Logging#

DecisionAI uses the loguru library for logging. By default, logging is disabled for the decision_ai namespace to avoid cluttering your console output.

Enabling Logging#

There are two ways to enable logging in DecisionAI:

1. Using an Environment Variable#

Set the DECISION_AI_LOG_LEVEL environment variable to your desired log level:

# Linux/MacOS
export DECISION_AI_LOG_LEVEL=DEBUG

# Windows (Command Prompt)
set DECISION_AI_LOG_LEVEL=DEBUG

# Windows (PowerShell)
$env:DECISION_AI_LOG_LEVEL = "DEBUG"

2. Programmatically#

You can also enable logging programmatically in your code:

from decision_ai import logger

# Enable logging for the decision_ai namespace
logger.enable("decision_ai")

# Set specific log level
logger.level("decision_ai", "DEBUG")

Available Log Levels#

Loguru provides the following log levels (in ascending order of severity):

Level	Method	Description
`TRACE`	`logger.trace()`	Most detailed debugging information
`DEBUG`	`logger.debug()`	Detailed debugging information
`INFO`	`logger.info()`	Confirmation that things are working as expected
`SUCCESS`	`logger.success()`	Successful operations (unique to Loguru)
`WARNING`	`logger.warning()`	Indication that something unexpected happened
`ERROR`	`logger.error()`	Error that prevented operation from completing
`CRITICAL`	`logger.critical()`	Very serious error that might cause program termination

In addition, the NONE level can be used for the environment variable DECISION_AI_LOG_LEVEL to disable logging.

Example usage:

from decision_ai import logger

logger.enable("decision_ai")

logger.trace("Very detailed debugging information")
logger.debug("Debugging information")
logger.info("Normal information message")
logger.success("Operation completed successfully")
logger.warning("Warning message")
logger.error("Error message")
logger.critical("Critical error message")

Custom Configuration#

Since DecisionAI uses loguru, you can apply any custom configuration supported by the loguru library:

from decision_ai import logger

# Add a file handler
logger.add("decision_ai.log", rotation="10 MB")

# Enable and configure
logger.enable("decision_ai")
logger.level("decision_ai", "DEBUG")

# Use colored output
logger.opt(colors=True).info("This message has <green>colored</green> parts")

# Log with exception information
try:
    1/0
except Exception:
    logger.exception("An error occurred")

Creating Custom Log Levels#

If needed, you can also create custom log levels:

from decision_ai import logger

# Create a custom level
logger.level("IMPORTANT", no=25, color="<yellow>", icon="⚠️")

# Use the custom level
logger.log("IMPORTANT", "This is an important message")

For more advanced configuration options, refer to the loguru documentation.