Skip to main content

Including Actor Tree in Report

How to include the actor hierarchy structure during workflow execution in reports.

Challenge​

The actor tree changes dynamically during execution. For example:

  • At workflow start: Only ROOT, loader, nodeGroup
  • After node connection: node-* actors are created under nodeGroup
  • After plugin load: Plugin actors are created under loader
  • At report generation: SectionBuilder actors are created under reportBuilder

Therefore, you need to explicitly specify "at which point you want to output the actor tree to the report."

Solution: Use % Notation​

Actor trees are collected by CheckResultsSection using % notation, not a dedicated SectionBuilder.

Flow​

  1. Call ROOT.printTree at the point you want to include in the report
  2. Output the result to logs using % notation
  3. CheckResultsSection collects % notation lines and includes them in the report

Implementation Method​

To output printTree results with % notation, use OutputMultiplexer:

- states: ["2", "3"]
  note: Output actor tree to report
  actions:
    # Get printTree result
    - actor: ROOT
      method: printTree
      store: actorTree

    # Output with % notation (add % to each line)
    - actor: outputMultiplexer
      method: writeLines
      arguments:
        prefix: "%[Actor Tree] "
        content: "${actorTree}"

Or, for simple single-line output:

- actor: outputMultiplexer
  method: writeLine
  arguments: ["%[Actor Tree]\n${actorTree}"]

Complete Workflow Example​

name: deployment-with-tree
description: Report with deployment results and actor tree

steps:
  - states: ["0", "1"]
    note: Create ReportBuilder with sections
    actions:
      - actor: loader
        method: createChild
        arguments: ["ROOT", "reportBuilder", "com.scivicslab.actoriac.report.ReportBuilderIIAR"]
      - actor: loader
        method: createChild
        arguments: ["reportBuilder", "wfName", "com.scivicslab.actoriac.report.sections.basic.WorkflowNameSectionIIAR"]
      - actor: loader
        method: createChild
        arguments: ["reportBuilder", "checkResults", "com.scivicslab.actoriac.report.sections.basic.CheckResultsSectionIIAR"]
      - actor: loader
        method: createChild
        arguments: ["reportBuilder", "trans:nodeGroup:children", "com.scivicslab.actoriac.report.sections.basic.TransitionHistorySectionIIAR"]

  - states: ["1", "2"]
    note: Deploy to all nodes
    actions:
      - actor: nodeGroup
        method: apply
        arguments:
          actor: "node-*"
          method: runWorkflow
          arguments: ["deploy.yaml"]

  - states: ["2", "3"]
    note: Capture actor tree after deployment
    actions:
      - actor: ROOT
        method: printTree
        store: actorTree
      - actor: outputMultiplexer
        method: writeLine
        arguments: ["%[Actor Tree]\n%${actorTree}"]

  - states: ["3", "end"]
    note: Generate report
    actions:
      - actor: reportBuilder
        method: report

Output Example​

================================================================================
                              WORKFLOW REPORT
================================================================================

[Workflow Name]
deployment-with-tree

[Check Results]
[Actor Tree]
ROOT
├── loader
├── nodeGroup
│   ├── node-192.168.5.13
│   ├── node-192.168.5.14
│   └── node-192.168.5.15
├── outputMultiplexer
└── reportBuilder
    ├── wfName
    ├── checkResults
    └── trans:nodeGroup:children

[Transition History: nodeGroup (with children)]
...

================================================================================

Actor Trees at Multiple Points​

You can also compare actor trees at different points.

- states: ["1", "2"]
  note: Capture initial tree
  actions:
    - actor: ROOT
      method: printTree
      store: initialTree
    - actor: outputMultiplexer
      method: writeLine
      arguments: ["%[Initial Actor Tree]\n%${initialTree}"]

- states: ["3", "4"]
  note: Capture tree after plugin load
  actions:
    - actor: ROOT
      method: printTree
      store: afterPluginTree
    - actor: outputMultiplexer
      method: writeLine
      arguments: ["%[After Plugin Load]\n%${afterPluginTree}"]

- states: ["5", "6"]
  note: Capture final tree
  actions:
    - actor: ROOT
      method: printTree
      store: finalTree
    - actor: outputMultiplexer
      method: writeLine
      arguments: ["%[Final Actor Tree]\n%${finalTree}"]

Output example:

[Check Results]
[Initial Actor Tree]
ROOT
├── loader
├── nodeGroup
└── outputMultiplexer

[After Plugin Load]
ROOT
├── loader
│   └── systemInfoAggregator
├── nodeGroup
│   ├── node-192.168.5.13
│   └── node-192.168.5.14
└── outputMultiplexer

[Final Actor Tree]
ROOT
├── loader
│   └── systemInfoAggregator
├── nodeGroup
│   ├── node-192.168.5.13
│   └── node-192.168.5.14
├── outputMultiplexer
└── reportBuilder
    ├── wfName
    └── checkResults

Why Not Make It a SectionBuilder?​

  1. Dynamic nature: The actor tree is constantly changing during workflow execution. The tree at the point when SectionBuilder's generate() is called (report generation time) is likely different from the point the user wants to see.

  2. Flexibility: Using % notation, you can output the tree at any point, any number of times. With a SectionBuilder, it would be fixed to "once at report generation time."

  3. Simplicity: It can be achieved by combining the existing printTree action with % notation, so there's no need to create a new SectionBuilder.

  • ActorTreeVisualization_260119_oo01 - Details on the printTree action
  • CheckResultsSection_260130_oo01 - % notation collection