blob: f3fef517f3caac50df326e946898f8722dfc3a05 [file] [log] [blame]
liamfallonf3981072020-07-10 15:59:25 +01001 .. This work is licensed under a Creative Commons Attribution 4.0 International License.
2 .. http://creativecommons.org/licenses/by/4.0
ramverma3b71c972019-07-10 11:25:37 +00003
liamfallonf3981072020-07-10 15:59:25 +01004#################
ramverma3b71c972019-07-10 11:25:37 +00005APEX Policy Guide
liamfallonf3981072020-07-10 15:59:25 +01006#################
7
ramverma3b71c972019-07-10 11:25:37 +00008
9.. contents::
liamfallonab5be042020-07-13 16:37:56 +010010 :depth: 5
ramverma3b71c972019-07-10 11:25:37 +000011
liamfallonf3981072020-07-10 15:59:25 +010012******************
ramverma3b71c972019-07-10 11:25:37 +000013APEX Policy Matrix
liamfallonf3981072020-07-10 15:59:25 +010014******************
ramverma3b71c972019-07-10 11:25:37 +000015
liamfallonf3981072020-07-10 15:59:25 +010016.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +000017
liamfallonf3981072020-07-10 15:59:25 +010018 APEX offers a lot of flexibility for defining, deploying, and executing policies. Based on a theoretic model, it
19 supports virtually any policy model and supports translation of legacy policies into the APEX execution format.
20 However, the most important aspect for using APEX is to decide what policy is needed, what underlying policy concepts
21 should be used, and how the decision logic should be realized. Once these aspects are decided, APEX can be used to
22 execute the policies. If the policy evolves, say from a simple decision table to a fully adaptable policy, only the
23 policy definition requires change. APEX supports all of that.
ramverma3b71c972019-07-10 11:25:37 +000024
liamfallonf3981072020-07-10 15:59:25 +010025.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +000026
liamfallonf3981072020-07-10 15:59:25 +010027 The figure below shows a (non-exhaustive) matrix, which will help to decide what policy is required to solve your
28 problem. Read the matrix from left to right choosing one cell in each column.
ramverma3b71c972019-07-10 11:25:37 +000029
liamfallonf3981072020-07-10 15:59:25 +010030.. container:: imageblock
ramverma3b71c972019-07-10 11:25:37 +000031
liamfallonf3981072020-07-10 15:59:25 +010032 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +000033
liamfallonf3981072020-07-10 15:59:25 +010034 |APEX Policy Matrix|
ramverma3b71c972019-07-10 11:25:37 +000035
liamfallonf3981072020-07-10 15:59:25 +010036 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +000037
liamfallonf3981072020-07-10 15:59:25 +010038 Figure 1. APEX Policy Matrix
ramverma3b71c972019-07-10 11:25:37 +000039
liamfallonf3981072020-07-10 15:59:25 +010040.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +000041
liamfallonf3981072020-07-10 15:59:25 +010042 The policy can support one of a number of stimuli with an associated purpose/model of the policy, for instance:
ramverma3b71c972019-07-10 11:25:37 +000043
liamfallonf3981072020-07-10 15:59:25 +010044.. container:: ulist
ramverma3b71c972019-07-10 11:25:37 +000045
liamfallonf3981072020-07-10 15:59:25 +010046 - Configuration, i.e. what should happen. An example is an event that states an intended network configuration
47 and the policy should provide the detailed actions for it. The policy can be realized for instance as an
48 obligation policy, a promise or an intent.
ramverma3b71c972019-07-10 11:25:37 +000049
liamfallonf3981072020-07-10 15:59:25 +010050 - Report, i.e. something did happen. An example is an event about an error or fault and the policy needs to
51 repair that problem. The policy would usually be an obligation, utility function, or goal policy.
ramverma3b71c972019-07-10 11:25:37 +000052
liamfallonf3981072020-07-10 15:59:25 +010053 - Monitoring, i.e. something does happen. An example is a notification about certain network conditions, to
54 which the policy might (or might not) react. The policy will mitigate the monitored events or permit (deny)
55 related actions as an obligation or authorization.
ramverma3b71c972019-07-10 11:25:37 +000056
liamfallonf3981072020-07-10 15:59:25 +010057 - Analysis, i.e. why did something happen. An example is an analytic component sends insights of a situation
58 requiring a policy to act on it. The policy can solve the problem, escalate it, or delegate it as a refrain or
59 delegation policy.
ramverma3b71c972019-07-10 11:25:37 +000060
liamfallonf3981072020-07-10 15:59:25 +010061 - Prediction, i.e. what will happen next. An example are events that a policy uses to predict a future network
62 condition. The policy can prevent or enforce the prediction as an adaptive policy, a utility function, or a goal.
ramverma3b71c972019-07-10 11:25:37 +000063
liamfallonf3981072020-07-10 15:59:25 +010064 - Feedback, i.e. why did something happen or not happen. Similar to analysis, but here the feedback will be in
65 the input event and the policy needs to something with that information. Feedback can be related to history or
66 experience, for instance a previous policy execution. The policy needs to be context-aware or be a meta-policy.
ramverma3b71c972019-07-10 11:25:37 +000067
liamfallonf3981072020-07-10 15:59:25 +010068.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +000069
liamfallonf3981072020-07-10 15:59:25 +010070 Once the purpose of the policy is decided, the next step is to look into what context information the policy will
71 require to do its job. This can range from very simple to a lot of different information, for instance:
ramverma3b71c972019-07-10 11:25:37 +000072
liamfallonf3981072020-07-10 15:59:25 +010073.. container:: ulist
ramverma3b71c972019-07-10 11:25:37 +000074
liamfallonf3981072020-07-10 15:59:25 +010075 - No context, nothing but a trigger event, e.g. a string or a number, is required
ramverma3b71c972019-07-10 11:25:37 +000076
liamfallonf3981072020-07-10 15:59:25 +010077 - Event context, the incoming event provides all information (more than a string or number) for the policy
ramverma3b71c972019-07-10 11:25:37 +000078
liamfallonf3981072020-07-10 15:59:25 +010079 - Policy context (read only), the policy has access to additional information related to its class but cannot
80 change/alter them
ramverma3b71c972019-07-10 11:25:37 +000081
liamfallonf3981072020-07-10 15:59:25 +010082 - Policy context (read and write), the policy has access to additional information related to its class and can
83 alter this information (for instance to record historic information)
ramverma3b71c972019-07-10 11:25:37 +000084
liamfallonf3981072020-07-10 15:59:25 +010085 - Global context (read only), the policy has access to additional information of any kind but cannot
86 change/alter them
ramverma3b71c972019-07-10 11:25:37 +000087
liamfallonf3981072020-07-10 15:59:25 +010088 - Global context (read and write), the policy the policy has access to additional information of any kind and
89 can alter this information (for instance to record historic information)
ramverma3b71c972019-07-10 11:25:37 +000090
liamfallonf3981072020-07-10 15:59:25 +010091.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +000092
liamfallonf3981072020-07-10 15:59:25 +010093 The next step is to decide how the policy should do its job, i.e. what flavor it has, how many states are needed,
94 and how many tasks. There are many possible combinations, for instance:
ramverma3b71c972019-07-10 11:25:37 +000095
liamfallonf3981072020-07-10 15:59:25 +010096.. container:: ulist
ramverma3b71c972019-07-10 11:25:37 +000097
liamfallonf3981072020-07-10 15:59:25 +010098 - Simple / God: a simple policy with 1 state and 1 task, which is doing everything for the decision-making. This
99 is the ideal policy for simple situation, e.g. deciding on configuration parameters or simple access control.
ramverma3b71c972019-07-10 11:25:37 +0000100
liamfallonf3981072020-07-10 15:59:25 +0100101 - Simple sequence: a simple policy with a number of states each having a single task. This is a very good policy
102 for simple decision-making with different steps. For instance, a classic action policy (ECA) would have 3 states
103 (E, C, and A) with some logic (1 task) in each state.
ramverma3b71c972019-07-10 11:25:37 +0000104
liamfallonf3981072020-07-10 15:59:25 +0100105 - Simple selective: a policy with 1 state but more than one task. Here, the appropriate task (and it’s logic)
106 will be selected at execution time. This policy is very good for dealing with similar (or the same) situation in
107 different contexts. For instance, the tasks can be related to available external software, or to current work load
108 on the compute node, or to time of day.
ramverma3b71c972019-07-10 11:25:37 +0000109
liamfallonf3981072020-07-10 15:59:25 +0100110 - Selective: any number of states having any number of tasks (usually more than 1 task). This is a combination
111 of the two policies above, for instance an ECA policy with more than one task in E, C, and A.
ramverma3b71c972019-07-10 11:25:37 +0000112
liamfallonf3981072020-07-10 15:59:25 +0100113 - Classic directed: a policy with more than one state, each having one task, but a non-sequential execution.
114 This means that the sequence of the states is not pre-defined in the policy (as would be for all cases above) but
115 calculated at runtime. This can be good to realize decision trees based on contextual information.
ramverma3b71c972019-07-10 11:25:37 +0000116
liamfallonf3981072020-07-10 15:59:25 +0100117 - Super Adaptive: using the full potential of the APEX policy model, states and tasks and state execution are
118 fully flexible and calculated at runtime (per policy execution). This policy is very close to a general
119 programming system (with only a few limitations), but can solve very hard problems.
ramverma3b71c972019-07-10 11:25:37 +0000120
liamfallonf3981072020-07-10 15:59:25 +0100121.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000122
liamfallonf3981072020-07-10 15:59:25 +0100123 The final step is to select a response that the policy creates. Possible responses have been discussed in the
124 literature for a very long time. A few examples are:
ramverma3b71c972019-07-10 11:25:37 +0000125
liamfallonf3981072020-07-10 15:59:25 +0100126.. container:: ulist
ramverma3b71c972019-07-10 11:25:37 +0000127
liamfallonf3981072020-07-10 15:59:25 +0100128 - Obligation (deontic for what should happen)
ramverma3b71c972019-07-10 11:25:37 +0000129
liamfallonf3981072020-07-10 15:59:25 +0100130 - Authorization (e.g. for rule-based or other access control or security systems)
ramverma3b71c972019-07-10 11:25:37 +0000131
liamfallonf3981072020-07-10 15:59:25 +0100132 - Intent (instead of providing detailed actions the response is an intent statement and a further system
133 processes that)
ramverma3b71c972019-07-10 11:25:37 +0000134
liamfallonf3981072020-07-10 15:59:25 +0100135 - Delegation (hand the problem over to someone else, possibly with some information or instructions)
ramverma3b71c972019-07-10 11:25:37 +0000136
liamfallonf3981072020-07-10 15:59:25 +0100137 - Fail / Error (the policy has encountered a problem, and reports it)
ramverma3b71c972019-07-10 11:25:37 +0000138
liamfallonf3981072020-07-10 15:59:25 +0100139 - Feedback (why did the policy make a certain decision)
ramverma3b71c972019-07-10 11:25:37 +0000140
liamfallonf3981072020-07-10 15:59:25 +0100141*****************
ramverma3b71c972019-07-10 11:25:37 +0000142APEX Policy Model
liamfallonf3981072020-07-10 15:59:25 +0100143*****************
ramverma3b71c972019-07-10 11:25:37 +0000144
liamfallonf3981072020-07-10 15:59:25 +0100145.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000146
liamfallonf3981072020-07-10 15:59:25 +0100147 The APEX policy model is shown in UML notation in the figure below. A policy model can be stored in JSON or XML
148 format in a file or can be held in a database. The APEX editor creates and modifies APEX policy models. APEX
149 deployment deploys policy models, and a policy model is loaded into APEX engines so that the engines can run the
150 policies in the policy model.
ramverma3b71c972019-07-10 11:25:37 +0000151
liamfallonf3981072020-07-10 15:59:25 +0100152.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000153
liamfallonf3981072020-07-10 15:59:25 +0100154 The figure shows four different views of the policy model:
ramverma3b71c972019-07-10 11:25:37 +0000155
liamfallonf3981072020-07-10 15:59:25 +0100156.. container:: ulist
ramverma3b71c972019-07-10 11:25:37 +0000157
liamfallonf3981072020-07-10 15:59:25 +0100158 - The general model view shows the main parts of a policy: state, state output, event, and task. A task can also
159 have parameters. Data types can be defined on a per-model basis using either standard atomic types (such as
160 character, string, numbers) or complex types from a policy domain.
ramverma3b71c972019-07-10 11:25:37 +0000161
liamfallonf3981072020-07-10 15:59:25 +0100162 - The logic model view emphasizes how decision-making logic is injected into a policy. There are essentially
163 three different types of logic: task logic (for decision making in a task), task selection logic (to select a task
164 if more than one is defined in a state), and state finalizer logic (to compute the final output event of a state
165 and select an appropriate next state from the policy model).
ramverma3b71c972019-07-10 11:25:37 +0000166
liamfallonf3981072020-07-10 15:59:25 +0100167 - The context model view shows how context is injected into a policy. States collect all context from their
168 tasks. A task can define what context it requires for the decision making, i.e. what context the task logic will
169 process. Context itself is a collection of items (individual context information) with data types. Context can be
170 templated.
ramverma3b71c972019-07-10 11:25:37 +0000171
liamfallonf3981072020-07-10 15:59:25 +0100172 - The event and field model view shows the events in the policy model. Tasks define what information they
173 consume (input) and produce (output). This information is modeled as fields, essentially a key/type tuple in the
174 model and a key/type/value triple at execution. Events then are collection of fields.
ramverma3b71c972019-07-10 11:25:37 +0000175
liamfallonf3981072020-07-10 15:59:25 +0100176.. container:: imageblock
ramverma3b71c972019-07-10 11:25:37 +0000177
liamfallonf3981072020-07-10 15:59:25 +0100178 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +0000179
liamfallonf3981072020-07-10 15:59:25 +0100180 |APEX Policy Model for Execution|
ramverma3b71c972019-07-10 11:25:37 +0000181
liamfallonf3981072020-07-10 15:59:25 +0100182 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +0000183
liamfallonf3981072020-07-10 15:59:25 +0100184 Figure 2. APEX Policy Model for Execution
ramverma3b71c972019-07-10 11:25:37 +0000185
186Concepts and Keys
liamfallonf3981072020-07-10 15:59:25 +0100187=================
ramverma3b71c972019-07-10 11:25:37 +0000188
liamfallonf3981072020-07-10 15:59:25 +0100189.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000190
liamfallonf3981072020-07-10 15:59:25 +0100191 Each element of the policy model is called a *concept*. Each *concept* is a subclass of the abstract *Concept*
192 class, as shown in the next figure. Every concept implements the following abstract methods:
ramverma3b71c972019-07-10 11:25:37 +0000193
liamfallonf3981072020-07-10 15:59:25 +0100194.. container:: imageblock
ramverma3b71c972019-07-10 11:25:37 +0000195
liamfallonf3981072020-07-10 15:59:25 +0100196 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +0000197
liamfallonf3981072020-07-10 15:59:25 +0100198 |Concepts and Keys|
ramverma3b71c972019-07-10 11:25:37 +0000199
liamfallonf3981072020-07-10 15:59:25 +0100200 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +0000201
liamfallonf3981072020-07-10 15:59:25 +0100202 Figure 3. Concepts and Keys
ramverma3b71c972019-07-10 11:25:37 +0000203
liamfallonf3981072020-07-10 15:59:25 +0100204.. container:: ulist
ramverma3b71c972019-07-10 11:25:37 +0000205
liamfallonf3981072020-07-10 15:59:25 +0100206 - ``getKey()`` - gets the unique key for this concept instance in the system
ramverma3b71c972019-07-10 11:25:37 +0000207
liamfallonf3981072020-07-10 15:59:25 +0100208 - ``validate()`` - validates the structure of this concept, its sub-concepts and its relationships
ramverma3b71c972019-07-10 11:25:37 +0000209
liamfallonf3981072020-07-10 15:59:25 +0100210 - ``clean()`` - carries out housekeeping on the concept such as trimming strings, remove any hanging references
ramverma3b71c972019-07-10 11:25:37 +0000211
liamfallonf3981072020-07-10 15:59:25 +0100212 - ``clone()`` - creates a deep copy of an instance of this concept
ramverma3b71c972019-07-10 11:25:37 +0000213
liamfallonf3981072020-07-10 15:59:25 +0100214 - ``equals()`` - checks if two instances of this concept are equal
ramverma3b71c972019-07-10 11:25:37 +0000215
liamfallonf3981072020-07-10 15:59:25 +0100216 - ``toString()`` - returns a string representation of the concept
ramverma3b71c972019-07-10 11:25:37 +0000217
liamfallonf3981072020-07-10 15:59:25 +0100218 - ``hashCode()`` - returns a hash code for the concept
ramverma3b71c972019-07-10 11:25:37 +0000219
liamfallonf3981072020-07-10 15:59:25 +0100220 - ``copyTo()`` - carries out a deep copy of one instance of the concept to another instance, overwriting the
221 target fields.
ramverma3b71c972019-07-10 11:25:37 +0000222
liamfallonf3981072020-07-10 15:59:25 +0100223.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000224
liamfallonf3981072020-07-10 15:59:25 +0100225 All concepts must have a *key*, which uniquely identifies a concept instance. The *key* of a subclass of an *Concept*
226 must either be an ``ArtifactKey`` or an ``ReferenceKey``. Concepts that have a stand-alone independent existence such
227 as *Policy*, *Task*, and *Event* must have an ``ArtifctKey`` key. Concepts that are contained in other concepts, that
228 do not exist as stand-alone concepts must have an ``ReferenceKey`` key. Examples of such concepts are *State* and
229 *EventParameter*.
ramverma3b71c972019-07-10 11:25:37 +0000230
liamfallonf3981072020-07-10 15:59:25 +0100231.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000232
liamfallonf3981072020-07-10 15:59:25 +0100233 An ``ArticactKey`` has two fields; the *Name* of the concept it is the key for and the concept’s *Version*. A
234 concept’s name must be unique in a given PolicyModel. A concept version is represented using the well known
235 *major.minor.path* scheme as used in semantic versioning.
ramverma3b71c972019-07-10 11:25:37 +0000236
liamfallonf3981072020-07-10 15:59:25 +0100237.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000238
liamfallonf3981072020-07-10 15:59:25 +0100239 A ``ReferenceKey`` has three fields. The *UserKeyName* and *UserKeyVersion* fields identify the ``ArtifactKey`` of
240 the concept in which the concept keyed by the ``ReferenceKey`` is contained. The *LocalName* field identifies the
241 contained concept instance. The *LocalName* must be unique in the concepts of a given type contained by a parent.
ramverma3b71c972019-07-10 11:25:37 +0000242
liamfallonf3981072020-07-10 15:59:25 +0100243.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000244
liamfallonf3981072020-07-10 15:59:25 +0100245 For example, a policy called ``SalesPolicy`` with a Version of ``1.12.4`` has a state called ``Decide``. The
246 ``Decide`` state is linked to the ``SalesPolicy`` with a ``ReferenceKey`` with fields *UserKeyName* of
247 ``SalesPolicy``, *UserKeyVersion* of ``1.12.4``, and *LocalName* of ``Decide``. There must not be another state
248 called ``Decide`` in the policy ``SalesPolicy``. However, there may well be a state called ``Decide`` in some other
249 policy called ``PurchasingPolicy``.
ramverma3b71c972019-07-10 11:25:37 +0000250
liamfallonf3981072020-07-10 15:59:25 +0100251.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000252
liamfallonf3981072020-07-10 15:59:25 +0100253 Each concept in the model is also a JPA
254 (`Java Persistence API <https://en.wikipedia.org/wiki/Java_Persistence_API>`__) Entity. This means that every concept
255 can be individually persisted or the entire model can be persisted en-bloc to any persistence mechanism using an JPA
256 framework such as `Hibernate <http://hibernate.org/>`__ or `EclipseLink <http://www.eclipse.org/eclipselink/>`__.
ramverma3b71c972019-07-10 11:25:37 +0000257
258Concept: PolicyModel
liamfallonf3981072020-07-10 15:59:25 +0100259====================
ramverma3b71c972019-07-10 11:25:37 +0000260
liamfallonf3981072020-07-10 15:59:25 +0100261.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000262
liamfallonf3981072020-07-10 15:59:25 +0100263 The *PolicyModel* concept is a container that holds the definition of a set of policies and their associated events,
264 context maps, and tasks. A *PolicyModel* is implemented as four maps for policies, events, context maps, and tasks.
265 Each map is indexed by the key of the policy, event, context map, or task. Any non-empty policy model must have at
266 least one entry in its policy, event, and task map because all policies must have at least one input and output event
267 and must execute at least one task.
ramverma3b71c972019-07-10 11:25:37 +0000268
liamfallonf3981072020-07-10 15:59:25 +0100269.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000270
liamfallonf3981072020-07-10 15:59:25 +0100271 A *PolicyModel* concept is keyed with an ``ArtifactKey key``. Because a *PolicyModel* is an ``AxConcept``, calling
272 the ``validate()`` method on a policy model validates the concepts, structure, and relationships of the entire policy
273 model.
ramverma3b71c972019-07-10 11:25:37 +0000274
275Concept: DataType
liamfallonf3981072020-07-10 15:59:25 +0100276=================
ramverma3b71c972019-07-10 11:25:37 +0000277
liamfallonf3981072020-07-10 15:59:25 +0100278.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000279
liamfallonf3981072020-07-10 15:59:25 +0100280 Data types are tightly controlled in APEX in order to provide a very high degree of consistency in policies and to
281 facilitate tracking of changes to context as policies execute. All context is modeled as a *DataType* concept. Each
282 DataType concept instance is keyed with an ``ArtifactKey`` key. The DataType field identifies the Java class of
283 objects that is used to represent concept instances that use this data type. All context has a *DataType*; incoming
284 and outgoing context is represented by *EventField* concepts and all other context is represented by *ContextItem*
285 concepts.
ramverma3b71c972019-07-10 11:25:37 +0000286
287Concept: Event
liamfallonf3981072020-07-10 15:59:25 +0100288==============
ramverma3b71c972019-07-10 11:25:37 +0000289
liamfallonf3981072020-07-10 15:59:25 +0100290.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000291
liamfallonf3981072020-07-10 15:59:25 +0100292 An *Event* defines the structure of a message that passes into or out of an APEX engine or that passes between two
293 states in an APEX engine. APEX supports message reception and sending in many formats and all messages are translated
294 into an *Event* prior to processing by an APEX engine. Event concepts are keyed with an ``ArtifactKey`` key. The
295 parameters of an event are held as a map of *EventField* concept instances with each parameter indexed by the
296 *LocalName* of its ``ReferenceKey``. An *Event* has three fields:
ramverma3b71c972019-07-10 11:25:37 +0000297
liamfallonf3981072020-07-10 15:59:25 +0100298.. container:: ulist
ramverma3b71c972019-07-10 11:25:37 +0000299
liamfallonf3981072020-07-10 15:59:25 +0100300 - The *NameSpace* identifies the domain of application of the event
ramverma3b71c972019-07-10 11:25:37 +0000301
liamfallonf3981072020-07-10 15:59:25 +0100302 - The *Source* of the event identifies the system that emitted the event
ramverma3b71c972019-07-10 11:25:37 +0000303
liamfallonf3981072020-07-10 15:59:25 +0100304 - The *Target* of the event identifies the system that the event was sent to
ramverma3b71c972019-07-10 11:25:37 +0000305
liamfallonf3981072020-07-10 15:59:25 +0100306.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000307
liamfallonf3981072020-07-10 15:59:25 +0100308 A *PolicyModel* contains a map of all the events known to a given policy model. Although an empty model may have no
309 events in its event map, any sane policy model must have at least one *Event* defined.
ramverma3b71c972019-07-10 11:25:37 +0000310
311Concept: EventField
liamfallonf3981072020-07-10 15:59:25 +0100312===================
ramverma3b71c972019-07-10 11:25:37 +0000313
liamfallonf3981072020-07-10 15:59:25 +0100314.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000315
liamfallonf3981072020-07-10 15:59:25 +0100316 The incoming context and outgoing context of an event are the fields of the event. Each field representing a single
317 piece of incoming or outgoing context. Each field of an *Event* is represented by an instance of the *EventField*
318 concept. Each *EventField* concept instance in an event is keyed with a ``ReferenceKey`` key, which references the
319 event. The *LocalName* field of the ``ReferenceKey`` holds the name of the field A reference to a *DataType* concept
320 defines the data type that values of this parameter have at run time.
ramverma3b71c972019-07-10 11:25:37 +0000321
322Concept: ContextMap
liamfallonf3981072020-07-10 15:59:25 +0100323===================
ramverma3b71c972019-07-10 11:25:37 +0000324
liamfallonf3981072020-07-10 15:59:25 +0100325.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000326
liamfallonf3981072020-07-10 15:59:25 +0100327 The set of context that is available for use by the policies of a *PolicyModel* is defined as *ContextMap* concept
328 instances. The *PolicyModel* holds a map of all the *ContextMap* definitions. A *ContextMap* is itself a container
329 for a group of related context items, each of which is represented by a *ContextItem* concept instance. *ContextMap*
330 concepts are keyed with an ``ArtifactKey`` key. A developer can use the APEX Policy Editor to create context maps for
331 their application domain.
ramverma3b71c972019-07-10 11:25:37 +0000332
liamfallonf3981072020-07-10 15:59:25 +0100333.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000334
liamfallonf3981072020-07-10 15:59:25 +0100335 A *ContextMap* uses a map to hold the context items. The ContextItem concept instances in the map are indexed by the
336 *LocalName* of their ``ReferenceKey``.
ramverma3b71c972019-07-10 11:25:37 +0000337
liamfallonf3981072020-07-10 15:59:25 +0100338.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000339
liamfallonf3981072020-07-10 15:59:25 +0100340 The *ContextMapType* field of a *ContextMap* defines the type of a context map. The type can have either of two
341 values:
ramverma3b71c972019-07-10 11:25:37 +0000342
liamfallonf3981072020-07-10 15:59:25 +0100343.. container:: ulist
ramverma3b71c972019-07-10 11:25:37 +0000344
liamfallonf3981072020-07-10 15:59:25 +0100345 - A *BAG* context map is a context map with fixed content. Each possible context item in the context map is
346 defined at design time and is held in the *ContextMap* context instance as *ContextItem* concept definitions and
347 only the values of the context items in the context map can be changed at run time. The context items in a *BAG*
348 context map have mixed types and distinct *ContextItem* concept instances of the same type can be defined. A *BAG*
349 context map is convenient for defining a group of context items that are diverse but are related by domain, such as
350 the characteristics of a device. A fully defined *BAG* context map has a fully populated *ContextItem* map but its
351 *ContextItemTemplate* reference is not defined.
ramverma3b71c972019-07-10 11:25:37 +0000352
liamfallonf3981072020-07-10 15:59:25 +0100353 - A *SAMETYPE* context map is used to represent a group of *ContextItem* instances of the same type. Unlike a
354 *BAG* context map, the *ContextItem* concept instances of a *SAMETYPE* context map can be added, modified, and
355 deleted at runtime. All *ContextItem* concept instances in a *SAMETYPE* context map must be of the same type, and
356 that context item is defined as a single *ContextItemTemplate* concept instances at design time. At run time, the
357 *ContextItemTemplate* definition is used to create new *ContextItem* concept instances for the context map on
358 demand. A fully defined *SAMETYPE context map has an empty ContextItem map and its ContextItemTemplate\_*
359 reference is defined.
ramverma3b71c972019-07-10 11:25:37 +0000360
liamfallonf3981072020-07-10 15:59:25 +0100361.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000362
liamfallonf3981072020-07-10 15:59:25 +0100363 The *Scope* of a *ContextMap* defines the range of applicability of a context map in APEX. The following scopes of
364 applicability are defined:
ramverma3b71c972019-07-10 11:25:37 +0000365
liamfallonf3981072020-07-10 15:59:25 +0100366.. container:: ulist
ramverma3b71c972019-07-10 11:25:37 +0000367
liamfallonf3981072020-07-10 15:59:25 +0100368 - *EPHEMERAL* scope means that the context map is owned, used, and modified by a single application but the
369 context map only exists while that application is running
ramverma3b71c972019-07-10 11:25:37 +0000370
liamfallonf3981072020-07-10 15:59:25 +0100371 - *APPLICATION* scope specifies that the context map is owned, used, and modified by a single application, the
372 context map is persistent
ramverma3b71c972019-07-10 11:25:37 +0000373
liamfallonf3981072020-07-10 15:59:25 +0100374 - *GLOBAL* scope specifies that the context map is globally owned and is used and modified by any application,
375 the context map is persistent
ramverma3b71c972019-07-10 11:25:37 +0000376
liamfallonf3981072020-07-10 15:59:25 +0100377 - *EXTERNAL* scope specifies that the context map is owned by an external system and may be used in a read-only
378 manner by any application, the context map is persistent
ramverma3b71c972019-07-10 11:25:37 +0000379
liamfallonf3981072020-07-10 15:59:25 +0100380.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000381
liamfallonf3981072020-07-10 15:59:25 +0100382 A much more sophisticated scoping mechanism for context maps is envisaged for Apex in future work. In such a
383 mechanism, the scope of a context map would work somewhat like the way roles work in security authentication systems.
ramverma3b71c972019-07-10 11:25:37 +0000384
385Concept: ContextItem
liamfallonf3981072020-07-10 15:59:25 +0100386====================
ramverma3b71c972019-07-10 11:25:37 +0000387
liamfallonf3981072020-07-10 15:59:25 +0100388.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000389
liamfallonf3981072020-07-10 15:59:25 +0100390 Each piece of context in a *ContextMap* is represented by an instance of the *ContextItem* concept. Each
391 *ContextItem* concept instance in a context map keyed with a ``ReferenceKey`` key, which references the context map
392 of the context item. The *LocalName* field of the ``ReferenceKey`` holds the name of the context item in the context
393 map A reference to a *DataType* concept defines the data type that values of this context item have at run time. The
394 *WritableFlag* indicates if the context item is read only or read-write at run time.
ramverma3b71c972019-07-10 11:25:37 +0000395
396Concept: ContextItemTemplate
liamfallonf3981072020-07-10 15:59:25 +0100397============================
ramverma3b71c972019-07-10 11:25:37 +0000398
liamfallonf3981072020-07-10 15:59:25 +0100399.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000400
liamfallonf3981072020-07-10 15:59:25 +0100401 In a *SAMETYPE* *ContextMap*, the *ContextItemTemplate* definition provides a template for the *ContextItem*
402 instances that will be created on the context map at run time. Each *ContextItem* concept instance in the context map
403 is created using the *ContextItemTemplate* template. It is keyed with a ``ReferenceKey`` key, which references the
404 context map of the context item. The *LocalName* field of the ``ReferenceKey``, supplied by the creator of the
405 context item at run time, holds the name of the context item in the context map. A reference to a *DataType* concept
406 defines the data type that values of this context item have at run time. The *WritableFlag* indicates if the context
407 item is read only or read-write at run time.
ramverma3b71c972019-07-10 11:25:37 +0000408
409Concept: Task
liamfallonf3981072020-07-10 15:59:25 +0100410=============
ramverma3b71c972019-07-10 11:25:37 +0000411
liamfallonf3981072020-07-10 15:59:25 +0100412.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000413
liamfallonf3981072020-07-10 15:59:25 +0100414 The smallest unit of logic in a policy is a *Task*. A task encapsulates a single atomic unit of logic, and is
415 designed to be a single indivisible unit of execution. A task may be invoked by a single policy or by many policies.
416 A task has a single trigger event, which is sent to the task when it is invoked. Tasks emit one or more outgoing
417 events, which carry the result of the task execution. Tasks may use or modify context as they execute.
ramverma3b71c972019-07-10 11:25:37 +0000418
liamfallonf3981072020-07-10 15:59:25 +0100419.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000420
liamfallonf3981072020-07-10 15:59:25 +0100421 The Task concept definition captures the definition of an APEX task. Task concepts are keyed with an ``ArtifactKey``
422 key. The Trigger of the task is a reference to the *Event* concept that triggers the task. The *OutgoingEvents* of a
423 task are a set of references to *Event* concepts that may be emitted by the task.
ramverma3b71c972019-07-10 11:25:37 +0000424
liamfallonf3981072020-07-10 15:59:25 +0100425.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000426
liamfallonf3981072020-07-10 15:59:25 +0100427 All tasks have logic, some code that is programmed to execute the work of the task. The *Logic* concept of the task
428 holds the definition of that logic.
ramverma3b71c972019-07-10 11:25:37 +0000429
liamfallonf3981072020-07-10 15:59:25 +0100430.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000431
liamfallonf3981072020-07-10 15:59:25 +0100432 The *Task* definition holds a set of *ContextItem* and *ContextItemTemplate* context items that the task is allow to
433 access, as defined by the task developer at design time. The type of access (read-only or read write) that a task has
434 is determined by the *WritableFlag* flag on the individual context item definitions. At run time, a task may only
435 access the context items specified in its context item set, the APEX engine makes only the context items in the task
436 context item set is available to the task.
ramverma3b71c972019-07-10 11:25:37 +0000437
liamfallonf3981072020-07-10 15:59:25 +0100438.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000439
liamfallonf3981072020-07-10 15:59:25 +0100440 A task can be configured with startup parameters. The set of parameters that can be configured on a task are defined
441 as a set of *TaskParameter* concept definitions.
ramverma3b71c972019-07-10 11:25:37 +0000442
443Concept: TaskParameter
liamfallonf3981072020-07-10 15:59:25 +0100444======================
ramverma3b71c972019-07-10 11:25:37 +0000445
liamfallonf3981072020-07-10 15:59:25 +0100446.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000447
liamfallonf3981072020-07-10 15:59:25 +0100448 Each configuration parameter of a task are represented as a *Taskparameter* concept keyed with a ``ReferenceKey``
449 key, which references the task. The *LocalName* field of the ``ReferenceKey`` holds the name of the parameter. The
450 *DefaultValue* field defines the default value that the task parameter is set to. The value of *TaskParameter*
451 instances can be overridden at deployment time by specifying their values in the configuration information passed to
452 APEX engines.
ramverma3b71c972019-07-10 11:25:37 +0000453
liamfallonf3981072020-07-10 15:59:25 +0100454.. container:: paragraph
a.sreekumarcc0e9172020-03-16 13:36:45 +0000455
liamfallonf3981072020-07-10 15:59:25 +0100456 The *taskParameters* field is specified under *engineParameters* in the ApexConfig. It can contain one or more task
457 parameters, where each item can contain the parameter key, value as well as the taskId to which it is associated. If
458 the taskId is not specified, then the parameters are added to all tasks.
a.sreekumarcc0e9172020-03-16 13:36:45 +0000459
ramverma3b71c972019-07-10 11:25:37 +0000460Concept: Logic
liamfallonf3981072020-07-10 15:59:25 +0100461==============
ramverma3b71c972019-07-10 11:25:37 +0000462
liamfallonf3981072020-07-10 15:59:25 +0100463.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000464
liamfallonf3981072020-07-10 15:59:25 +0100465 The *Logic* concept instance holds the actual programmed task logic for a task defined in a *Task* concept or the
466 programmed task selection logic for a state defined in a *State* concept. It is keyed with a ``ReferenceKey`` key,
467 which references the task or state that owns the logic. The *LocalName* field of the Logic concept is the name of the
468 logic.
ramverma3b71c972019-07-10 11:25:37 +0000469
liamfallonf3981072020-07-10 15:59:25 +0100470.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000471
liamfallonf3981072020-07-10 15:59:25 +0100472 The *LogicCode* field of a Logic concept definition is a string that holds the program code that is to be executed
473 at run time. The *LogicType* field defines the language of the code. The standard values are the logic languages
rameshiyer27b629de12024-08-06 23:20:49 +0100474 supported by APEX: `JAVASCRIPT <https://en.wikipedia.org/wiki/JavaScript>`__, `JAVA <https://www.oracle.com/java/>`__,
liamfallonf3981072020-07-10 15:59:25 +0100475 `JYTHON <http://www.jython.org/>`__, `JRUBY <http://jruby.org/>`__, or
476 `MVEL <https://en.wikibooks.org/wiki/Transwiki:MVEL_Language_Guide>`__.
ramverma3b71c972019-07-10 11:25:37 +0000477
liamfallonf3981072020-07-10 15:59:25 +0100478.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000479
liamfallonf3981072020-07-10 15:59:25 +0100480 The APEX engine uses the *LogicType* field value to decide which language interpreter to use for a task and then
481 sends the logic defined in the *LogicCode* field to that interpreter.
ramverma3b71c972019-07-10 11:25:37 +0000482
483Concept: Policy
liamfallonf3981072020-07-10 15:59:25 +0100484===============
ramverma3b71c972019-07-10 11:25:37 +0000485
liamfallonf3981072020-07-10 15:59:25 +0100486.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000487
liamfallonf3981072020-07-10 15:59:25 +0100488 The *Policy* concept defines a policy in APEX. The definition is rather straightforward. A policy is made up of a
489 set of states with the flavor of the policy determining the structure of the policy states and the first state
490 defining what state in the policy executes first. *Policy* concepts are keyed with an ``ArtifactKey`` key.
ramverma3b71c972019-07-10 11:25:37 +0000491
liamfallonf3981072020-07-10 15:59:25 +0100492.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000493
liamfallonf3981072020-07-10 15:59:25 +0100494 The *PolicyFlavour* of a *Policy* concept specifies the structure that will be used for the states in the policy. A
495 number of commonly used policy patterns are supported as APEX policy flavors. The standard policy flavors are:
ramverma3b71c972019-07-10 11:25:37 +0000496
liamfallonf3981072020-07-10 15:59:25 +0100497.. container:: ulist
ramverma3b71c972019-07-10 11:25:37 +0000498
liamfallonf3981072020-07-10 15:59:25 +0100499 - The *MEDA* flavor supports policies written to the
FrancescoFioraEstad9a1e82023-04-14 14:05:31 +0100500 `MEDA policy pattern https://ieeexplore.ieee.org/document/7367357`
liamfallonf3981072020-07-10 15:59:25 +0100501 and require a sequence of four states: namely *Match*, *Establish*, *Decide* and *Act*.
ramverma3b71c972019-07-10 11:25:37 +0000502
liamfallonf3981072020-07-10 15:59:25 +0100503 - The *OODA* flavor supports policies written to the
504 `OODA loop pattern <https://en.wikipedia.org/wiki/OODA_loop>`__ and require a sequence of four states: namely
505 *Observe*, *Orient*, *Decide* and *Act*.
ramverma3b71c972019-07-10 11:25:37 +0000506
liamfallonf3981072020-07-10 15:59:25 +0100507 - The *ECA* flavor supports policies written to the
508 `ECA active rule pattern <https://en.wikipedia.org/wiki/Event_condition_action>`__ and require a sequence of three
509 states: namely *Event*, *Condition* and *Action*
ramverma3b71c972019-07-10 11:25:37 +0000510
liamfallonf3981072020-07-10 15:59:25 +0100511 - The *XACML* flavor supports policies written in `XACML <https://en.wikipedia.org/wiki/XACML>`__ and require a
512 single state: namely *XACML*
ramverma3b71c972019-07-10 11:25:37 +0000513
liamfallonf3981072020-07-10 15:59:25 +0100514 - The *FREEFORM* flavor supports policies written in an arbitrary style. A user can define a *FREEFORM* policy
515 as an arbitrarily long chain of states.
ramverma3b71c972019-07-10 11:25:37 +0000516
liamfallonf3981072020-07-10 15:59:25 +0100517.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000518
liamfallonf3981072020-07-10 15:59:25 +0100519 The *FirstState* field of a *Policy* definition is the starting point for execution of a policy. Therefore, the
520 trigger event of the state referenced in the *FirstState* field is also the trigger event for the entire policy.
ramverma3b71c972019-07-10 11:25:37 +0000521
522Concept: State
liamfallonf3981072020-07-10 15:59:25 +0100523==============
ramverma3b71c972019-07-10 11:25:37 +0000524
liamfallonf3981072020-07-10 15:59:25 +0100525.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000526
liamfallonf3981072020-07-10 15:59:25 +0100527 The *State* concept represents a phase or a stage in a policy, with a policy being composed of a series of states.
528 Each state has at least one but may have many tasks and, on each run of execution, a state executes one and only one
529 of its tasks. If a state has more than one task, then its task selection logic is used to select which task to
530 execute. Task selection logic is programmable logic provided by the state designer. That logic can use incoming,
531 policy, global, and external context to select which task best accomplishes the purpose of the state in a give
532 situation if more than one task has been specified on a state. A state calls one and only one task when it is
533 executed.
ramverma3b71c972019-07-10 11:25:37 +0000534
liamfallonf3981072020-07-10 15:59:25 +0100535.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000536
liamfallonf3981072020-07-10 15:59:25 +0100537 Each state is triggered by an event, which means that all tasks of a state must also be triggered by that same
538 event. The set of output events for a state is the union of all output events from all tasks for that task. In
539 practice at the moment, because a state can only have a single input event, a state that is not the final state of a
540 policy may only output a single event and all tasks of that state may also only output that single event. In future
541 work, the concept of having a less restrictive trigger pattern will be examined.
ramverma3b71c972019-07-10 11:25:37 +0000542
liamfallonf3981072020-07-10 15:59:25 +0100543.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000544
a.sreekumar98b59692021-07-23 13:38:06 +0100545 A state that is the final state of a policy may output multiple events, and the task associated with the final state
546 outputs those events.
547
548.. container:: paragraph
549
liamfallonf3981072020-07-10 15:59:25 +0100550 A *State* concept is keyed with a ``ReferenceKey`` key, which references the *Policy* concept that owns the state.
551 The *LocalName* field of the ``ReferenceKey`` holds the name of the state. As a state is part of a chain of states,
552 the *NextState* field of a state holds the ``ReferenceKey`` key of the state in the policy to execute after this
553 state.
ramverma3b71c972019-07-10 11:25:37 +0000554
liamfallonf3981072020-07-10 15:59:25 +0100555.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000556
liamfallonf3981072020-07-10 15:59:25 +0100557 The *Trigger* field of a state holds the ``ArtifactKey`` of the event that triggers this state. The *OutgoingEvents*
558 field holds the ``ArtifactKey`` references of all possible events that may be output from the state. This is a set
559 that is the union of all output events of all tasks of the state.
ramverma3b71c972019-07-10 11:25:37 +0000560
liamfallonf3981072020-07-10 15:59:25 +0100561.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000562
liamfallonf3981072020-07-10 15:59:25 +0100563 The *Task* concepts that hold the definitions of the task for the state are held as a set of ``ArtifactKey``
564 references in the state. The *DefaultTask* field holds a reference to the default task for the state, a task that is
565 executed if no task selection logic is specified. If the state has only one task, that task is the default task.
ramverma3b71c972019-07-10 11:25:37 +0000566
liamfallonf3981072020-07-10 15:59:25 +0100567.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000568
liamfallonf3981072020-07-10 15:59:25 +0100569 The *Logic* concept referenced by a state holds the task selection logic for a state. The task selection logic uses
570 the incoming context (parameters of the incoming event) and other context to determine the best task to use to
571 execute its goals. The state holds a set of references to *ContextItem* and *ContextItemTemplate* definitions for the
572 context used by its task selection logic.
ramverma3b71c972019-07-10 11:25:37 +0000573
liamfallonf3981072020-07-10 15:59:25 +0100574*************
ramverma3b71c972019-07-10 11:25:37 +0000575Writing Logic
liamfallonf3981072020-07-10 15:59:25 +0100576*************
ramverma3b71c972019-07-10 11:25:37 +0000577
578Writing APEX Task Logic
liamfallonf3981072020-07-10 15:59:25 +0100579=======================
ramverma3b71c972019-07-10 11:25:37 +0000580
liamfallonf3981072020-07-10 15:59:25 +0100581.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000582
liamfallonf3981072020-07-10 15:59:25 +0100583 Task logic specifies the behavior of an Apex Task. This logic can be specified in a number of ways, exploiting
584 Apex’s plug-in architecture to support a range of logic executors. In Apex scripted Task Logic can be written in any
585 of these languages:
ramverma3b71c972019-07-10 11:25:37 +0000586
liamfallonf3981072020-07-10 15:59:25 +0100587.. container:: ulist
ramverma3b71c972019-07-10 11:25:37 +0000588
liamfallonf3981072020-07-10 15:59:25 +0100589 - ```MVEL`` <https://en.wikipedia.org/wiki/MVEL>`__,
ramverma3b71c972019-07-10 11:25:37 +0000590
liamfallonf3981072020-07-10 15:59:25 +0100591 - ```JavaScript`` <https://en.wikipedia.org/wiki/JavaScript>`__,
ramverma3b71c972019-07-10 11:25:37 +0000592
liamfallonf3981072020-07-10 15:59:25 +0100593 - ```JRuby`` <https://en.wikipedia.org/wiki/JRuby>`__ or
ramverma3b71c972019-07-10 11:25:37 +0000594
liamfallonf3981072020-07-10 15:59:25 +0100595 - ```Jython`` <https://en.wikipedia.org/wiki/Jython>`__.
ramverma3b71c972019-07-10 11:25:37 +0000596
liamfallonf3981072020-07-10 15:59:25 +0100597.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000598
liamfallonf3981072020-07-10 15:59:25 +0100599 These languages were chosen because the scripts can be compiled into Java bytecode at runtime and then efficiently
600 executed natively in the JVM. Task Logic an also be written directly in Java but needs to be compiled, with the
601 resulting classes added to the classpath. There are also a number of other Task Logic types (e.g. Fuzzy Logic), but
602 these are not supported as yet. This guide will focus on the scripted Task Logic approaches, with MVEL and JavaScript
603 being our favorite languages. In particular this guide will focus on the Apex aspects of the scripts. However, this
604 guide does not attempt to teach you about the scripting languages themselves …​ that is up to you!
ramverma3b71c972019-07-10 11:25:37 +0000605
liamfallonf3981072020-07-10 15:59:25 +0100606.. tip::
607 JVM-based scripting languages For more more information on scripting for the Java platform see:
608 https://docs.oracle.com/javase/8/docs/technotes/guides/scripting/prog_guide/index.html
ramverma3b71c972019-07-10 11:25:37 +0000609
liamfallonf3981072020-07-10 15:59:25 +0100610.. note::
611 What do Tasks do? The function of an Apex Task is to provide the logic that can be executed for an Apex State as one
612 of the steps in an Apex Policy. Each task receives some *incoming fields*, executes some logic (e.g: make a decision
613 based on *shared state* or *context*, *incoming fields*, *external context*, etc.), perhaps set some *shared state*
a.sreekumar98b59692021-07-23 13:38:06 +0100614 or *context* and then emits *outgoing fields* (in case of a single outgoing event), or a set of *outgoing fields*
615 (in case of multiple outgoing events). The state that uses the task is responsible for extracting the
liamfallonf3981072020-07-10 15:59:25 +0100616 *incoming fields* from the state input event. The state also has an *output mapper* associated with the task, and
617 this *output mapper* is responsible for mapping the *outgoing fields* from the task into an appropriate output event
618 for the state.
ramverma3b71c972019-07-10 11:25:37 +0000619
liamfallonf3981072020-07-10 15:59:25 +0100620.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000621
liamfallonf3981072020-07-10 15:59:25 +0100622 First lets start with a sample task, drawn from the "My First Apex Policy" example: The task "MorningBoozeCheck"
623 from the "My First Apex Policy" example is available in both MVEL and JavaScript:
ramverma3b71c972019-07-10 11:25:37 +0000624
liamfallonf3981072020-07-10 15:59:25 +0100625.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +0000626
liamfallonf3981072020-07-10 15:59:25 +0100627 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +0000628
liamfallonf3981072020-07-10 15:59:25 +0100629 Javascript code for the ``MorningBoozeCheck`` task
ramverma3b71c972019-07-10 11:25:37 +0000630
liamfallonf3981072020-07-10 15:59:25 +0100631 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +0000632
liamfallonf3981072020-07-10 15:59:25 +0100633 .. code:: javascript
634 :number-lines:
ramverma3b71c972019-07-10 11:25:37 +0000635
liamfallonf3981072020-07-10 15:59:25 +0100636 /*
637 * ============LICENSE_START=======================================================
638 * Copyright (C) 2016-2018 Ericsson. All rights reserved.
639 * Modifications Copyright (C) 2020 Nordix Foundation.
640 * ================================================================================
641 * Licensed under the Apache License, Version 2.0 (the "License");
642 * you may not use this file except in compliance with the License.
643 * You may obtain a copy of the License at
644 *
645 * http://www.apache.org/licenses/LICENSE-2.0
646 *
647 * Unless required by applicable law or agreed to in writing, software
648 * distributed under the License is distributed on an "AS IS" BASIS,
649 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
650 * See the License for the specific language governing permissions and
651 * limitations under the License.
652 *
653 * SPDX-License-Identifier: Apache-2.0
654 * ============LICENSE_END=========================================================
655 */
ramverma3b71c972019-07-10 11:25:37 +0000656
liamfallonf3981072020-07-10 15:59:25 +0100657 executor.logger.info("Task Execution: '"+executor.subject.id+"'. Input Fields: '"+executor.inFields+"'");
ramverma3b71c972019-07-10 11:25:37 +0000658
liamfallonf3981072020-07-10 15:59:25 +0100659 executor.outFields.put("amount" , executor.inFields.get("amount"));
660 executor.outFields.put("assistant_ID", executor.inFields.get("assistant_ID"));
661 executor.outFields.put("notes" , executor.inFields.get("notes"));
662 executor.outFields.put("quantity" , executor.inFields.get("quantity"));
663 executor.outFields.put("branch_ID" , executor.inFields.get("branch_ID"));
664 executor.outFields.put("item_ID" , executor.inFields.get("item_ID"));
665 executor.outFields.put("time" , executor.inFields.get("time"));
666 executor.outFields.put("sale_ID" , executor.inFields.get("sale_ID"));
ramverma3b71c972019-07-10 11:25:37 +0000667
liamfallonf3981072020-07-10 15:59:25 +0100668 item_id = executor.inFields.get("item_ID");
ramverma3b71c972019-07-10 11:25:37 +0000669
liamfallonf3981072020-07-10 15:59:25 +0100670 //All times in this script are in GMT/UTC since the policy and events assume time is in GMT.
671 var timenow_gmt = new Date(Number(executor.inFields.get("time")));
ramverma3b71c972019-07-10 11:25:37 +0000672
liamfallonf3981072020-07-10 15:59:25 +0100673 var midnight_gmt = new Date(Number(executor.inFields.get("time")));
674 midnight_gmt.setUTCHours(0,0,0,0);
ramverma3b71c972019-07-10 11:25:37 +0000675
liamfallonf3981072020-07-10 15:59:25 +0100676 var eleven30_gmt = new Date(Number(executor.inFields.get("time")));
677 eleven30_gmt.setUTCHours(11,30,0,0);
ramverma3b71c972019-07-10 11:25:37 +0000678
liamfallonf3981072020-07-10 15:59:25 +0100679 var timeformatter = new java.text.SimpleDateFormat("HH:mm:ss z");
ramverma3b71c972019-07-10 11:25:37 +0000680
liamfallonf3981072020-07-10 15:59:25 +0100681 var itemisalcohol = false;
682 if(item_id != null && item_id >=1000 && item_id < 2000)
683 itemisalcohol = true;
ramverma3b71c972019-07-10 11:25:37 +0000684
liamfallonf3981072020-07-10 15:59:25 +0100685 if( itemisalcohol
686 && timenow_gmt.getTime() >= midnight_gmt.getTime()
687 && timenow_gmt.getTime() < eleven30_gmt.getTime()) {
ramverma3b71c972019-07-10 11:25:37 +0000688
liamfallonf3981072020-07-10 15:59:25 +0100689 executor.outFields.put("authorised", false);
690 executor.outFields.put("message", "Sale not authorised by policy task " +
691 executor.subject.taskName+ " for time " + timeformatter.format(timenow_gmt.getTime()) +
692 ". Alcohol can not be sold between " + timeformatter.format(midnight_gmt.getTime()) +
693 " and " + timeformatter.format(eleven30_gmt.getTime()));
694 }
695 else{
696 executor.outFields.put("authorised", true);
697 executor.outFields.put("message", "Sale authorised by policy task " +
698 executor.subject.taskName + " for time "+timeformatter.format(timenow_gmt.getTime()));
699 }
ramverma3b71c972019-07-10 11:25:37 +0000700
liamfallonf3981072020-07-10 15:59:25 +0100701 /*
702 This task checks if a sale request is for an item that is an alcoholic drink.
703 If the local time is between 00:00:00 GMT and 11:30:00 GMT then the sale is not
704 authorised. Otherwise the sale is authorised.
705 In this implementation we assume that items with item_ID value between 1000 and
706 2000 are all alcoholic drinks :-)
707 */
ramverma3b71c972019-07-10 11:25:37 +0000708
liamfallonf3981072020-07-10 15:59:25 +0100709 true;
ramverma3b71c972019-07-10 11:25:37 +0000710
liamfallonf3981072020-07-10 15:59:25 +0100711.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +0000712
liamfallonf3981072020-07-10 15:59:25 +0100713 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +0000714
liamfallonf3981072020-07-10 15:59:25 +0100715 MVEL code for the ``MorningBoozeCheck`` task
ramverma3b71c972019-07-10 11:25:37 +0000716
liamfallonf3981072020-07-10 15:59:25 +0100717 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +0000718
liamfallonf3981072020-07-10 15:59:25 +0100719 .. code:: javascript
720 :number-lines:
ramverma3b71c972019-07-10 11:25:37 +0000721
liamfallonf3981072020-07-10 15:59:25 +0100722 /*
723 * ============LICENSE_START=======================================================
724 * Copyright (C) 2016-2018 Ericsson. All rights reserved.
725 * Modifications Copyright (C) 2020 Nordix Foundation.
726 * ================================================================================
727 * Licensed under the Apache License, Version 2.0 (the "License");
728 * you may not use this file except in compliance with the License.
729 * You may obtain a copy of the License at
730 *
731 * http://www.apache.org/licenses/LICENSE-2.0
732 *
733 * Unless required by applicable law or agreed to in writing, software
734 * distributed under the License is distributed on an "AS IS" BASIS,
735 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
736 * See the License for the specific language governing permissions and
737 * limitations under the License.
738 *
739 * SPDX-License-Identifier: Apache-2.0
740 * ============LICENSE_END=========================================================
741 */
742 import java.util.Date;
743 import java.util.Calendar;
744 import java.util.TimeZone;
745 import java.text.SimpleDateFormat;
ramverma3b71c972019-07-10 11:25:37 +0000746
liamfallonf3981072020-07-10 15:59:25 +0100747 logger.info("Task Execution: '"+subject.id+"'. Input Fields: '"+inFields+"'");
ramverma3b71c972019-07-10 11:25:37 +0000748
liamfallonf3981072020-07-10 15:59:25 +0100749 outFields.put("amount" , inFields.get("amount"));
750 outFields.put("assistant_ID", inFields.get("assistant_ID"));
751 outFields.put("notes" , inFields.get("notes"));
752 outFields.put("quantity" , inFields.get("quantity"));
753 outFields.put("branch_ID" , inFields.get("branch_ID"));
754 outFields.put("item_ID" , inFields.get("item_ID"));
755 outFields.put("time" , inFields.get("time"));
756 outFields.put("sale_ID" , inFields.get("sale_ID"));
ramverma3b71c972019-07-10 11:25:37 +0000757
liamfallonf3981072020-07-10 15:59:25 +0100758 item_id = inFields.get("item_ID");
ramverma3b71c972019-07-10 11:25:37 +0000759
liamfallonf3981072020-07-10 15:59:25 +0100760 //The events used later to test this task use GMT timezone!
761 gmt = TimeZone.getTimeZone("GMT");
762 timenow = Calendar.getInstance(gmt);
763 df = new SimpleDateFormat("HH:mm:ss z");
764 df.setTimeZone(gmt);
765 timenow.setTimeInMillis(inFields.get("time"));
ramverma3b71c972019-07-10 11:25:37 +0000766
liamfallonf3981072020-07-10 15:59:25 +0100767 midnight = timenow.clone();
768 midnight.set(
769 timenow.get(Calendar.YEAR),timenow.get(Calendar.MONTH),
770 timenow.get(Calendar.DATE),0,0,0);
771 eleven30 = timenow.clone();
772 eleven30.set(
773 timenow.get(Calendar.YEAR),timenow.get(Calendar.MONTH),
774 timenow.get(Calendar.DATE),11,30,0);
ramverma3b71c972019-07-10 11:25:37 +0000775
liamfallonf3981072020-07-10 15:59:25 +0100776 itemisalcohol = false;
777 if(item_id != null && item_id >=1000 && item_id < 2000)
778 itemisalcohol = true;
ramverma3b71c972019-07-10 11:25:37 +0000779
liamfallonf3981072020-07-10 15:59:25 +0100780 if( itemisalcohol
781 && timenow.after(midnight) && timenow.before(eleven30)){
782 outFields.put("authorised", false);
783 outFields.put("message", "Sale not authorised by policy task "+subject.taskName+
784 " for time "+df.format(timenow.getTime())+
785 ". Alcohol can not be sold between "+df.format(midnight.getTime())+
786 " and "+df.format(eleven30.getTime()));
787 return true;
788 }
789 else{
790 outFields.put("authorised", true);
791 outFields.put("message", "Sale authorised by policy task "+subject.taskName+
792 " for time "+df.format(timenow.getTime()));
793 return true;
794 }
ramverma3b71c972019-07-10 11:25:37 +0000795
liamfallonf3981072020-07-10 15:59:25 +0100796 /*
797 This task checks if a sale request is for an item that is an alcoholic drink.
798 If the local time is between 00:00:00 GMT and 11:30:00 GMT then the sale is not
799 authorised. Otherwise the sale is authorised.
800 In this implementation we assume that items with item_ID value between 1000 and
801 2000 are all alcoholic drinks :-)
802 */
ramverma3b71c972019-07-10 11:25:37 +0000803
liamfallonf3981072020-07-10 15:59:25 +0100804.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000805
liamfallonf3981072020-07-10 15:59:25 +0100806 The role of the task in this simple example is to copy the values in the incoming fields into the outgoing
807 fields, then examine the values in some incoming fields (``item_id`` and ``time``), then set the values in some
808 other outgoing fields (``authorised`` and ``message``).
ramverma3b71c972019-07-10 11:25:37 +0000809
liamfallonf3981072020-07-10 15:59:25 +0100810.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000811
liamfallonf3981072020-07-10 15:59:25 +0100812 Both MVEL and JavaScript like most JVM-based scripting languages can use standard Java libraries to perform
813 complex tasks. Towards the top of the scripts you will see how to import Java classes and packages to be used
814 directly in the logic. Another thing to notice is that Task Logic should return a ``java.lang.Boolean`` value
815 ``true`` if the logic executed correctly. If the logic fails for some reason then ``false`` can be returned, but
816 this will cause the policy invoking this task will fail and exit.
ramverma3b71c972019-07-10 11:25:37 +0000817
liamfallonf3981072020-07-10 15:59:25 +0100818.. note::
819 How to return a value from task logic
820 Some languages explicitly support returning values from the script (e.g. MVEL and JRuby) using an explicit
821 return statement (e.g. ``return true``), other languages do not (e.g. Jython). For
822 languages that do not support the ``return`` statement, a special field called ``returnValue`` must be
823 created to hold the result of the task logic operation (i.e. assign a ``java.lang.Boolean``
824 value to the ``returnValue`` field before completing the task).
825 Also, in MVEL if there is no explicit return statement then the return value of the last executed statement will
826 return (e.g. the statement a=(1+2) will return the value 3).
ramverma3b71c972019-07-10 11:25:37 +0000827
liamfallonf3981072020-07-10 15:59:25 +0100828 For Javascript, the last statement of a script must be a statement that evaluates to *true* or *false*, indicating
829 whether the script executed correctly or not. In the case where the script always executes to compeletion
830 sucessfully, simply add a last line with the statement *true'*. In cases where success or failure is assessed in the
831 script, create a boolean
832 local variable with a name such as ``returnvalue``. In the execution of the script, set ``returnValue`` to be ``true``
833 or ``false`` as appropriate. The last line of the scritp tehn should simply be ``returnValue;``, which returns the
834 value of ``returnValue``.
ramverma3b71c972019-07-10 11:25:37 +0000835
liamfallonf3981072020-07-10 15:59:25 +0100836.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000837
liamfallonf3981072020-07-10 15:59:25 +0100838 Besides these imported classes and normal language features Apex provides some natively available parameters
839 and functions that can be used directly. At run-time these parameters are populated by the Apex execution
840 environment and made natively available to logic scripts each time the logic script is invoked. (These can be
841 accessed using the ``executor`` keyword for most languages, or can be accessed directly without the
842 ``executor`` keyword in MVEL):
ramverma3b71c972019-07-10 11:25:37 +0000843
liamfallonf3981072020-07-10 15:59:25 +0100844Table 1. The ``executor`` Fields / Methods
ramverma760cce92019-07-11 12:57:49 +0000845
liamfallonf3981072020-07-10 15:59:25 +0100846 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
847 | Name | Type | Java type | Description |
848 +=====================================================+==========================================================================+===============================+==================================================================================+
849 | inFields | Fields | java.util.Map <String,Object> |The incoming task fields, implemented as a standard Java (unmodifiable) Map |
850 | | | | |
851 | | | |**Example:** |
852 | | | | |
853 | | | |.. code:: javascript |
854 | | | | |
855 | | | | executor.logger.debug("Incoming fields: " +executor.inFields.entrySet()); |
856 | | | | var item_id = executor.incomingFields["item_ID"]; |
857 | | | | if (item_id >=1000) { ... } |
858 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
859 | outFields | Fields | java.util.Map <String,Object> |The outgoing task fields. This is implemented as a standard initially empty Java |
860 | | | |(modifiable) Map. To create a new schema-compliant instance of a field object |
a.sreekumar98b59692021-07-23 13:38:06 +0100861 | | | |see the utility method subject.getOutFieldSchemaHelper() below that takes the |
862 | | | |fieldName as an argument. |
liamfallonf3981072020-07-10 15:59:25 +0100863 | | | | |
864 | | | |**Example:** |
865 | | | | |
866 | | | |.. code:: javascript |
867 | | | | |
868 | | | | executor.outFields["authorised"] = false; |
869 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
a.sreekumar98b59692021-07-23 13:38:06 +0100870 | outFieldsList | Fields | java.util.Collection |The collection of outgoing task fields when there are multiple outputs from the |
871 | | | <Map<String, Object>> |final state. To create a new schema-compliant instance of a field, see the |
872 | | | |utility method subject.getOutFieldSchemaHelper() below that takes eventName and |
873 | | | |fieldName as arguments. |
874 | | | |To add the set of output fields to the outFieldsList, the utility method |
875 | | | |executor.addFieldsToOutput can be used as shown below. |
876 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
877 | void addFieldsToOutput(Map<String, Object> fields) |A utility method to add fields to outgoing fields. | | |
878 | |When there are multiple output events emitted from the task associated | | |
879 | |with a final state, this utility method can be used to add the | | |
880 | |corresponding fields to the outFieldsList. | | |
881 | | | | |
882 | |**Example:** | | |
883 | | | | |
884 | |.. code:: javascript | | |
885 | | | | |
886 | | var cdsRequestEventFields = java.util.HashMap(); | | |
887 | | var actionIdentifiers = executor.subject.getOutFieldSchemaHelper | | |
888 | | ("CDSRequestEvent","actionIdentifiers").createNewInstance(); | | |
889 | | cdsRequestEventFields.put("actionIdentifiers", actionIdentifiers); | | |
890 | | executor.addFieldsToOutput(cdsRequestEventFields); | | |
891 | | | | |
892 | | var logEventFields = java.util.HashMap(); | | |
893 | | logEventFields.put("status", "FINAL_SUCCESS"); | | |
894 | | executor.addFieldsToOutput(logEventFields); | | |
895 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
liamfallonf3981072020-07-10 15:59:25 +0100896 | logger | Logger | org.slf4j.ext.XLogger |A helpful logger |
897 | | | | |
898 | | | |**Example:** |
899 | | | | |
900 | | | |.. code:: javascript |
901 | | | | |
902 | | | | executor.logger.info("Executing task: " +executor.subject.id); |
903 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
904 | TRUE/FALSE | boolean | java.lang.Boolean |2 helpful constants. These are useful to retrieve correct return values for the |
905 | | | |task logic |
906 | | | | |
907 | | | |**Example:** |
908 | | | | |
909 | | | |.. code:: javascript |
910 | | | | |
911 | | | | var returnValue = executor.isTrue; |
912 | | | | var returnValueType = Java.type("java.lang.Boolean"); |
913 | | | | var returnValue = new returnValueType(true); |
914 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
915 | subject | Task | TaskFacade |This provides some useful information about the task that contains this task |
916 | | | |logic. This object has some useful fields and methods : |
917 | | | | |
918 | | | |.. container:: ulist |
919 | | | | |
920 | | | | - **AxTask task** to get access to the full task definition of the host task |
921 | | | | |
922 | | | | - **String getTaskName()** to get the name of the host task |
923 | | | | |
924 | | | | - **String getId()** to get the ID of the host task |
925 | | | | |
926 | | | | - **SchemaHelper getInFieldSchemaHelper( String fieldName )** to |
927 | | | | get a ``SchemaHelper`` helper object to manipulate incoming |
928 | | | | task fields in a schema-aware manner |
929 | | | | |
930 | | | | - **SchemaHelper getOutFieldSchemaHelper( String fieldName )** to |
931 | | | | get a ``SchemaHelper`` helper object to manipulate outgoing |
932 | | | | task fields in a schema-aware manner, e.g. to instantiate new |
933 | | | | schema-compliant field objects to populate the |
a.sreekumar98b59692021-07-23 13:38:06 +0100934 | | | | ``executor.outFields`` outgoing fields map. This can be used only when there |
935 | | | | is a single outgoing event from a task. |
936 | | | | |
937 | | | | - **SchemaHelper getOutFieldSchemaHelper( String eventname, String fieldName )**|
938 | | | | to get a ``SchemaHelper`` helper object to manipulate outgoing |
939 | | | | task fields in a schema-aware manner, e.g. to instantiate new |
940 | | | | schema-compliant field objects to populate the |
941 | | | | ``executor.outFieldsList`` collection of outgoing fields map. This must be |
942 | | | | used in case of multiple outgoing events from a task, as the intention is to |
943 | | | | fetch the schema of a field associated to one of the expected events. |
944 | | | | This method works fine in case of single outgoing event too, but the previous |
945 | | | | method is enough as the field anyway belongs to the single event. |
liamfallonf3981072020-07-10 15:59:25 +0100946 | | | | |
947 | | | |**Example:** |
948 | | | | |
949 | | | |.. code:: javascript |
950 | | | | |
951 | | | | executor.logger.info("Task name: " + executor.subject.getTaskName()); |
952 | | | | executor.logger.info("Task id: " + executor.subject.getId()); |
liamfallonf3981072020-07-10 15:59:25 +0100953 | | | | executor.outFields["authorised"] = executor.subject |
954 | | | | .getOutFieldSchemaHelper("authorised").createNewInstance("false"); |
a.sreekumar98b59692021-07-23 13:38:06 +0100955 | | | | |
956 | | | | var actionIdentifiers = executor.subject.getOutFieldSchemaHelper |
957 | | | | ("CDSRequestEvent","actionIdentifiers").createNewInstance(); |
958 | | | | actionIdentifiers.put("blueprintName", "sample-bp"); |
959 | | | | var cdsRequestEventFields = java.util.HashMap(); |
960 | | | | cdsRequestEventFields.put("actionIdentifiers", actionIdentifiers); |
961 | | | | executor.addFieldsToOutput(cdsRequestEventFields); |
liamfallonf3981072020-07-10 15:59:25 +0100962 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
963 | ContextAlbum getContextAlbum(String ctxtAlbumName ) |A utility method to retrieve a ``ContextAlbum`` for use in the task. | | |
964 | |This is how you access the context used by the task. The returned | | |
965 | |``ContextAlbum`` implements the ``java.util.Map <String,Object>`` | | |
966 | |interface to get and set context as appropriate. The returned | | |
967 | |``ContextAlbum`` also has methods to lock context albums, get | | |
968 | |information about the schema of the items to be stored in a context | | |
969 | |album, and get a ``SchemaHelper`` to manipulate context album items. How | | |
970 | |to define and use context in a task is described in the Apex | | |
971 | |Programmer’s Guide and in the My First Apex Policy guide. | | |
972 | | | | |
973 | |**Example:** | | |
974 | | | | |
975 | |.. code:: javascript | | |
976 | | | | |
977 | | var bkey = executor.inFields.get("branch_ID"); | | |
978 | | var cnts = executor.getContextMap("BranchCounts"); | | |
979 | | cnts.lockForWriting(bkey); | | |
980 | | cnts.put(bkey, cnts.get(bkey) + 1); | | |
981 | | cnts.unlockForWriting(bkey); | | |
982 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
ramverma3b71c972019-07-10 11:25:37 +0000983
984Writing APEX Task Selection Logic
liamfallonf3981072020-07-10 15:59:25 +0100985=================================
ramverma3b71c972019-07-10 11:25:37 +0000986
liamfallonf3981072020-07-10 15:59:25 +0100987.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +0000988
liamfallonf3981072020-07-10 15:59:25 +0100989 The function of Task Selection Logic is to choose which task should be executed for an Apex State as one of
990 the steps in an Apex Policy. Since each state must define a default task there is no need for Task Selection
991 Logic unless the state uses more than one task. This logic can be specified in a number of ways, exploiting
992 Apex’s plug-in architecture to support a range of logic executors. In Apex scripted Task Selection Logic can be
993 written in any of these languages:
ramverma3b71c972019-07-10 11:25:37 +0000994
liamfallonf3981072020-07-10 15:59:25 +0100995.. container:: ulist
ramverma3b71c972019-07-10 11:25:37 +0000996
liamfallonf3981072020-07-10 15:59:25 +0100997 - ```MVEL`` <https://en.wikipedia.org/wiki/MVEL>`__,
ramverma3b71c972019-07-10 11:25:37 +0000998
liamfallonf3981072020-07-10 15:59:25 +0100999 - ```JavaScript`` <https://en.wikipedia.org/wiki/JavaScript>`__,
ramverma3b71c972019-07-10 11:25:37 +00001000
liamfallonf3981072020-07-10 15:59:25 +01001001 - ```JRuby`` <https://en.wikipedia.org/wiki/JRuby>`__ or
ramverma3b71c972019-07-10 11:25:37 +00001002
liamfallonf3981072020-07-10 15:59:25 +01001003 - ```Jython`` <https://en.wikipedia.org/wiki/Jython>`__.
ramverma3b71c972019-07-10 11:25:37 +00001004
liamfallonf3981072020-07-10 15:59:25 +01001005.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001006
liamfallonf3981072020-07-10 15:59:25 +01001007 These languages were chosen because the scripts can be compiled into Java bytecode at runtime and then
1008 efficiently executed natively in the JVM. Task Selection Logic an also be written directly in Java but needs to
1009 be compiled, with the resulting classes added to the classpath. There are also a number of other Task Selection
1010 Logic types but these are not supported as yet. This guide will focus on the scripted Task Selection Logic
1011 approaches, with MVEL and JavaScript being our favorite languages. In particular this guide will focus on the
1012 Apex aspects of the scripts. However, this guide does not attempt to teach you about the scripting languages
1013 themselves …​ that is up to you!
ramverma3b71c972019-07-10 11:25:37 +00001014
liamfallonf3981072020-07-10 15:59:25 +01001015.. tip::
1016 JVM-based scripting languages
1017 For more more information on Scripting for the Java platform see:
1018 https://docs.oracle.com/javase/8/docs/technotes/guides/scripting/prog_guide/index.html
ramverma3b71c972019-07-10 11:25:37 +00001019
liamfallonf3981072020-07-10 15:59:25 +01001020.. note::
1021 What does Task Selection Logic do?
1022 When an Apex state references multiple tasks, there must be a way to dynamically decide
1023 which task should be chosen and executed. This can depend on the many factors, e.g. the
1024 *incoming event for the state*, *shared state* or *context*, *external context*,
1025 etc.. This is the function of a state’s Task Selection Logic. Obviously, if there is
1026 only one task then Task only one task then Task Selection Logic is not needed.
1027 Each state must also select one of the tasks a the *default state*. If the Task
1028 Selection Logic is unable to select an appropriate task, then it should select the
1029 *default task*. Once the task has been selected the Apex Engine will then execute that task.
ramverma3b71c972019-07-10 11:25:37 +00001030
liamfallonf3981072020-07-10 15:59:25 +01001031.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001032
liamfallonf3981072020-07-10 15:59:25 +01001033 First lets start with some simple Task Selection Logic, drawn from the "My First Apex Policy" example: The Task
1034 Selection Logic from the "My First Apex Policy" example is specified in JavaScript here:
ramverma3b71c972019-07-10 11:25:37 +00001035
liamfallonf3981072020-07-10 15:59:25 +01001036.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001037
liamfallonf3981072020-07-10 15:59:25 +01001038 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001039
liamfallonf3981072020-07-10 15:59:25 +01001040 Javascript code for the "My First Policy" Task Selection Logic
ramverma3b71c972019-07-10 11:25:37 +00001041
liamfallonf3981072020-07-10 15:59:25 +01001042 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001043
liamfallonf3981072020-07-10 15:59:25 +01001044 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001045
liamfallonf3981072020-07-10 15:59:25 +01001046 /*
1047 * ============LICENSE_START=======================================================
1048 * Copyright (C) 2016-2018 Ericsson. All rights reserved.
1049 * Modifications Copyright (C) 2020 Nordix Foundation.
1050 * ================================================================================
1051 * Licensed under the Apache License, Version 2.0 (the "License");
1052 * you may not use this file except in compliance with the License.
1053 * You may obtain a copy of the License at
1054 *
1055 * http://www.apache.org/licenses/LICENSE-2.0
1056 *
1057 * Unless required by applicable law or agreed to in writing, software
1058 * distributed under the License is distributed on an "AS IS" BASIS,
1059 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
1060 * See the License for the specific language governing permissions and
1061 * limitations under the License.
1062 *
1063 * SPDX-License-Identifier: Apache-2.0
1064 * ============LICENSE_END=========================================================
1065 */
ramverma3b71c972019-07-10 11:25:37 +00001066
liamfallonf3981072020-07-10 15:59:25 +01001067 executor.logger.info("Task Selection Execution: '"+executor.subject.id+
1068 "'. Input Event: '"+executor.inFields+"'");
ramverma3b71c972019-07-10 11:25:37 +00001069
liamfallonf3981072020-07-10 15:59:25 +01001070 branchid = executor.inFields.get("branch_ID");
1071 taskorig = executor.subject.getTaskKey("MorningBoozeCheck");
1072 taskalt = executor.subject.getTaskKey("MorningBoozeCheckAlt1");
1073 taskdef = executor.subject.getDefaultTaskKey();
ramverma3b71c972019-07-10 11:25:37 +00001074
liamfallonf3981072020-07-10 15:59:25 +01001075 if(branchid >=0 && branchid <1000){
1076 taskorig.copyTo(executor.selectedTask);
1077 }
1078 else if (branchid >=1000 && branchid <2000){
1079 taskalt.copyTo(executor.selectedTask);
1080 }
1081 else{
1082 taskdef.copyTo(executor.selectedTask);
1083 }
ramverma3b71c972019-07-10 11:25:37 +00001084
liamfallonf3981072020-07-10 15:59:25 +01001085 /*
1086 This task selection logic selects task "MorningBoozeCheck" for branches with
1087 0<=branch_ID<1000 and selects task "MorningBoozeCheckAlt1" for branches with
1088 1000<=branch_ID<2000. Otherwise the default task is selected.
1089 In this case the default task is also "MorningBoozeCheck"
1090 */
ramverma3b71c972019-07-10 11:25:37 +00001091
liamfallonf3981072020-07-10 15:59:25 +01001092 true;
ramverma3b71c972019-07-10 11:25:37 +00001093
liamfallonf3981072020-07-10 15:59:25 +01001094.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001095
liamfallonf3981072020-07-10 15:59:25 +01001096 The role of the Task Selection Logic in this simple example is to examine the value in one incoming field
1097 (``branchid``), then depending on that field’s value set the value for the selected task to the appropriate task
1098 (``MorningBoozeCheck``, ``MorningBoozeCheckAlt1``, or the default task).
ramverma3b71c972019-07-10 11:25:37 +00001099
liamfallonf3981072020-07-10 15:59:25 +01001100.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001101
liamfallonf3981072020-07-10 15:59:25 +01001102 Another thing to notice is that Task Selection Logic should return a ``java.lang.Boolean`` value ``true`` if
1103 the logic executed correctly. If the logic fails for some reason then ``false`` can be returned, but this will
1104 cause the policy invoking this task will fail and exit.
ramverma3b71c972019-07-10 11:25:37 +00001105
liamfallonf3981072020-07-10 15:59:25 +01001106.. note::
1107 How to return a value from Task Selection Logic
1108 Some languages explicitly support returning values from the script (e.g. MVEL and
1109 JRuby) using an explicit return statement (e.g. ``return true``), other languages do not (e.g.
1110 JavaScript and Jython). For languages that do not support the ``return`` statement, a special field called
1111 ``returnValue`` must be created to hold the result of the task logic operation (i.e. assign a ``java.lang.Boolean``
1112 value to the ``returnValue`` field before completing the task).
1113 Also, in MVEL if there is not explicit return statement then the return value of the last executed statement will
1114 return (e.g. the statement a=(1+2) will return the value 3).
ramverma3b71c972019-07-10 11:25:37 +00001115
liamfallonf3981072020-07-10 15:59:25 +01001116.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001117
liamfallonf3981072020-07-10 15:59:25 +01001118 Each of the scripting languages used in Apex can import and use standard Java libraries to perform complex tasks.
1119 Besides imported classes and normal language features Apex provides some natively available parameters and functions
1120 that can be used directly. At run-time these parameters are populated by the Apex execution environment and made
1121 natively available to logic scripts each time the logic script is invoked. (These can be accessed using the
1122 ``executor`` keyword for most languages, or can be accessed directly without the ``executor`` keyword in MVEL):
ramverma3b71c972019-07-10 11:25:37 +00001123
liamfallonf3981072020-07-10 15:59:25 +01001124Table 2. The ``executor`` Fields / Methods
1125 +-----------------------------------+------------------------------------+
1126 | Unix, Cygwin | Windows |
1127 +===================================+====================================+
1128 |.. container:: content |.. container:: content |
1129 | | |
1130 | .. code:: bash | .. code:: bash |
1131 | :number-lines: | :number-lines: |
1132 | | |
1133 | >c: | # cd /usr/local/src/apex-pdp |
1134 | >cd \dev\apex | # mvn clean install -DskipTests |
1135 | >mvn clean install -DskipTests | |
1136 +-----------------------------------+------------------------------------+
ramverma3b71c972019-07-10 11:25:37 +00001137
liamfallonf3981072020-07-10 15:59:25 +01001138 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
1139 | Name | Type | Java type | Description |
1140 +=====================================================+==========================================================================+===============================+==================================================================================+
1141 | inFields | Fields | java.util.Map <String,Object> | All fields in the state’s incoming event. This is implemented as a standard Java |
1142 | | | | Java (unmodifiable) Map |
1143 | | | | |
1144 | | | | **Example:** |
1145 | | | | |
1146 | | | | .. code:: javascript |
1147 | | | | |
1148 | | | | executor.logger.debug("Incoming fields: " + executor.inFields.entrySet()); |
1149 | | | | var item_id = executor.incomingFields["item_ID"]; |
1150 | | | | if (item_id >=1000) { ... } |
1151 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
1152 | outFields | Fields | java.util.Map <String,Object> | The outgoing task fields. This is implemented as a standard initially empty Java |
1153 | | | | (modifiable) Map. To create a new schema-compliant instance of a field object |
1154 | | | | see the utility method subject.getOutFieldSchemaHelper() below |
1155 | | | | |
1156 | | | | **Example:** |
1157 | | | | |
1158 | | | | .. code:: javascript |
1159 | | | | |
1160 | | | | executor.outFields["authorised"] = false; |
1161 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
1162 | logger | Logger | org.slf4j.ext.XLogger | A helpful logger |
1163 | | | | |
1164 | | | | **Example:** |
1165 | | | | |
1166 | | | | .. code:: javascript |
1167 | | | | |
1168 | | | | executor.logger.info("Executing task: " |
1169 | | | | +executor.subject.id); |
1170 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
1171 | TRUE/FALSE | boolean | java.lang.Boolean | 2 helpful constants. These are useful to retrieve correct return values for the |
1172 | | | | task logic |
1173 | | | | |
1174 | | | | **Example:** |
1175 | | | | |
1176 | | | | .. code:: javascript |
1177 | | | | |
1178 | | | | var returnValue = executor.isTrue; |
1179 | | | | var returnValueType = Java.type("java.lang.Boolean"); |
1180 | | | | var returnValue = new returnValueType(true); |
1181 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
1182 | subject | Task | TaskFacade | This provides some useful information about the task that contains this task |
1183 | | | | logic. This object has some useful fields and methods : |
1184 | | | | |
1185 | | | | .. container:: ulist |
1186 | | | | |
1187 | | | | - **AxTask task** to get access to the full task definition of the host task |
1188 | | | | |
1189 | | | | - **String getTaskName()** to get the name of the host task |
1190 | | | | |
1191 | | | | - **String getId()** to get the ID of the host task |
1192 | | | | |
1193 | | | | - **SchemaHelper getInFieldSchemaHelper( String fieldName )** to |
1194 | | | | get a ``SchemaHelper`` helper object to manipulate incoming |
1195 | | | | task fields in a schema-aware manner |
1196 | | | | |
1197 | | | | - **SchemaHelper getOutFieldSchemaHelper( String fieldName )** to |
1198 | | | | get a ``SchemaHelper`` helper object to manipulate outgoing |
1199 | | | | task fields in a schema-aware manner, e.g. to instantiate new |
1200 | | | | schema-compliant field objects to populate the |
1201 | | | | ``executor.outFields`` outgoing fields map |
1202 | | | | |
1203 | | | | **Example:** |
1204 | | | | |
1205 | | | | .. code:: javascript |
1206 | | | | |
1207 | | | | executor.logger.info("Task name: " + executor.subject.getTaskName()); |
1208 | | | | executor.logger.info("Task id: " + executor.subject.getId()); |
liamfallonf3981072020-07-10 15:59:25 +01001209 | | | | executor.outFields["authorised"] = executor.subject |
1210 | | | | .getOutFieldSchemaHelper("authorised") |
1211 | | | | .createNewInstance("false"); |
1212 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
1213 | parameters | Fields | java.util.Map <String,String> | All parameters in the current task. This is implemented as a standard Java Map. |
1214 | | | | |
1215 | | | | **Example:** |
1216 | | | | |
1217 | | | | .. code:: javascript |
1218 | | | | |
1219 | | | | executor.parameters.get("ParameterKey1")) |
1220 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
1221 | ContextAlbum getContextAlbum(String ctxtAlbumName ) | A utility method to retrieve a ``ContextAlbum`` for use in the task. | | |
1222 | | This is how you access the context used by the task. The returned | | |
1223 | | ``ContextAlbum`` implements the ``java.util.Map <String,Object>`` | | |
1224 | | interface to get and set context as appropriate. The returned | | |
1225 | | ``ContextAlbum`` also has methods to lock context albums, get | | |
1226 | | information about the schema of the items to be stored in a context | | |
1227 | | album, and get a ``SchemaHelper`` to manipulate context album items. How | | |
1228 | | to define and use context in a task is described in the Apex | | |
1229 | | Programmer’s Guide and in the My First Apex Policy guide. | | |
1230 | | | | |
1231 | | **Example:** | | |
1232 | | | | |
1233 | | .. code:: javascript | | |
1234 | | | | |
1235 | | var bkey = executor.inFields.get("branch_ID"); | | |
1236 | | var cnts = executor.getContextMap("BranchCounts"); | | |
1237 | | cnts.lockForWriting(bkey); | | |
1238 | | cnts.put(bkey, cnts.get(bkey) + 1); | | |
1239 | | cnts.unlockForWriting(bkey); | | |
1240 +-----------------------------------------------------+--------------------------------------------------------------------------+-------------------------------+----------------------------------------------------------------------------------+
ramverma3b71c972019-07-10 11:25:37 +00001241
liamfallonf3981072020-07-10 15:59:25 +01001242Logic Cheat Sheet
1243=================
ramverma3b71c972019-07-10 11:25:37 +00001244
liamfallonf3981072020-07-10 15:59:25 +01001245.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001246
liamfallonf3981072020-07-10 15:59:25 +01001247 Examples given here use Javascript (if not stated otherwise), other execution environments will be similar.
ramverma3b71c972019-07-10 11:25:37 +00001248
ramverma3b71c972019-07-10 11:25:37 +00001249Finish Logic with Success or Error
liamfallonf3981072020-07-10 15:59:25 +01001250----------------------------------
ramverma3b71c972019-07-10 11:25:37 +00001251
liamfallonf3981072020-07-10 15:59:25 +01001252.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001253
liamfallonf3981072020-07-10 15:59:25 +01001254 To finish logic, i.e. return to APEX, with success use the following line close to the end of the logic.
ramverma3b71c972019-07-10 11:25:37 +00001255
liamfallonf3981072020-07-10 15:59:25 +01001256.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001257
liamfallonf3981072020-07-10 15:59:25 +01001258 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001259
liamfallonf3981072020-07-10 15:59:25 +01001260 JS Success
ramverma3b71c972019-07-10 11:25:37 +00001261
liamfallonf3981072020-07-10 15:59:25 +01001262 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001263
liamfallonf3981072020-07-10 15:59:25 +01001264 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001265
liamfallonf3981072020-07-10 15:59:25 +01001266 true;
ramverma3b71c972019-07-10 11:25:37 +00001267
liamfallonf3981072020-07-10 15:59:25 +01001268.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001269
liamfallonf3981072020-07-10 15:59:25 +01001270 To notify a problem, finish with an error.
ramverma3b71c972019-07-10 11:25:37 +00001271
liamfallonf3981072020-07-10 15:59:25 +01001272 .. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001273
liamfallonf3981072020-07-10 15:59:25 +01001274 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001275
liamfallonf3981072020-07-10 15:59:25 +01001276 JS Fail
ramverma3b71c972019-07-10 11:25:37 +00001277
liamfallonf3981072020-07-10 15:59:25 +01001278 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001279
liamfallonf3981072020-07-10 15:59:25 +01001280 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001281
liamfallonf3981072020-07-10 15:59:25 +01001282 false;
ramverma3b71c972019-07-10 11:25:37 +00001283
1284Logic Logging
liamfallonf3981072020-07-10 15:59:25 +01001285-------------
ramverma3b71c972019-07-10 11:25:37 +00001286
liamfallonf3981072020-07-10 15:59:25 +01001287.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001288
liamfallonf3981072020-07-10 15:59:25 +01001289 Logging can be made easy using a local variable for the logger. Line 1 below does that. Then we start
1290 with a trace log with the task (or task logic) identifier followed by the infields.
ramverma3b71c972019-07-10 11:25:37 +00001291
liamfallonf3981072020-07-10 15:59:25 +01001292.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001293
liamfallonf3981072020-07-10 15:59:25 +01001294 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001295
liamfallonf3981072020-07-10 15:59:25 +01001296 JS Logging
ramverma3b71c972019-07-10 11:25:37 +00001297
liamfallonf3981072020-07-10 15:59:25 +01001298 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001299
liamfallonf3981072020-07-10 15:59:25 +01001300 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001301
liamfallonf3981072020-07-10 15:59:25 +01001302 var logger = executor.logger;
1303 logger.trace("start: " + executor.subject.id);
1304 logger.trace("-- infields: " + executor.inFields);
ramverma3b71c972019-07-10 11:25:37 +00001305
liamfallonf3981072020-07-10 15:59:25 +01001306.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001307
liamfallonf3981072020-07-10 15:59:25 +01001308 For larger logging blocks you can use the standard logging API to detect log levels, for instance:
ramverma3b71c972019-07-10 11:25:37 +00001309
liamfallonf3981072020-07-10 15:59:25 +01001310 .. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001311
liamfallonf3981072020-07-10 15:59:25 +01001312 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001313
liamfallonf3981072020-07-10 15:59:25 +01001314 JS Logging Blocks
ramverma3b71c972019-07-10 11:25:37 +00001315
liamfallonf3981072020-07-10 15:59:25 +01001316 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001317
liamfallonf3981072020-07-10 15:59:25 +01001318 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001319
liamfallonf3981072020-07-10 15:59:25 +01001320 if(logger.isTraceEnabled()){
1321 // trace logging block here
1322 }
ramverma3b71c972019-07-10 11:25:37 +00001323
liamfallonf3981072020-07-10 15:59:25 +01001324.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001325
liamfallonf3981072020-07-10 15:59:25 +01001326 Note: the shown logger here logs to ``org.onap.policy.apex.executionlogging``. The behavior of the actual logging can
1327 be specified in the ``$APEX_HOME/etc/logback.xml``.
ramverma3b71c972019-07-10 11:25:37 +00001328
liamfallonf3981072020-07-10 15:59:25 +01001329.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001330
liamfallonf3981072020-07-10 15:59:25 +01001331 If you want to log into the APEX root logger (which is sometimes necessary to report serious logic errors to the top),
1332 then import the required class and use this logger.
ramverma3b71c972019-07-10 11:25:37 +00001333
liamfallonf3981072020-07-10 15:59:25 +01001334.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001335
liamfallonf3981072020-07-10 15:59:25 +01001336 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001337
liamfallonf3981072020-07-10 15:59:25 +01001338 JS Root Logger
ramverma3b71c972019-07-10 11:25:37 +00001339
liamfallonf3981072020-07-10 15:59:25 +01001340 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001341
liamfallonf3981072020-07-10 15:59:25 +01001342 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001343
liamfallonf3981072020-07-10 15:59:25 +01001344 var rootLogger = LoggerFactory.getLogger(logger.ROOT_LOGGER_NAME);
1345 rootLogger.error("Serious error in logic detected: " + executor.subject.id);
ramverma3b71c972019-07-10 11:25:37 +00001346
a.sreekumarcc0e9172020-03-16 13:36:45 +00001347Accessing TaskParameters
liamfallonf3981072020-07-10 15:59:25 +01001348------------------------
a.sreekumarcc0e9172020-03-16 13:36:45 +00001349
liamfallonf3981072020-07-10 15:59:25 +01001350.. container:: paragraph
a.sreekumarcc0e9172020-03-16 13:36:45 +00001351
liamfallonf3981072020-07-10 15:59:25 +01001352 TaskParameters available in a Task can be accessed in the logic. The parameters in each task are made
1353 available at the executor level. This example assumes a parameter with key ``ParameterKey1``.
a.sreekumarcc0e9172020-03-16 13:36:45 +00001354
liamfallonf3981072020-07-10 15:59:25 +01001355 .. container:: listingblock
a.sreekumarcc0e9172020-03-16 13:36:45 +00001356
liamfallonf3981072020-07-10 15:59:25 +01001357 .. container:: title
a.sreekumarcc0e9172020-03-16 13:36:45 +00001358
liamfallonf3981072020-07-10 15:59:25 +01001359 JS TaskParameter value
a.sreekumarcc0e9172020-03-16 13:36:45 +00001360
liamfallonf3981072020-07-10 15:59:25 +01001361 .. container:: content
a.sreekumarcc0e9172020-03-16 13:36:45 +00001362
liamfallonf3981072020-07-10 15:59:25 +01001363 .. code:: javascript
a.sreekumarcc0e9172020-03-16 13:36:45 +00001364
liamfallonf3981072020-07-10 15:59:25 +01001365 executor.parameters.get("ParameterKey1"))
a.sreekumarcc0e9172020-03-16 13:36:45 +00001366
liamfallonf3981072020-07-10 15:59:25 +01001367.. container:: paragraph
a.sreekumarcc0e9172020-03-16 13:36:45 +00001368
liamfallonf3981072020-07-10 15:59:25 +01001369 Alternatively, the task parameters can also be accessed from the task object.
a.sreekumarcc0e9172020-03-16 13:36:45 +00001370
liamfallonf3981072020-07-10 15:59:25 +01001371 .. container:: listingblock
a.sreekumarcc0e9172020-03-16 13:36:45 +00001372
liamfallonf3981072020-07-10 15:59:25 +01001373 .. container:: title
a.sreekumarcc0e9172020-03-16 13:36:45 +00001374
liamfallonf3981072020-07-10 15:59:25 +01001375 JS TaskParameter value using task object
a.sreekumarcc0e9172020-03-16 13:36:45 +00001376
liamfallonf3981072020-07-10 15:59:25 +01001377 .. container:: content
a.sreekumarcc0e9172020-03-16 13:36:45 +00001378
liamfallonf3981072020-07-10 15:59:25 +01001379 .. code:: javascript
a.sreekumarcc0e9172020-03-16 13:36:45 +00001380
liamfallonf3981072020-07-10 15:59:25 +01001381 executor.subject.task.getTaskParameters.get("ParameterKey1").getTaskParameterValue()
a.sreekumarcc0e9172020-03-16 13:36:45 +00001382
ramverma3b71c972019-07-10 11:25:37 +00001383Local Variable for Infields
liamfallonf3981072020-07-10 15:59:25 +01001384---------------------------
ramverma3b71c972019-07-10 11:25:37 +00001385
liamfallonf3981072020-07-10 15:59:25 +01001386.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001387
liamfallonf3981072020-07-10 15:59:25 +01001388 It is a good idea to use local variables for ``infields``. This avoids long code lines and policy
1389 evolution. The following example assumes infields named ``nodeName`` and ``nodeAlias``.
ramverma3b71c972019-07-10 11:25:37 +00001390
liamfallonf3981072020-07-10 15:59:25 +01001391 .. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001392
liamfallonf3981072020-07-10 15:59:25 +01001393 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001394
liamfallonf3981072020-07-10 15:59:25 +01001395 JS Infields Local Var
ramverma3b71c972019-07-10 11:25:37 +00001396
liamfallonf3981072020-07-10 15:59:25 +01001397 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001398
liamfallonf3981072020-07-10 15:59:25 +01001399 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001400
liamfallonf3981072020-07-10 15:59:25 +01001401 var ifNodeName = executor.inFields["nodeName"];
1402 var ifNodeAlias = executor.inFields["nodeAlias"];
ramverma3b71c972019-07-10 11:25:37 +00001403
1404Local Variable for Context Albums
liamfallonf3981072020-07-10 15:59:25 +01001405---------------------------------
ramverma3b71c972019-07-10 11:25:37 +00001406
liamfallonf3981072020-07-10 15:59:25 +01001407.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001408
liamfallonf3981072020-07-10 15:59:25 +01001409 Similar to the ``infields`` it is good practice to use local variables for context albums as well. The
1410 following example assumes that a task can access a context album ``albumTopoNodes``. The second line gets a
1411 particular node from this context album.
ramverma3b71c972019-07-10 11:25:37 +00001412
liamfallonf3981072020-07-10 15:59:25 +01001413.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001414
liamfallonf3981072020-07-10 15:59:25 +01001415 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001416
liamfallonf3981072020-07-10 15:59:25 +01001417 JS Infields Local Var
ramverma3b71c972019-07-10 11:25:37 +00001418
liamfallonf3981072020-07-10 15:59:25 +01001419 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001420
liamfallonf3981072020-07-10 15:59:25 +01001421 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001422
liamfallonf3981072020-07-10 15:59:25 +01001423 var albumTopoNodes = executor.getContextAlbum("albumTopoNodes");
1424 var ctxtNode = albumTopoNodes.get(ifNodeName);
ramverma3b71c972019-07-10 11:25:37 +00001425
1426Set Outfields in Logic
liamfallonf3981072020-07-10 15:59:25 +01001427----------------------
ramverma3b71c972019-07-10 11:25:37 +00001428
liamfallonf3981072020-07-10 15:59:25 +01001429.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001430
liamfallonf3981072020-07-10 15:59:25 +01001431 The task logic needs to set outfields with content generated. The exception are outfields that are a
1432 direct copy from an infield of the same name, APEX does that autmatically.
ramverma3b71c972019-07-10 11:25:37 +00001433
liamfallonf3981072020-07-10 15:59:25 +01001434.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001435
liamfallonf3981072020-07-10 15:59:25 +01001436 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001437
liamfallonf3981072020-07-10 15:59:25 +01001438 JS Set Outfields
ramverma3b71c972019-07-10 11:25:37 +00001439
liamfallonf3981072020-07-10 15:59:25 +01001440 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001441
liamfallonf3981072020-07-10 15:59:25 +01001442 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001443
liamfallonf3981072020-07-10 15:59:25 +01001444 executor.outFields["report"] = "node ctxt :: added node " + ifNodeName;
ramverma3b71c972019-07-10 11:25:37 +00001445
1446Create a instance of an Outfield using Schemas
liamfallonf3981072020-07-10 15:59:25 +01001447----------------------------------------------
ramverma3b71c972019-07-10 11:25:37 +00001448
liamfallonf3981072020-07-10 15:59:25 +01001449.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001450
liamfallonf3981072020-07-10 15:59:25 +01001451 If an outfield is not an atomic type (string, integer, etc.) but uses a complex schema (with a Java or
1452 Avro backend), APEX can help to create new instances. The ``executor`` provides a field called ``subject``,
liamfallon688a2dd2022-11-01 13:16:21 +00001453 which provides a schem helper with an API for this.
ramverma3b71c972019-07-10 11:25:37 +00001454
liamfallonf3981072020-07-10 15:59:25 +01001455.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001456
liamfallonf3981072020-07-10 15:59:25 +01001457 If the backend is Java, then the Java class implementing the schema needs to be imported.
ramverma3b71c972019-07-10 11:25:37 +00001458
a.sreekumar98b59692021-07-23 13:38:06 +01001459*Single outgoing event*
1460
liamfallonf3981072020-07-10 15:59:25 +01001461.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001462
a.sreekumar98b59692021-07-23 13:38:06 +01001463 When there is a single outgoing event associated with a task, the fieldName alone is enough to fetch its schema.
liamfallonf3981072020-07-10 15:59:25 +01001464 The following example assumes an outfield ``situation``. The ``subject`` method ``getOutFieldSchemaHelper()`` is used
1465 to create a new instance.
ramverma3b71c972019-07-10 11:25:37 +00001466
liamfallonf3981072020-07-10 15:59:25 +01001467.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001468
liamfallonf3981072020-07-10 15:59:25 +01001469 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001470
liamfallonf3981072020-07-10 15:59:25 +01001471 JS Outfield Instance with Schema
ramverma3b71c972019-07-10 11:25:37 +00001472
liamfallonf3981072020-07-10 15:59:25 +01001473 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001474
liamfallonf3981072020-07-10 15:59:25 +01001475 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001476
liamfallonf3981072020-07-10 15:59:25 +01001477 var situation = executor.subject.getOutFieldSchemaHelper("situation").createNewInstance();
ramverma3b71c972019-07-10 11:25:37 +00001478
liamfallonf3981072020-07-10 15:59:25 +01001479.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001480
liamfallonf3981072020-07-10 15:59:25 +01001481 If the schema backend is Java, the new instance will be as implemented in the Java class. If the schema backend is
1482 Avro, the new instance will have all fields from the Avro schema specification, but set to ``null``. So any entry here
1483 needs to be done separately. For instance, the ``situation`` schema has a field ``problemID`` which we set.
ramverma3b71c972019-07-10 11:25:37 +00001484
liamfallonf3981072020-07-10 15:59:25 +01001485.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001486
liamfallonf3981072020-07-10 15:59:25 +01001487 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001488
liamfallonf3981072020-07-10 15:59:25 +01001489 JS Outfield Instance with Schema, set
ramverma3b71c972019-07-10 11:25:37 +00001490
liamfallonf3981072020-07-10 15:59:25 +01001491 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001492
liamfallonf3981072020-07-10 15:59:25 +01001493 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001494
liamfallonf3981072020-07-10 15:59:25 +01001495 situation.put("problemID", "my-problem");
ramverma3b71c972019-07-10 11:25:37 +00001496
a.sreekumar98b59692021-07-23 13:38:06 +01001497*Multiple outgoing events*
1498
1499.. container:: paragraph
1500
1501 When there are multiple outgoing events associated with a task, the fieldName along with the eventName it belongs to
1502 are needed to fetch its schema.
1503 The following example assumes an outfield ``actionIdentifiers`` which belongs to ``CDSRequestEvent``.
1504 The ``subject`` method ``getOutFieldSchemaHelper()`` is used to create a new instance.
1505
1506.. container:: listingblock
1507
1508 .. container:: content
1509
1510 .. code:: javascript
1511
1512 var actionIdentifiers = executor.subject.getOutFieldSchemaHelper("CDSRequestEvent", "actionIdentifiers").createNewInstance();
1513
ramverma3b71c972019-07-10 11:25:37 +00001514Create a instance of an Context Album entry using Schemas
liamfallonf3981072020-07-10 15:59:25 +01001515---------------------------------------------------------
ramverma3b71c972019-07-10 11:25:37 +00001516
liamfallonf3981072020-07-10 15:59:25 +01001517.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001518
liamfallonf3981072020-07-10 15:59:25 +01001519 Context album instances can be created using very similar to the outfields. Here, the schema helper
liamfallon688a2dd2022-11-01 13:16:21 +00001520 comes from the context album directly. The API of the schema helper is the same as for outfields.
ramverma3b71c972019-07-10 11:25:37 +00001521
liamfallonf3981072020-07-10 15:59:25 +01001522.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001523
liamfallonf3981072020-07-10 15:59:25 +01001524 If the backend is Java, then the Java class implementing the schema needs to be imported.
ramverma3b71c972019-07-10 11:25:37 +00001525
liamfallonf3981072020-07-10 15:59:25 +01001526.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001527
liamfallonf3981072020-07-10 15:59:25 +01001528 The following example creates a new instance of a context album instance named ``albumProblemMap``.
ramverma3b71c972019-07-10 11:25:37 +00001529
liamfallonf3981072020-07-10 15:59:25 +01001530.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001531
liamfallonf3981072020-07-10 15:59:25 +01001532 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001533
liamfallonf3981072020-07-10 15:59:25 +01001534 JS Outfield Instance with Schema
ramverma3b71c972019-07-10 11:25:37 +00001535
liamfallonf3981072020-07-10 15:59:25 +01001536 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001537
liamfallonf3981072020-07-10 15:59:25 +01001538 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001539
liamfallonf3981072020-07-10 15:59:25 +01001540 var albumProblemMap = executor.getContextAlbum("albumProblemMap");
1541 var linkProblem = albumProblemMap.getSchemaHelper().createNewInstance();
ramverma3b71c972019-07-10 11:25:37 +00001542
liamfallonf3981072020-07-10 15:59:25 +01001543.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001544
liamfallonf3981072020-07-10 15:59:25 +01001545 This can of course be also done in a single call without the local variable for the context album.
ramverma3b71c972019-07-10 11:25:37 +00001546
liamfallonf3981072020-07-10 15:59:25 +01001547.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001548
liamfallonf3981072020-07-10 15:59:25 +01001549 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001550
liamfallonf3981072020-07-10 15:59:25 +01001551 JS Outfield Instance with Schema, one line
ramverma3b71c972019-07-10 11:25:37 +00001552
liamfallonf3981072020-07-10 15:59:25 +01001553 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001554
liamfallonf3981072020-07-10 15:59:25 +01001555 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001556
liamfallonf3981072020-07-10 15:59:25 +01001557 var linkProblem = executor.getContextAlbum("albumProblemMap").getSchemaHelper().createNewInstance();
ramverma3b71c972019-07-10 11:25:37 +00001558
liamfallonf3981072020-07-10 15:59:25 +01001559.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001560
liamfallonf3981072020-07-10 15:59:25 +01001561 If the schema backend is Java, the new instance will be as implemented in the Java class. If the schema backend is
1562 Avro, the new instance will have all fields from the Avro schema specification, but set to ``null``. So any entry here
1563 needs to be done separately (see above in outfields for an example).
ramverma3b71c972019-07-10 11:25:37 +00001564
1565Enumerates
liamfallonf3981072020-07-10 15:59:25 +01001566----------
ramverma3b71c972019-07-10 11:25:37 +00001567
liamfallonf3981072020-07-10 15:59:25 +01001568.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001569
liamfallonf3981072020-07-10 15:59:25 +01001570 When dealing with enumerates (Avro or Java defined), it is sometimes and in some execution
1571 environments necessary to convert them to a string. For example, assume an Avro enumerate schema as:
ramverma3b71c972019-07-10 11:25:37 +00001572
liamfallonf3981072020-07-10 15:59:25 +01001573.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001574
liamfallonf3981072020-07-10 15:59:25 +01001575 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001576
liamfallonf3981072020-07-10 15:59:25 +01001577 Avro Enumerate Schema
ramverma3b71c972019-07-10 11:25:37 +00001578
liamfallonf3981072020-07-10 15:59:25 +01001579 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001580
liamfallonf3981072020-07-10 15:59:25 +01001581 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001582
liamfallonf3981072020-07-10 15:59:25 +01001583 {
1584 "type": "enum", "name": "Status", "symbols" : [
1585 "UP", "DOWN"
1586 ]
1587 }
ramverma3b71c972019-07-10 11:25:37 +00001588
liamfallonf3981072020-07-10 15:59:25 +01001589.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001590
liamfallonf3981072020-07-10 15:59:25 +01001591 Using a switch over a field initialized with this enumerate in Javascript will fail. Instead, use the ``toString`` method, for example:
ramverma3b71c972019-07-10 11:25:37 +00001592
liamfallonf3981072020-07-10 15:59:25 +01001593.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001594
liamfallonf3981072020-07-10 15:59:25 +01001595 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001596
liamfallonf3981072020-07-10 15:59:25 +01001597 JS Outfield Instance with Schema, one line
ramverma3b71c972019-07-10 11:25:37 +00001598
liamfallonf3981072020-07-10 15:59:25 +01001599 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001600
liamfallonf3981072020-07-10 15:59:25 +01001601 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001602
liamfallonf3981072020-07-10 15:59:25 +01001603 var switchTest = executor.inFields["status"]; switch(switchTest.toString()){
1604 case "UP": ...; break; case "DOWN": ...; break; default: ...;
1605 }
ramverma3b71c972019-07-10 11:25:37 +00001606
1607MVEL Initialize Outfields First!
liamfallonf3981072020-07-10 15:59:25 +01001608--------------------------------
ramverma3b71c972019-07-10 11:25:37 +00001609
liamfallonf3981072020-07-10 15:59:25 +01001610.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001611
liamfallonf3981072020-07-10 15:59:25 +01001612 In MVEL, we observed a problem when accessing (setting) outfields without a prior access to them. So
1613 in any MVEL task logic, before setting any outfield, simply do a get (with any string), to load the outfields
1614 into the MVEL cache.
ramverma3b71c972019-07-10 11:25:37 +00001615
liamfallonf3981072020-07-10 15:59:25 +01001616.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001617
liamfallonf3981072020-07-10 15:59:25 +01001618 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001619
liamfallonf3981072020-07-10 15:59:25 +01001620 MVEL Outfield Initialization
ramverma3b71c972019-07-10 11:25:37 +00001621
liamfallonf3981072020-07-10 15:59:25 +01001622 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001623
liamfallonf3981072020-07-10 15:59:25 +01001624 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001625
liamfallonf3981072020-07-10 15:59:25 +01001626 outFields.get("initialize outfields");
ramverma3b71c972019-07-10 11:25:37 +00001627
1628Using Java in Scripting Logic
liamfallonf3981072020-07-10 15:59:25 +01001629-----------------------------
ramverma3b71c972019-07-10 11:25:37 +00001630
liamfallonf3981072020-07-10 15:59:25 +01001631.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001632
liamfallonf3981072020-07-10 15:59:25 +01001633 Since APEX executes the logic inside a JVM, most scripting languages provide access to all standard
1634 Java classes. Simply add an import for the required class and then use it as in actual Java.
ramverma3b71c972019-07-10 11:25:37 +00001635
liamfallonf3981072020-07-10 15:59:25 +01001636.. container:: paragraph
ramverma3b71c972019-07-10 11:25:37 +00001637
liamfallonf3981072020-07-10 15:59:25 +01001638 The following example imports ``java.util.arraylist`` into a Javascript logic, and then creates a new
1639 list.
ramverma3b71c972019-07-10 11:25:37 +00001640
liamfallonf3981072020-07-10 15:59:25 +01001641.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001642
liamfallonf3981072020-07-10 15:59:25 +01001643 .. container:: title
ramverma3b71c972019-07-10 11:25:37 +00001644
liamfallonf3981072020-07-10 15:59:25 +01001645 JS Import ArrayList
ramverma3b71c972019-07-10 11:25:37 +00001646
liamfallonf3981072020-07-10 15:59:25 +01001647 .. container:: content
ramverma3b71c972019-07-10 11:25:37 +00001648
liamfallonf3981072020-07-10 15:59:25 +01001649 .. code:: javascript
ramverma3b71c972019-07-10 11:25:37 +00001650
liamfallonf3981072020-07-10 15:59:25 +01001651 var myList = new ArrayList();
1652
1653Converting Javascript scripts from Nashorn to Rhino dialects
1654------------------------------------------------------------
1655
1656The Nashorn Javascript engine was removed from Java in the Java 11 release. Java 11 was introduced into
1657the Policy Framework in the Frankfurt release, so from Frankfurt on, APEX Javascript scripts use the Rhino
1658Javascript engine and scripts must be in the Rhino dialect.
1659
1660There are some minor but important differences between the dialects that users should be aware of so
1661that they can convert their scripts into the Rhino dialect.
1662
1663Return Values
1664^^^^^^^^^^^^^
1665
1666APEX scripts must always return a value of ``true`` indicating that the script executed correctly or ``false``
1667indicating that there was an error in script execution.
1668
1669*Pre Frankfurt*
1670
1671In Nashorn dialect scripts, the user had to create a special variable called ``returnValue`` and set the value of
1672that variable to be the return value for the script.
1673
1674*Frankfurt and Later*
1675
1676In Rhino dialect scripts, the return value of the script is the logical result of the last statement. Therefore the
1677last line of the script must evaluate to either ``true`` or ``false``.
1678
1679.. container:: listingblock
1680
1681 .. container:: title
1682
1683 JS Rhino script last executed line examples
1684
1685 .. container:: content
1686
1687 .. code:: javascript
1688
1689 true;
1690
1691 returnValue; // Where returnValue is assigned earlier in the script
1692
1693 someValue == 1; // Where the value of someValue is assigned earlier in the script
1694
1695return statement
1696^^^^^^^^^^^^^^^^
1697
1698The ``return`` statement is not supported from the main script called in the Rhino interpreter.
1699
1700*Pre Frankfurt*
1701
1702In Nashorn dialect scripts, the user could return a value of ``true`` or ``false`` at any point in their script.
1703
1704.. container:: listingblock
1705
1706 .. container:: title
1707
1708 JS Nashorn main script returning ``true`` and ``false``
1709
1710 .. container:: content
1711
1712 .. code:: javascript
1713
1714 var n;
1715
1716 // some code assigns n a value
1717
1718 if (n < 2) {
1719 return false;
1720 } else {
1721 return true;
1722 }
1723
1724*Frankfurt and Later*
1725
1726In Rhino dialect scripts, the ``return`` statement cannot be used in the main method, but it can still be used in
1727functions. If you want to have a ``return`` statement in your code prior to the last statement, encapsulate your code
1728in a function.
1729
1730.. container:: listingblock
1731
1732 .. container:: title
1733
1734 JS Rhino script with ``return`` statements in a function
1735
1736 .. container:: content
1737
1738 .. code:: javascript
1739
1740 someFunction();
1741
1742 function someFunction() {
1743 var n;
1744
1745 // some code assigns n a value
1746
1747 if (n < 2) {
1748 return false;
1749 } else {
1750 return true;
1751 }
1752 }
1753
1754Compatibility Script
1755^^^^^^^^^^^^^^^^^^^^
1756
1757For Nashorn, the user had to call a compatibility script at the beginning of their Javascript script. This is not
1758required in Rhino.
1759
1760*Pre Frankfurt*
1761
1762In Nashorn dialect scripts, the compatibility script must be loaded.
1763
1764.. container:: listingblock
1765
1766 .. container:: title
1767
1768 Nashorn compatability script loading
1769
1770 .. container:: content
1771
1772 .. code:: javascript
1773
1774 load("nashorn:mozilla_compat.js");
1775
1776*Frankfurt and Later*
1777
1778Not required.
1779
1780Import of Java classes
1781^^^^^^^^^^^^^^^^^^^^^^
1782
1783For Nashorn, the user had explicitly import all the Java packages and classes they wished to use in their Javascript
1784script. In Rhino, all Java classes on the classpath are available for use.
1785
1786*Pre Frankfurt*
1787
1788In Nashorn dialect scripts, Java classes must be imported.
1789
1790.. container:: listingblock
1791
1792 .. container:: title
1793
1794 Importation of Java packages and classes
1795
1796 .. container:: content
1797
1798 .. code:: javascript
1799
1800 importPackage(java.text);
1801 importClass(java.text.SimpleDateFormat);
1802
1803*Frankfurt and Later*
1804
1805Not required.
1806
1807Using Java Classes and Objects as Variables
1808^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1809
1810Setting a Javascript variable to hold a Java class or a Java object is more straightforward in Rhino than it is in
1811Nashorn. The examples below show how to instantiate a Javascript variable as a Java class and how to use that variable
1812to create an instance of the Java class in another Javascript variable in both dialects.
ramverma3b71c972019-07-10 11:25:37 +00001813
ramverma3b71c972019-07-10 11:25:37 +00001814
liamfallonf3981072020-07-10 15:59:25 +01001815*Pre Frankfurt*
ramverma3b71c972019-07-10 11:25:37 +00001816
liamfallonf3981072020-07-10 15:59:25 +01001817.. container:: listingblock
ramverma3b71c972019-07-10 11:25:37 +00001818
liamfallonf3981072020-07-10 15:59:25 +01001819 .. container:: title
1820
1821 Create Javascript variables to hold a Java class and instance
1822
1823 .. container:: content
1824
1825 .. code:: javascript
1826
1827 var webClientClass = Java.type("org.onap.policy.apex.examples.bbs.WebClient");
1828 var webClientObject = new webClientClass();
1829
1830*Frankfurt and Later*
1831
1832.. container:: listingblock
1833
1834 .. container:: title
1835
1836 Create Javascript variables to hold a Java class and instance
1837
1838 .. container:: content
1839
1840 .. code:: javascript
1841
1842 var webClientClass = org.onap.policy.apex.examples.bbs.WebClient;
1843 var webClientObject = new webClientClass();
1844
1845Equal Value and Equal Type operator ``===``
1846^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1847
1848The *Equal Value and Equal Type* operator ``===`` is not supported in Rhino. Developers must use the Equal To
1849operator ``==`` instead. To check types, they may need to explicitly find and check the type of the variables
1850they are using.
ramverma3b71c972019-07-10 11:25:37 +00001851
a.sreekumar98b59692021-07-23 13:38:06 +01001852*************************************************
1853Writing Multiple Output Events from a Final State
1854*************************************************
1855
1856.. container:: paragraph
1857
1858 APEX-PDP now supports sending multiple events from a final state in a Policy. The task assocaiated with the final
1859 state can populate the fields of multiple events, and then they can be passed over as the output events from the final
1860 state of a policy.
1861
1862.. note::
1863 inputfields and outputfields are not needed as part of the task definition anymore. Fields of an event are already
1864 defined as part of the event definition. Input event (single trigger event) and output event/events can be populated
1865 to a task as part of the policy/state definition because the event tagging is done there anyway.
1866
1867.. container:: paragraph
1868
1869 Consider a simple example where a policy *CDSActionPolicy* has a state *MakeCDSRequestState* which is also a final
1870 state. The state is triggered by an event *AAIEvent*. A task called *HandleCDSActionTask* is associated with
1871 *MakeCDSRequestState*.There are two output events expected from *MakeCDSRequestState* which are *CDSRequestEvent*
1872 (request event sent to CDS) and *LogEvent* (log event sent to DMaaP).
1873 Writing an APEX policy with this example will involve the below changes.
1874
1875*Command File:*
1876
1877.. container:: listingblock
1878
1879 .. container:: title
1880
1881 Define all the concepts in the Policy. Only relevant parts for the multiple output support are shown.
1882
1883 .. container:: content
1884
1885 .. code::
1886
1887 ## Define Events
1888 event create name=AAIEvent version=0.0.1 nameSpace=org.onap.policy.apex.test source=AAI target=APEX
1889 ..
1890 event create name=CDSRequestEvent version=0.0.1 nameSpace=org.onap.policy.apex.test source=APEX target=CDS
1891 event parameter create name=CDSRequestEvent parName=actionIdentifiers schemaName=CDSActionIdentifiersType
1892 ..
1893 event create name=LogEvent version=0.0.1 nameSpace=org.onap.policy.apex.test source=APEX target=DMaaP
1894 event parameter create name=LogEvent parName=status schemaName=SimpleStringType
1895 ..
1896
1897 ## Define Tasks
1898 task create name=HandleCDSActionTask
1899 task contextref create name=HandleCDSActionTask albumName=EventDetailsAlbum
1900 task logic create name=HandleCDSActionTask logicFlavour=JAVASCRIPT logic=LS
1901 #MACROFILE:"src/main/resources/logic/HandleCDSActionTask.js"
1902 LE
1903 ..
1904
1905 ## Define Policies and States
1906 policy create name=CDSActionPolicy template=Freestyle firstState=MakeCDSRequestState
1907 policy state create name=CDSActionPolicy stateName=MakeCDSRequestState triggerName=AAIEvent defaultTaskName=HandleCDSActionTask
1908 # Specify CDSRequestEvent as output
1909 policy state output create name=CDSActionPolicy stateName=MakeCDSRequestState outputName=CDSActionStateOutput eventName=CDSRequestEvent
1910 # Specify LogEvent as output
1911 policy state output create name=CDSActionPolicy stateName=MakeCDSRequestState outputName=CDSActionStateOutput eventName=LogEvent
1912 policy state taskref create name=CDSActionPolicy stateName=MakeCDSRequestState taskName=HandleCDSActionTask outputType=DIRECT outputName=CDSActionStateOutput
1913
1914*Task Logic File:*
1915
1916.. container:: listingblock
1917
1918 .. container:: title
1919
1920 Create outfields' instance if required, populate and add them the output events
1921
1922 .. container:: content
1923
1924 .. code:: javascript
1925
1926 ..
1927 var cdsRequestEventFields = java.util.HashMap();
1928 var actionIdentifiers = executor.subject.getOutFieldSchemaHelper("CDSRequestEvent","actionIdentifiers").createNewInstance();
1929 actionIdentifiers.put("blueprintName", "sample-bp");
1930 cdsRequestEventFields.put("actionIdentifiers", actionIdentifiers);
1931 executor.addFieldsToOutput(cdsRequestEventFields);
1932
1933 var logEventFields = java.util.HashMap();
1934 logEventFields.put("status", "FINAL_SUCCESS");
1935 executor.addFieldsToOutput(logEventFields);
1936
1937.. container:: paragraph
1938
1939 With the above changes, the task populates the fields for both the expected events, and the corresponding state which
1940 is *MakeCDSRequestState* outputs both *CDSRequestEvent* and *LogEvent*
1941
ramverma3b71c972019-07-10 11:25:37 +00001942.. |APEX Policy Matrix| image:: images/apex-intro/ApexPolicyMatrix.png
1943.. |APEX Policy Model for Execution| image:: images/apex-policy-model/UmlPolicyModels.png
1944.. |Concepts and Keys| image:: images/apex-policy-model/ConceptsKeys.png
1945