Logging in tests¶

How to add logging to your tests using Python's built-in logging package.

Python's built-in logging package provides a flexible way to add logging to your tests, which can help with debugging and monitoring test execution.

Logging in tests is valuable because it:

Provides visibility into test execution flow
Helps diagnose failures by capturing context
Creates an audit trail of test operations
Enables performance monitoring
Facilitates troubleshooting in CI/CD pipelines

The logs will be written to:

Console output
CI/CD system logs if running in a pipeline

While Python's logging package is versatile, HardPy offers additional specialized logging methods:

How to implement logging¶

Basic setup¶

Import the logging module where needed:
```
import logging
```

Add a fixture to conftest.py:

@pytest.fixture(scope="module")
def module_log(request: pytest.FixtureRequest):
    log_name = request.module.__name__
    yield logging.getLogger(log_name)

Using the logger¶

You can call logging commands anywhere in your tests:

self._log.warning("Saving user credentials for registration test case")

Example of logging in a test function:

def test_log(module_log: logging.Logger):
    module_log.info("DUT serial number logged")
    assert True

Configuration¶

To configure logging parameters, add these settings to your pytest.ini file:

[pytest]
log_cli = true
log_cli_level = INFO
log_cli_format = %(asctime)s [%(levelname)s] %(message)s
log_cli_date_format = %H:%M:%S

You can find the description in How to manage logging

Explanation of configuration options¶

log_cli = true: Enables logging output during test execution
log_cli_level = INFO: Sets the minimum logging level to display
log_cli_format: Defines the format of log messages:
%(asctime)s: Timestamp
[%(levelname)s]: Log level (INFO, WARNING, etc.)
%(message)s: The actual log message
log_cli_date_format = %H:%M:%S: Sets the timestamp format to hours:minutes:seconds