Origin Programming Language · v1.7.31

Origin is my programming language — English-like syntax, strict typing, hardware-first (GPIO/servo primitives for Raspberry Pi), and the full Python ecosystem underneath. this repo is the stable release: pip install origin-or and you're writing .or files. dev chaos happens in origin-dev; this is the polished snapshot.

docs live at docs-origin.onrender.com.

what's new in v1.7.31

• JS-shaped default dialect — the bundled origin.toml now maps JavaScript-style lines (console.log(x);, x++;, a === b;, Math.floor(x);, arr.push(x);, // comments, trailing ;) to real Origin. keyword-led lines (let, if, while, ...) still parse natively, semicolon-free.
• dialect preset pack — tomls/ ships js/py/ts/c/cpp/rs/java_or.toml shorthands plus origin_dialect.md, the authoring spec (rules, keyword guard, debugging recipe).

what's new in v1.7.30

• dialect support (TOML) — mix foreign-language lines straight into .or programs. each line is looked up in the bundled origin.toml ("<origin source>" _ "<dialect line>") and compiled to real Origin before execution, including $name template captures (e.g. "let $n:int = $v" _ "$n:$v" turns x:5 into let x:int = 5). expansion is inline, sharing constants and control flow.
• dialect-aware runner — origin/runner.py preloads origin.toml, precompiles exact mappings to origin_bytecode.toml, resolves templates at parse time. interpreter mode (origin i) skips dialect lines; use default VM mode.
• open ... as statement — open "data.txt", "r" as f binds a file handle (new OPEN_FILE opcode 0x53 on the VM, name = open(path, mode) in the interpreter). path/mode accept plain strings, f-strings, and variables.
• dialect guard — templated mappings (e.g. "$n:$v") no longer hijack lines starting with a core keyword, so open f"C:/..." paths and typed declarations compile as real statements.
• fresh origin.exe + OriginInstallerV1.7.30.zip, carried-over modular layout (lexer/parser/interpreter/classes subpackages) with the bytecode VM as default backend.

quickstart

git clone https://github.com/boblio-max/origin.git
pip install -r requirements.txt

let x: int = 10
let name: str = "Origin"
let flag: bool = true
const e: float = 2.71828

let a, b = 1, 2
print x + a

python -m origin main.or    # bytecode VM (default)
python -m origin i main.or  # AST interpreter

how it actually works

origin/ package pipeline: lexer.py tokenizes → parser.py builds the AST → either bc/to_byte.py compiles to bytecode for bc/svm.py (the stack VM, default) or the interpreter generates Python and execs it. dialect.py + origin.toml handle the foreign-line expansion before parsing. errors.py gives friendly diagnostics with suggestions instead of raw tracebacks.

features

English-like syntax · strict (recommended) type annotations with inference · classes/OOP · try/except/else · parallel thread blocks · module system (import math as m, from lib import x) · .or library imports (calc.or, graph.or) · hardware I/O (GPIO, ServoKit/PCA9685 with 0–180° servo clamping) · binary builder (PyInstaller .exe) · binary + install.sh/secure_install.ps1 installers.

stack

Python 3.10+, zero runtime deps (extras: Pi/servo libs, pyinstaller for builds). see requirements.txt, setup.py, helloworld.or.