logger boilerplate with dataclass models for parsing
Project description
loggi
logger boilerplate with dataclass models for parsing
Installation
Install with:
pip install loggi
Usage
>>> import loggi >>> logger = loggi.getLogger("demo", "logs")
The file "demo.log" will be created inside a folder named "logs" in the current directory.
loggi wraps the logging level mapping so the log level can be set without importing the logging module
>>> logger.setLevel(loggi.DEBUG)
Also loggi imports the logging module when it's initialized, so all logging module features can be utilized without explicity importing logging yourself
>>> print(loggi.logging.getLevelName(loggi.INFO)) INFO
loggi uses the format {level}|-|{date}|-|{message}
where date has the format %x %X
>>> logger.info("yeehaw")
produces the log
INFO|-|10/26/23 18:48:30|-|yeehaw
loggi also contains two dataclasses: Log
and Event
.
A Log
object contains a list of Event
objects that can be loaded from a log file (that uses the above format).
Each Event
contains level: str
, date: datetime
, message: str
fields.
>>> log = loggi.load_log("logs/demo.log") >>> print(log) INFO|-|10/26/23 18:48:30|-|yeehaw >>> print(log.num_events) 1 >>> print(log.events[0].level) INFO
Log
objects can be added together.
Useless examples:
>>> log += log >>> print(log) INFO|-|2023-10-26 18:48:30|-|yeehaw INFO|-|2023-10-26 18:48:30|-|yeehaw
New, filtered Log
objects can be created using the filter_dates
, filter_levels
, and filter_messages
functions.
>>> from datetime import datetime, timedelta >>> log = loggi.load_log("realistic_log.log")
Filtering for events between 24 and 48 hours ago:
>>> filtered_log = log.filter_dates(datetime.now() - timedelta(days=2), datetime.now() - timedelta(days=1))
Filtering for events with critical and error levels:
>>> filtered_log = log.filter_levels(["CRITICAL", "ERROR"])
Filtering for events whose message contains "yeehaw", but not "double yeehaw" or "oopsie":
>>> filtered_log = log.filter_messages(["*yeehaw*"], ["*double yeehaw*", "*oopsie*"])
The filtering methods can be chained:
>>> log_slice = log.filter_dates(datetime.now() - timedelta(days=2), datetime.now() - timedelta(days=1)).filter_levels(["CRITICAL", "ERROR"])
When adding Log
objects, the chronosort()
function can be used to reorder the events by date:
>>> log = filtered_log + log_slice >>> log.chronosort()
log
now contains all critical and error level events between 24 and 48 hours ago,
as well as events from anytime of any level with "yeehaw" in the message, but not "double yeehaw" or "oopsie".
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.