caplog for log messages in PyTest - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the output of this pytest caplog test?

Consider this pytest test using caplog to capture logs. What will be the value of caplog.records[0].levelname after running the test?

PyTest

import logging

def test_log_level(caplog):
    logger = logging.getLogger('mylogger')
    with caplog.at_level(logging.WARNING):
        logger.error('Error occurred')
    print(caplog.records[0].levelname)

A "WARNING"

B "INFO"

C "ERROR"

D "DEBUG"

💡 Hint

Remember that logger.error logs at ERROR level, which is higher than WARNING.

The logger logs an error message, which has level ERROR. caplog.records captures the actual log records, so the first record's levelname is 'ERROR'.

❓ assertion

intermediate

Which assertion correctly checks for a log message using caplog?

You want to assert that the log message 'User created' was logged during the test. Which assertion is correct?

PyTest

import logging

def test_user_creation_logs(caplog):
    logger = logging.getLogger('app')
    logger.info('User created')

A assert caplog.records[0].msg == 'User created'

B assert 'User created' in caplog.text

C assert caplog.records[0].message == 'User created'

D assert caplog.messages[0] == 'User created'

💡 Hint

caplog.messages is a list of log messages as strings.

caplog.messages contains the formatted log messages as strings. So checking caplog.messages[0] equals 'User created' is correct.

🔧 Debug

advanced

Why does this caplog test fail to capture logs?

This test does not capture any logs. What is the reason?

PyTest

import logging

def test_no_logs(caplog):
    logger = logging.getLogger('mylogger')
    logger.setLevel(logging.ERROR)
    with caplog.at_level(logging.WARNING):
        logger.info('Info message')
    assert len(caplog.records) == 1

A Logger level is ERROR, so INFO messages are ignored even if caplog level is WARNING

B caplog.at_level does not change the logger level, so no logs are captured

C caplog only captures WARNING and above by default, so INFO is filtered

D The assertion is wrong; caplog.records length is 0, not 1

💡 Hint

Logger's own level controls which messages it emits.

Logger's level is ERROR, so it ignores INFO messages. caplog.at_level sets capture level but does not override logger's level.

❓ framework

advanced

How to capture logs from a specific logger using caplog?

You want to capture logs only from the logger named 'myapp.database'. Which caplog usage is correct?

A with caplog.at_level(logging.INFO, logger='myapp.database'):

B with caplog.at_level(logging.INFO): # then filter logs manually

C caplog.set_level(logging.INFO, logger='myapp.database')

D caplog.set_level(logging.INFO) # captures all loggers

💡 Hint

caplog.at_level accepts a logger parameter to filter logs.

Using caplog.at_level with logger='myapp.database' captures logs only from that logger at INFO level.

🧠 Conceptual

expert

What is the main difference between caplog.records and caplog.messages?

Choose the best explanation of the difference between caplog.records and caplog.messages in pytest.

A caplog.records contains strings; caplog.messages contains LogRecord objects

B caplog.records contains LogRecord objects; caplog.messages contains formatted log strings

C caplog.records is a list of logger names; caplog.messages is a list of log levels

D caplog.records and caplog.messages are identical lists of log messages

💡 Hint

Think about raw log data vs formatted text.

caplog.records holds LogRecord objects with detailed info; caplog.messages holds just the formatted message strings.