Log level filtering in PyTest - Framework Patterns

Framework Mode - Log level filtering

Folder Structure

project-root/
├── tests/
│   ├── test_example.py
│   └── conftest.py
├── src/
│   └── app_code.py
├── logs/
│   └── test.log
├── pytest.ini
└── requirements.txt
    

Test Framework Layers

• Tests: Located in tests/, contain test cases using pytest.
• Fixtures & Config: conftest.py manages setup, including log level filtering configuration.
• Application Code: src/ holds the main code under test.
• Logging Output: logs/ stores log files generated during tests.
• Configuration File: pytest.ini sets pytest options including log level filters.

Configuration Patterns

Use pytest.ini to set log level filtering globally for tests. Example:

[pytest]
log_cli = true
log_cli_level = INFO
log_file = logs/test.log
log_file_level = WARNING
    

This config means:

• Show logs of level INFO and above in the console during test runs.
• Write logs of level WARNING and above to the file logs/test.log.

In conftest.py, you can add fixtures to customize or override log levels per test if needed.

Test Reporting and CI/CD Integration

Logs filtered by level help keep test output clean and focused.

• CI systems can collect logs/test.log for detailed error investigation.
• Console logs show only important info, reducing noise in CI logs.
• Combine with pytest reports (e.g., JUnit XML) for full test results and logs.

Best Practices for Log Level Filtering

1. Set sensible default log levels: Use INFO for console, WARNING or ERROR for files.
2. Use pytest.ini for global settings: Centralize log config for consistency.
3. Keep logs readable: Avoid too verbose DEBUG logs in CI unless troubleshooting.
4. Use fixtures to customize: Override log levels for specific tests if needed.
5. Separate logs by purpose: Console for quick feedback, files for detailed analysis.

Self Check

Where in this folder structure would you add a fixture to change the log level to DEBUG only for a specific test?

Key Result

Use pytest.ini for global log level filtering and conftest.py fixtures for test-specific log control.

Practice

1. What does caplog.at_level() do in pytest?

easy

A. It sets the minimum log level to capture during a test.

B. It clears all captured logs before the test runs.

C. It disables logging for the entire test session.

D. It changes the log message format.

Solution

1. Step 1: Understand caplog usage

   caplog is a pytest fixture used to capture log messages during tests.

2. Step 2: Purpose of at_level()

   The method at_level() sets the minimum log level to capture, filtering out lower-level messages.

3. Final Answer:

   It sets the minimum log level to capture during a test. -> Option A

4. Quick Check:

   caplog.at_level() sets log capture level [OK]

Hint: Remember: at_level sets minimum log level to capture [OK]

Common Mistakes:

• Thinking it clears logs instead of setting level
• Confusing it with disabling logging
• Assuming it changes log format

2. Which of the following is the correct syntax to capture only WARNING and above logs in pytest using caplog?

easy

A. with caplog.at_level(logging.WARNING):

B. with caplog.set_level(logging.WARNING):

C. caplog.at_level = logging.WARNING

D. caplog.capture_level(logging.WARNING)

Solution

1. Step 1: Identify correct method usage

   The correct method to set log level context is caplog.at_level() used with with statement.

2. Step 2: Check syntax correctness

   with caplog.at_level(logging.WARNING): uses with caplog.at_level(logging.WARNING): which is the documented and correct syntax.

3. Final Answer:

   with caplog.at_level(logging.WARNING): -> Option A

4. Quick Check:

   Use with caplog.at_level(level) for filtering [OK]

Hint: Use 'with caplog.at_level(level):' to filter logs [OK]

Common Mistakes:

• Using assignment instead of context manager
• Using non-existent methods like set_level or capture_level
• Missing the with statement

3. Given the code below, what log messages will be captured?

import logging

def test_logs(caplog):
    with caplog.at_level(logging.ERROR):
        logging.warning('Warning message')
        logging.error('Error message')
    return caplog.records

medium

A. Both 'Warning message' and 'Error message' are captured.

B. Only 'Warning message' is captured.

C. Only 'Error message' is captured.

D. No messages are captured.

Solution

1. Step 1: Understand log level filtering

   Setting caplog.at_level(logging.ERROR) captures logs at ERROR level or higher only.

2. Step 2: Analyze logged messages

   Warning is below ERROR level, so it is ignored. Error message is at ERROR level, so it is captured.

3. Final Answer:

   Only 'Error message' is captured. -> Option C

4. Quick Check:

   ERROR level filters out WARNING [OK]

Hint: Only logs at or above set level appear [OK]

Common Mistakes:

• Assuming WARNING logs appear when level is ERROR
• Thinking all logs are captured regardless of level
• Confusing ERROR and WARNING levels

4. Identify the error in the following pytest test code using caplog:

def test_example(caplog):
    caplog.at_level(logging.INFO)
    logging.debug('Debug message')
    logging.info('Info message')
    assert 'Info message' in caplog.text

medium

A. The assert statement syntax is incorrect.

B. caplog.at_level() must be used as a context manager with 'with'.

C. caplog.text does not exist; use caplog.records instead.

D. Logging messages must be uppercase strings.

Solution

1. Step 1: Check caplog.at_level usage

   The method at_level() must be used with a with statement to set the log level context.

2. Step 2: Identify the error

   Calling caplog.at_level(logging.INFO) without with does not set the level properly, so logs may not be filtered as expected.

3. Final Answer:

   caplog.at_level() must be used as a context manager with 'with'. -> Option B

4. Quick Check:

   Use 'with caplog.at_level(level):' always [OK]

Hint: Always use 'with' when calling caplog.at_level() [OK]

Common Mistakes:

• Calling at_level() without 'with' statement
• Assuming caplog.text is invalid (it exists)
• Thinking assert syntax is wrong

5. You want to test a function that logs messages at DEBUG, INFO, and ERROR levels. You only want to capture INFO and ERROR logs but exclude DEBUG. Which pytest code snippet correctly sets this up?

hard

A. with caplog.at_level(logging.ERROR):
func()
assert 'INFO message' in caplog.text

B. with caplog.at_level(logging.DEBUG):
func()
assert 'INFO message' in caplog.text

C. with caplog.at_level(logging.INFO):
func()
assert 'DEBUG message' in caplog.text

D. with caplog.at_level(logging.INFO):
func()
assert 'DEBUG message' not in caplog.text

Solution

1. Step 1: Determine log levels to capture

   Setting caplog.at_level(logging.INFO) captures INFO and ERROR logs but excludes DEBUG logs.

2. Step 2: Verify assertions

   with caplog.at_level(logging.INFO): func() assert 'DEBUG message' not in caplog.text asserts DEBUG messages are not in captured logs, which matches the requirement.

3. Final Answer:

   with caplog.at_level(logging.INFO): func() assert 'DEBUG message' not in caplog.text -> Option D

4. Quick Check:

   Set level INFO to exclude DEBUG logs [OK]

Hint: Set level to INFO to exclude DEBUG logs [OK]

Common Mistakes:

• Setting level too low and capturing DEBUG logs
• Setting level too high and missing INFO logs
• Wrong assertion about presence of DEBUG logs