caltib 0.3.0

Tibetan calendar toolkit (traditional + reform engines).

caltib High-Precision Tibetan Calendrical Math

caltib is a rigorous Python library designed to unify 15th-century historical algorithms with 21st-century astrophysics. It provides a polymorphic framework for evaluating traditional lunisolar calendars alongside modern mathematical reforms.

Live Demonstration The library can be evaluated directly in the browser without local installation. The web calendar runs completely client-side via WebAssembly: Launch the caltib Web App (Note: Insert your actual link here)

Key Features Unified API: Evaluate traditional engines (Phugpa, Mongol, Tsurphu, etc.) and modern reforms (L0–L6) using a single, harmonized interface.

Progressive Reforms: A tiered sequence of engines advancing from pure-integer mean motion (L0) and fixed-iteration Picard solvers (L1–L3) to strictly reproducible floating-point models (L4–L5) and JPL-powered numerical integration (L6).

Strict Reproducibility: L0–L5 engines utilize integer sine-tables or Chebyshev minimax polynomials to ensure bit-for-bit consistency across all CPU architectures.

Localized Physics: Support for geographical location injection to accurately handle spherical sunrise boundaries and civil day-numbering conventions.

Design & Diagnostics Laboratory: A professional-grade toolkit for designing new calendar parameters via continued fractions and analyzing secular drift over millennia.

Installation caltib is designed to be lightweight. The core package has zero external dependencies, making it highly suitable for embedded environments or web deployment.

Bash

Core: Traditional engines & Reform Tiers L0-L5

pip install caltib

Tools: Diagnostic charting & parameter design (requires NumPy & Matplotlib)

pip install "caltib[tools]"

Ephemeris: JPL DE422 integration & L6 engine (requires Skyfield & jplephem)

pip install "caltib[ephemeris]" Quickstart Basic Day Query Python from datetime import date import caltib

Standard traditional lookup

info = caltib.day_info(date(2026, 2, 21), engine="mongol") print(f"Mongol Tithi: {info.tibetan.tithi}") Advanced Localized Reform Python from caltib.engines.specs import LOC_MONTREAL import caltib

Build an L3 engine localized to Montreal coordinates.

L3 utilizes fixed-iteration rational Picard solvers and spherical sunrise geometry.

engine = caltib.get_calendar("l3", location=LOC_MONTREAL) info = engine.day_info(date(2026, 2, 21)) Diagnostics & Design The library includes a comprehensive diagnostic suite to rigorously analyze calendar accuracy and historical deviations.

Secular Drift: Track how sidereal calendars decouple from the tropical equinox over 1,500-year timescales.

Variance Mapping: Visualize the error propagation and variance of traditional anomaly approximations.

Rational Design: Derive optimal convergents for constructing new calendar architectures.

Example Usage:

Bash

Analyze secular acceleration against JPL DE422 numerical integration

caltib diag analysis --engine mongol --ephem de422

License Distributed under the MIT License. Created by Tsogtgerel Gantumur.