Fix SQLAlchemy "index ix_users_email already exists" error in CI

[Copilot]

Pull Request

Description

CI test runs fail with sqlalchemy.exc.OperationalError: index ix_users_email already exists because the PostgreSQL test database persists between runs. Indexes created in previous runs conflict with subsequent create_all() calls.

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Code refactoring
• Performance improvement
• Test coverage improvement

Related Issues

Changes Made

tests/conftest.py (new)

• Pytest session hooks (pytest_sessionstart, pytest_sessionfinish) drop all tables before/after test sessions
• Skips cleanup for in-memory SQLite (:memory:) to avoid overhead
• Provides informative logging for persistent database cleanup

tests/test_database_integration.py

• test_db fixture now respects DATABASE_URL env var (PostgreSQL in CI, SQLite locally)
• Explicit drop_all() before create_all() in fixture setup/teardown
• Engine configuration handles both SQLite and PostgreSQL properly

src/database.py

• Added reset_db() utility for manual database reset with logging

Testing

• Unit tests pass locally
• Integration tests pass locally
• Manual testing completed
• Docker build succeeds

Test Commands

# Local (SQLite)
pytest tests/test_database_integration.py -v

# CI simulation (PostgreSQL)
DATABASE_URL=postgresql://user:pass@localhost:5432/testdb pytest tests/test_database_integration.py -v

Screenshots

N/A

Checklist

• My code follows the project's code style
• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• Any dependent changes have been merged and published

Additional Notes

The fix operates at two levels:

1. Session-level: conftest.py hooks ensure clean state between CI runs
2. Function-level: test_db fixture ensures clean state between individual tests

Example of the cleanup logic:

def _cleanup_database(test_db_url, phase="session"):
    # Skip in-memory databases
    if not test_db_url.startswith('sqlite:///:memory:'):
        engine = _create_cleanup_engine(test_db_url)
        Base.metadata.drop_all(bind=engine)
        engine.dispose()

CodeQL security scan: 0 alerts

Deployment Notes

• Database migrations required
• Environment variables added/changed
• Configuration changes needed
• Dependencies updated

---

Reviewer Notes:

Original prompt

The failure is caused by the error:

sqlalchemy.exc.OperationalError: (sqlite3.OperationalError) index ix_users_email already exists

This indicates your test suite (using SQLite) tries to create the unique index "ix_users_email" on the "users" table multiple times. This typically happens when test database schema setup doesn't properly drop and recreate tables/indexes between tests, or model definitions/migrations are being applied repeatedly.

Solution:

1. Ensure Test Database Cleanup

   • Before each test or test class, drop all tables or run a teardown so the database is fresh.

   • With SQLAlchemy, you can do (example in pytest fixture):

     import pytest
     from sqlalchemy import create_engine, MetaData
     from sqlalchemy.orm import sessionmaker
     
     @pytest.fixture(autouse=True, scope='function')
     def setup_and_teardown_db():
         engine = create_engine('sqlite:///:memory:')
         # metadata is your SQLAlchemy MetaData object
         metadata.create_all(engine)
         yield
         metadata.drop_all(engine)

   • Alternatively, if using file-based SQLite file, delete the DB file before each test run.

2. Do Not Reapply Models/Indexes If Already Exists

   • If you manage schema manually, guard index creation logic:

     from sqlalchemy import inspect
     
     def create_indexes(engine, metadata):
         inspector = inspect(engine)
         if 'ix_users_email' not in [i['name'] for i in inspector.get_indexes('users')]:
             # code to create index

3. Verify Test Setup

   • Double-check any migration/initialization code to ensure it doesn’t create indexes multiple times.

Code Fix Example:
If you use Base.metadata.create_all(engine) in tests, be sure to pair it with Base.metadata.drop_all(engine) before re-creating to ensure a clean state:

# At the beginning of your test setup
Base.metadata.drop_all(engine)
Base.metadata.create_all(engine)

Summary:

• Clean up the test DB schema (drop tables) before creating, or use an in-memory SQLite DB initialized once per test.
• Avoid creating indexes/tables repeatedly without cleanup.

Making these changes should resolve the repeated “index ... already exists” errors and fix your CI job.

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

[netlify]

❌ Deploy Preview for spectacular-strudel-d144b2 failed.

Name	Link
🔨 Latest commit	bfdcddc
🔍 Latest deploy log	https://app.netlify.com/projects/spectacular-strudel-d144b2/deploys/6978d971185d7a00086fd449