typesht 0.1.2

Python-syntax to TypeScript/JavaScript compiler

TypeSht

A statically-typed programming language that uses Python-like syntax and compiles to TypeScript/JavaScript. Designed for Python developers who want to write type-safe code that runs in JavaScript environments (browsers, Node.js) without learning TypeScript or JavaScript syntax.

If it works in Python, it works the same in TypeSht. No surprises.

---

Why TypeSht?

Python developers often need to target JavaScript environments — web frontends, Node.js servers, serverless functions. The usual options are:

• Learn JavaScript/TypeScript (steep learning curve)
• Use a transpiler that loses Python semantics (silent bugs)

TypeSht takes a different approach: compile Python-syntax source files to TypeScript, and ship a runtime package (typesht-runtime) that preserves Python behavior faithfully in JavaScript environments. Floor division, negative indexing, Python-style dicts, print() with sep/end — it all works exactly as you'd expect.

---

Requirements

Compiler:

• Python 3.10+

Runtime & output:

• Node.js 16+
• TypeScript (npm install -g typescript)

---

Installation

1. Install the TypeSht compiler (Python):

pip install typesht

2. Install the TypeSht runtime (Node.js):

npm install typesht-runtime

---

Usage

# Compile to JavaScript (default)
typesht compile myprogram.tsht

# Compile to TypeScript only
typesht compile myprogram.tsht --target ts

# Compile to both JS and TS
typesht compile myprogram.tsht --target both

# Specify output directory
typesht compile myprogram.tsht --output dist/

Output goes to an output/ folder next to your source file by default.

To run the compiled output:

cd output/
npm init -y
# add "type": "module" to package.json
npm install typesht-runtime
node myprogram.js

---

Language Features

Variables & Type Annotations

name: str = "Alice"
age: int = 30
score: float = 9.5
is_active: bool = True
x = 42  # untyped

F-Strings

msg: str = f"Hello, {name}! You are {age} years old."
result: str = f"Next year: {age + 1}"

Functions

def greet(name: str, age: int) -> str:
    return f"Hi, I am {name}"

def add(a: int, b: int) -> int:
    return a + b

Lambda Functions

double = lambda x: x * 2
add = lambda x, y: x + y
no_args = lambda: 42

Classes

class Person:
    name: str
    age: int

    def __init__(self, name: str, age: int):
        self.name = name
        self.age = age

    def greet(self) -> str:
        return f"Hi, I am {self.name}"

person = Person("Alice", 30)
print(person.greet())

Decorators

class MathHelper:
    @staticmethod
    def add(x: int, y: int) -> int:
        return x + y

    @classmethod
    def create(cls, base: int) -> int:
        return base

    @property
    def value(self) -> int:
        return self._value

# Custom decorators compile to higher-order function calls
@my_decorator
def process(data: str) -> str:
    return data
# → process = myDecorator(process)

Generics

numbers: list[int] = [1, 2, 3]
words: list[str] = ["hello", "world"]
data: dict[str, int] = {"a": 1}

def get_first(items: list[int]) -> int:
    return items[0]

Lists & Dicts

items: list = [10, 20, 30, 40, 50]
print(items[0])     # → 10
print(items[-1])    # → 50  (negative indexing)
print(items[1:3])   # → [20, 30]  (slicing)

data: dict = {"lang": "TypeSht", "version": "0.1"}

Control Flow

if age > 18:
    print("Adult")
elif age == 18:
    print("Just turned 18")
else:
    print("Minor")

for i in range(10):
    print(i)

for item in items:
    print(item)

for k, v in pairs:
    print(f"{k} = {v}")

while x > 0:
    x -= 1

Operators

result = 10 // 3    # floor division → 3
remainder = 10 % 3  # modulo → 1
power = 2 ** 8      # power → 256
x = a and b         # logical and
y = a or b          # logical or
z = not a           # logical not
found = item in items       # membership test
missing = item not in items

Triple-Quoted Strings

msg = """
This is a
multi-line string
"""

Imports

# Standard library (provided by typesht-runtime)
import math
import random
import json
import re
import sys
import os

# With alias
import random as rng

# From imports
from json import dumps, loads
from math import sqrt, pi

# User modules
import mymodule
from mymodule import my_func

---

Standard Library

Module	Available Functions
math	pi, e, sqrt, floor, ceil, abs, pow, log, log2, log10, sin, cos, tan, exp, trunc, inf, nan, isnan, isinf, gcd, factorial
random	random, randint, uniform, choice, shuffle, sample, seed
json	dumps, loads
re	match, search, findall, sub, split, compile
sys	argv, exit, version, platform
os	path.join, path.dirname, path.basename, path.abspath, path.normpath, path.splitext, path.split, path.exists, path.isfile, path.isdir, path.sep, getcwd, getenv

---

Type Mappings

TypeSht	TypeScript
str	string
int	number
float	number
bool	boolean
list	any[]
list[int]	number[]
list[str]	string[]
dict	any (backed by PyDict runtime class)
None	null

---

Python Semantics Preserved

TypeSht ships typesht-runtime — an npm package that ensures Python behavior is preserved at runtime:

• print() with sep and end keyword arguments
• Negative indexing on strings and lists (items[-1])
• Slicing on strings and lists (items[1:3])
• Floor division (//) toward negative infinity
• Python-style modulo (%) for negative numbers
• dict with ordered keys and Python iteration behavior (for k in d, .items(), .values())
• in / not in membership testing
• Python-style error types (ValueError, TypeError, IndexError, KeyError, ZeroDivisionError)

---

Project Structure

typesht/
├── src/typesht/       — pip-installable compiler package
│   ├── lexer.py       — Tokenizer
│   ├── parser.py      — AST + Parser
│   ├── codegen.py     — TypeScript code generator
│   └── main.py        — CLI entry point
│
├── compiler/          — Development copy (used by the test runner)
│   ├── lexer.py
│   ├── parser.py
│   ├── codegen.py
│   └── main.py
│
├── runtime/
│   ├── typesht-runtime.ts   — Python behavior runtime (source)
│   └── package.json
│
├── tests/
│   └── test_compiler.py     — 136 compiler unit tests
│
└── examples/
    └── hello.tsht           — Comprehensive example program

---

Running Tests

python -m pytest tests/

---

Roadmap

• f-string format specs (f"{value:.2f}")
• Exception handling (try / except / finally)
• Multiple inheritance
• Type aliases
• Standard library expansion (itertools, functools, collections)
• Optional JS extras (explicit opt-in JavaScript-specific features)
• Package manager integration (pip install typesht, npm install typesht-runtime)

---

Author

Jeiel K Libert

License

MIT