Custom Log Levels

RichColorLog extends Python’s standard logging with additional syslog-compatible log levels.

Level Hierarchy

From lowest to highest severity:

Level	Value	Icon	Description
DEBUG	10	🐛	Detailed diagnostic information
INFO	20	🔔	General informational messages
NOTICE	25	📢	Normal but significant conditions
WARNING	30	⛔	Warning conditions
ERROR	40	❌	Error conditions
CRITICAL	50	💥	Critical conditions
FATAL	55	💀	Fatal errors
ALERT	59	🚨	Action must be taken immediately
EMERGENCY	60	🆘	System is unusable

Using Custom Levels

All custom levels are available as logger methods:

from richcolorlog import setup_logging

logger = setup_logging()

# Standard Python levels
logger.debug("Debugging information")
logger.info("Informational message")
logger.warning("Warning message")
logger.error("Error occurred")
logger.critical("Critical error")

# Custom levels
logger.notice("Notice: User logged in")
logger.fatal("Fatal: Cannot recover from error")
logger.alert("Alert: Disk space critical")
logger.emergency("Emergency: System failure")

Level Constants

Import level constants for programmatic use:

from richcolorlog import (
    DEBUG_LEVEL,
    INFO_LEVEL,
    NOTICE_LEVEL,
    WARNING_LEVEL,
    ERROR_LEVEL,
    CRITICAL_LEVEL,
    FATAL_LEVEL,
    ALERT_LEVEL,
    EMERGENCY_LEVEL,
)

# Use in configuration
logger = setup_logging(level=NOTICE_LEVEL)

# Check if level is enabled
if logger.isEnabledFor(ALERT_LEVEL):
    logger.alert("This will be logged")

Syslog Severity Mapping

Custom levels map to RFC 5424 syslog severities:

RichColorLog Level	Syslog Severity	Syslog Name
EMERGENCY	0	Emergency
ALERT	1	Alert
FATAL	1	Alert (no standard)
CRITICAL	2	Critical
ERROR	3	Error
WARNING	4	Warning
NOTICE	5	Notice
INFO	6	Informational
DEBUG	7	Debug

This mapping ensures compatibility when using the Syslog handler.

Filtering by Level

Set minimum level for handlers:

from richcolorlog import setup_logging, NOTICE_LEVEL

# Console shows NOTICE and above
# File logs everything including DEBUG
logger = setup_logging(
    level='DEBUG',           # Logger accepts all
    log_file=True,
    log_file_level='DEBUG', # File gets everything
)

# The console handler level is set by 'level' parameter

Per-Handler Levels

Different handlers can have different levels:

logger = setup_logging(
    level='DEBUG',
    log_file=True,
    log_file_level='DEBUG',      # File: all messages
    syslog=True,
    syslog_level='WARNING',      # Syslog: warnings and above
    rabbitmq=True,
    rabbitmq_level='ERROR',      # RabbitMQ: errors only
)

Level-Based Table Names (Database)

When using the database handler, logs are stored in separate tables by level:

Level	Table Name
EMERGENCY	log_emergency
ALERT	log_alert
FATAL	log_fatal
CRITICAL	log_critical
ERROR	log_error
WARNING	log_warning
NOTICE	log_notice
INFO	log_info
DEBUG	log_debug

All logs are also stored in log_syslog table for combined querying.

Icon Customization

Icons are provided by the Icon class:

from richcolorlog.logger import Icon

# Access icons
print(Icon.debug)      # 🐛
print(Icon.info)       # 🔔
print(Icon.notice)     # 📢
print(Icon.warning)    # ⛔
print(Icon.error)      # ❌
print(Icon.critical)   # 💥
print(Icon.fatal)      # 💀
print(Icon.alert)      # 🚨
print(Icon.emergency)  # 🆘

# Uppercase aliases also work
print(Icon.DEBUG)      # 🐛
print(Icon.EMERGENCY)  # 🆘

Best Practices

Use the right level for the right situation:

# DEBUG: Detailed info for debugging
logger.debug(f"Processing item {item_id} with data: {data}")

# INFO: Normal operations
logger.info(f"User {user_id} logged in successfully")

# NOTICE: Noteworthy but normal
logger.notice(f"Configuration reloaded from {config_path}")

# WARNING: Something unexpected but recoverable
logger.warning(f"Rate limit approaching: {current}/{limit}")

# ERROR: Operation failed
logger.error(f"Failed to save file: {filename}")

# CRITICAL: Serious problem
logger.critical(f"Database connection pool exhausted")

# FATAL: Unrecoverable error
logger.fatal(f"Cannot initialize required service: {service}")

# ALERT: Needs immediate attention
logger.alert(f"Security breach detected from IP: {ip}")

# EMERGENCY: System unusable
logger.emergency("System memory exhausted, shutting down")