[P3] E2E: Seed data creates projects in wrong phase (active vs planning)

[frankbria]

Problem

E2E test seed data creates projects in active phase when tests expect planning phase. This causes 10+ test failures across Firefox, WebKit, Mobile Chrome, and Mobile Safari.

Failing Tests

Late-Joining User Tests:

• test_late_joining_user.spec.ts:129 - should show "Tasks Ready" section @smoke
• test_late_joining_user.spec.ts:201 - should not show "Generate Tasks" button

State Reconciliation Tests:

• test_state_reconciliation.spec.ts:132 - should show "Review Tasks" @smoke
• test_state_reconciliation.spec.ts:657 - should maintain correct state after page refresh @smoke

Error

Error: expect(received).toBe(expected) // Object.is equality
Expected: "planning"
Received: "active"

At line:

// Project must be in planning phase with tasks to test late-joining scenario
expect(progressData.phase).toBe('planning');

Root Cause

The seed script (scripts/seed-test-data.py) creates projects with tasks, but the phase transitions to active either:

1. During seeding when tasks are added
2. Due to automatic phase progression logic
3. Because the seed explicitly sets active phase

Location

• scripts/seed-test-data.py - Seed data creation
• codeframe/persistence/repositories/ - Phase transition logic

Expected Behavior

For late-joining user tests, the seeded project should be:

1. In planning phase
2. Have tasks already generated
3. Be awaiting user review/approval (not auto-progressed)

Suggested Fix

Option 1: Update seed script to explicitly set phase after task creation:

# After creating tasks
db.update_project_phase(project_id, "planning")

Option 2: Add a flag to prevent auto-phase-progression during seeding:

# Seed tasks without triggering phase change
db.create_tasks(tasks, auto_progress=False)

Impact

• 10+ test failures across 4 browsers
• Blocks late-joining user and state reconciliation test suites

Related

• [P3] E2E Tests: Fix mobile browser and Firefox compatibility issues #230 - E2E Tests: Fix mobile browser and Firefox compatibility issues
• [P1] Discovery enters stuck state on new project creation - no current question #255 - Discovery enters stuck state (different issue but related to project state)

[traycerai]

Plan

Observations

The E2E seed script (file:tests/e2e/seed-test-data.py) creates 7 test projects with different phases. Project 2 (id=2) is explicitly created with phase='planning' at line 1317 and is intended for late-joining user tests. However, at line 1285, there's an UPDATE statement that sets the phase to 'discovery' for the project_id parameter passed to the seed function. This UPDATE is intended only for Project 1 but incorrectly affects whichever project ID is passed as a parameter. The tests expect Project 2 to remain in 'planning' phase but are receiving 'active' phase instead.

Approach

The root cause is that the seed script has a hardcoded UPDATE statement (line 1285) that changes the phase to 'discovery' for the project_id parameter, when it should only update Project 1. The fix is to change this UPDATE statement to explicitly target Project 1 (id=1) instead of using the parameter. This ensures that Project 2 and other projects maintain their intended phases. The approach aligns with the issue's "Option 1" suggestion to explicitly set the phase after task creation, but more specifically, to fix the incorrect UPDATE that's overwriting the correct phase.

Implementation Steps

1. Fix the Phase Update Logic in Seed Script

File: file:tests/e2e/seed-test-data.py

Location: Lines 1275-1287 (Section "8. Update Project Phase for Discovery Tests")

Change Required:

The UPDATE statement at line 1285 currently uses the project_id parameter:

cursor.execute(
    "UPDATE projects SET phase = 'discovery' WHERE id = ?",
    (project_id,),
)

This should be changed to explicitly target Project 1:

cursor.execute(
    "UPDATE projects SET phase = 'discovery' WHERE id = 1"
)

Rationale: This section is specifically for setting up Project 1 for discovery tests. It should not affect other projects. By hardcoding id = 1, we ensure that only Project 1's phase is updated to 'discovery', and Project 2 retains its 'planning' phase as intended.

2. Verify Project 2 Phase Remains 'planning'

File: file:tests/e2e/seed-test-data.py

Location: After line 1590 (after Project 2 setup completes)

Optional Enhancement:

Add a verification query to ensure Project 2's phase is still 'planning' after all seed operations:

cursor.execute("SELECT phase FROM projects WHERE id = 2")
phase = cursor.fetchone()[0]
if phase != 'planning':
    print(f"⚠️  WARNING: Project 2 phase is '{phase}', expected 'planning'")

This helps catch any future regressions where Project 2's phase might be inadvertently changed.

3. Add Comment to Prevent Future Bugs

File: file:tests/e2e/seed-test-data.py

Location: Line 1275 (before the UPDATE statement)

Add Documentation:

Add a clear comment explaining why this UPDATE only targets Project 1:

# ========================================
# 8. Update Project Phase for Discovery Tests (Project 1 ONLY)
# ========================================
# IMPORTANT: This UPDATE must ONLY affect Project 1 (id=1).
# Other projects (2-7) have their phases set during creation and must not be modified here.
# Project 2 must remain in 'planning' phase for late-joining user tests.

4. Verify Test Expectations

Files to Review:

• file:tests/e2e/test_late_joining_user.spec.ts (line 169)
• file:tests/e2e/test_state_reconciliation.spec.ts (line 139)

Verification: After fixing the seed script, run these tests to confirm they now pass:

npx playwright test test_late_joining_user.spec.ts:129
npx playwright test test_late_joining_user.spec.ts:201
npx playwright test test_state_reconciliation.spec.ts:132
npx playwright test test_state_reconciliation.spec.ts:657

All tests should now see Project 2 in 'planning' phase as expected.

5. Consider Adding Phase Validation

File: file:tests/e2e/seed-test-data.py

Location: End of seed_test_data function (before line 2398)

Optional Enhancement:

Add validation to verify all projects have their expected phases:

# Validate project phases match expectations
expected_phases = {
    1: 'discovery',
    2: 'planning',
    3: 'active',
    4: 'review',
    5: 'complete',
    6: 'planning',
    7: 'active'
}
for proj_id, expected_phase in expected_phases.items():
    cursor.execute("SELECT phase FROM projects WHERE id = ?", (proj_id,))
    row = cursor.fetchone()
    if row:
        actual_phase = row[0]
        if actual_phase != expected_phase:
            print(f"⚠️  WARNING: Project {proj_id} phase is '{actual_phase}', expected '{expected_phase}'")

This provides early detection of phase-related seed data issues.

Testing

After implementing the fix:

1. Run the seed script directly:

   python tests/e2e/seed-test-data.py tests/e2e/.test-db.sqlite 1

2. Verify Project 2 phase in database:

   sqlite3 tests/e2e/.test-db.sqlite "SELECT id, name, phase FROM projects WHERE id = 2;"

   Expected output: 2|e2e-planning-project|planning

3. Run failing E2E tests:

   npx playwright test test_late_joining_user.spec.ts --grep "@smoke"
   npx playwright test test_state_reconciliation.spec.ts --grep "@smoke"

4. Run full E2E test suite:

   npx playwright test

All 10+ previously failing tests across Firefox, WebKit, Mobile Chrome, and Mobile Safari should now pass.

Impact

• Fixes: 10+ test failures across 4 browsers
• Affected Tests: Late-joining user tests and state reconciliation tests
• Root Cause: Incorrect use of parameterized project_id in UPDATE statement
• Solution: Hardcode Project 1's ID in the UPDATE statement to prevent affecting other projects

Import In IDE

🤖 Prompt for AI Agents

## Observations

The E2E seed script (`file:tests/e2e/seed-test-data.py`) creates 7 test projects with different phases. Project 2 (id=2) is explicitly created with `phase='planning'` at line 1317 and is intended for late-joining user tests. However, at line 1285, there's an UPDATE statement that sets the phase to 'discovery' for the `project_id` parameter passed to the seed function. This UPDATE is intended only for Project 1 but incorrectly affects whichever project ID is passed as a parameter. The tests expect Project 2 to remain in 'planning' phase but are receiving 'active' phase instead.

## Approach

The root cause is that the seed script has a hardcoded UPDATE statement (line 1285) that changes the phase to 'discovery' for the `project_id` parameter, when it should only update Project 1. The fix is to change this UPDATE statement to explicitly target Project 1 (id=1) instead of using the parameter. This ensures that Project 2 and other projects maintain their intended phases. The approach aligns with the issue's "Option 1" suggestion to explicitly set the phase after task creation, but more specifically, to fix the incorrect UPDATE that's overwriting the correct phase.

## Implementation Steps

### 1. Fix the Phase Update Logic in Seed Script

**File:** `file:tests/e2e/seed-test-data.py`

**Location:** Lines 1275-1287 (Section "8. Update Project Phase for Discovery Tests")

**Change Required:**

The UPDATE statement at line 1285 currently uses the `project_id` parameter:
```python
cursor.execute(
    "UPDATE projects SET phase = 'discovery' WHERE id = ?",
    (project_id,),
)
```

This should be changed to explicitly target Project 1:
```python
cursor.execute(
    "UPDATE projects SET phase = 'discovery' WHERE id = 1"
)
```

**Rationale:** This section is specifically for setting up Project 1 for discovery tests. It should not affect other projects. By hardcoding `id = 1`, we ensure that only Project 1's phase is updated to 'discovery', and Project 2 retains its 'planning' phase as intended.

### 2. Verify Project 2 Phase Remains 'planning'

**File:** `file:tests/e2e/seed-test-data.py`

**Location:** After line 1590 (after Project 2 setup completes)

**Optional Enhancement:**

Add a verification query to ensure Project 2's phase is still 'planning' after all seed operations:
```python
cursor.execute("SELECT phase FROM projects WHERE id = 2")
phase = cursor.fetchone()[0]
if phase != 'planning':
    print(f"⚠️  WARNING: Project 2 phase is '{phase}', expected 'planning'")
```

This helps catch any future regressions where Project 2's phase might be inadvertently changed.

### 3. Add Comment to Prevent Future Bugs

**File:** `file:tests/e2e/seed-test-data.py`

**Location:** Line 1275 (before the UPDATE statement)

**Add Documentation:**

Add a clear comment explaining why this UPDATE only targets Project 1:
```python
# ========================================
# 8. Update Project Phase for Discovery Tests (Project 1 ONLY)
# ========================================
# IMPORTANT: This UPDATE must ONLY affect Project 1 (id=1).
# Other projects (2-7) have their phases set during creation and must not be modified here.
# Project 2 must remain in 'planning' phase for late-joining user tests.
```

### 4. Verify Test Expectations

**Files to Review:**
- `file:tests/e2e/test_late_joining_user.spec.ts` (line 169)
- `file:tests/e2e/test_state_reconciliation.spec.ts` (line 139)

**Verification:** After fixing the seed script, run these tests to confirm they now pass:
```bash
npx playwright test test_late_joining_user.spec.ts:129
npx playwright test test_late_joining_user.spec.ts:201
npx playwright test test_state_reconciliation.spec.ts:132
npx playwright test test_state_reconciliation.spec.ts:657
```

All tests should now see Project 2 in 'planning' phase as expected.

### 5. Consider Adding Phase Validation

**File:** `file:tests/e2e/seed-test-data.py`

**Location:** End of seed_test_data function (before line 2398)

**Optional Enhancement:**

Add validation to verify all projects have their expected phases:
```python
# Validate project phases match expectations
expected_phases = {
    1: 'discovery',
    2: 'planning',
    3: 'active',
    4: 'review',
    5: 'complete',
    6: 'planning',
    7: 'active'
}
for proj_id, expected_phase in expected_phases.items():
    cursor.execute("SELECT phase FROM projects WHERE id = ?", (proj_id,))
    row = cursor.fetchone()
    if row:
        actual_phase = row[0]
        if actual_phase != expected_phase:
            print(f"⚠️  WARNING: Project {proj_id} phase is '{actual_phase}', expected '{expected_phase}'")
```

This provides early detection of phase-related seed data issues.

## Testing

After implementing the fix:

1. **Run the seed script directly:**
   ```bash
   python tests/e2e/seed-test-data.py tests/e2e/.test-db.sqlite 1
   ```

2. **Verify Project 2 phase in database:**
   ```bash
   sqlite3 tests/e2e/.test-db.sqlite "SELECT id, name, phase FROM projects WHERE id = 2;"
   ```
   Expected output: `2|e2e-planning-project|planning`

3. **Run failing E2E tests:**
   ```bash
   npx playwright test test_late_joining_user.spec.ts --grep "@smoke"
   npx playwright test test_state_reconciliation.spec.ts --grep "@smoke"
   ```

4. **Run full E2E test suite:**
   ```bash
   npx playwright test
   ```

All 10+ previously failing tests across Firefox, WebKit, Mobile Chrome, and Mobile Safari should now pass.

## Impact

- **Fixes:** 10+ test failures across 4 browsers
- **Affected Tests:** Late-joining user tests and state reconciliation tests
- **Root Cause:** Incorrect use of parameterized project_id in UPDATE statement
- **Solution:** Hardcode Project 1's ID in the UPDATE statement to prevent affecting other projects

---

Execution Information

Branch: main
Commit: 9b4790f

---

💡 Tips

Supported Commands (Inside Comments)

• Use @traycerai generate to iterate on the previous version of the implementation plan.

Supported Commands (Inside Description)

• Add @traycerai ignore anywhere in the ticket description to prevent this ticket from being processed.
• Add @traycerai branch:<branch-name> anywhere in the ticket description to specify the target branch for the implementation plan.

Community

• Join our Discord Community to get help, request features, and share feedback.
• Follow us on X/Twitter for updates and announcements.

[coderabbitai]

📝 CodeRabbit Plan Mode

Generate an implementation plan and prompts that you can use with your favorite coding agent.

• Create Plan

Examples

• Example 1
• Example 2

---

🔗 Related PRs

#90 - fix(e2e): E2E test database isolation - Fixes #89 [merged]
#191 - feat(e2e): Enable conditional tests and fix discovery startup [merged]
#211 - feat(dashboard): Add PhaseProgress component for phase visualization [merged]
#229 - feat(e2e): add state reconciliation tests for late-joining users [merged]
#234 - fix(ui): make TaskStats phase-aware to fix late-joining user bug [merged]

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!

[frankbria]

Fixed in commit 2da47ba.

Root cause: The UPDATE statement setting Project 1's phase to 'discovery' was using a parameterized WHERE id = ? with (project_id,), which incorrectly affected whichever project was passed. This caused Project 2's 'planning' phase to be overwritten.

Fix:

• Hardcoded WHERE id = 1 to ensure only Project 1 is updated
• Added phase validation at end of seed function to catch future regressions

Test results:

• Late-joining user tests: 15/15 passed ✅
• State reconciliation tests: 60/60 passed ✅