WARNING: This document has been automatically Deferred after 12 months of inactivity in its previous Experimental state. Implementation of the protocol described herein is not recommended for production systems. However, exploratory implementations are encouraged to resume the standards process.
Publish-Subscribe (XEP-0060) [1] defines an XMPP protocol extension for generic publish-subscribe features. However, it only allows notifications from nodes to which an entity is directly subscribed. It is useful in some circumstances to describe a relationship between nodes so that a publish on one node may be delivered via another node. For instance, if an entity is interested in notifications from a set of nodes the entity would discover each node somehow and then subscribe to them. With collection nodes, the entity would subscribe only to the collection which links the desired nodes, simplifying the subscription process.
In addition to simplifying the subscriber's usage, collection nodes also allow the owner to describe almost any type of relationship between nodes. Using various access models on different nodes the owner can also create almost any desired authorization semantics on a set of leaf nodes.
Note: Any use cases not described herein are described in XEP-0060. Also, this document does not show error flows related to the generic publish-subscribe use cases referenced herein, since they are exhaustively defined in XEP-0060. The reader is referred to XEP-0060 for all relevant protocol details related to the XMPP publish-subscribe extension.
Collection nodes link nodes together to unify notifications from a set of collection or leaf nodes. An entity can subscribe to the collection and receive notifications of any associated leaf nodes.
A collection node can link with any other node in order to create a directed acyclic graph (DAG). Collection nodes MUST NOT be linked in such a way as to produce a cyclic graph (i.e., they cannot link to nodes that eventually link back to the initial node).
Collection nodes only contain other nodes and MUST NOT contain published items (therefore a collection MUST NOT support the "publish" feature or related features such as "persistent-items").
An entity might wish to discover if a service implements collection nodes; in order to do so, it sends a service discovery information ("disco#info") query to the component's JID using Service Discovery (XEP-0030) [2]. If a service supports collection nodes it MUST return a "pubsub#collections" feature. In addition, if the service supports associating a node with more than one collection it MUST return a feature of "pubsub#multi-collections".
The service discovery items ("disco#items") protocol enables an entity to query a service for a list of associated items, which, in the case of collection nodes would consist of the children associated with a given node.
If a notification on a child node is created and then delivered via the collection then the notifications generated by the service MUST contain additional information. The 'node' attribute of the <item/> or <node/> element contained in the notification message MUST specify the node identifier of the node that generated the notification (not the collection) and the <message/> stanza MUST contain Stanza Headers and Internet Metadata (XEP-0131) [3] that specifies the node identifier of the collection.
Note: The delivery options (such as "pubsub#deliver_payloads") are determined by the publishing leaf node, not by the collection node. If the owner of a collection node sets delivery options for a collection node, the service SHOULD ignore those options and apply the options set for the leaf node that publishes an item.
5.3.1.1 Notifications about Items
Item notifications are notifications about the contents of a leaf node, and are generated by a publish, retract, or purge request.
5.3.1.2 Notifications about Nodes
Node notifications are notifications about nodes themselves, and are generated by a create, delete, or configure request.
If a collection node is configured to send notification of node associations and disassociations, the service shall send an event that contains a <collection/> element whose 'node' attribute specifies the NodeID of the collection; this element in turn contains an <associate/> or <dissociate/> element whose 'node' attribute specifies the NodeID of node that has been associated with the collection.
5.3.2.1 Including Node Meta-Data
The notification event MAY also include the node metadata, formatted using the Data Forms (XEP-0004) [4] protocol.
A service that implements collection nodes SHOULD allow entities to subscribe to collection nodes (subject to access models and local security policies).
In addition to the subscription configuration options already defined in XEP-0060, there are two subscription configuration options specific to collection nodes:
pubsub#subscription_type
This subscription option enables the subscriber to subscribe either to notifications about items or notifications about nodes.
If the subscription type is "items", the subscriber shall be notified whenever any node contained in the collection has an item published to it, retracted from it, or the node is purged, as modified by the value of the "pubsub#subscription_depth" option.
If the subscription type is "nodes", the subscriber shall be notified whenever a new node is added to the collection, removed from the collection, or the configuration of a node within the collection has changed, as modified by the value of the "pubsub#subscription_depth" option.
If the subscription type is "all", the subscriber shall be notified about both "items" and "nodes" types of events, as modified by the value of the "pubsub#subscription_depth" option.
The default value of this subscription option SHOULD be "nodes".
pubsub#subscription_depth
This subscription option enables the subscriber to specify how far to traverse the node graph when determining whether a notification will be sent. It may be any integer value, 0 or greater, or the value "all" which means that any node within the collection will generate a notification.
The default value of this subscription option SHOULD be "1".
In order to subscribe to a collection node, an entity MUST send a subscription request to the node; the subscription request MAY include subscription options, but this is not strictly necessary (especially if the entity does not wish to override the default settings for the "pubsub#subscription_type" and "pubsub#subscription_depth" options).
A service MAY allow an entity to subscribe to a collection node in two ways, once with a subscription of type "nodes" (to receive notification of any new nodes added to the collection or the entire tree) and once with a subscription of type "items" (to receive all items published within the tree). However, a service SHOULD NOT allow an entity to subscribe twice to a collection node (once with a subscription depth of "1" and once with a subscription depth of "all") for the same subscription type, since two such subscriptions are unnecessary (a depth of "all" includes by definition a depth of "1"); in this case the service SHOULD return a <conflict/> error to the requesting entity.
Depending on the nature of the node graph, a subscription type of "items" and depth of "all" may result in an extremely large number of notifications. Therefore, a service MAY disallow such a combination of subscription options, in which case it MUST return a <not-allowed/> error to the requesting entity.
When an entity requests items on a collection node the service SHOULD return the items on any leaf nodes associated with it subject to the access model of the collection node.
Depending on the nature of the node graph it may be expensive to allow item retrieval from a collection node. Therefore the service MAY disallow item retrieval via collection nodes, in which case it MUST return a <feature-not-implemented/> error to the requesting entity.
To create a new collection node, the requesting entity MUST include a Data Form containing a "pubsub#node_type" field whose <value/> element contains "collection". The default value for "pubsub#node_type" SHOULD be "leaf".
In addition to the errors already defined for leaf node creation, there are several reasons why the collection node creation request might fail:
The service does not support collection nodes.
The service does not support creation of collection nodes.
The requesting entity does not have sufficient privileges to create collection nodes.
These error cases are described more fully in the following sections.
7.1.3.1 Collection Nodes Unsupported
If the service does not support collection nodes, it MUST respond with a <feature-not-implemented/> error, specifying a pubsub-specific error condition of <unsupported/> and a feature of "collections".
7.1.3.2 Collection Nodes Can't be Created
If the service supports collection nodes but does not allow new collection nodes to be created, it MUST respond with a <not-allowed/> error.
7.1.3.3 Entity is not Authorized
If the requesting entity has insufficient privileges to create new collections, the service MUST respond with a <forbidden/> error.
In addition to the node configuration options specified in XEP-0060, there are three additional node configuration options that a service which supports collection nodes MUST supply.
pubsub#node_type
Whether this is a "leaf" or "collection" node.
pubsub#collection
The parents of this node.
pubsub#children
The children of this node.
To associate the root node to the collection the <value/> element MUST be empty.
A service MAY offer some node configuration options that are specific to collection nodes and SHOULD NOT be provided in configuration forms related to leaf nodes. The following are RECOMMENDED:
pubsub#children_association_policy
The policy regarding who may associate child nodes with the collection (values: all, owner, whitelist).
pubsub#children_association_whitelist
The whitelist of entities that may associate child nodes with the collection.
pubsub#children_max
The maximum number of child nodes that may be associated with a collection.
Leaf nodes only contain published items and MUST NOT have any children. If an entity attempts to add children to a leaf node (either via "pubsub#children" on the leaf node or "pubsub#collection" on another node) the service MUST return a <not-allowed/> error with a pubsub-specific error condition of <invalid-options/>.
7.2.3.2 Entity is not Authorized
If the requesting entity is not authorized to add the node to a collection then the service MUST return a <forbidden/> error.
7.2.3.3 Maximum Number of Children Exceeded
If the configuration would exceed the maximum number of children allowed on a node, either because the node's "pubsub#children" exceeds its own "pubsub#children_max" value or because adding this node to a parent via "pubsub#collection" would exceed the parent's "pubsub#children_max" value, the service MUST return a <not-allowed/> error with a pubsub-specific error condition of <max-nodes-exceeded/>.
7.2.3.4 Changing Node Types
The service MUST NOT allow the node type to be changed. If it is attempted the service MUST return a <not-allowed/> error, specifying a pubsub-specific error condition of <invalid-options/>
7.2.3.5 Creating a Cycle in the Collection
The service MUST NOT allow a cycle to be created in the node graph (e.g., node A to B to C to A). If an entity attempts to submit a configuration that would create a cycle the service MUST return a <not-allowed/> error, specifying a pubsub-specific error condition of <invalid-options/>.
A service MAY allow collection nodes to have children associated with them without changing the rest of the configuration. If the service allows this an entity can send and <associate/> element with a 'node' attribute that contains the child node within a <collection/> element that posesses a 'node' attribute containing the parent node to the service.
If the configuration would exceed the maximum number of children allowed on a node, either because the node's "pubsub#children" exceeds its own "pubsub#children_max" value or because adding this node to a parent via "pubsub#collection" would exceed the parent's "pubsub#children_max" value, the service MUST return a <not-allowed/> error with a pubsub-specific error condition of <max-nodes-exceeded/>.
7.4.3.3 Creating a Cycle in the Collection
The service MUST NOT allow a cycle to be created in the node graph (e.g., node A to B to C to A). If an entity attempts to submit a configuration that would create a cycle the service MUST return a <not-allowed/> error, specifying a pubsub-specific error condition of <invalid-options/>.
A service MAY allow collection nodes to have children dissociated from them without changing the rest of the configuration. If the service allows this an entity can send and <dissociate/> element with a 'node' attribute that contains the child node within a <collection/> element that posesses a 'node' attribute containing the parent node to the service.
To provide a starting point for service discovery a service SHOULD support a root node. A root node represents the node belonging to a given service and MUST be identified by the lack of a node identifier (i.e., the address of the pubsub service itself, such as "pubsub.shakespeare.lit"). Because the root node is owned by the service itself an entity SHOULD NOT be allowed create, delete, or configure the root node.
If a node is created or configured without any parents specified, a service MAY automatically associate otherwise orphaned nodes directly to the root node. If a service automatically associates a node with the root it MUST reflect that in the node configuration data form.
Deletion of collection nodes can have a number of side effects due to implementation. Depending on the nature of the collection any of the following MAY happen when a collection node is deleted:
When the collection node is deleted and the child nodes have no other parents the child nodes are orphaned; meaning that they will have no parent node, but continue to exist.
When the collection node is deleted and the child nodes have no other parents the child nodes are re-assigned as children of the root node.
When the collection node is deleted and the child nodes have no other parents the child nodes are also deleted.
When the collection node is deleted and the child nodes have at least one other node the child nodes MUST remain associated with remaining parent nodes.
8.3 Updating Node Configuration When Associating or Dissociating Nodes¶
Node configuration MUST always reflect the current state of the node graph. Because node configuration contains both a pointer to its parents as well as its children an update to a primary node's "pubsub#collection" value will change the value of the secondary node's "pubsub#children" value, and vice-versa. A service MAY send a notification of the configuration change on the secondary node to subscribers if "pubsub#notify_config" is enabled on the secondary node.
Collection nodes can be used to associate almost any node within the service, but only the access model of the collection node itself is used to determine what an entity is allowed to see. Therefore care should be taken that nodes are not linked in such a way as to leak private data (e.g., from a "closed" leaf node through an "open" collection) unless that behavior is specifically desired.
Permission is hereby granted, free of charge, to any person obtaining a copy of this specification (the "Specification"), to make use of the Specification without restriction, including without limitation the rights to implement the Specification in a software program, deploy the Specification in a network service, and copy, modify, merge, publish, translate, distribute, sublicense, or sell copies of the Specification, and to permit persons to whom the Specification is furnished to do so, subject to the condition that the foregoing copyright notice and this permission notice shall be included in all copies or substantial portions of the Specification. Unless separate permission is granted, modified works that are redistributed shall not contain misleading information regarding the authors, title, number, or publisher of the Specification, and shall not claim endorsement of the modified works by the authors, any organization or project to which the authors belong, or the XMPP Standards Foundation.
Disclaimer of Warranty
## NOTE WELL: This Specification is provided on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, express or implied, including, without limitation, any warranties or conditions of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A PARTICULAR PURPOSE. ##
Limitation of Liability
In no event and under no legal theory, whether in tort (including negligence), contract, or otherwise, unless required by applicable law (such as deliberate and grossly negligent acts) or agreed to in writing, shall the XMPP Standards Foundation or any author of this Specification be liable for damages, including any direct, indirect, special, incidental, or consequential damages of any character arising from, out of, or in connection with the Specification or the implementation, deployment, or other use of the Specification (including but not limited to damages for loss of goodwill, work stoppage, computer failure or malfunction, or any and all other commercial damages or losses), even if the XMPP Standards Foundation or such author has been advised of the possibility of such damages.
IPR Conformance
This XMPP Extension Protocol has been contributed in full conformance with the XSF's Intellectual Property Rights Policy (a copy of which can be found at <https://xmpp.org/about/xsf/ipr-policy> or obtained by writing to XMPP Standards Foundation, P.O. Box 787, Parker, CO 80134 USA).
Visual Presentation
The HTML representation (you are looking at) is maintained by the XSF. It is based on the YAML CSS Framework, which is licensed under the terms of the CC-BY-SA 2.0 license.
The Extensible Messaging and Presence Protocol (XMPP) is defined in the XMPP Core (RFC 6120) and XMPP IM (RFC 6121) specifications contributed by the XMPP Standards Foundation to the Internet Standards Process, which is managed by the Internet Engineering Task Force in accordance with RFC 2026. Any protocol defined in this document has been developed outside the Internet Standards Process and is to be understood as an extension to XMPP rather than as an evolution, development, or modification of XMPP itself.
The following requirements keywords as used in this document are to be interpreted as described in RFC 2119: "MUST", "SHALL", "REQUIRED"; "MUST NOT", "SHALL NOT"; "SHOULD", "RECOMMENDED"; "SHOULD NOT", "NOT RECOMMENDED"; "MAY", "OPTIONAL".
5. The Internet Assigned Numbers Authority (IANA) is the central coordinator for the assignment of unique parameter values for Internet protocols, such as port numbers and URI schemes. For further information, see <http://www.iana.org/>.