Add comprehensive test coverage for untested backend functions and web routes

[Copilot]

Added 47 tests (32 backend + 15 web) covering all previously untested backend functions and web application routes, increasing total test count from 4 to 51.

Coverage Added

Backend Functions (32 tests):

user_handler.py (13 tests)

• Password hashing/verification and validation logic
• User retrieval by ID and username
• User deletion and team membership management
• Input validation and duplicate handling

team_handler.py (6 tests)

• Team CRUD operations (create, get by ID/name, delete)
• Not-found scenarios

table_operations.py (8 tests)

• Table lifecycle (create, delete) with error cases
• Cell value operations (set, get, update, delete)

table_permissions.py (5 tests)

• Permission management (set, update, get, delete by range/all)

Web Application (15 tests):

test_web_routes.py

• Index/homepage rendering
• Login/logout functionality (GET and POST)
• Dashboard access control
• Error pages (403, 404, 500)
• Static file serving (CSS, images)
• Session middleware functionality
• Custom request headers and middleware

Backend API Improvements

Fixed UUID Handling:

• Updated create_table, set_cell_value, and get_cell_value in table_handler.py to accept project_id: UUID directly
• Eliminated need for str(project_id) conversions throughout codebase
• Functions now handle UUID-to-string conversion internally where needed for SQL queries
• Makes the API more type-safe and consistent with other backend functions

Implementation

All tests follow existing patterns:

• Use @pytest.mark.user_creation, @pytest.mark.data_db, and @pytest.mark.web markers for CI test selection
• Leverage transaction fixtures for test isolation
• Handle async/await properly with @pytest.mark.asyncio
• Web tests use AsyncClient from httpx for FastAPI endpoint testing

Test Organization & Clarity Improvements

Structured Test Files:

• Split large test file into focused modules: test_table_operations.py and test_table_permissions.py
• Added test_web_routes.py for comprehensive web application testing
• Each test file has clear, single responsibility

Enhanced Debugging:

• Added descriptive assertion messages to all tests for easier failure diagnosis
• Example: assert user is not None, f"User {user_id} should be found"
• Failed tests now provide specific context about what went wrong

Helper Fixtures:

• Added test_user, test_project, and test_table fixtures in conftest.py
• Reduces code duplication and improves maintainability

New Pytest Markers:

• @pytest.mark.table_operations - For table and cell operation tests
• @pytest.mark.table_permissions - For permission management tests
• @pytest.mark.web - For web route and UI tests
• Enables selective test execution: pytest -m table_operations or pytest -m web

Documentation:

• Comprehensive tests/README.md documenting test structure, markers, best practices, and debugging guidance
• Includes examples for running specific test groups

Example test structure:

@pytest.mark.table_operations
@pytest.mark.data_db
@pytest.mark.asyncio
async def test_set_and_get_cell_value(user_db_transaction, data_db_transaction):
    user_id = await create_user(...)
    project_id = await create_project(...)
    await create_table(data_db_transaction, "test_table", project_id)
    
    test_value = "Test Value"
    await set_cell_value(data_db_transaction, project_id, "test_table", 1, 1, test_value)
    value = await get_cell_value(data_db_transaction, project_id, "test_table", 1, 1)
    
    assert value == test_value, f"Cell value should be '{test_value}', got '{value}'"

Tests are ready for CI execution with existing database setup.

Original prompt

Add tests for all untested code please

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[realMorgon]

@copilot You may also create new test markers or new testfiles. Make it a priority that tests are easy to understand and if failed the reason is easy to find. You may also edit my tests if there is a need for that.

[Copilot]

@copilot You may also create new test markers or new testfiles. Make it a priority that tests are easy to understand and if failed the reason is easy to find. You may also edit my tests if there is a ...

I've improved the test organization and clarity in commit d785bee:

Test Organization:

• Split the 490-line test_table_handler.py into two focused files: test_table_operations.py (table/cell operations) and test_table_permissions.py (permission management)
• Added new pytest markers: @pytest.mark.table_operations and @pytest.mark.table_permissions for selective test execution
• Created tests/README.md documenting test structure, markers, and best practices

Improved Clarity:

• Added descriptive assertion messages to all tests. Example: assert user is not None, f"User {user_id} should be found"
• Added helper fixtures (test_user, test_project, test_table) in conftest.py to reduce duplication
• Enhanced docstrings to clearly explain what each test does

Easier Failure Diagnosis:
When tests fail, the assertion messages now provide specific context:

• What was expected
• What was received
• Relevant IDs and values

You can now run focused test groups: pytest -m table_operations or pytest -m table_permissions

[realMorgon]

@copilot Are the tests still in a way the db is reset after?

[Copilot]

@copilot Are the tests still in a way the db is reset after?

Yes, the database is still reset after each test. All tests use the user_db_transaction and data_db_transaction fixtures which automatically rollback all changes in their finally block (lines 36 and 66 in conftest.py).

This includes:

• All original tests
• All new tests I added (test_user_handler.py, test_team_handler.py, test_table_operations.py, test_table_permissions.py)
• The new helper fixtures (test_user, test_project, test_table) - they also use the transaction fixtures, so their changes are rolled back too

Each test runs in complete isolation with a clean database state.

[realMorgon]

@copilot I see you are using much "str(project_id)". It seems i didn't change all the functions to use UUIDs as parameters. Please take care of this.
Also: Is there a way to test the website part of the project? If so implement it please

[Copilot]

@copilot I see you are using much "str(project_id)". It seems i didn't change all the functions to use UUIDs as parameters. Please take care of this.
Also: Is there a way to test the website part of t...

Fixed in commit d3deb5f:

UUID Handling:

• Updated create_table, set_cell_value, and get_cell_value in table_handler.py to accept project_id: UUID directly
• Removed all str(project_id) calls from test files
• Functions now handle UUID-to-string conversion internally where needed for SQL queries
• This makes the API more type-safe and consistent

Web Tests:

• Added test_web_routes.py with 15 tests covering:
  • Homepage and login pages
  • Login/logout functionality
  • Dashboard access control
  • Error pages (403, 404, 500)
  • Static file serving
  • Session middleware and custom headers
• Added @pytest.mark.web marker for selective test execution
• Run with: pytest -m web

Total test count increased from 36 to 51 tests.