daily-notification-plugin/test-apps/android-test-app/docs/CONSOLE-REMAINING-WORK.md

Console Implementation - Remaining Work

✅ Completed

1. Test Plan JSON (plan.json) - All phases/tests/steps defined
2. Console UI (console/index.html, console.js, console.css) - Full Textual-style interface
3. TestEventsPlugin - Android plugin to receive ADB broadcasts
4. Event System - Library functions emit events (ui_prompt, capture_*, verdict_*)
5. Helper Functions - set_test_context(), step_start(), step_pass(), step_warn(), step_fail()
6. Example Wiring - Test 0 and partial Test 1 in test-phase1.sh demonstrate pattern
7. Plugin Registration - TestEventsPlugin added to capacitor.plugins.json
8. Documentation - Usage guide created

🔧 Critical Fix Applied

✅ TestEventsPlugin Registration - Added to capacitor.plugins.json (required for plugin to work)

📋 Remaining Work

1. Complete Wiring of test-phase1.sh

Status: Partially complete (Test 0 done, Test 1 started, Tests 2-4 not wired)

Pattern to apply:

# At test start
set_test_context "phase1" "phase1_testX" ""

# For each step
set_test_context "phase1" "phase1_testX" "p1_tX_sY"
step_start "p1_tX_sY" "Step description"
# ... do work ...
step_pass "p1_tX_sY" "Step completed"

Tests to wire:

• ✅ Test 0: Daily Rollover Verification (complete)
• ⚠️ Test 1: Force-Stop Recovery (partially done, needs completion)
• ❌ Test 2: Schedule Update Verification (not wired)
• ❌ Test 3: Recovery Timeout (not wired)
• ❌ Test 4: Invalid Data Handling (not wired)

Step IDs from plan.json:

• Test 1: p1_t1_s1 through p1_t1_s5
• Test 2: p1_t2_s1 through p1_t2_s4
• Test 3: p1_t3_s1 through p1_t3_s2
• Test 4: p1_t4_s1 through p1_t4_s4

2. Wire test-phase2.sh

Status: Not started

Tests to wire:

• Test 1: Force Stop – Alarms Cleared (phase2_test1)
• Test 2: Force Stop – Alarms Intact (phase2_test2)
• Test 3: First Launch / No Schedules (phase2_test3)

Step IDs from plan.json:

• Test 1: p2_t1_s1 through p2_t1_s5
• Test 2: p2_t2_s1 through p2_t2_s5
• Test 3: p2_t3_s1 through p2_t3_s4

3. Wire test-phase3.sh

Status: Not started

Tests to wire:

• Test 1: Boot with Future Alarms (phase3_test1)
• Test 2: Boot with Past Alarms (phase3_test2)
• Test 3: Boot with No Schedules (phase3_test3)
• Test 4: Silent Boot Recovery (phase3_test4)

Step IDs from plan.json:

• Test 1: p3_t1_s1 through p3_t1_s5
• Test 2: p3_t2_s1 through p3_t2_s5
• Test 3: p3_t3_s1 through p3_t3_s4
• Test 4: p3_t4_s1 through p3_t4_s4

4. Testing & Validation

Steps:

1. Build Android app: cd test-apps/android-test-app && ./gradlew assembleDebug
2. Install on emulator/device
3. Navigate to console: Open app → should redirect to /console/
4. Verify console loads: Check browser console for errors
5. Test Phase A (UI-only):
   • Select a test
   • Manually mark steps as done/fail
   • Verify state persists
6. Test Phase B (Live events):
   • Set export DNP_UI_EVENTS=1
   • Run ./test-phase1.sh 0 (just test 0)
   • Verify events appear in console
   • Verify step status updates automatically

🎯 Quick Start: Wiring a Test Function

Here's a complete example for wiring a test:

test_example() {
  section "TEST: Example Test"
  
  # Set test context (no step yet)
  set_test_context "phase1" "phase1_example" ""
  
  # Step 1: Setup
  set_test_context "phase1" "phase1_example" "p1_ex_s1"
  step_start "p1_ex_s1" "Setting up test"
  capture_alarms "example_initial"
  step_pass "p1_ex_s1" "Setup complete"
  
  # Step 2: Execute
  set_test_context "phase1" "phase1_example" "p1_ex_s2"
  step_start "p1_ex_s2" "Executing test"
  # ... do work ...
  if [ success ]; then
    step_pass "p1_ex_s2" "Execution complete"
  else
    step_fail "p1_ex_s2" "Execution failed"
  fi
  
  # Step 3: Verify
  set_test_context "phase1" "phase1_example" "p1_ex_s3"
  step_start "p1_ex_s3" "Verifying results"
  # ... verify ...
  step_pass "p1_ex_s3" "Verification passed"
  
  # Verdict (automatically emits event)
  set_test_context "phase1" "phase1_example" "p1_ex_s4"
  verdict_pass "example_test" "Test passed"
}

📝 Notes

• Step IDs must match plan.json - Check console/plan.json for exact step IDs
• Context must be set before step events - Call set_test_context() before step_start()
• Verdict functions auto-emit events - No need to manually emit verdict events
• Evidence capture auto-emits events - capture_alarms(), capture_logcat(), capture_screenshot() emit events automatically
• Operator prompts auto-emit events - ui_prompt() emits operator_required events automatically

🔍 Verification Checklist

After wiring tests, verify:

• All test functions have set_test_context() at start
• All major steps have step_start() and step_pass/fail/warn()
• Step IDs match plan.json exactly
• Verdict functions are called (they emit events automatically)
• Evidence capture functions are called (they emit events automatically)
• Scripts run without errors
• Console receives events when DNP_UI_EVENTS=1 is set

🚀 Priority Order

1. Complete test-phase1.sh (highest priority - most used)
2. Wire test-phase2.sh (medium priority)
3. Wire test-phase3.sh (medium priority)
4. Test integration (validate everything works)

📚 Reference

• Usage Guide: docs/CONSOLE-USAGE.md
• Test Plan: app/src/main/assets/public/console/plan.json
• Event Functions: alarm-test-lib.sh (functions: emit_event, set_test_context, step_*)