tiferet-ly

A Lex/Yacc Wrapper built with Tiferet

Overview

tiferet-ly lets you describe a small language — its words and the sentence patterns it accepts — as a declared list instead of hand-written code. Hand tiferet-ly that declaration and get back a working reader for text written in that language, built on Python's well-established Lex/Yacc engine (PLY). Adding a new keyword or sentence pattern becomes an edit to the declaration rather than a new function buried in a growing file, and the pieces around it can be tested without invoking the real parsing engine at all.

Documentation

Declare a language's words and sentence patterns as data. Translate that declaration into the form the Lex/Yacc engine expects. Assemble those translated rules into a reader. Read text written in the declared language and hand back the result.

• Quick-Start — run the checked-in calculator from the repository root.
• Domain Vision Statement — what tiferet-ly is for and the value it aims to provide.
• Core Domain Distillation — the technical breakdown of its intended vocabulary, behaviors, and design commitments.
• AGENTS.md — orientation for contributors and AI agents working in this repository.