xeps/inbox/entityversioning.xml

<?xml version='1.0' encoding='UTF-8'?>
<!DOCTYPE xep SYSTEM 'xep.dtd' [
  <!ENTITY % ents SYSTEM 'xep.ent'>
%ents;
]>
<?xml-stylesheet type='text/xsl' href='xep.xsl'?>
<xep>
<header>
  <title>Entity Versioning</title>
  <abstract>
    A method by which rosters and disco items may be versioned so that servers
    will not need to send the entire list if it has not been modified, saving
    bandwidth and time during session initialization with minimal state being
    stored by the server and client.
  </abstract>
  &LEGALNOTICE;
  <number>xxxx</number>
  <status>ProtoXEP</status>
  <type>Standards Track</type>
  <sig>Standards</sig>
  <approver>Council</approver>
  <dependencies>
    <spec>RFC 6120</spec>
    <spec>RFC 6121</spec>
  </dependencies>
  <supersedes />
  <supersededby />
  <shortname>EV</shortname>
  <author>
    <firstname>Sam</firstname>
    <surname>Whited</surname>
    <email>swhited@atlassian.com</email>
    <jid>sam@samwhited.com</jid>
  </author>
  <author>
    <firstname>Doug</firstname>
    <surname>Keen</surname>
    <email>dkeen@atlassian.com</email>
  </author>
  <revision>
    <version>0.0.1</version>
    <date>2015-08-25</date>
    <initials>ssw</initials>
    <remark><p>First draft</p></remark>
  </revision>
</header>

<section1 topic='Introduction' anchor='intro'>
  <p>
    This problem of "downloading the world" (downloading the entire roster
    every time a session is initialized) was partially addressed by &xep0237;
    which was later merged into &rfc6121; §2.6. While this solved the problem
    for the roster, it didn't account for other entities (eg. disco items).
    Furthermore, roster versioning requires that the server maintain a great
    deal of state (roster items which should be pushed for each entity on
    reconnect, or monotonically increasing counters, etc.) which can be
    difficult to store or synchronize in a large, distributed system. This XEP
    defines a method by which the roster and entities other than the roster can
    be versioned and cached, and which is optimized for distributed systems
    with large enity lists (but works equally well on small, single server
    deployments).
  </p>
</section1>

<section1 topic='Requirements' anchor='reqs'>
  <ul>
    <li>
      An extra round trip MUST NOT be required to initiate entity versioning.
    </li>
    <li>
      Clients that do not implement the protocol (but which use servers that
      do) MUST still be able to request and receive entities normally.
    </li>
    <li>
      Servers which implement this protocol MUST NOT be required to store
      multiple versions of an entity list or maintain other redundant state.
    </li>
    <li>
      Inconsistant state between servers in a cluster should not cause cache
      invalidation for the entire entity list.
    </li>
    <li>
      Large changes SHOULD NOT be required for existing servers / clients.
    </li>
  </ul>
</section1>

<section1 topic='Glossary' anchor='glossary'>
  <dl>
    <dt>Aggregate Token</dt>
    <dd>
      A hash which represents the state of a list of entities, and changes if
      any of those entities changes.
    </dd>
    <dt>Versioned Entity</dt>
    <dd>Any abstract object which may be versioned (eg. rooms, users).</dd>
    <dt>Version Token</dt>
    <dd>
      A generally short, case sensitive string which represents an entity and
      changes if that entity changes.
    </dd>
  </dl>
</section1>

<section1 topic='Use Cases' anchor='use_cases'>
  <section2 topic='Clients' anchor='client_use_cases'>
    <ul>
      <li>
        A client on a mobile device where bandwidth and throughput are limited
        has a very large roster which cause connections to take an unacceptable
        amount of time. With entity versioning, connections after the first
        connection do not take as long, and use less bandwidth.
      </li>
      <li>
        A client often wants to view the list of multi-user chat rooms
        available on a servers MUC service. However, the list is very long and
        takes a long time to download. After enabling entity versioning the
        client can fetch the list, and then poll for changes at a later date
        without re-requesting the entire list.
      </li>
      <li>
        A client wishes to cache the features supported by servers of the
        contacts in their roster since their disco items is not likely to
        change often.
      </li>
    </ul>
  </section2>
  <section2 topic='Servers' anchor='server_use_cases'>
    <ul>
      <li>
        A server is running in an environment where storing multiple versions
        of each users roster may put too much pressure on the storage backend.
        After enabling entity versioning, they only have to store a small token
        per user and can calculate the diffs to send to the client afterwards.
      </li>
      <li>
        A server maintains an out-of-band HTTP API for fetching information
        about MUC rooms to display on their web page. They wish to use a
        reverse proxy to cache API requests based on etags. Instead of
        attempting to check if the backend page has changed and generate etags,
        the room's entity version token is used as a weakly-validated ETag.
      </li>
    </ul>
  </section2>
</section1>

<section1 topic='Discovering Support' anchor='disco'>
    <p>
      If a server supports entity versioning, it MUST inform the connecting
      client when returning stream features during the stream negotiation
      process. This is done by including a &lt;ver/&gt; element, qualified by
      the 'urn:xmpp:entityver:0' namespace. At the latest, this SHOULD
      be done when informing a client that resource binding is required. For
      example:
    </p>
    <example caption="Stream Features"><![CDATA[
<stream:features>
  <bind xmlns='urn:ietf:params:xml:ns:xmpp-bind'>
    <required/>
  </bind>
  <ver xmlns='urn:xmpp:entityver:0'/>
</stream:features>
    ]]></example>
  <p>
    The entity versioning stream feature is merely informative and therefore
    is never mandatory-to-negotiate.
  </p>
</section1>

<section1 topic='Version Tokens' anchor='version_tokens'>
  <p>
    Version tokens are short case-sensitive strings which are generated by the
    server. Their format is not defined in this spec, but a recommendation may
    be found in the <link url="#impl">Implementation Notes</link>. Version
    tokens are akin to a weakly-validated etag for the entity in question.
  </p>
  <p>
    Servers that implement this protocol must assign such a version token to
    each entity that is controlled by the server. The server MUST then update
    this version every time any mutable property of the entity changes (eg.
    when the subscription status of a user changes). The server MAY choose to
    update this token at any time (to force the clients to invalidate their
    cached representation fo the object). This version token MUST then be
    included with every object representation of that entity sent down in the
    stream. This is done by including a sub-node called "version" qualified
    by the entity versioning XML namespace defined in this document.
    Similarly, clients MAY also add version nodes for each version token they
    possess to the request for a list (not specifying a version token will
    force the server to send information on that entity to the client). If a
    server sends up a list of version tokens, the server MUST then check to
    see if those tokens correspond to any entity which it knows about, and
    not send down any entities with matching version tokens in the response.
  </p>
  <p>For example, a roster request might look like:</p>
  <example caption="Roster Request"><![CDATA[
<!-- Client -->
<iq from='romeo@montague.lit/home' id='56' to='romeo@montague.lit' type='get'>
  <query xmlns='jabber:iq:roster'>
    <item jid='bill@shakespeare.lit'>
      <version xmlns='urn:xmpp:entityver:0'>25P2A7H8</version>
    </item>
    <item jid='anne@shakespeare.lit'>
      <version xmlns='urn:xmpp:entityver:0'>VIZSVF0D</version>
    </item>
  </query>
</iq>

<!-- Server -->
<iq from='romeo@montague.lit' id='56' to='romeo@montague.lit/home' type='result>
  <query xmlns='jabber:iq:roster'>
    <item subscription='both' jid='bill@shakespeare.lit'>
      <version xmlns='urn:xmpp:entityver:0'>9ZFZXVP9</version>
    </item>
  </query>
</iq>
    ]]></example>
    <p>
      Note that in this case there may be three roster items total (and the
      client only knows about two of them), or there may be two total roster
      items and the server is informing the client about a change to
      "bill@shakespeare.lit". Version tokens MUST also be present in roster
      pushes:
    </p>
    <example caption="Roster Push"><![CDATA[
<!-- Server -->
<iq from='romeo@montague.lit' id='ah382g678jka7' to='romeo@montague.lit/home' type='set'>
    <query xmlns='jabber:iq:roster' ver='ver34'>
      <item jid='tybalt@shakespeare.lit' subscription='remove'>
        <version xmlns='urn:xmpp:entityver:0'>XWE4MUUP</version>
      </item>
    </query>
</iq>
    ]]></example>
    <p>A disco request for rooms (as defined in &xep0045;) might look like:</p>
    <example caption="MUC Disco"><![CDATA[
<!-- Client -->
<iq from='hag66@shakespeare.lit/phone'
    id='zb8q41fas6yn4'
    to='chat.shakespeare.lit'
    type='get'>
  <query xmlns='http://jabber.org/protocol/disco#items>
    <item jid='coven@chat.shakespeare.lit'>
      <version xmlns='urn:xmpp:entityver:0'>25P2A7H8</version>
    </item>
    <item jid='inverness@chat.shakespeare.lit'>
      <version xmlns='urn:xmpp:entityver:0'>4OLGSVNY</version>
    </item>
  </query>
</iq>

<!-- Server -->
<iq from='chat.shakespeare.lit'
  id='zb8q41fas6yn4'
  to='hag66@shakespeare.lit/phone'
  type='result'>
  <query xmlns='http://jabber.org/protocol/disco#items'>
    <item jid='coven@chat.shakespeare.lit'
          name='A Dark Cave'>
      <version xmlns='urn:xmpp:entityver:0'>VIZSVF0D</version>
    </item>
  </query>
</iq>
    ]]></example>
  <p>
    In this example coven@chat.shakespeare.lit has been modified (eg. the
    room name might have been changed), but inverness@chat.shakespeare.lit
    has not changed, therefore no update is sent down.
  </p>
  <p>
    Clients that implement this protocol SHOULD then cache the entity in
    question when a version token is received.
  </p>

  <section2 topic='Cache Invalidation' anchor='deletes'>
    <p>
      When a client syncs with the server and indicates that it has a version
      token in its cache that does not match any entity on the server (or when
      the server wants to remove an entity from the clients cache for any other
      reason), the server MUST reply with an empty &lt;version/&gt; node. When
      the client receives such an empty version node it SHOULD purge the entity
      from its cache. For example, the following exchange would trigger the
      removal of 'inverness@chat.shakespeare.lit' from the cached MUC list:

      <example caption="MUC deletion"><![CDATA[
  <!-- Client -->
  <iq from='hag66@shakespeare.lit/phone'
      id='zb8q41fas6yn4'
      to='chat.shakespeare.lit'
      type='get'>
    <query xmlns='http://jabber.org/protocol/disco#items>
      <item jid='coven@chat.shakespeare.lit'>
        <version xmlns='urn:xmpp:entityver:0'>25P2A7H8</version>
      </item>
      <item jid='inverness@chat.shakespeare.lit'>
        <version xmlns='urn:xmpp:entityver:0'>4OLGSVNY</version>
      </item>
    </query>
  </iq>

  <!-- Server -->
  <iq from='chat.shakespeare.lit'
    id='zb8q41fas6yn4'
    to='hag66@shakespeare.lit/phone'
    type='result'>
    <query xmlns='http://jabber.org/protocol/disco#items'>
      <item jid='inverness@chat.shakespeare.lit'>
        <version xmlns='urn:xmpp:entityver:0'/>
      </item>
    </query>
  </iq>
      ]]></example>
    </p>
    <p>
      If the client receives an indication that it should delete an item from a
      list by any other means (eg. via a roster push), it SHOULD remove the
      version token associated with that entity from its cache.
    </p>
  </section2>
</section1>
<section1 topic='Aggregate Tokens' anchor='agg_tokens'>
  <p>
    While the version token approach to caching does not require a great deal
    of state to be stored on the client or the server, it does require a lot
    more information to be sent by the client when requesting a list of
    entities. For a very large list which is not likely to have changed, it may
    be useful know in advance if the roster has changed or not (so that we can
    avoid sending the large request entirely). To do this, we can request an
    aggregate version token from the server. This aggregate token is calculated
    by constructing a string of comma separated "JID:version" pairs sorted in
    byte-wise order (because the JID:version pair is constructed before
    sorting, if two items in the list have the same JID they can still be
    sorted by the version token), and taking the MD5 hash of the constructed
    string. For example, if the server is calculating the aggregate version
    token for a roster, it might end up with the following string:
  </p>
  <example caption="Aggregate token list"><![CDATA[
anne@shakespeare.lit:VIZSVF0D,bill@shakespeare.lit:25P2A7H8
  ]]></example>
  <p>
    Which results in the aggregate token:
  </p>
  <example caption="Aggregate token"><![CDATA[
0514fc90e6c7981b06bbb2173bb8ef03
  ]]></example>
  <p>
    The actual request is an IQ sent to the server, or entity handling the
    versioned list which contains a query that specifies the namespace of the
    list we want to fetch. Eg. to fetch the aggregate token for the roster one
    would query the server with the type set to the `jabber:iq:roster`
    namespace:
  </p>
  <example caption="Roster aggregate token request"><![CDATA[
<!-- Client -->
<iq to='bill@shakespeare.lit' type='get' id='bill1'>
  <query xmlns='urn:xmpp:entityver:0' type='jabber:iq:roster' />
</iq>

<!-- Server -->
<iq to='bill@shakespeare.lit/home' type='result' id='bill1'>
  <query xmlns='urn:xmpp:entityver:0' type='jabber:iq:roster'>
    0514fc90e6c7981b06bbb2173bb8ef03
  </query>
</iq>
  ]]></example>
  <p>
    Similarly, to fetch the aggregate token for a list of MUC rooms, one would
    query the MUC component directly with the type set to the 'disco#items'
    namespace:
  </p>
  <example caption="MUC rooms aggregate token request"><![CDATA[
    <!-- Client -->
    <iq to='chat.shakespeare.lit' type='get' id='bill2'>
      <query xmlns='urn:xmpp:entityver:0' type='http://jabber.org/protocol/disco#items' />
    </iq>

    <!-- Server -->
    <iq to='bill@shakespeare.lit/home' type='result' id='bill2'>
      <query xmlns='urn:xmpp:entityver:0' type='http://jabber.org/protocol/disco#items'>
        32151d1d01440d5536a7f106afd3f4d8
      </query>
    </iq>
  ]]></example>
  <p>
    Because aggregate tokens are OPTIONAL to implement, clients MUST fall back
    to a normal request if any error is returned in response to an aggregate
    token IQ.
  </p>
  <p>
    If an aggregate token is requested for a list that may contain more than
    one type of entity (eg. MUC rooms and pubsub nodes that live on the same
    component), then the server MUST return the aggregate token constructed
    with the entire list (rooms and pubsub nodes).
  </p>
  <p>
    Clients are also NOT REQUIRED to check aggregate tokens. However, clients
    MAY wish to check aggregate tokens before making a roster or MUC request
    when the cached roster or MUC list is very large. When to check aggregate
    tokens (if at all) is left up to the implementation.
  </p>
</section1>

<section1 topic='Implementation Notes' anchor='impl'>
  <p>
    Version tokens may not provide enough collision resistance across versioned
    entities (hereafter simply called "entities"), and may vary from server to
    server, and therefore they MUST NOT be used as an entity identifier.
  </p>
  <p>
    Version tokens SHOULD always be considered opaque to the client (eg. even
    if the version token is a derivable and consistent hash on the server side,
    clients should not need to know how the server is calculating the token).
  </p>
  <p>
    The author RECOMMENDS using 8 character (32-bit) random alphanumeric ASCII
    strings (eg. AABd7z9T) for version tokens.
  </p>
  <p>
    If a server which supports this XEP provides an HTTP API which can be used
    to fetch information about entities (eg. for listing information about MUC
    rooms that a server provides on the providers web page), the entities
    version token MAY be used as a weakly validated ETag for any API requests
    for that entity.
  </p>
</section1>

<section1 topic='Security Considerations' anchor='security'>
  <p>
    Client-side caching of entity information across sessions (rather than
    holding them in memory only for the life of a session) could pose a privacy
    risk, especially on shared systems. Implementations SHOULD protect cached
    entity data with strong encryption or other appropriate means.
  </p>
</section1>

<section1 topic='IANA Considerations' anchor='iana'>
  <p>This document requires no interaction with &IANA;.</p> 
</section1>

<section1 topic='XMPP Registrar Considerations' anchor='registrar'>
  <section2 topic='Protocol Namespaces' anchor='registrar-ns'>
    <p>This specification defines the following XML namespace:</p>
    <ul>
      <li>urn:xmpp:entityver:0</li>
    </ul>
    <p>
      Upon advancement of this specification from a status of Experimental to a
      status of Draft, the &REGISTRAR; shall add the foregoing namespace to the
      registry located at &STREAMFEATURES;, as described in Section 4 of
      &xep0053;.
    </p>
  </section2>
</section1>

<section1 topic='XML Schema' anchor='schema'>
TODO
</section1>

<section1 topic='Acknowledgements' anchor='ack'>
  <p>
    The original entity versioning proposal was engineered and written by
    HipChat's Doug Keen.
  </p>
</section1>

</xep>