scitex-logging 0.1.5

Logging utilities for the SciTeX ecosystem

scitex-logging

Logging utilities for SciTeX — SUCCESS/FAIL log levels, typed exception hierarchy, Tee context-manager.

Full Documentation · uv pip install scitex-logging[all]

---

Problem and Solution

#	Problem	Solution
1	stdlib logging has only 5 levels -- experiment scripts want a distinct SUCCESS and FAIL signal that stands out in grep	SUCCESS + FAIL levels added -- color-coded, level-aware handlers; drop-in compatible with getLogger(__name__)
2	raise ValueError("shape mismatch") loses context -- every package rolls its own exception hierarchy	30+ typed exceptions -- SciTeXError root + ShapeError, DTypeError, ConfigKeyError, PDFDownloadError, ...; isinstance(e, DataError) catches the whole class
3	Tee stdout-to-file is a recipe -- every script implements it differently	Tee("run.log") context-manager -- one import, no boilerplate

Installation

pip install scitex-logging

1 Interfaces

Python API

import scitex_logging

# Configure logging
scitex_logging.configure(level=scitex_logging.INFO, enable_file=True)

# Get a logger
import logging
logger = logging.getLogger(__name__)
logger.info("Hello from SciTeX logging")

# Tee stdout/stderr to log files
import sys
sys.stdout, sys.stderr = scitex_logging.tee(sys)

# Custom error classes
from scitex_logging import SciTeXError, SaveError

# Warning utilities
from scitex_logging import warn_deprecated, warn_performance

# LLM session log parsing
log = scitex_logging.llm.load("session.jsonl")
log.summary()
log.render("out.html")

Architecture

scitex_logging/
├── _config.py            ← `configure()` entry point + level registration
├── _levels.py            ← SUCCESS / FAIL custom log levels
├── _tee.py               ← stdout/stderr Tee context-manager
├── exceptions/           ← 30+ typed `SciTeXError` subclasses
│   ├── _data.py          ← ShapeError, DTypeError, ...
│   ├── _config.py        ← ConfigKeyError, ...
│   └── _network.py       ← PDFDownloadError, ...
├── warnings/             ← `warn_deprecated`, `warn_performance`
└── llm/                  ← Claude / LLM session-log parsers

Demo

flowchart LR
    A[script.py] -->|getLogger| B[scitex_logging]
    B --> C{level}
    C -->|SUCCESS| D[green log + Tee → run.log]
    C -->|FAIL| E[red log + Tee → run.log]
    C -->|ERROR| F[typed SciTeXError raised]

import logging, scitex_logging as sxl

sxl.configure(level=sxl.INFO, enable_file=True)
log = logging.getLogger(__name__)

log.success("model converged")    # green
log.fail("validation failed")     # red, distinct from ERROR

2026-05-07 12:00:00 SUCCESS  model converged
2026-05-07 12:00:01 FAIL     validation failed

Part of SciTeX

scitex-logging is part of SciTeX. Install via the umbrella with pip install scitex[logging] to use as scitex.logging (Python).

Four Freedoms for Research

0. The freedom to run your research anywhere — your machine, your terms.
1. The freedom to study how every step works — from raw data to final manuscript.
2. The freedom to redistribute your workflows, not just your papers.
3. The freedom to modify any module and share improvements with the community.

AGPL-3.0 — because we believe research infrastructure deserves the same freedoms as the software it runs on.

License

AGPL-3.0-only — see LICENSE.

---