Skip to main content

UI Log JSON (json/)

The json/ directory is the log viewer contract.
Bublik’s UI loads these files directly by URL to render the run log tree and per-node log pages.

Data Contract​

Node JSON files should conform to the schema maintained in the Test Environment (TE) repository

Files​

json/tree.json​

  • Purpose: navigation index for the UI.
  • Key fields:
    • main_package: filename of the root node JSON (commonly node_1_0.json).
    • tree: mapping of filename -> entry, where package entries may include children (filenames of child nodes).

json/node_*.json​

  • Purpose: per-node log payload (root package, packages, and tests).
  • File envelope (top-level shape):
{
  "version": "v1",
  "root": [
    {
      "type": "te-log",
      "content": [
        { "type": "te-log-meta", "entity_model": {}, "meta": {} },
        { "type": "te-log-entity-list", "items": [] },
        { "type": "te-log-table", "data": [] }
      ]
    }
  ]
}
  • Blocks:
    • te-log-meta: entity id/name/result/duration and optional params/verdicts/artifacts.
    • te-log-entity-list: present when the node has child iterations (nested entities for tree navigation). May be omitted when there are no children.
    • te-log-table: hierarchical log rows in data.

Table rows (minimum fields)​

Each row in te-log-table.data should include:

  • line_number
  • level
  • entity_name
  • user_name
  • timestamp
  • log_content

Rows may contain nested children (recursively).

Levels (common)​

Typical levels used by the UI:

  • ERROR
  • WARN
  • INFO
  • VERB
  • PACKET
  • RING

Custom string levels are allowed; MI is commonly used for measurement rows (parents with an MI descendant are expanded automatically).

Common user_name values used in the ecosystem:

  • TAPI Jumps (main test navigation marker)
  • Step (scenario action)
  • StepPush / StepPop (step nesting)
  • Artifact (artifact/measurement summary rows)
  • Verdict (verdict rows)
  • Self (ordinary lines for the current entity)
  • Run (often used together with entity_name: "Tester" for run-level summary rows)

#Scenario and #Test mapping​

The UI detects the main test entity from the first log row where:

user_name == "TAPI Jumps"

The entity_name of that row becomes the main test entity for the page.

How #Test works​

#Test shows rows where:

entity_name == <main test entity>

How #Scenario works​

#Scenario commonly includes rows such as:

<main test entity>:Step
<main test entity>:Artifact
<main test entity>:Self
<main test entity>:Verdict
Tester:Run
  • Emit at least one row with user_name: "TAPI Jumps" for each test page where #Scenario / #Test should work.
  • Set that row's entity_name to the logical main test entity for the page.
  • Keep ordinary test lines under that same entity with user_name: "Self".
  • Keep scenario actions under that same entity with user_name values such as Step, StepPush, and StepPop.
  • Keep verdict and artifact rows under that same entity with user_name values such as Verdict and Artifact.
  • Keep run-level summary rows under entity_name: "Tester" and user_name: "Run" so they are available as Tester:Run.
  • Keep helper or secondary logs under their real source entities instead of moving everything under the main test entity just to satisfy filters.

Common failure mode​

If no TAPI Jumps row is present, the UI may not detect the main test entity, and #Scenario / #Test can remain disabled or incomplete.