pino.py

Json natural logger for python inspired by pino.js :evergreen_tree:

In building port of pinojs logging library to python :snake:

:warning: This is a in building prototype, it's API is subject to change. A CHANGELOG will be introduced once it's stable enough and publicized. Use it at you own risk, but feel free to reach with an issue.

Basic Example

from pino import pino

logger = pino(
    bindings={"apptype": "prototype", "context": "main"}
)

logger.info("Hello, I just started")
logger.debug({"details": 42}, "Some details that won't be seen")

child_logger = logger.child(context="some_job")
child_logger.info("Job started")
child_logger.info({"duration": 4012}, "Job completed %s", "NOW")

logger.info("Program completed")

Which would output:

{"level": "info", "time": 1587740056952, "message": "Hello, I just started", "host": "SomeHost", "apptype": "prototype", "context": "main", "millidiff": 0}
{"level": "info", "time": 1587740056952, "message": "Job started", "host": "SomeHost", "context": "some_job", "apptype": "prototype", "millidiff": 0}
{"level": "info", "time": 1587740056952, "message": "Job completed NOW", "host": "SomeHost", "duration": 4012, "context": "some_job", "apptype": "prototype", "millidiff": 0}
{"level": "info", "time": 1587740056952, "message": "Program completed", "host": "SomeHost", "apptype": "prototype", "context": "main", "millidiff": 0}

API

pino() constructor arguments

• bindings: meta attached to the messages by default
• level: minimal level to output logs, default to info
• enabled: is logger enabled, default to true
• millidiff: whether a millidiff is added to message, ms since last message, enabled by default.
• stream : stream to write logs to, default to sys.stdout
• dump_function: function to be used to serialise object to JSON, default json.dumps
• messagekey: key for message entry, default message

pino logger instance

• log methods: critical, error, warn, info, debug: (extra_bindings?), message, template value

• .level: access or update current log level

• child(metas): create a child logger instance with new metas/bindings attached to it. (metas can be provided either as dict or kwargs)

Complex examples

You can see more detailed examples in the examples folder, notably complex.py

Development :hammer_and_wrench:

This library use Poetry for management of the project, it's dependencies, build and else.

You can run test on all supported python version with poetry run task test (which will run tox), or you can run on your current python version with poetry run task pytest.