DataBeak

# Test Fixtures This directory contains sample CSV files used for integration testing. ## Files ### `sample_data.csv` Basic employee data with: - name, age, city, salary columns - 5 rows of clean data - Good for basic loading and operations testing ### `sales_data.csv` Sales transaction data with: - date, product, quantity, price, customer_id columns - 6 rows of sales data - Useful for date/time operations and aggregation testing ### `empty.csv` Empty CSV with headers only: - id, name, value columns - No data rows - Tests handling of empty datasets ### `missing_values.csv` Data with missing values: - name, age, city, score columns - 6 rows with various missing values (empty strings and missing fields) - Tests data quality and missing value handling ## Usage Use the `get_fixture_path()` helper function from `conftest.py` to get absolute paths: ```python from tests.integration.conftest import get_fixture_path # Get real absolute path to fixture csv_path = get_fixture_path("sample_data.csv") result = await server.call_tool("load_csv", {"file_path": csv_path}) ``` This ensures that DataBeak receives the full absolute path, avoiding any confusion about relative paths during testing. ### Security Features The `get_fixture_path()` function includes security validations to prevent directory traversal attacks: - **Path separator validation**: Rejects fixture names containing `/`, `\`, or other path separators - **Relative path protection**: Blocks `..`, `.`, and other relative path components - **Directory containment**: Ensures resolved paths stay within the `tests/fixtures/` directory - **Input sanitization**: Rejects empty strings and whitespace-only names **Examples of rejected inputs:** - `../../../etc/passwd` (directory traversal) - `subdir/file.csv` (path separators) - `.hidden` (relative path component) - `""` (empty string)