.

For the "online" notification-capability the notify_method_capability test can match one of the following key-list values:

• "yes" - the entity identified by the notification-uri can receive a notify notification immediately. Note that even after this value is returned, there is no guaranty that the entity would actually be able to receive any notification immediately or even receive it at all. Transport errors, recipient policy, etc. can prevent that.
• "no" - the entity identified by the notification-uri is not currently available to receive an immediate notification.
• "maybe" - Sieve interpreter can't determine if the the entity identified by the notification-uri is online or not.

The "relational" extension [Relational] (Segmuller, W. and B. Leiba, “Sieve Extension: Relational Tests,” December 2005.) adds a match type called ":count". The count of an notify_method_capability test is 0 if the returned information is the empty string, or 1 otherwise.

Example 5:
          require ["enotify"];

          if notify_method_capability
                 "xmpp:tim@example.com?message;subject=SIEVE"
                 "Online"
                 "yes" {
              notify :importance "1" :message "You got mail"
                   "xmpp:tim@example.com?message;subject=SIEVE";
          } else {
              notify :message "You got mail" "sms:+14085551212";
          }

6.  Modifier encodeurl to the 'set' action

Usage:

":encodeurl"

When the Sieve script specifies both "variables" [Variables] (Homme, K., “Sieve Extension: Variables,” December 2005.) and "enotify" capabilities in the "require", a new "set" action modifier (see [Variables] (Homme, K., “Sieve Extension: Variables,” December 2005.)) ":encodeurl" becomes available to Sieve scripts. This modifier performs percent-encoding of any octet in the string which doesn't belong to the "unreserved" set (see [URI] (Berners-Lee, T., Fielding, R., and L. Masinter, “Uniform Resource Identifier (URI): Generic Syntax,” January 2005.)). The percent-encoding procedure is described in [URI] (Berners-Lee, T., Fielding, R., and L. Masinter, “Uniform Resource Identifier (URI): Generic Syntax,” January 2005.).

The ":encodeurl" modifier has precedence 15.

Example 6:
    require ["enotify", "variables"];

    set :encodeurl "body_param" "Safe body&evil=evilbody";

    notify "mailto:tim@example.com?body=${body_param}";

7.  Interactions with Other Sieve Actions

The notify action is compatible with all other actions, and does not affect the operation of other actions. In particular, the notify action MUST NOT cancel the implicit keep.

Multiple executed notify actions are allowed. Specific notification methods MAY allow multiple notifications from the same script to be collapsed into one.

8.  Security Considerations

Security considerations are discussed in [Sieve] (Guenther, P. and T. Showalter, “Sieve: An Email Filtering Language,” August 2006.). Additionally, implementations must be careful to follow the security considerations of the specific notification methods.

The notify action is potentially very dangerous. The path the notification takes through the network may not be secure. An error in the options string may cause the message to be transmitted to someone it was not intended for, or may expose information to eavesdroppers.

Just because a notification is received doesn't mean that it was sent by the Sieve implementation. It might be possible to forge notifications with some notification methods.

An organization may have a policy about the forwarding of classified information to unclassified networks. Unless the policy is also enforced in the module responsible for generating (or sending) of notifications, users can use the extension defined in this document to extract classified information and bypass the policy.

Notifications can result in loops and bounces. In particular, a notification to an email address will not contain necessary Received header fields that might be otherwise used to prevent mail loops. All notification methods MUST take care to provide mechanisms for avoiding notification loops.

Implementations that construct URIs internally from various notify parameters MUST make sure that all components of such URIs are properly percent-encoded (see [URI] (Berners-Lee, T., Fielding, R., and L. Masinter, “Uniform Resource Identifier (URI): Generic Syntax,” January 2005.)). In particular this applies to values of the :from and the :message tagged arguments and may apply to the :options values.

9.  IANA Considerations

9.1.  Registration of Sieve extension

The following template specifies the IANA registration of the notify Sieve extension specified in this document:

To: iana@iana.org
Subject: Registration of new Sieve extension
Capability name: enotify
Description: adds the 'notify' action for notifying user about the received message. It also provides two new test: valid_notify_method checks notification URIs for validity; notify_method_capability can check recipients capabilities.
RFC number: this RFC
Contact address:
    The Sieve discussion list <ietf-mta-filters@imc.org>

This information should be added to the list of sieve extensions given on http://www.iana.org/assignments/sieve-extensions.

9.2.  New registry for Sieve notification mechanisms

This defines the template for a new registry for Sieve notification mechanisms, to be created as http://www.iana.org/assignments/sieve-notification. There are no initial entries for this registry.

To: iana@iana.org
Subject: Registration of new Sieve notification mechanism
Mechanism name: [the name of the mechanism]
Mechanism URI: [the RFC number of the document that defines the URI used by this mechanism]
Mechanism-specific tags: [the names of any Sieve notify tags that are specific to this mechanism, or "none"]
Standards Track/IESG-approved experimental RFC number: [the RFC number of the document that defines this notification mechanism]
Person and email address to contact for further information: [the name and email address of the technical contact for information about this mechanism]

10.  Acknowledgements

Thanks to Larry Greenfield, Sarah Robeson, Tim Showalter, Cyrus Daboo, Nigel Swinson, Kjetil Torgrim Homme, Michael Haardt, Mark E. Mallett, Ned Freed, Lisa Dusseault, Dilyan Palauzov, Arnt Gulbrandsen and Peter Saint-Andre for help with this document.

11.  References

11.1. Normative References

11.2. Informative References

Authors' Addresses

Full Copyright Statement

Copyright © The IETF Trust (2007).

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, THE IETF TRUST 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.

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.