stdlogkit 0.1.0

A small stdlib logging auto-configuration kit with colored multiline logs and log-once helpers.

stdlogkit

stdlogkit is a tiny, vLLM-independent wrapper around Python's standard logging package.

Importing it configures standard logging immediately:

import logging
import stdlogkit  # noqa: F401

logger = logging.getLogger(__name__)
logger.info("hello")
logger.warning_once("this warning appears once")

Default output:

INFO 06-24 12:00:01 [example.py:5] hello
WARNING 06-24 12:00:01 [example.py:6] this warning appears once

Features

• Uses only Python's standard logging machinery.
• Auto-configures logging on import stdlogkit.
• Configures the root logger by default, so normal logging.getLogger(...) calls work immediately.
• Adds debug_once, info_once, and warning_once to standard logger instances.
• Supports multiline log alignment.
• Supports ANSI colors with auto, forced-on, and forced-off modes.
• Supports custom JSON config through logging.config.dictConfig.
• Includes optional uvicorn access-log filtering helpers.
• Has no runtime dependency on vLLM or any other third-party package.

Environment Variables

Variable	Default	Description
STDLOGKIT_CONFIGURE_LOGGING	1	Set 0 to disable auto configuration.
STDLOGKIT_LOGGING_LEVEL	INFO	Root or named logger level.
STDLOGKIT_LOGGING_STREAM	ext://sys.stdout	Handler stream.
STDLOGKIT_LOGGING_PREFIX	empty	Prefix prepended to every log line.
STDLOGKIT_LOGGING_COLOR	auto	auto, 1, or 0.
STDLOGKIT_LOGGING_CONFIG_PATH	unset	Path to a JSON dictConfig file.
STDLOGKIT_LOGGER_NAME	empty	Configure a named logger instead of root.
STDLOGKIT_ROOT_DIR	current working directory	Base path for relative file display.
STDLOGKIT_SHOW_REL_PATH	debug	debug, always, or never.
NO_COLOR	0	Standard flag to disable ANSI colors.

Custom JSON Config

{
  "version": 1,
  "disable_existing_loggers": false,
  "formatters": {
    "plain": {
      "class": "stdlogkit.formatter.NewLineFormatter",
      "format": "%(levelname)s %(asctime)s [%(fileinfo)s:%(lineno)d] %(message)s",
      "datefmt": "%m-%d %H:%M:%S"
    }
  },
  "handlers": {
    "console": {
      "class": "logging.StreamHandler",
      "formatter": "plain",
      "level": "INFO",
      "stream": "ext://sys.stdout"
    }
  },
  "root": {
    "handlers": ["console"],
    "level": "INFO"
  }
}

Run with:

STDLOGKIT_LOGGING_CONFIG_PATH=/path/to/logging.json python app.py

Named Logger Mode

By default, stdlogkit configures the root logger. If you want behavior closer to a framework-specific logger, configure only a named logger:

STDLOGKIT_LOGGER_NAME=my_app python app.py

Then loggers under my_app.* propagate to that logger:

import logging
import stdlogkit  # noqa: F401

logger = logging.getLogger("my_app.service")
logger.info("hello")

Development

cd standalone_logging_pkg
uv venv --python 3.12
uv pip install -e ".[dev]"
.venv/bin/python -m pytest -q
uv build

Publish to PyPI

Before publishing, make sure the project name in pyproject.toml is available on PyPI and replace the placeholder author/repository metadata.

cd standalone_logging_pkg
uv build
uv publish --token "$PYPI_TOKEN"

For TestPyPI:

uv publish --publish-url https://test.pypi.org/legacy/ --token "$TEST_PYPI_TOKEN"