1
0
mirror of https://github.com/moparisthebest/xeps synced 2024-11-21 16:55:07 -05:00

Merge branches 'feature/protoxep-udt' and 'feature/protoxep-mfc'

This commit is contained in:
Jonas Schäfer 2019-12-30 14:12:55 +01:00
commit 1d3bcba3b4
2 changed files with 400 additions and 0 deletions

203
inbox/mamfc.xml Normal file
View File

@ -0,0 +1,203 @@
<?xml version='1.0' encoding='UTF-8'?>
<!DOCTYPE xep SYSTEM 'xep.dtd' [
<!ENTITY % ents SYSTEM 'xep.ent'>
%ents;
<!ENTITY ns "urn:xmpp:mamfc:0">
<!ENTITY nsx "urn:example:">
]>
<?xml-stylesheet type='text/xsl' href='xep.xsl'?>
<xep>
<header>
<title>MAM Fastening Collation</title>
<abstract>This specification proposes a mechanism by which MAM results containing fastenings can be collated effectively.</abstract>
&LEGALNOTICE;
<number>XXXX</number>
<status>ProtoXEP</status>
<type>Standards Track</type>
<sig>Standards</sig>
<dependencies>
<spec>XMPP Core</spec>
<spec>XEP-0422</spec>
<spec>XEP-0313</spec>
</dependencies>
<supersedes/>
<supersededby/>
<shortname>mamfc</shortname>
&dcridland;
&ksmithisode;
<revision>
<version>0.0.1</version>
<date>2019-12-19</date>
<initials>dwd</initials>
<remark>
<ul>
<li>Initial Revision</li>
</ul>
</remark>
</revision>
</header>
<section1 topic='Introduction' anchor='intro'>
<p>In XMPP, all messages are not equal. Some are semantically actual human messages; these are referred to in this
document as "instant messages". Others are ancillary messages - reactions, receipts, and so on - that are useful
and important, but do not conform to the concept of an instant message in the sense that a user might reasonably
expect.</p>
<p>Fastenings, &xep0422;, provides a generic mechanism for a sending entity to indicate that a particular message is
associated closely to an instant message. But in doing so, this presents the problem that if many messages are not in fact
actual human messages, an archive query might end up downloading dozens or even hundreds of messages in order to
present just a handful of actual instant messages to the user. Much of the information downloaded is not required.</p>
<p>For example, to display a message, a client may need to download all the "likes" for it - whereas a simply number of
likes might be sufficient for most users' needs.</p>
<p>This specification tackles the problem by allowing these to be filtered, collated, or presented in full depending
on the needs of the client. The client now downloads just the instant messages from the archive, and any likes,
reactions, or receipts are simply summarized.</p>
<section2 topic="Terminology">
<p>Because this document defines mechanisms for dealing generically with potential types of fastenings, it does not
offer any real examples of actual fastenings that might be used.</p>
<p>Instead, example fastenings are used within an XML namespace prefixed by <tt>&nsx;</tt></p>
<p>Pseudo-fastenings are messages that are semantically equivalent to fastenings, but use a different syntax,
see <link url="#pseudo">Pseudo Fastenings</link>.</p>
<p>Nomenclature used for instant messages versus ancillary messages will need to be adjusted to make it consistent
with &xep0422; et al.</p>
</section2>
</section1>
<section1 topic='Overview' anchor="overview">
<section2 topic="Discovering Support" anchor="feature">
<p>Support for this protocol is advertised by the Service Discovery protocol defined in &xep0030; using a feature
of <tt>&ns;</tt>.</p>
</section2>
<section2 topic="Summarizing">
<p>This specification provides for four types of summary listing.</p>
<p>The first form, "<tt>simplified</tt>", is the default, and essentially represents the status quo. MAM queries
behave as if the archive contains only traditional IM traffic. No summary is provided.</p>
<p>The second form, "<tt>full</tt>", presents every message stanza in the results, including all fastenings,
errors, and so on.</p>
<p>The third form, "<tt>collate</tt>", presents each traditional IM message, as "<tt>simplified</tt>", but within
the result includes summary information about the fastenings (and pseudo-fastenings) encountered.</p>
<p>Finally a fourth form, "<tt>fastenings</tt>", returns only the fastenings for a particular message.</p>
<p>The "collate" form is the bulk of the specification presented herein.</p>
<section3 topic="Summary Information">
<p>The &lt;apply-to/> element of &xep0422; contains a number of fastening elements. These in turn consist of a
qualified element, with a number of attributes, and potentially some content within the element.</p>
<p>This specification refers to the qualified name (the tuple of XML namespace and local-name) as the "fastening
type" (represented as an XML element herein), and the top-level element (including attributes and their
values), as the "fastening summary".</p>
<p>For example, a hypothetical edit fastening type might be <tt>&lt;edit xmlns="&nsx;edit:0"/></tt>, and that would
be the fastening summary as well. The full fastening would include any children (text or further XML elements)
of the top-level element. But a hypothetical reaction fastening type might be
<tt>&lt;reaction xmlns="&nsx;reaction:0"/></tt>, but the fastening summary could be
<tt>&lt;reaction xmlns="&nsx;reaction:0" emoji="&#x1F49;"/></tt></p>
<p>The summary information against each parent message consists of, for each fastening summary:</p>
<ul>
<li>The fastening summary itself.</li>
<li>A count of the number of fastenings with this summary fastened to the parent message.</li>
<li>The full fastening for the last fastening received for the parent message.</li>
</ul>
</section3>
</section2>
</section1>
<section1 topic="Protocol Elements">
<section2 topic="Querying">
<p>This specification adds an additional field to the form defined in &xep0313; with the field name
"<tt>{&ns;}summary</tt>". This may have only the following values (unless of course further values are advertised
by a future specification):</p>
<ul>
<li>simplified</li>
<li>full</li>
<li>collate</li>
<li>fastenings</li>
</ul>
</section2>
<section2 topic="Results">
<p>The <tt>&lt;result/></tt> element defined in &xep0313; is extended by adding zero or more additional elements with
a local name of "<tt>applied</tt>", qualified by the "<tt>&ns;</tt>" namespace.</p>
<p>Each <tt>&lt;applied/></tt> element is associated with precisely one fastening summary.</p>
<p>This element contains, as its first child element, the full fastening for the last fastening received by the
server for the parent message. This is not included for "<tt>shell</tt>" fastenings, which are untyped.</p>
<p>There is a "<tt>count</tt>" attribute, consisting of an integral count of the fastenings with the same summary
as the first child element (or the count of shell fastenings when this is not included). This count, if absent,
defaults to 1. For "<tt>shell</tt>" fastenings, the attribute "<tt>shell</tt>" is set to true (or another value
with the same semantics for an XML boolean).</p>
<p>The <tt>&lt;applied/></tt> elements are only included in the <tt>&lt;result/></tt> element when the requested
summary type is "<tt>collate</tt>".</p>
</section2>
<section2 topic="Latest Archive ID">
<p>The latest archive id can usually be deduced either from the last message stanza received (and its stanza-id,
see &xep0359;) or by the id attribute of the last <tt>&lt;result/></tt> element from a query extending to the
latest message.</p>
<p>Since this specification can cause the latest message to be only in a summarized form when presented in the
archive, it also adds an additional element to the <tt>&lt;fin/></tt> element which terminates the query, to
indicate the latest id held in the archive (which may be that of a fastening).</p>
<p>This element, qualified by the "<tt>&ns;</tt>", has the local name of "<tt>latest</tt>" and a single attribute,
"<tt>id</tt>", containing the latest archive id.</p>
</section2>
<section2 topic="Incremental queries">
<p>A MAM query where the MAM summary type is "<tt>collate</tt>", and where "<tt>start</tt>" and "<tt>end</tt>" (or
the RSM <tt>&lt;after/></tt> element) would exclude the parent message but include the fastening, then the MAM
result is sent with the <tt>&lt;forwarded/></tt> element omitted but the summary present (including all
fastenings, not just those that have changed).</p>
</section2>
</section1>
<section1 topic="Pseudo-Fastenings" anchor="pseudo">
<p>A number of previous specifications exist that - if they were rewritten today - might use fastenings.</p>
<p>For the purposes of this specification, it is convenient to treat these as pseudo-fastenings, behaving as if they
were a fastening for the purposes of the "collate" and "fastenings" summary types.</p>
<p>This specification defines two such types. Both MUST be supported by servers:</p>
<ul>
<li>Message Delivery Receipts: &xep0184; "ack messages" - those containing a <tt>&lt;received/></tt> element - are
considered to be equivalent to a fastening containing just the &lt;received/> element, applying to the message
given by the "<tt>id</tt>" attribute.</li>
<li>Chat Markers: &xep0333; A Chat Marker is similarly equivalent to a fastening containing the Chat Marker, but
applying to all previous messages (since previous messages can be assumed to have been read and or displayed,
etc).</li>
</ul>
<p>In both cases, the fastening summary SHOULD omit the <tt>id</tt> attribute.</p>
</section1>
<section1 topic="Examples">
<p>A firm TODO; contributions are welcome - and a good test of whether I've written the rest right!</p>
</section1>
<section1 topic="Schema">
<code>
<![CDATA[
<?xml version="1.0" encoding="UTF-8"?>
<xs:schema attributeFormDefault="unqualified" elementFormDefault="qualified" targetNamespace="]]>&ns;<![CDATA[" xmlns:xs="http://www.w3.org/2001/XMLSchema">
<xs:element name="applied" type="mamfc:appliedType" xmlns:mamfc="]]>&ns;<![CDATA["/>
<xs:complexType name="appliedType">
<xs:sequence>
<xs:any minOccurs="0" maxOccurs="1"/>
<!-- Fastening summary, absent when shell is true -->
</xs:sequence>
<xs:attribute type="xs:positiveInteger" name="count" default="1"/>
<xs:attribute type="xs:boolean" name="shell" default="false"/>
</xs:complexType>
<xs:element name="latest" type="mamfc:latestType" xmlns:mamfc="]]>&ns;<![CDATA["/>
<xs:complexType name="latestType">
<xs:attribute type="xs:string"/>
</xs:complexType>
</xs:schema>
]]>
</code>
</section1>
<section1 topic='Security Considerations' anchor='security'>
<p>This specification imposes substantial workload for servers.</p>
</section1>
<section1 topic='IANA Considerations' anchor='iana'>
<p>This XEP requires no interaction with &IANA;. </p>
</section1>
<section1 topic='XMPP Registrar Considerations' anchor='registrar'>
<p>None.</p>
</section1>
<section1 topic='Acknowledgements' anchor='ack'>
<p>The authors wish to share any credit with many members of the community, including Marvin Wissfield.</p>
</section1>
</xep>

197
inbox/udt.xml Normal file
View File

@ -0,0 +1,197 @@
<?xml version='1.0' encoding='UTF-8'?>
<!DOCTYPE xep SYSTEM 'xep.dtd' [
<!ENTITY % ents SYSTEM 'xep.ent'>
%ents;
<!ENTITY ns "urn:xmpp:udt:0">
<!ENTITY nsx "urn:example:">
]>
<?xml-stylesheet type='text/xsl' href='xep.xsl'?>
<xep>
<header>
<title>User-defined Data Transfer</title>
<abstract>This specification proposes a simple mechanism by which applications can transfer data safely, without
needing additional protocol design work. It is intended to provide a protocol that is trivial to implement and can
be driven with a simple API.</abstract>
&LEGALNOTICE;
<number>XXXX</number>
<status>ProtoXEP</status>
<type>Standards Track</type>
<sig>Standards</sig>
<dependencies>
<spec>XMPP Core</spec>
</dependencies>
<supersedes/>
<supersededby/>
<shortname>udt</shortname>
&dcridland;
<revision>
<version>0.0.1</version>
<date>2019-12-30</date>
<initials>dwd</initials>
<remark>
<ul>
<li>Initial Revision</li>
</ul>
</remark>
</revision>
</header>
<section1 topic='Introduction' anchor='intro'>
<p>Applications written on top of XMPP often need to exchange data that has no existing standard. Such applications are
often written by developers unfamiliar with best practise in designing new extensions for XMPP, making it hard to achieve
this simple design goal without causing longer term problems.</p>
<p>This leads to "solutions" such as stuffing JSON directly in the &lt;body/> element, for example, and recognising
this at the receiver either by heuristics or by a special &lt;subject/>. While this works, it's difficult to then
migrate to something else, and enforces that custom clients are always used.</p>
<p>Therefore this document proposes a very simple (and simplistic) framework for sending such data which - while
very light on features - nevertheless conforms to best practice. Unusually, this specification SHOULD NOT be used
as a base upon which to build other standards, and suggests an API for library developers to implement.</p>
<section2 topic="Terminology">
<p>Data transferred using this specification is encoded using JSON. The type of the data is given by a URI under
the same rules as an XML namespace, and this specification refers to this as the datatype.</p>
<p>Because this document defines mechanisms for sending essentially arbitrary data, no real-world examples are
given.</p>
<p>Instead, example namespaces are used within an XML namespace prefixed by <tt>&nsx;</tt></p>
</section2>
</section1>
<section1 topic='Overview' anchor="overview">
<section2 topic="Discovering Support" anchor="feature">
<p>Support for this protocol is advertised by the Service Discovery protocol defined in &xep0030; using a feature
of <tt>&ns;</tt>.</p>
<p>Support for a particular datatype is given by concatenating the <tt>&ns;</tt> feature with a hash character
('<tt>#</tt>') and the datatype, for example <tt>&ns;#&nsx;foo</tt>.</p>
</section2>
<section2 topic="Data Transfers">
<p>This specification provides for two types of user-defined data transfers. Each uses a similar payload syntax,
the UDT data payload.</p>
<p>Requests are carried within &lt;iq/> stanzas, either of type "get" or "set", and result in a "result" optionally
containing a second UDT data payload.</p>
<p>UDT payloads may also be placed within a &lt;message/> stanza. &lt;message/> stanzas MAY contain multiple UDT
payloads, but typical usage is expected to be that there will be only one. The UDT payload may be ancillary data
to another message, or a standalone message in its own right.</p>
<section3 topic="Protocol Syntax">
<p>A UDT payload consists of a single element, <tt>&lt;payload/></tt>, qualified by the XML namespace
<tt>&ns;</tt>. It has a single, mandatory attribute of <tt>datatype</tt>, which MUST contain a string conformant
to the requirements for XML namespaces (typically a URI under the control of the application developer).</p>
<p>As with XML namespaces, this URI is never expected to be resolved, and is used solely as an indentifier.
Different strings are considered entirely different datatypes, and common prefixes etc MUST be considered
irrelevant for the purposes of interpreting the data. There are no common or standard datatypes.</p>
<p>The <tt>&lt;payload</tt> element contains exactly one mandatory child element, the <tt>&lt;json/></tt> element
defined in &xep0335;. This in turns contains the JSON data.</p>
<example><![CDATA[
<message from="gamer@game-company.example"
to="match-maker.game-company.example"
id="12345">
<payload xmlns="]]>&ns;<![CDATA[" datatype="]]>&nsx;foo<![CDATA[">
<json xmlns="urn:xmpp:json:0">
{
"annoying-teenager-level": 11
}
</json>
</payload>
</message>
]]></example>
<p>Note that the suggested custom &IQ; query payload of &xep0335; is not used as this would generally require
custom handlers within client libraries:</p>
<example><![CDATA[
<iq from="gamer@game-company.example"
to="match-maker.game-company.example"
id="12345"
type="set">
<payload xmlns="]]>&ns;<![CDATA[" datatype="]]>&nsx;foo<![CDATA[">
<json xmlns="urn:xmpp:json:0">
{
"annoying-teenager-percentage": 101
}
</json>
</payload>
</iq>
]]></example>
</section3>
</section2>
</section1>
<section1 topic="API Requirements">
<p>In order to satisfy the goals of this protocol, it is necessary to define an API that can be consistently
implemented across APIs. This allows application developers to use the protocol easily, and encourages this
over using the ad-hoc techniques described in the introduction.</p>
<p>Therefore, while not imposing hard and fast API definitions, this specification proposes naming conventions
for APIs that will hopefully guide application developers toward consistent usage.</p>
<p>While names are specified in "snake_case", API developers are free to use their own naming. This specification
also defines APIs as taking session arguments which will often be implied by method calls, and omits types.</p>
<p>Library developers SHOULD make these calls as simple to use as possible. If these are significantly harder to use
for inexperienced developers than ad-hoc techniques, then ad-hoc techniques will be used instead.</p>
<section2 topic="Arguments and Types">
<ul>
<li><tt>session</tt> - The session, connection or stream object or handle for the library. In cases where the
library uses a single global connection this is omitted. An OO library is likely to elide this and have the call
be a method call on the object instead.</li>
<li><tt>message</tt> - The message object or handle for the library. An OO library is likely to elide this and have the call
be a method call on the object instead.</li>
<li><tt>datatype</tt> - Typically a string containing an XML-style namespace.</li>
<li><tt>jid</tt> - A string or JID object. Any form of legal jid might be used here.</li>
<li><tt>data</tt> - Either a JSON-encoded string, or an object or data structure which can be converted by the
library.</li>
<li><tt>get_or_set</tt> - A flag for indicating whether this is a "get" or "set" request.</li>
<li><tt>callback</tt> - Some callable object or similar as meets the idiom of the language.</li>
</ul>
</section2>
<section2 topic="Advertising Support">
<p>Support is advertised on a session by calling <tt>udt_advertise(session, datatype)</tt>. Calling this MUST
explicitly advertise both the <tt>&ns;</tt> feature and that of the datatype support.</p>
<p>APIs are free to (and encouraged to) implicitly advertise support when other calls are made.</p>
</section2>
<section2 topic="Requests">
<p>Requests are sent to a particular jid by calling <tt>udt_request(session, jid, get_or_set, datatype, data)</tt>.
This might return a UDT payload, or have an additional <tt>callback</tt> argument to call with the response.</p>
<p>Applications may register to handle such requests by calling <tt>udt_request_callback(session, get_or_set,
datatype, callback)</tt>. The callback should be called as <tt>callback(session, jid, get_or_set, datatype,
data)</tt>, and returning a payload should send a <tt>result</tt> containing it. Applications registering this
way SHOULD implicitly advertise support for the datatype.</p>
</section2>
<section2 topic="Messages">
<p>Messages are sent to a particular jid by calling <tt>udt_message(session, jid, get_or_set, datatype, data)</tt>.</p>
<p>Applications may register to handle such requests by calling <tt>udt_message_callback(session, get_or_set,
datatype, callback)</tt>. The callback should be called as <tt>callback(session, jid, get_or_set, datatype,
data)</tt>. Applications registering this way SHOULD implicitly advertise support for the datatype.</p>
<p>Applications may check to see if any message stanza contains a UDT payload by calling <tt>udt_payload(message,
datatype)</tt>, which returns the payload if it exists.</p>
</section2>
</section1>
<section1 topic="Schema">
<code>
<![CDATA[
<?xml version="1.0" encoding="UTF-8"?>
<xs:schema attributeFormDefault="unqualified" elementFormDefault="qualified" targetNamespace="]]>&ns;<![CDATA[" xmlns:xs="http://www.w3.org/2001/XMLSchema">
<xs:element name="payload" type="udt:payloadType" xmlns:udt="]]>&ns;<![CDATA["/>
<xs:complexType name="payloadType">
<xs:sequence>
<xs:any minOccurs="1" maxOccurs="1"/>
<!-- Always a XEP-0335 json element, but I can't figure that out. -->
</xs:sequence>
<xs:attribute type="xs:string" name="datatype"/>
</xs:complexType>
</xs:schema>
]]>
</code>
</section1>
<section1 topic='Security Considerations' anchor='security'>
<p>All security implications herein are those of the payload.</p>
</section1>
<section1 topic='IANA Considerations' anchor='iana'>
<p>This XEP requires no interaction with &IANA;. </p>
</section1>
<section1 topic='XMPP Registrar Considerations' anchor='registrar'>
<p>None.</p>
</section1>
<section1 topic='Acknowledgements' anchor='ack'>
<p>The authors wish to share any credit with many members of the community, including Florian Schmaus.</p>
</section1>
</xep>