Create a new named topic view.
If a view with the same name already exists the new view will update the existing view.
the name of the view
the specification of the view using the DSL
a Result that completes when a response is received from the server, returning the topic view created by the operation.
If the task fails, the Result will resolve with an error. Common reasons for failure, include:
specification
is invalid;
Get a named Topic View.
If the named view does not exist the Result will resolve with null result.
the name of the view
a Result that resolves when a response is received from the server, returning a named view if it exists
If the task fails, the Result will resolve with an Error. Common reasons for failure include:
List all the topic views that have been created.
a Result that resolves when a response is received from the server, returning a list of views sorted by their creation order.
If the task fails, the Result will resolve with an Error. Common reasons for failure include:
Remove a named topic view if it exists.
If the named view does not exist the completable future will complete successfully.
the name of the view
a Result that resolves when a response is received from the server.
If the task fails, the Result will resolve with an Error. Common reasons for failure include:
Topic view feature.
This feature allows a client session to manage topic views.
A topic view maps one part of a server's topic tree to another. It dynamically creates a set of reference topics from a set of source topics, based on a declarative topic view specification. The capabilities of topic views range from simple mirroring of topics within the topic tree to advanced capabilities including publication of partial values, expanding a single topic value into many topics, changing topic values, inserting values from other topics, throttling the rate of publication, and applying a fixed delay to the publication.
A topic view can also map topics from another server (in a different cluster). This capability is referred to as 'remote topic views'. The view can specify the server that the source topics are hosted on in terms of a remote server (see RemoteServers for details of how to create and maintain remote servers).
Each reference topic has a single source topic and has the same topic type as its source topic. Reference topics are read-only (they cannot be updated), nor can they be created or removed directly. Otherwise, they behave just like standard topics. A client session can subscribe to a reference topic, and can fetch the reference topic's current value if it has one.
The source topics of a topic view are defined by a topic selector. One or more reference topics are created for each source topic, according to the topic view. If a source topic is removed, reference topics that are derived from it will automatically be removed. If a topic is added that matches the source topic selector of a topic view, corresponding reference topics will be created. Removing a topic view will remove all of its reference topics.
Topic view specifications
The following is a simple topic view specification that mirrors all topics below the path
a
to reference topics below the pathb
.A topic view with this specification will map a source topic at the path
a/x/y/z
to a reference topic at the pathb/x/y/z
. The specification is simple, so the reference topic will exactly mirror the source topic.A topic view specification comprises three main parts:
Mapping comprises:
Mapping
Source topic clause
The source topic clause begins with themap
keyword and is followed by a topic selector. These topic selectors follow the same parsing rules as other topic selectors.When evaluating a topic view, all topics in the topic tree that match the source topic selector are considered (excluding ROUTING topics). However, if a view specification uses some feature that can only be applied to JSON topics then only JSON topics will be selected.
Reference topics are valid source topics. In particular, chaining of topic views is supported; that is, a reference topic created by one topic view can be the source topic of another topic view. Additionally, a reference topic can be the source topic of a routing topic subscription.
From clause
The 'from' clause optionally follows the source topic clause. It begins with the
from
keyword and is followed by a remote server name. The name refers to a remote server created using the RemoteServers feature.The presence of the clause indicates that the source topics will be selected from the specified server and not from the local server.
Further details regarding the processing of remote topic views are given below.
Path mapping clause
The paths of reference topics are derived from the source topic according to the path mapping clause. The path mapping allows the source topic path and the value of the source topic to determine the path of the reference topic. In addition the path mapping can include expand directives which allow objects and arrays in JSON source topic values to be expanded to produce many reference topics.
A path mapping clause begins with the
to
keyword and is followed by a path mapping template. A path mapping template is a topic path with embedded directives. Directives are evaluated when creating the topic reference and substituted into the topic path. Directives are delimited by angle brackets (<
,>
) and consist of the name of the directive and a list of parameters. The parameter list is comma-separated and surrounded by parentheses ((
,)
).The following path mapping directives are supported:
<path(start, number)>
, or<path(start)>
when the number of parts parameter is omitted.For example, given the source path
a/b/c/d
, the source path directive<path(1, 2)>
is mapped to the reference topic pathb/c
, and the source path directive<path(2)>
is mapped to the reference topic pathc/d
.scalar
and are parameterized by a single JSON pointer that extracts a scalar value from the source (or current) value. A scalar value is a string, a number,true
,false
, ornull
, that is, anything other than an array or a object. If the JSON pointer does not refer to a scalar value in the source (or current) value, no reference topic will be created. This includes cases where the JSON pointer refers to an array or an object), or when no part of the source value is selected.Deriving the reference topic paths from part of the source topic value effectively creates a secondary index on the value. For source value directives to work efficiently, the selected scalar values should be relatively stable. If an update to the source topic changes the selected scalar value, the corresponding reference topic will be removed and a new reference topic will be created.
For example, given a source value of
and the source value directive
currency/<scalar(/balance/currency)>/account/<scalar(/account)>
, the reference topic path will becurrency/USD/account/1234
.If the extracted value is a string, it is copied literally to the reference topic path. A value that contains path separators (
/
) will create a reference topic path with more levels than the path mapping template. Use the separator directive to replace path separators with an alternative string.An extracted value of
null
will be copied to the reference topic path as the string"null"
.Expand value directives use the keyword
expand
and are parameterized by one or two JSON pointers.The first pointer indicates the element within the value to be expanded, and if omitted, the value is expanded from the root. Expansion of a source topic indicates that every direct child of the element pointed to by the expand pointer will be used to create a new reference topic (or provide input to later expand or scalar directives). For example
<expand()>
would expand every child item in the source value and<expand(/account)>
would expand every child of theaccount
value in the source value. The specified value could be an object, an array or even a scalar value, but a scalar value would expand to only a single new value.The optional second parameter of the expand directive specifies a pointer to a scalar value within the expanded value which will be used to derive the path fragment of the reference topic path. If the second pointer is not specified or no scalar value is found for the pointer, the path fragment is taken from the key (if the child value is an object) or the index (if the child value is an array). Scalar child values will expand to a reference topic but will not add anything to the generated path. For example
<expand(,/name)>
would expand from the root of the source value and each child value path fragment would be obtained from the scalar value with the keyname
.So if a source topic had a value of
a path mapping of
value<expand(/values)>
would expand the value to the following reference topics:-path
value0
with a value of1
path
value1
with a value of5
path
value2
with a value of7
Expand directives can be nested (i.e. there can be more than one expand directive in a path mapping). In this case a second expand directive will use the value from the previous expand as its source (root) value and not the value of the source topic. This also applies to scalar directives that follow an expand directive.
If expansion causes more than one mapping to the same topic path, only the first encountered will be created and updated.
Expanding source topic values effectively creates secondary indices on the value. For expanded value directives to work efficiently, the value selected for expansion should be relatively stable in terms of the children it contains. If an update to the source topic changes the children of the expanded value, then corresponding reference topics will be removed and created. Updates should generally be limited to changing values within the expanded values.
Transformations
Transformations are specified after the mapping and before any options. Transformations can only be applied to JSON topics.
Transformations are applied to the value extracted from the source topic in the order specified. There can be any number of transformations interspersed with one another and the value from one will be that which is input to the next. The only restriction is that all insert transformations must occur at the end, after any others.
A transformation is applied to the current value within a view processing chain, so if a transformation occurs after an expand then it will be applied to each expanded value.
Process transformations
Process transformations may be used to apply conditional processing to a value (optionally determining whether a reference topic is created) and/or change the value in some way (for example, by applying some calculation to a field within the value).
The format of a process transformation is:-
process {statement}
Where the statement can be:-
One or more operations separated by ';'.
Comprising one or more conditions with operations to perform if they are satisfied.
For example, the following topic view specification could be used to write a field into the value of the reference topic :
The following example shows a simple conditional statement which would only generate reference topics if the value of field
/Price
was greater than 50 :And the following shows a more complex statement which would set a field according to the value of the input field
/Price
:The following operations are supported :
For example :
If a hierarchic pointer is specified, the parent object or array must exist.
For example :
If a hierarchic pointer is specified, the parent object or array must exist.
In this case the operations are all performed on the original value, creating a chain of deltas which are only applied to the original value at the end. If the 'set' operation fails no reference topic would be generated, however, processing will continue if the fields specified in the 'remove's are not present.
A calculation may be specified as the value of a set operation. A calculation is a simple arithmetic calculation upon integer fields. If applied to a non integer field the evaluation will not proceed. Floating point calculations are not supported.
Arithmetic operators supported are
+
,-
,*
and/
.Examples of calculations are:-
Standard operator precedence is applied, so in the last example above we have
(/Salary + 1000 + (/Age * 10))
not((/Salary + 1000 + /Age) * 10)
. Brackets may be used to override this.A conditional statement is made up of an if clause, optionally followed by one or more elseif clauses and an optional final else clause.
The if clause takes the form :
Where the condition is a quoted string as described in detail below and the operation(s) is as described previously. If the condition is satisfied, the operations are applied to the value and the process is complete. If the condition is not satisfied, processing moves on to any elseif or else clauses that follow, but if there are none, the topic view evaluation does not proceed and no reference topic is created.
An elseif (else if) clause takes the form :
If the condition is satisfied, the operations are applied to the value and the process is complete. If the condition is not satisfied, processing moves on to any elseif or else clauses that follow, but if there are none, the topic view evaluation does not proceed and no reference topic is created. Note that elseif can be abbreviated to elsf.
An else clause takes the form :
And will only be reached if no previous if or elseif conditions were satisfied. If reached then the operations are applied to the value and the topic view evaluation proceeds. The continue operation may be used to proceed with an unchanged value.
A condition is of the form:
Where pointer is a JSON pointer, operator is a relational operator and constant is a string, integer, or boolean value.
for example:-
Operators allowed are:
Compound conditions are supported by means of boolean operators:
|
oror
&
orand
For example:
Normal boolean precedence applies but brackets can be used to control precedence. For example:
Boolean 'not' is also allowed :
Patch transformations
Patch transformations indicate that a JSON patch is to be applied to the value.
The format of a patch transformation is
The patch string should be formatted according to the JSON Patch standard (see RFC 6902: JavaScript Object Notation (JSON) Patch).
Patches are a sequence of JSON Patch operations contained in an array. They are applied as an atomic update to the previous value if the resulting update is successfully calculated. The following patch will check the value at a specific key and update if the expected value is correct:
The available operations are:
{"op": "add", "path": "/a/b/c", "value": [ "foo", "bar" ]}
{"op": "remove", "path": "/a/b/c"}
{"op": "replace", "path": "/a/b/c", "value": 43}
{"op": "move", "from": "/a/b/c", "path": "/a/b/d"}
{"op": "copy", "from": "/a/b/c", "path": "/a/b/e"}
{"op": "test", "path": "/a/b/c", "value": "foo"}
The test operation checks that the CBOR representation of the value of a topic is identical to the value provided in the patch after converting it to CBOR. If the value is represented differently as CBOR, commonly due to different key ordering, then the patch will return the index of the failed operation. e.g the values
{"foo": "bar", "count": 43}
and{"count": 43, "foo": "bar"}
are unequal despite semantic equality due to the differences in a byte for byte comparison.The following patch clause would add the 'price' field and remove the 'name' field from an input JSON object.
Patches can only be applied to JSON arrays or objects and if they fail to apply, no resulting reference topic will be created by the view. If an update patch fails, any previously created reference topic would be removed.
Insert transformations
Insert transformations are used to insert a value from another topic into the current value.
The name of the topic to insert from can be specified in a similar way to the path mapping in that constants, path directives, and scalar directives (but not expand directives) may be used.
The value from the insertion topic (if found) is inserted into the 'current' value at a specified key position. The current value may be the source topic value, the value output from expand directives (in which case the insertion applies to each value), or the value from a previous transformation. Insertion topics may be JSON, STRING, INT64, or DOUBLE.
If, when evaluating a topic view, the insertion topic is not found (or the specified key within it is not found), or it is of an unsupported topic type, an optional default value may be inserted, but if no default is specified then no insertion takes place and the value is passed to the derived reference topic unchanged.
The values of the insertion topics are only taken at the point when the source topic is evaluated against the topic view (i.e. when the source topic is updated). Changes to the value of the insertion topic that occur in the interim are not applied to the derived reference topics.
The format of an insert transformation is
The path is specified in exactly the same way as for the path mapping clause, except it may not contain expand directives. path directives operate on the path of the source topic, whereas scalar directives operate on the 'current' value as defined previously.
key is optional and can specify a fromKey which is a JSON pointer indicating the data within the insertion topic that is to be inserted. If no key is specified then the whole of the insertion topic value is inserted.
at specifies the insertionKey which is a JSON pointer indicating where to insert the insertion topic value. If a value already exists at the specified key then it is replaced otherwise it is inserted. Insertion can only occur if the parent of the key exists in the value and is of a compatible type. Array pointers may only be used to replace existing entries or append one greater than the last entry. The special pointer value
/-
may be used to append to the end of an existing array.default is optional and may be used to specify a string defaultValue to be inserted if the insertion topic does not exist, it is of an unsupported topic type, or the specified key within it could not be found.
The following insert transformation would cause the whole value of the topic named AnyTopic to be inserted into the current value at key /T, assuming that the current value is an object.
The following insert transformation would cause the whole value of the topic named AnyTopic to be inserted into the current value at key /T/MyKey, assuming that an object with the key T exists in the current value.
The following insert transformation would cause the whole value of the topic named AnyTopic to be appended to the array at the key T in the current value.
The following insert transformation would cause the value at the key name within the topic named AnyTopic to be appended to the array at the key T in the current value.
In the above examples no insertion would take place if the insertion value was not found, but a default value can be specified to insert into the current value in this case.
The path of the topic to insert from can be built from parts of the source topic path and/or scalar values within the current value. For example:
Options
Options are specified after the mapping and any transformations. Any number of options may be specified but any particular option may only be specified once.
Topic property mapping
The topic specification of a reference topic is derived from the topic specification of the source topics. A reference topic has the same topic type as its source topic.
The topic properties of a reference topic are derived from the source topic. Some topic properties can be tuned using the topic property mapping option. The following table describes the behavior for each topic property.
TIME_SERIES_RETAINED_RANGE
property. The retained range of a reference time series topic will be constrained to be no greater than that of its source topic.A topic property option begins with the keywords
with properties
and consists of a comma-separated list of topic property keys and values, each separated by a colon. For example, the following topic view specification maps all topics below the patha
to reference topics below the pathb
, and disables both conflation and compression for the reference topics.Topic value option
By default, a reference topic's value is a copy of the source topic value, or part of the source value produced by an expand path mapping directive and/or modified by transformations. For JSON source topics or TIME_SERIES topics with a
JSON
event type, the value option can be applied to extract part of the resulting value (the latest value in the case of TIME_SERIES topics).A topic value option begins with the keyword
as
and is followed by a value directive. A value directive is delimited by angle brackets (<
,>
), and consists of thevalue
keywords and a single JSON pointer parameter. The JSON pointer selects the part of the current value to copy.For example, given a current value of
and the value option
as <value(/balance)>
, the reference topic value will beValue mappings that follow expand directives and/or transformations apply to the current derived value and not the source topic value.
Topic value mappings only alter the reference topic value; only the path mapping determines whether a reference topic should exist. If the topic value mapping's JSON pointer fails to select anything from the source topic value, the reference topic will have the JSON value
null
.Topic value mappings are often used with path value mappings to avoid repeating information in the path and the value. For example:
Throttle option
The throttle option can be used to constrain the rate at which a reference topic is updated when its source topic is updated. The primary application of a throttle option is to restrict the number of updates sent to reference topic subscribers, reducing network utilization or the processing each subscriber must do. Throttling also restricts the rate at which client sessions can observe changes to reference topic values using the fetch API.
The throttle option has the form
throttle to X updates every period
, where X is a positive integer, and period is a positive integer followed by a time unit which is one ofseconds
,minutes
, orhours
.For example, the following topic view specification maps all topics below the path
a
to reference topics below the pathb
, but updates the value of each reference topic at most twice every five seconds:To improve readability, the throttle option allows
1 update
as an alternative to1 updates
, andevery second
as an alternative toevery 1 seconds
(and so on, for other time units). For example, the following topic view specification maps all topics below the patha
to reference topics below the pathb
, but updates the value of each reference topic at most once every hour:The throttle option is only applied when a source topic is updated more frequently than the configured rate. If a source topic is updated less frequently, updates are passed on unconstrained. If the rate is exceeded, a reference topic will not be updated again until the configured period has expired. At this time, the reference topic will be updated based on the source topic updates that happened in the interim, and a single value will be published. Thus, the throttle option provides topic-scoped conflation.
The throttle option is ignored for time series topics because time series updates do not support efficient conflation. Updates to source time series topics are passed on immediately to the corresponding reference topics, regardless of any throttle clause.
Delay option
The delay option causes a change to a view's source topic to be delayed by a fixed time before it is reflected in reference topics. Topic additions, updates, and removals are all delayed. Delays can range from one second to many days​.
Such a publication delay is a useful way to devalue topic data so it can be given away to non-paying users.
The delay option has the form
delay by duration
, where duration is a positive integer followed by a time unit which is one ofseconds
,minutes
, orhours
.For example, the following topic view specification maps all topics below the path
a
to reference topics below the pathb
, but changes to a source topic are delayed by five minutes before they are reflected in the corresponding reference topic.Views with the delay option specified initially create reference topics in an unpublished state. The topics are published once the delay time has expired. A topic in the unpublished state prevents a lower priority topic view from creating a reference topic with the same path. Sessions with the rights to read the source topic can browse unpublished topics using the withUnpublishedDelayedTopics fetch request option.
Separator option
Views can use the scalar and expand directives in path mappings to extract text from the source value. By default, any
/
characters in the text are interpreted as path separators and will introduce extra levels in reference topic paths. If this is undesirable, the separator option can be used to replace any/
characters produced by the scalar and expand directives with a constant string. The replacement can contain path separators but must have no empty path segments (//
).Here is an example replacing all path separators with an alternate character:
If the value at '/x/y' in the source topic was foo/bar, this would cause the reference topic to be created at
/b/foo%bar
instead of/b/foo/bar
.Preserve topics option
The default behavior for a topic view is that only the reference topics that can be derived from the current value of the source topic are maintained. This applies to views using directives that derive the path of the reference topic(s) from a value within the source topic (e.g. scalar or expand). For example if a view uses a scalar directive to determine the path of the reference topic and the source topic is updated such that the selected scalar value changes then the previously created reference topic will be removed and a new one created. Similarly an expand directive may create many reference topics from the source value but if the source is updated then only those reference topics that still have a corresponding value within the source value will be updated (or new ones added if not in the previous value), whilst others will be removed.
There are situations where an application may not want reference topics to be removed because of source value changes. In such cases the 'preserve topics' option may be used to indicate that all reference topics created by a view should be retained (and updated by later updates that apply to it) until either the source topic or the creating view are removed.
For example:
Topic type option
The 'type' option can be used to specify the TopicType of the target reference topic. If the current source value of type indicated by the source topic's type can be converted to the target type, a reference topic of the specified type will be created.
For example:
The specified type must be one of the supported target types (STRING, INT64, DOUBLE, JSON, TIME_SERIES, or BINARY), case insensitive.
The following table describes the supported conversions from the source topic type (the left column) to the supported target types. The number in brackets indicates a note at the foot of the table describing exactly how the conversion is processed. Where there is no number, no conversion is necessary and the derived source value is simply mapped to the target reference topic as if the 'type' option was not specified. Where there is no entry for the source topic type in the left column, no conversion is supported and topics of unsupported types will not be processed by the topic view. An 'x' indicates that the conversion is not supported.
In all cases the value being processed will be the 'current' value as derived from other mappings within the topic view (e.g. expand) which is not necessarily the value of the source topic.
Conversion notes:-
Escaping and quoting special characters
Each part of a topic view expression has characters with special significance. Source topic clauses and path mapping clauses are delimited by white space. Directives in path and topic property mapping clauses are delimited by the<
and>
characters, and each directive parameter is terminated by,
or)
. Topic property mapping clauses are delimited by white space, and the:
and,
characters.Sometimes a topic view must refer to or generate topics with paths that containing special characters, or use a JSON pointer containing special characters. The escape sequence
\x
can be used to literally insert any characterx
, with a one exception:/
cannot be used in path fragments since the path delimiter/
is always significant.Here is an example topic view expression containing escape sequences. It maps the topic path
a topic
a reference topic with the pathanother topic
.Here is an example with a source value directive that uses the JSON pointer
/x()/y
to extract the target path from the source value. The)
character in the JSON pointer must be escaped so it is not treated as the end of the parameter list.To insert
\
, the escape sequence\\
must be used.There is no need to escape white space in JSON pointers directive parameters. However, white space is significant. For example, the following expressions have different topic value mapping clauses since the JSON pointer in the second expression is
/x
; that is, it has a trailing space:Instead of using escape sequences, white space characters can be included in source topic clauses and path mapping clauses using quotes. A clause is quoted by wrapping it in single quote (
'
) or double quote ("
) characters. For example:Within a quoted clause, quotes of the same type must be escaped:
For consistency, the values in topic property mapping clauses can be escaped or quoted. However, there is no need to do so because none of the valid values for the mappable properties contain special characters.
Dealing with topic path conflicts
Reference topics have a lower priority than normal topics created through the API, including replicas of normal topics created by topic replication or fan-out. A reference topic will only be created if no topic or reference topic is already bound to its derived topic path.
Topic views have a precedence based on order of creation. If two topic views define mappings the same topic path, the earliest-created topic view will create a reference topic. If a topic view is updated, it retains its original precedence.
Remote topic views
A remote topic view is one that specifies another server as the location of the source topics using the from clause as shown in the example below:
map ?a// from server1 to b/<path(1)>
The server name (
server1
in this example) refers to the name of a remote server created using the RemoteServers feature.In this case, upon establishing a successful connection with the remote server indicated the topic view will create reference topics locally based upon the topics selected by the topic view's selector at the remote server. It is important to note that the selector only refers to topics that match it at the remote server and not on the local server and there is no reason why there could not be a source topic at the remote server that has the same path as an entirely different topic on the local server.
More than one topic view can specify the same remote server.
A remote server only makes a physical connection when it is in use, therefore the first topic view that specifies a remote server will cause it to establish a connection. Similarly, if the last topic view that uses a remote server is removed then the connection will be closed.
It is not necessary for the named remote server definition to exist before creating the topic view, as if it does not then the topic view will simply remain dormant until the remote server is created and a successful connection to the server specified in its URL is established. Similarly, if a remote server that is in use by remote topic views is removed then all of the reference topics created by the topic views will be removed and the topic views will become dormant until the named remote server is created again or the views are changed to name a different remote server.
If a remote topic view selects a ROUTING topic at the remote server then local mappings will only be performed if the routing topic mapping at the remote server is able to establish a mapping for the remote server connection. The mapping will be done as if from the resolved routing topic.
The rules of precedence for remote topic views are the same as for other topic views. If the remote server for a remote topic view does not exist or does not have an established connection then the remote topic view is not evaluated (i.e. it is as if the source topics for the view did not exist), but if the remote server later connects then the view will be evaluated and rules of precedence will determine whether reference topic will replace those created by earlier views.
Topic view persistence and replication
Reference topics are neither replicated nor persisted. They are created and removed based on their source topics. However, topic views are replicated and persisted. A server that restarts will restore topic views during recovery. Each topic view will then create reference topics based on the source topics that have been recovered.
The server records all changes to topic views in a persistent store. Topic views are restored if the server is started.
If a server belongs to a cluster, topic views (and remote servers) will be replicated to each server in the cluster. Topic views are evaluated locally within a server. Replicated topic views that select non-replicated source topics can create different reference topics on each server in the cluster. When remote topic views are in use, each server in the cluster will make a connection to the specified remote server and will separately manage their remote topic views.
A view with a delay clause uses temporary storage to record delayed events. If there is a high volume of updates, temporary per-server disk files will be used to save server memory. The storage is per-server, and does not survive server restart. When a server is started, no data will be published by a view with a delay clause until the delay time has expired.
Access control
The following access control restrictions are applied:
Accessing the feature
This feature may be obtained from a session as follows:
6.3