Spliit App Tests

QA_ARCHITECT_PROMPT.md

Lead QA Architect - Spliit Testing Implementation

Role

You are the Lead QA Architect implementing the Spliit test plan. Your task: Implement ONE test case per execution from file @TEST_PLAN_STRUCTURED.md.

Test Selection Strategy

1. Review: Analyze file @TEST_PLAN_STRUCTURED.md for incomplete tests (Priority, Complexity, Effort)
2. Prioritize: Select the highest-value test using YOUR judgment:
   • Prioritize 🍏 Low Hanging Fruit in 🔴 P0 categories
   • NOT necessarily the first in the list
   • Balance quick wins with critical coverage

Implementation Protocol

1. Define Task

State the specific test case clearly (e.g., "Implement: Expense creation validation")

2. Gather Context

Required reading:

• Existing tests in the same test file or related suite
• Source code for the feature being tested (use semantic_search or grep_search)
• Related component/page implementations
• Use playwright-mcp if page structure is unclear after code review

Pattern recognition:

• Identify existing test patterns (setup, assertions, cleanup)
• Note how similar features are tested
• Check for shared fixtures or helpers
  • (E2E Helpers: tests/helpers/index.ts)

3. Implement Test

• Follow existing patterns: Match style, structure, and naming conventions
• Precise assertions: Especially for money values - verify exact amounts, not approximations
• Edge cases: Consider boundary conditions explicitly mentioned in test requirements
• No magic numbers: Use descriptive constants or variables
• Mocking Prisma/DB: Not allowed, when there is a need to interact with the DB, use the existing real Prisma client instance. DO NOT TRY to mock or stub it, you will fail.
• Module mocking: Jest has bad support for ES Modules mocking. Individual class/function mocking is allowed, but module-level mocking is FORBIDDEN. If you need to mock a module, STOP and think for solution alternatives.

4. Handle Uncertainties

STOP and ask if you encounter:

• Ambiguous business logic requirements
• Unclear expected behavior for edge cases
• Missing information about data structures
• DO NOT guess or make assumptions about money calculations

5. Validate Test

Run: npm run test-e2e <file.spec.ts>

Validation checklist:

• Test passes
• Test runs successfully in isolation
• No console errors or warnings
• Assertions are specific and meaningful

If test fails:

• Analyze error output carefully
• Check for race conditions or timing issues
• Verify selectors match actual DOM structure
• Retry once after fixes
• If still failing after one retry, escalate with details

6. Update Documentation

Mark test as done in TEST_PLAN_STRUCTURED.md (and TEST_PLAN.md if it exists)

7. Commit Changes

Git commit your changes, use clear commit message: Implement: <Test Case Description>

Critical Constraints

🚨 TIMEOUT RULES (MOST IMPORTANT)

NEVER add timeouts exceeding 2 seconds!

❌ FORBIDDEN:

await page.waitForTimeout(3000)
await page.waitForTimeout(5000)
page.setDefaultTimeout(10000)

✅ CORRECT - Use Playwright's auto-waiting:

await page.waitForSelector('.element')
await page.getByRole('button', { name: 'Submit' }).click()
await expect(page.getByText('Success')).toBeVisible()

Other Constraints

• Efficiency: Rely on Playwright's built-in auto-waiting mechanisms
• Patterns: Match existing test architecture and coding style
• Selectors: Prefer role-based selectors over CSS/XPath when possible
• Isolation: Each test should be independent and not rely on execution order

Completion Report

Upon successful completion, output:

<promise>COMPLETE</promise>

DO NOT start another test - report completion and stop.

Environment

• Dev server: http://localhost:3000 (already running)
• Framework: Playwright
• Tools: playwright-mcp
• Test command: npm run test-e2e <file.spec.ts>

---

Begin now: Select your highest-priority incomplete test from TEST_PLAN_STRUCTURED.md and implement it following this protocol.

TEST_PLAN.md

Spliit Test Plan

This document provides a comprehensive testing strategy for the Spliit expense-splitting application. Tests are categorized by type (Jest unit/integration tests vs Playwright E2E tests), priority, complexity, and implementation effort.

Legend

Priority Levels:

• 🔴 P0 (Critical): Core functionality, must have
• 🟡 P1 (High): Important features, should have
• 🟢 P2 (Medium): Nice to have, can defer
• 🔵 P3 (Low): Edge cases, optional

Complexity:

• 🟢 Low: Simple, straightforward tests
• 🟡 Medium: Moderate setup or logic required
• 🔴 High: Complex scenarios, multiple dependencies

Effort:

• 🍏 Low Hanging Fruit: Easy wins, test immediately
• 🍊 Medium Effort: Reasonable investment
• 🍎 High Effort: Significant time investment

---

1. JEST UNIT TESTS (Business Logic)

These tests focus on pure business logic functions in src/lib/ that are critical to the application's correctness. No external dependencies, fast execution.

1.1 Balance Calculations (src/lib/balances.ts)

Test Case	Priority	Complexity	Effort	Notes
getBalances() - evenly split expenses	🔴 P0	🟢 Low	✅ Done	Critical - verify equal splits work correctly
getBalances() - BY_SHARES split mode	🔴 P0	🟡 Medium	✅ Done	Test weighted splits (1:2:3 ratios)
getBalances() - BY_PERCENTAGE split mode	🔴 P0	🟡 Medium	✅ Done	Test percentage splits summing to 100%
getBalances() - BY_AMOUNT split mode	🔴 P0	🟡 Medium	✅ Done	Test specific amounts per person
getBalances() - handles rounding correctly	🟡 P1	🟡 Medium	✅ Done	Verify no floating point errors, totals balance
getBalances() - avoids negative zeros	🟡 P1	🟢 Low	✅ Done	Check for -0 values being normalized to 0
getBalances() - multiple participants, mixed expenses	🔴 P0	🟡 Medium	✅ Done	Integration test with realistic scenario
getBalances() - last participant gets remaining amount	🔴 P0	🟡 Medium	✅ Done	Verify remainder distribution logic
getBalances() - handles empty expense list	🟢 P2	🟢 Low	✅ Done	Edge case - empty array
getBalances() - single expense, single participant	🟢 P2	🟢 Low	✅ Done	Edge case - simplest scenario
getSuggestedReimbursements() - minimizes transactions	🔴 P0	🟡 Medium	✅ Done	Core feature - verify greedy algorithm works
getSuggestedReimbursements() - stable sorting	🟡 P1	🟡 Medium	✅ Done	Verify same balances yield same suggestions
getSuggestedReimbursements() - filters zero amounts	🟡 P1	🟢 Low	✅ Done	Ensure zero-value reimbursements excluded
getSuggestedReimbursements() - handles balanced group	🟢 P2	🟢 Low	✅ Done	Edge case - all balances zero
getSuggestedReimbursements() - complex 5+ person scenario	🟡 P1	🔴 High	✅ Done	Realistic multi-person settlement
getPublicBalances() - converts reimbursements to balances	🟡 P1	🟢 Low	✅ Done	Test reimbursement → balance conversion
compareBalancesForReimbursements() - sorts correctly	🟡 P1	🟢 Low	✅ Done	Verify sorting logic (positive before negative)

Total: 17 tests | P0: 6, P1: 8, P2: 3 | Low effort: 11, Medium: 5, High: 1

---

1.2 Totals & Statistics (src/lib/totals.ts)

Test Case	Priority	Complexity	Effort	Notes
getTotalGroupSpending() - sums all expenses	🔴 P0	🟢 Low	✅ Done	Core metric calculation
getTotalGroupSpending() - excludes reimbursements	🔴 P0	🟢 Low	✅ Done	Critical - reimbursements shouldn't count
getTotalGroupSpending() - handles empty array	🟢 P2	🟢 Low	✅ Done	Edge case
getTotalActiveUserPaidFor() - sums user's payments	🔴 P0	🟢 Low	✅ Done	Personalization feature
getTotalActiveUserPaidFor() - excludes reimbursements	🔴 P0	🟢 Low	✅ Done	Consistency with group total
getTotalActiveUserPaidFor() - returns 0 for null user	🟡 P1	🟢 Low	✅ Done	Handle no active user case
calculateShare() - EVENLY mode correct calculation	🔴 P0	🟢 Low	✅ Done	Core splitting logic
calculateShare() - BY_AMOUNT mode uses exact shares	🔴 P0	🟢 Low	✅ Done	Direct amount passthrough
calculateShare() - BY_PERCENTAGE mode (shares/10000)	🔴 P0	🟡 Medium	✅ Done	Percentage calculation with basis points
calculateShare() - BY_SHARES weighted correctly	🔴 P0	🟡 Medium	✅ Done	Ratio-based splitting
calculateShare() - returns 0 for reimbursements	🔴 P0	🟢 Low	✅ Done	Reimbursement exclusion
calculateShare() - returns 0 if participant not in paidFor	🟡 P1	🟢 Low	✅ Done	Participant not involved in expense
getTotalActiveUserShare() - sums across expenses	🔴 P0	🟡 Medium	✅ Done	Total owed calculation
getTotalActiveUserShare() - rounds to 2 decimals	🟡 P1	🟢 Low	✅ Done	Currency precision handling

Total: 14 tests | P0: 10, P1: 3, P2: 1 | Low effort: 14

---

1.3 Currency Utilities (src/lib/currency.ts, src/lib/utils.ts)

Test Case	Priority	Complexity	Effort	Notes
formatCurrency() - USD formatting (existing)	🔴 P0	🟢 Low	✅ Done	Already tested in utils.test.ts
formatCurrency() - EUR formatting (existing)	🔴 P0	🟢 Low	✅ Done	Already tested
formatCurrency() - custom currency symbols	🟡 P1	🟢 Low	✅ Done	Test non-ISO currency
formatCurrency() - zero decimal currencies (JPY)	🟡 P1	🟢 Low	✅ Done	Test decimal_digits = 0
amountAsDecimal() - converts cents to decimal	🔴 P0	🟢 Low	✅ Done	Core conversion
amountAsDecimal() - handles rounding when requested	🟡 P1	🟢 Low	✅ Done	Optional rounding param
amountAsMinorUnits() - converts decimal to cents	🔴 P0	🟢 Low	✅ Done	Inverse of above
amountAsMinorUnits() - rounds correctly	🟡 P1	🟢 Low	✅ Done	No floating point issues
formatAmountAsDecimal() - formats with correct decimals	🟡 P1	🟢 Low	✅ Done	String formatting
getCurrency() - returns currency by code	🔴 P0	🟢 Low	✅ Done	Lookup function
getCurrency() - returns custom for empty code	🟡 P1	🟢 Low	✅ Done	Fallback behavior
getCurrency() - handles locale variations	🟢 P2	🟢 Low	✅ Done	i18n currency names
getCurrencyFromGroup() - extracts from group object	🟡 P1	🟢 Low	✅ Done	Helper function
defaultCurrencyList() - includes custom currency	🟢 P2	🟢 Low	✅ Done	Custom currency in list

Total: 14 tests | P0: 5, P1: 7, P2: 2 | Low effort: 12, Done: 2

---

1.4 Utility Functions (src/lib/utils.ts)

Test Case	Priority	Complexity	Effort	Notes
formatDate() - formats with locale	🟡 P1	🟢 Low	✅ Done	Date display
formatDateOnly() - avoids timezone shifts	🔴 P0	🟡 Medium	✅ Done	Critical for DATE fields, UTC handling
formatDateOnly() - handles month boundaries	🟡 P1	🟡 Medium	✅ Done	Edge case - dates near midnight
formatFileSize() - formats bytes correctly	🟢 P2	🟢 Low	✅ Done	Utility function
formatFileSize() - handles GB, MB, KB, B units	🟢 P2	🟢 Low	✅ Done	Unit conversion
normalizeString() - removes accents	🟡 P1	🟢 Low	✅ Done	Search functionality
normalizeString() - lowercases	🟡 P1	🟢 Low	✅ Done	Case-insensitive search
formatCategoryForAIPrompt() - formats correctly	🟢 P2	🟢 Low	✅ Done	AI feature helper
delay() - resolves after ms	🔵 P3	🟢 Low	✅ Done	Simple utility
cn() - merges class names	🔵 P3	🟢 Low	✅ Done	TailwindCSS helper

Total: 10 tests | P0: 1, P1: 4, P2: 3, P3: 2 | Low effort: 9, Medium: 1

---

1.5 Schemas & Validation (src/lib/schemas.ts)

Test Case	Priority	Complexity	Effort	Notes
expenseFormSchema - validates required fields	🔴 P0	🟢 Low	✅ Done	Form validation
expenseFormSchema - rejects invalid split mode	🔴 P0	🟢 Low	✅ Done	Enum validation
expenseFormSchema - validates percentage sums to 100%	🔴 P0	🟡 Medium	✅ Done	Business rule
expenseFormSchema - validates amount sum equals total	🔴 P0	🟡 Medium	✅ Done	BY_AMOUNT validation
expenseFormSchema - allows valid recurring rules	🟡 P1	🟢 Low	✅ Done	Enum validation
groupFormSchema - validates group creation	🔴 P0	🟢 Low	✅ Done	Group validation
groupFormSchema - requires at least 2 participants	🔴 P0	🟢 Low	✅ Done	Business rule
groupFormSchema - validates currency format	🟡 P1	🟢 Low	✅ Done	Currency validation

Total: 8 tests | P0: 6, P1: 2 | Low effort: 6, Medium: 2

---

2. JEST INTEGRATION TESTS (API & Database)

These tests require database setup but test important business logic flows. Use test database or mocks.

2.1 Recurring Expense Logic (src/lib/api.ts)

Test Case	Priority	Complexity	Effort	Notes
createRecurringExpenses() - creates daily recurring	🔴 P0	🔴 High	✅ Done	Complex, requires DB
createRecurringExpenses() - creates weekly recurring	🔴 P0	🔴 High	✅ Done	Date arithmetic
createRecurringExpenses() - creates monthly recurring	🔴 P0	🔴 High	✅ Done	Edge cases: 29-31st dates
createRecurringExpenses() - handles month boundaries	🟡 P1	🔴 High	✅ Done	Feb 29, 30, 31 edge cases
createRecurringExpenses() - stops at correct time	🔴 P0	🟡 Medium	✅ Done	Verify loop termination
createRecurringExpenses() - uses transactions	🟡 P1	🔴 High	✅ Done	Race condition prevention
createPayloadForNewRecurringExpenseLink() - daily	🟡 P1	🟡 Medium	✅ Done	Payload generation
createPayloadForNewRecurringExpenseLink() - weekly	🟡 P1	🟡 Medium	✅ Done	Payload generation
createPayloadForNewRecurringExpenseLink() - monthly	🟡 P1	🟡 Medium	✅ Done	Payload generation

Total: 9 tests | P0: 4, P1: 5 | Medium: 3, High: 6

⚠️ Note: These are complex and require careful DB setup. Consider mocking Prisma for faster tests.

---

2.2 Activity Logging (src/lib/api.ts)

Test Case	Priority	Complexity	Effort	Notes
logActivity() - logs CREATE_EXPENSE	🟡 P1	🟡 Medium	✅ Done	Audit trail
logActivity() - logs UPDATE_EXPENSE	🟡 P1	🟡 Medium	✅ Done	Audit trail
logActivity() - logs DELETE_EXPENSE	🟡 P1	🟡 Medium	✅ Done	Audit trail
logActivity() - logs UPDATE_GROUP	🟡 P1	🟡 Medium	✅ Done	Audit trail
logActivity() - stores participant ID	🟢 P2	🟡 Medium	✅ Done	User tracking
logActivity() - stores expense data	🟢 P2	🟡 Medium	✅ Done	Metadata storage

Total: 6 tests | P1: 4, P2: 2 | Medium: 6

---

3. PLAYWRIGHT E2E TESTS

These tests verify user-facing functionality through the browser. Focus on critical user journeys.

3.1 Group Management Flows

Test Case	Priority	Complexity	Effort	Notes
Create group - happy path	🔴 P0	🟢 Low	✅ Done	Core user flow
Create group - with custom currency	🟡 P1	🟢 Low	✅ Done	Custom currency selection
Create group - validation errors	🟡 P1	🟡 Medium	✅ Done	Form validation
Edit group - update name and info	🟡 P1	🟢 Low	✅ Done	Edit flow
Edit group - add participant	🔴 P0	🟡 Medium	✅ Done	Important feature
Edit group - remove participant	🟡 P1	🟡 Medium	✅ Done	Important feature
Edit group - rename participant	🟢 P2	🟢 Low	✅ Done	Edit participant
View group information page	🟢 P2	🟢 Low	✅ Done	Read-only view
Share group - copy URL	🟡 P1	🟡 Medium	✅ Done	Collaboration feature
Recent groups list - persists	🟢 P2	🟡 Medium	✅ Done	LocalStorage test
Navigate between groups	🟢 P2	🟢 Low	✅ Done	Navigation

Total: 11 tests | P0: 2, P1: 5, P2: 4 | Low effort: 6, Medium: 5

---

3.2 Expense Management Flows

Test Case	Priority	Complexity	Effort	Notes
Create expense - evenly split	🔴 P0	🟡 Medium	✅ Done	Most common flow
Create expense - by shares	🔴 P0	🟡 Medium	✅ Done	Weighted split
Create expense - by percentage	🔴 P0	🟡 Medium	✅ Done	Percentage split
Create expense - by amount	🔴 P0	🟡 Medium	✅ Done	Specific amounts
Create expense - with category	🟡 P1	🟢 Low	✅ Done	Category selection
Create expense - with notes	🟢 P2	🟢 Low	✅ Done	Optional field
Create expense - with custom date	🟡 P1	🟢 Low	✅ Done	Date picker
Create expense - with currency conversion	🟡 P1	🟡 Medium	✅ Done	Multi-currency
Create expense - as reimbursement	🟡 P1	🟢 Low	✅ Done	Reimbursement flag
Create expense - validation errors	🟡 P1	🟡 Medium	✅ Done	Form validation
Edit expense - update all fields	🟡 P1	🟡 Medium	✅ Done	Edit flow
Edit expense - change split mode	🟡 P1	🟡 Medium	✅ Done	Mode switching
Delete expense - confirmation flow	🔴 P0	🟢 Low	✅ Done	Deletion
List expenses - pagination	🟡 P1	🟡 Medium	✅ Done	Large lists
List expenses - filter by text	🟡 P1	🟡 Medium	✅ Done	Search functionality
Expense displays correct date	🟡 P1	🟢 Low	✅ Done	Date rendering
Expense displays correct amount	🔴 P0	🟢 Low	✅ Done	Amount display
Expense shows category	🟢 P2	🟢 Low	✅ Done	Category display

Total: 18 tests | P0: 4, P1: 11, P2: 3 | Low effort: 10, Medium: 8

---

3.3 Recurring Expenses Flow

Test Case	Priority	Complexity	Effort	Notes
Create daily recurring expense	🔴 P0	🟡 Medium	✅ Done	Fill form, select Daily, submit, verify recurring indicator UI
Create weekly recurring expense	🔴 P0	🟡 Medium	✅ Done	Fill form, select Weekly, submit, verify UI
Create monthly recurring expense	🔴 P0	🟡 Medium	✅ Done	Fill form, select Monthly, submit, verify UI
Verify recurring instances created	🔴 P0	🔴 High	✅ Done	Create recurring; trigger backend job/mock time; verify new expense in list
Edit recurring expense - stops future (missing feature #492)	🟡 P1	🟡 Medium	N/A	Missing feature spliit-app/spliit#492, do not test
Delete recurring expense - only current	🟡 P1	🟡 Medium	✅ Done	Delete a recurring instance; assert other instances in chain remain
Recurring expense shows indicator	🟢 P2	🟢 Low	✅ Done	UI element

Total: 7 tests | P0: 4, P1: 2, P2: 1 | Low effort: 1, Medium: 4, High: 2

---

3.4 Balance & Reimbursement Flows

Test Case	Priority	Complexity	Effort	Notes
View balances page - calculates correctly	🔴 P0	🟡 Medium	✅ Done	Core feature
Balances match expected from expenses	🔴 P0	🟡 Medium	✅ Done	Verification test
Suggested reimbursements displayed	🔴 P0	🟢 Low	✅ Done	Suggestions shown
Suggested reimbursements minimized	🟡 P1	🟡 Medium	✅ Done	Algorithm verification
Active user balance highlighted	🟡 P1	🟢 Low	✅ Done	Personalization
Create reimbursement expense	🟡 P1	🟡 Medium	✅ Done	Settle debt flow
Reimbursement excludes from totals	🔴 P0	🟡 Medium	✅ Done	Important logic
Zero balances display correctly	🟢 P2	🟢 Low	✅ Done	Edge case

Total: 8 tests | P0: 4, P1: 3, P2: 1 | Low effort: 4, Medium: 4

---

3.5 Statistics & Export Flows

Test Case	Priority	Complexity	Effort	Notes
View statistics page	🟡 P1	🟢 Low	✅ Done	Stats display
Total group spending correct	🔴 P0	🟡 Medium	✅ Done	Calculation verification
User total paid correct	🔴 P0	🟡 Medium	✅ Done	Personalization
User total share correct	🔴 P0	🟡 Medium	✅ Done	Personalization
Export to JSON - downloads file	🟡 P1	🟡 Medium	✅ Done	Export functionality
Export to JSON - correct data	🟡 P1	🟡 Medium	✅ Done	Data verification
Export to CSV - downloads file	🟡 P1	🟡 Medium	✅ Done	Export functionality
Export to CSV - correct format	🟡 P1	🟡 Medium	✅ Done	CSV structure

Total: 8 tests | P0: 3, P1: 5 | Low effort: 1, Medium: 7

---

3.6 Active User Selection

Test Case	Priority	Complexity	Effort	Notes
Select active user - persists	🟡 P1	🟡 Medium	✅ Done	LocalStorage test
Active user changes balance view	🔴 P0	🟡 Medium	✅ Done	Personalization
Active user changes stats	🔴 P0	🟡 Medium	✅ Done	Personalization
Clear active user - neutral view	🟢 P2	🟢 Low	✅ Done	Reset feature

Total: 4 tests | P0: 2, P1: 1, P2: 1 | Low effort: 1, Medium: 3

---

3.7 Activity Log

Test Case	Priority	Complexity	Effort	Notes
View activity page	🟢 P2	🟢 Low	✅ Done	Activity list
Activity shows expense creation	🟢 P2	🟡 Medium	✅ Done	Audit log
Activity shows expense update	🟢 P2	🟡 Medium	✅ Done	Audit log
Activity shows expense deletion	🟢 P2	🟡 Medium	✅ Done	Audit log
Activity pagination works	🟢 P2	🟡 Medium	✅ Done	Long lists

Total: 5 tests | P2: 5 | Low effort: 1, Medium: 4

---

3.8 Category Management

Test Case	Priority	Complexity	Effort	Notes
Select category when creating expense	🟡 P1	🟢 Low	✅ Done	Dropdown selection
Category displays on expense	🟡 P1	🟢 Low	✅ Done	Display test
Default category (General) selected	🟢 P2	🟢 Low	✅ Done	Default behavior

Total: 3 tests | P1: 2, P2: 1 | Low effort: 3

---

3.9 Internationalization (i18n)

Test Case	Priority	Complexity	Effort	Notes
Change language - UI updates	🟢 P2	🟡 Medium	🍊 Medium	i18n test
Currency formats per locale	🟡 P1	🟡 Medium	✅ Done	Locale formatting
Date formats per locale	🟡 P1	🟡 Medium	✅ Done	Locale formatting

Total: 3 tests | P1: 2, P2: 1 | Medium: 3

---

3.10 Theme & UI

Test Case	Priority	Complexity	Effort	Notes
Toggle dark mode - persists	🟢 P2	🟢 Low	✅ Done	Theme switching
Mobile responsive - drawer opens	🟡 P1	🟡 Medium	✅ Done	Mobile testing
Desktop responsive - dialog opens	🟡 P1	🟡 Medium	✅ Done	Desktop testing

Total: 3 tests | P1: 2, P2: 1 | Low effort: 1, Medium: 2

---

3.11 Document Upload (if S3 enabled)

Test Case	Priority	Complexity	Effort	Notes
Upload document to expense	🟡 P1	🔴 High	🍎 High	S3 integration
Multiple documents on expense	🟢 P2	🔴 High	🍎 High	S3 integration
Document displays in expense	🟡 P1	🔴 High	🍎 High	S3 integration

Total: 3 tests | P1: 2, P2: 1 | High: 3

⚠️ Note: Requires S3 setup or mocking. Skip if feature flag disabled.

---

3.12 Health Endpoints

Test Case	Priority	Complexity	Effort	Notes
/api/health/liveness returns 200	🟡 P1	🟢 Low	✅ Done	Health check
/api/health/readiness checks DB	🟡 P1	🟡 Medium	✅ Done	DB health

Total: 2 tests | P1: 2 | Low effort: 1, Medium: 1

---

4. TEST PRIORITY SUMMARY

Note: Status in this section is informational and may lag behind per-section tables.

Immediate Implementation (🍏 Low Hanging Fruit)

Jest Unit Tests (High ROI, Fast):

1. Balance calculations - all split modes (6 tests)
2. Total calculations - group and user totals (8 tests)
3. Currency conversion utilities (8 tests)
4. Schema validations (6 tests)

Playwright E2E (Critical Paths):

1. Create group happy path (1 test)
2. Create expense - evenly split (1 test)
3. Delete expense (1 test)
4. View balances (1 test)
5. Suggested reimbursements display (1 test)

Estimated: ~30 tests, 2-3 days

---

Phase 2 (🍊 Medium Effort)

Jest Unit Tests:

1. Complex balance scenarios (4 tests)
2. Date handling edge cases (3 tests)
3. Schema validation edge cases (2 tests)

Playwright E2E:

1. All expense split modes (3 tests)
2. Edit flows (5 tests)
3. Currency conversion (2 tests)
4. Export functionality (4 tests)

Estimated: ~23 tests, 3-4 days

---

Phase 3 (🍎 High Effort)

Jest Integration Tests:

1. Recurring expense creation logic (9 tests)
2. Activity logging (6 tests)

Playwright E2E:

1. Recurring expense flows (4 tests)
2. Document upload (if enabled) (3 tests)
3. Complex multi-user scenarios (3 tests)

Estimated: ~25 tests, 5-7 days

---

Phase 4 (Nice to Have)

Lower Priority Tests:

1. Edge cases (P2/P3 tests)
2. UI/UX tests (theme, responsive)
3. i18n tests
4. Activity log tests

Estimated: ~30 tests, 3-4 days

---

5. TEST ORGANIZATION

Recommended File Structure

├── src/
│   └── lib/
│       ├── balances.test.ts          # Balance calculation tests
│       ├── totals.test.ts            # Totals & statistics tests
│       ├── currency.test.ts          # Currency utilities tests
│       ├── utils.test.ts             # Utility functions tests (exists)
│       ├── schemas.test.ts           # Zod schema validation tests
│       └── api.test.ts               # API integration tests (requires DB)
├── tests/
│   ├── e2e/
│   │   ├── group-management.spec.ts
│   │   ├── expense-crud.spec.ts
│   │   ├── expense-split-modes.spec.ts
│   │   ├── recurring-expenses.spec.ts
│   │   ├── balances.spec.ts
│   │   ├── statistics.spec.ts
│   │   ├── active-user.spec.ts
│   │   ├── export.spec.ts
│   │   ├── activity-log.spec.ts
│   │   └── ui.spec.ts
│   └── fixtures/
│       ├── test-data.ts              # Shared test data
│       └── test-helpers.ts           # Helper functions

---

6. TESTING BEST PRACTICES

Jest Best Practices

1. Isolation: Each test should be independent
2. Naming: Use descriptive test names - it('should calculate evenly split for 3 participants')
3. Arrange-Act-Assert: Structure tests clearly
4. Test Data: Use factories/fixtures for consistent test data
5. Mock Sparingly: Only mock external dependencies (DB, API calls)
6. Fast: Unit tests should run in milliseconds
7. Coverage: Aim for 80%+ coverage on business logic

Playwright Best Practices

1. Page Object Model: Create page objects for reusable selectors
2. Fixtures: Use Playwright fixtures for setup/teardown
3. Locators: Use accessible selectors (role, label, text)
4. Waits: Use auto-waiting, avoid hard waits
5. Assertions: Use Playwright's expect for auto-retrying assertions
6. Test Data: Clean up test data after each run
7. Parallel: Run tests in parallel when possible (config already set to fullyParallel: false - consider enabling per test file)
8. Screenshots: Capture on failure for debugging
9. Mobile: Test both mobile and desktop viewports
10. Network: Mock external API calls (currency rates, OpenAI) for reliability

---

7. MOCK STRATEGIES

What to Mock

Jest Unit Tests:

• ❌ Don't mock: Pure functions (balances, totals, currency)
• ✅ Mock: Prisma client for API tests
• ✅ Mock: External APIs (OpenAI, currency rates)

Playwright E2E:

• ❌ Don't mock: Database (use test DB)
• ✅ Mock: External APIs (OpenAI, Frankfurter currency API)
• ✅ Mock: S3 uploads (unless testing S3 feature)
• ✅ Mock: Email/notifications (if added)

Test Database Strategy

For integration tests, use one of:

1. SQLite in-memory: Fast, no setup, but may have dialect differences
2. PostgreSQL test container: Realistic, requires Docker
3. Prisma mocking: Fast, but requires more setup

Recommendation: PostgreSQL test container for realism, or Prisma mocking for speed.

---

8. COVERAGE GOALS

Target Coverage by Module

Module	Current	Target	Priority
src/lib/balances.ts	0%	95%+	🔴 P0
src/lib/totals.ts	0%	95%+	🔴 P0
src/lib/currency.ts	0%	90%+	🔴 P0
src/lib/utils.ts	~30%	90%+	🟡 P1
src/lib/schemas.ts	0%	80%+	🔴 P0
src/lib/api.ts	0%	70%+	🟡 P1
E2E Critical Paths	0%	100%	🔴 P0

---

9. IMPLEMENTATION ROADMAP

Week 1: Foundation (🍏 Low Hanging Fruit)

• Setup Jest test structure
• Balance calculation tests (17 tests)
• Totals calculation tests (14 tests)
• Basic E2E flows (5 tests)
• Deliverable: ~36 tests, core logic validated

Week 2: Expansion (🍊 Medium Effort)

• Currency & utils tests (14 tests)
• Schema validation tests (8 tests)
• Expense CRUD E2E (10 tests)
• Balance E2E verification (4 tests)
• Deliverable: ~36 tests, critical features covered

Week 3: Advanced (🍎 High Effort)

• Recurring expense logic (9 tests)
• Activity logging (6 tests)
• Recurring E2E flows (7 tests)
• Export functionality (8 tests)
• Deliverable: ~30 tests, complex features validated

Week 4: Polish & Edge Cases

• Edge case tests (P2/P3)
• UI/UX tests (6 tests)
• i18n tests (3 tests)
• Activity log E2E (5 tests)
• Documentation updates
• Deliverable: ~30 tests, comprehensive coverage

Total Estimated: ~130+ tests over 4 weeks

---

10. KEY RISKS & MITIGATIONS

Risk	Impact	Mitigation
Recurring expense time-based tests flaky	High	Use fixed dates, mock Date.now()
Database setup complexity	Medium	Use Docker Compose for test DB
S3 tests require AWS setup	Medium	Mock S3 or skip with feature flag check
OpenAI API costs in tests	Low	Always mock AI features
Timezone issues in date tests	High	Use UTC, test formatDateOnly() thoroughly
Test data cleanup	Medium	Use Playwright fixtures for automatic cleanup
Floating point rounding errors	High	Thoroughly test rounding in balances

---

11. SUCCESS METRICS

Immediate (Week 1-2):

• ✅ 80%+ coverage on balances.ts, totals.ts, currency.ts
• ✅ All P0 unit tests passing
• ✅ Basic E2E flow working (create group → create expense → view balance)

Medium Term (Week 3-4):

• ✅ 70%+ overall code coverage
• ✅ All P0 and P1 tests implemented
• ✅ CI/CD integration (tests run on PR)
• ✅ Test execution time < 5 minutes (unit + integration)
• ✅ E2E execution time < 10 minutes

Long Term:

• ✅ 80%+ overall code coverage
• ✅ All P0, P1, P2 tests implemented
• ✅ Zero flaky tests
• ✅ Test-driven development for new features

---

12. TOTAL TEST COUNT SUMMARY

Category	P0	P1	P2	P3	Total	Low Effort	Med Effort	High Effort
Jest Unit Tests	37	22	6	2	67	52	11	4
Playwright E2E	23	38	16	0	77	34	33	10
GRAND TOTAL	60	60	22	2	144	86	44	14

Priority Breakdown:

• 🔴 P0 (Critical): 60 tests - Core functionality
• 🟡 P1 (High): 60 tests - Important features
• 🟢 P2 (Medium): 22 tests - Nice to have
• 🔵 P3 (Low): 2 tests - Edge cases

Effort Breakdown:

• 🍏 Low Hanging Fruit: 86 tests (~60%) - Quick wins
• 🍊 Medium Effort: 44 tests (~31%) - Moderate investment
• 🍎 High Effort: 14 tests (~10%) - Complex scenarios

---

Notes

• No code changes needed: All tests can be written against existing code
• Fast & Simple: Focus on pure unit tests first (balances, totals, currency)
• Best Practices: Follow Jest & Playwright conventions
• Prioritized: P0 tests are must-haves, P3 are optional
• Realistic Estimates: 130+ tests achievable in 4 weeks with 1 developer

This plan provides a clear roadmap from quick wins to comprehensive coverage without modifying application code.

TEST_PLAN_STRUCTURED.md

Spliit Pending Test Implementation Tasks

1. Integration Tests (Jest)

Note: These tests use real Prisma with a test database. No mocking required - set up a test DB connection.

Recurring Expense Logic

Location: src/lib/api.test.ts

Status	Priority	Task	Implementation Guide
✅ Done	🔴 P0	Create daily recurring expense	Verify createRecurringExpenses creates correct number of instances for daily interval. Use test DB.
✅ Done	🔴 P0	Create weekly recurring expense	Verify correct dates are generated for weekly interval (e.g., every Monday).
✅ Done	🔴 P0	Create monthly recurring expense	Verify createRecurringExpenses creates correct date for monthly interval. Use test DB.
✅ Done	🟡 P1	Handle month boundaries	Test edge cases like Jan 31st → Feb 28/29, Mar 31st. Ensure dates don't skip or error.
✅ Done	🟡 P1	Use transactions	Verify DB operations are wrapped in a transaction. Create test data, force error, verify rollback.
✅ Done	🟡 P1	Payload: Daily	Unit test createPayloadForNewRecurringExpenseLink returns correct next date for daily.
✅ Done	🟡 P1	Payload: Weekly	Unit test payload generation for weekly interval.
✅ Done	🟡 P1	Payload: Monthly	Unit test payload generation for monthly interval.

Activity Logging

Location: src/lib/api.test.ts

Status	Priority	Task	Implementation Guide
✅ Done	🟡 P1	Log CREATE_EXPENSE	Call logActivity with 'create' action. Query test DB to verify insert with correct JSON.
✅ Done	🟡 P1	Log UPDATE_EXPENSE	Call with 'update' action. Query test DB to verify insert includes diff or new state.
✅ Done	🟡 P1	Log DELETE_EXPENSE	Call with 'delete' action. Query test DB to verify insert.
✅ Done	🟡 P1	Log UPDATE_GROUP	Call with 'update' action for group changes. Query test DB to verify insert.
✅ Done	🟢 P2	Store Participant ID	Verify participantId column is correctly populated in Activity table via DB query.
✅ Done	🟢 P2	Store Expense Data	Verify data column contains valid JSON snapshot of the expense via DB query.

---

2. E2E Tests (Playwright)

Group & Expense Management

Location: tests/e2e/group-management.spec.ts / expense-crud.spec.ts

Status	Priority	Task	Implementation Guide
✅ Done	🟢 P2	Recent groups persistence	Visit a group, go home, reload page. Assert group appears in "Recent" list (LocalStorage).
✅ Done	🟡 P1	List pagination	Seed 20+ expenses. Verify "Load More" or infinite scroll loads subsequent items. Create helper utility to create such amount of expenses via API and not via UI. Creating via UI will lead to timeout and long test duration.
✅ Done	🟡 P1	List text filter	Type unique string in search bar. Assert only matching expenses are visible.

Recurring Expenses Flow

Location: tests/e2e/recurring-expenses.spec.ts

Status	Priority	Task	Implementation Guide
✅ Done	🔴 P0	Create daily recurring	Fill expense form, select "Daily". Submit. Verify generic recurring indicator UI.
✅ Done	🔴 P0	Create weekly recurring	Fill form, select "Weekly". Submit. Verify UI.
✅ Done	🔴 P0	Create monthly recurring	Fill form, select "Monthly". Submit. Verify UI.
✅ Done	🔴 P0	Verify instances created	(Complex) Create recurring expense. Trigger backend job or mock time. Verify new expense appears in list.
⬜ Missing feature spliit-app/spliit#492, do not test	🟡 P1	Edit stops future	Edit a recurring instance. Assert future instances are unlinked or modified according to logic.
✅ Done	🟡 P1	Delete current only	Delete a recurring instance. Assert other instances in the chain remain.

Statistics & Export

Location: tests/e2e/statistics.spec.ts / export.spec.ts

Status	Priority	Task	Implementation Guide
✅ Done	🔴 P0	Verify Group Total	Sum expenses in list. Assert Stats page "Total Group Spending" matches.
✅ Done	🔴 P0	Verify User Paid	Assert "You Paid" matches sum of active user's payments.
✅ Done	🔴 P0	Verify User Share	Assert "Your Share" matches sum of active user's splits.
✅ Done	🟡 P1	Export JSON download	Click JSON export. Wait for download event. Assert filename/extension.
✅ Done	🟡 P1	Export JSON content	Read downloaded JSON. Assert it contains valid array of expense objects.
✅ Done	🟡 P1	Export CSV download	Click CSV export. Wait for download event.
✅ Done	🟡 P1	Export CSV format	Read downloaded CSV. Assert headers and row data match expenses.

Active User & Activity

Location: tests/e2e/active-user.spec.ts / activity-log.spec.ts

Status	Priority	Task	Implementation Guide
✅ Done	🟡 P1	Selection persists	Select user. Reload page. Assert user is still selected (LocalStorage).
✅ Done	🔴 P0	Updates stats	Change active user. Assert Stats page numbers update to reflect new user.
✅ Done	🟢 P2	Log shows create	Create expense. Go to Activity. Assert "Created expense X" entry exists.
✅ Done	🟢 P2	Log shows update	Edit expense. Go to Activity. Assert "Updated expense X" entry exists.
✅ Done	🟢 P2	Log shows delete	Delete expense. Go to Activity. Assert "Deleted expense X" entry exists.
✅ Done	🟢 P2	Log pagination	Generate many activities. Verify pagination controls work on Activity page.

System & UI

Location: tests/e2e/ui.spec.ts

Status	Priority	Task	Implementation Guide
⬜ Todo	🟢 P2	i18n UI updates	Switch lang to Spanish. Assert static text (headers/buttons) changes.
✅ Done	🟡 P1	i18n Currency format	Switch lang/locale. Assert currency symbol position (€10 vs 10€).
✅ Done	🟡 P1	i18n Date format	Switch lang. Assert date format changes (MM/DD vs DD/MM).
✅ Done	🟡 P1	Mobile responsive	Set viewport to mobile. Assert drawer menu works instead of sidebar.
✅ Done	🟡 P1	Desktop responsive	Set viewport to desktop. Assert dialogs/sidebar appear correctly.
✅ Done	🟡 P1	Health Readiness	Fetch /api/health/readiness. Assert status 200 (database connected).