prettylogx 0.1.0

Beautiful, production-ready Python logging with colors and JSON output

prettylogx

Beautiful, production-ready Python logging — zero external dependencies.

prettylogx wraps Python's built-in logging module and gives you:

• Colored, human-readable logs for your terminal
• Structured JSON logs for production log shippers
• Automatic request ID tracing (per-thread)
• Clean exception / traceback output
• Custom log levels (VERBOSE, SUCCESS)
• A single function — setup_logger() — to set it all up

---

Installation (local development)

# 1. Clone the repo
git clone https://github.com/yourusername/prettylogx.git
cd prettylogx

# 2. Create and activate a virtual environment
python -m venv .venv
# Windows:
.venv\Scripts\activate
# macOS / Linux:
source .venv/bin/activate

# 3. Install in editable mode (changes to source take effect immediately)
pip install -e ".[dev]"

---

Quick start

from prettylogx import setup_logger

logger = setup_logger("my-app")

logger.info("Server started on port 8000")
logger.warning("Database response time is high")
logger.error("Payment failed for order #9921")

Terminal output:

2024-01-15 10:30:45 | INFO     | my-app | Server started on port 8000
2024-01-15 10:30:46 | WARNING  | my-app | Database response time is high
2024-01-15 10:30:47 | ERROR    | my-app | Payment failed for order #9921

Each level is printed in its own color (green / yellow / red / etc.).

---

Colored logs (terminal)

from prettylogx import setup_logger

logger = setup_logger(
    name="my-app",
    level="DEBUG",       # show all levels including DEBUG
    show_time=True,      # include timestamp (default: True)
    show_file=True,      # include filename:lineno
    use_colors=True,     # ANSI colors (default: True)
)

logger.debug("Connecting to DB")
logger.info("User logged in")
logger.warning("Rate limit at 80%")
logger.error("Auth token expired")
logger.critical("Service unreachable")

---

JSON logs (production)

Switch to JSON output by passing json_logs=True. Every line is a self-contained JSON object you can pipe to jq or ship to Datadog / Loki / CloudWatch.

from prettylogx import setup_logger

logger = setup_logger("api", json_logs=True)

logger.info("Request received")
logger.error("DB connection refused")

Output:

{"timestamp": "2024-01-15T10:30:45Z", "level": "INFO", "logger": "api", "message": "Request received", "module": "app", "function": "handle", "line": 12}
{"timestamp": "2024-01-15T10:30:46Z", "level": "ERROR", "logger": "api", "message": "DB connection refused", "module": "app", "function": "connect", "line": 34}

---

Request ID tracing

Track a single request across many log lines without passing the ID to every function.

from prettylogx import setup_logger, set_request_id, clear_request_id

logger = setup_logger("api")

# At the start of a request:
set_request_id("req-abc123")

logger.info("Validating input")     # → includes [req-abc123]
logger.info("Querying database")    # → includes [req-abc123]
logger.info("Returning response")   # → includes [req-abc123]

# At the end of the request:
clear_request_id()

Or pass it explicitly per call:

logger.info("Order placed", extra={"request_id": "req-xyz"})

---

Exception logging

from prettylogx import setup_logger

logger = setup_logger("app")

try:
    result = 1 / 0
except ZeroDivisionError:
    logger.exception("Unexpected error in calculation")

The full traceback is captured and formatted cleanly below the log line. In JSON mode it appears as the "exception" field.

---

setup_logger() parameters

Parameter	Type	Default	Description
name	str	"app"	Logger name shown in every log line
level	str|int	"INFO"	Minimum level: DEBUG / INFO / WARNING / ERROR / CRITICAL
json_logs	bool	False	True → JSON output, False → colored output
show_time	bool	True	Include timestamp in colored output
show_file	bool	False	Include filename:lineno in colored output
use_colors	bool	True	False strips ANSI codes (e.g. for file output)

---

Why not just use the raw logging module?

Feature	logging	prettylogx
Colored terminal output	No	Yes
JSON / structured output	No	Yes
Request ID tracing	Manual	Built-in
Zero-config setup	No	Yes
No external dependencies	Yes	Yes
Works with existing code	—	Yes

prettylogx is a thin wrapper — you can still use every standard logging feature (handlers, filters, log levels) alongside it.

---

Running locally

# Run examples
python examples/basic_usage.py
python examples/json_logs.py
python examples/request_id.py
python examples/exception_logging.py

# Run tests
pytest

# Run tests with coverage report
pytest --cov=prettylogx --cov-report=term-missing

# Build the distributable package
pip install build
python -m build
# → creates dist/prettylogx-0.1.0.tar.gz and dist/prettylogx-0.1.0-py3-none-any.whl

---

Roadmap

• File handler support (write logs to a rotating file)
• Async support (asyncio-safe context variables via contextvars)
• Middleware helpers for FastAPI / Django / Flask
• Log sampling (emit only N% of DEBUG logs in production)
• PyPI publication