Quick Start Guide¶

An LCL Module stores named expressions without evaluating them. A Frame supplies host values and evaluates definitions lazily, resolving references to other definitions as needed.

import asyncio

from lclang.lang import define_frame, define_module


async def main() -> None:
    module = define_module(
        "welcome",
        {
            "greeting": 'f"Hello, {name}!"',
            "result": "greeting + ' Welcome to lclang.'",
        },
    )
    async with define_frame(module, preset={"name": "Ada"}) as frame:
        assert await frame.get("result") == "Hello, Ada! Welcome to lclang."
        assert await frame.get("greeting") == "Hello, Ada!"


asyncio.run(main())

define_module parses all definitions up front. define_frame creates an independent cache. Asking for result evaluates greeting first; the second lookup returns its cached snapshot. Leaving the async with block closes the Frame and releases its owned tasks and resources.

For one expression, name its result in a Module. The canonical Frame includes reviewed builtins such as int and len, alongside your supplied inputs:

import asyncio

from lclang.lang import define_frame, define_module


TOTAL = define_module("total", {"result": "int(subtotal) + len(taxes)"})


async def main() -> None:
    async with define_frame(
        TOTAL,
        preset={"subtotal": "40", "taxes": ["state", "local"]},
    ) as frame:
        assert await frame.get("result") == 42


asyncio.run(main())

int converts the host string "40", and len counts the two supplied taxes. The same Module and Frame APIs handle both a single calculation and a set of related definitions. Use asyncio.run once at the outer boundary of a synchronous application; async applications await their coroutine in the existing loop.

Next, use the tutorials and examples, or consult the reference index.