Skip to content
BranchesTags

Repository files navigation

lang-mini

Lang-mini is a lightweight, zero-runtime-dependency JavaScript toolkit that powers the core utilities, type system, and event infrastructure used across the wider jsgui3 ecosystem. It ships as a single file (lang-mini.js), works identically in Node.js and modern browsers, and focuses on highly-polymorphic utilities that remain ergonomic in plain JavaScript.

✨ Key idea: Lang-mini combines small, composable helpers (iteration, cloning, truth maps) with advanced primitives (multi-function polymorphism, functional data types, grammar-based type inference) so that larger frameworks can build rich behavior without carrying heavy dependencies.

Why lang-mini?

  • Cross-environment – lang-mini.js auto-detects whether it's running in Node or the browser and exports the same API in both contexts via lib-lang-mini.js.
  • Polymorphic by default – mfp() and fp() route calls using runtime signatures, letting you author concise yet type-aware functions.
  • Typed without TypeScript – Functional_Data_Type, Type_Signifier, and Grammar provide runtime validation, parsing, and structural signatures that plug into UI and data layers.
  • Event-first architecture – Evented_Class, eventify, field, and prop enable reactive data binding with tiny objects.
  • Battle-tested – A combined Jest + legacy test suite (118+ assertions) covers the core behaviors and prevents regressions.

Table of contents

  1. Quick start
  2. Installation
  3. Core concepts
  4. Reactive models & events
  5. Detailed feature guide
  6. API highlights
  7. Usage patterns & examples
  8. Asynchronous coordination
  9. Project layout
  10. Testing & development
  11. Status & roadmap notes
  12. License

Quick start

// CommonJS (Node.js ≥ 15)
const lang = require('lang-mini');
const { each, mfp } = lang;

each([1, 2, 3], (value, index) => {
	console.log(index, value * 2);
});

const describe = mfp({ name: 'describe', return_type: 'string' }, {
	's': (str) => `string(${str.length})`
});

console.log(describe('hello'));
// ESM / bundlers (Vite, Webpack, etc.)
import lang from 'lang-mini';

const pointDistance = lang.distance_between_points([[0, 0], [3, 4]]);
console.log(pointDistance); // → 5
<!-- Browser (UMD-style) -->
<script src="/path/to/lang-mini.js"></script>
<script>
  const { each, eventify } = window.lang;
  const state = eventify({});
  each(['a', 'b'], (item) => console.log(item));
</script>

Installation

Lang-mini is published to npm as a CommonJS module with no production dependencies.

npm install lang-mini

Node / bundlers

  • Entry point: lib-lang-mini.js re-exports everything from lang-mini.js.
  • Works with Node.js ≥ 15 (per package.json engines field).
  • When using ESM, import the default export: import lang from 'lang-mini';.
  • Tree-shaking is limited because the build is a single module; destructure the functions you need.

Browser via script tag

  1. Copy lang-mini.js into your project or serve it from your build pipeline.
  2. Include it before your scripts. It attaches a lang global.
  3. For bundlers, treat it as a standard CommonJS module and rely on your bundler's CJS support.

Core concepts

Collection helpers

  • each(collection, fn, [context]) – Unified iterator for arrays and objects. Returns a new array/object of collected results. Supports early exit via a third stop callback parameter.
  • clone(value) – Shallow clone for primitives, arrays, and objects. (Deep cloning is out of scope.)
  • arr_like_to_arr(arrayLike) – Converts arguments, DOM collections, and other array-like objects into real arrays.
  • Truth-map utilities:
    • get_truth_map_from_arr(array) converts an array into { value: true } object.
    • get_arr_from_truth_map(map) converts it back (stringifying keys).
    • get_map_from_arr(array) indexes values to their latest position.
  • is_array(value) and is_defined(value) provide robust guards that match the library's type semantics.

Functional polymorphism

  • mfp(config, handlers) – Multi-function polymorphism. Dispatches on signature strings built from runtime argument types (e.g. 'n,n', 's,[n]'). Supports verbs, nouns, and grammar metadata for descriptive error messages.
  • fp(handlers) – Lightweight version for simple polymorphic dispatch.
  • vectorify(fn) – Lifts a binary numeric function so it works element-wise over vectors/arrays.
  • distance_between_points([[x1, y1], [x2, y2]]) – Example of a vector-aware helper included in the library.

Type detection & signatures

  • tof(value) – Enhanced typeof returning array, null, undefined, etc.
  • tf(value) – Abbreviated signatures ('n', 's', 'b', 'a', 'o', 'u', 'N'). Used internally by mfp.
  • deep_sig(value, [maxDepth]) – Generates deterministic structural signatures such as '[{"a":n},{"b":n}]'. Useful for memoization and grammar inference.

Runtime type system

  • Functional_Data_Type – Define runtime types with validate, optional parse_string, and metadata. Frequently combined with field for model validation.
  • Type_Signifier & Type_Representation – Experimental utilities (exported via lib-lang-mini.js) for distinguishing what a value is from how it is represented. Useful when modelling complex domains (colors, dates, binary blobs).

Events & data binding

  • Evented_Class – Minimal event emitter with .on(event, handler) and .raise(event, data).
  • eventify(obj) – Mixes event methods into plain objects so they can emit (raise) and listen (on) to events.
  • field(obj, name, [typeOrDefault], [defaultOrTransform]) – Declares data-bound properties on eventified objects. Supports:
    • default values (field(obj, 'age', 21))
    • validation via Functional_Data_Type
    • transformation functions (field(obj, 'upperName', value => value.toUpperCase()))
    • change notifications (obj.raise('change', { name, old, value })).
  • prop – For binding properties across objects (used alongside eventify).

⚠️ Requirement: Always pass an object through eventify() before calling field() or prop(). Otherwise a TypeError is thrown (raise is not a function).

Grammar-driven inference (WIP)

  • Grammar – Describes structured data with singular/plural forms and resolves runtime signatures.
  • grammar.tof(value) – Infers the matching definition name (e.g. 'user_login').
  • grammar.sig(value) – Resolves plural forms ('users', 'locations').
  • Some deeper methods in Grammar are still marked NYI; expect evolution in future releases.

Reactive models & events

Reactive behavior is built into the library through Evented_Class, eventify, field, prop, and Functional_Data_Type. The combination lets you turn plain objects into observable models, add typed fields, and hook change events without pulling in a heavier framework. Check docs/Reactive_Models.md for the full guide and usage examples.

Detailed feature guide

The helpers highlighted under “core concepts” are unpacked at docs/Core_Features.md. That guide includes short examples for the collection utilities, polymorphic dispatch helpers, signature detectors, runtime type layer, and grammar/combinator building blocks.

API highlights

Area Key exports Notes
Collections each, clone, arr_like_to_arr, get_truth_map_from_arr, get_map_from_arr, get_arr_from_truth_map Consistent semantics between arrays and objects; helpers return transformed copies.
Type detection tof, tf, deep_sig Signatures integrate with grammar and polymorphism features.
Functional utilities mfp, fp, vectorify, distance_between_points mfp supports default handlers ('default' key) and grammar-aware error messages.
Events Evented_Class, eventify Underpins field and prop; Evented_Class can be subclassed for richer components.
Data binding field, prop, Functional_Data_Type Compose validation, parsing, defaults, and transformations with change events.
Advanced types Type_Signifier, Type_Representation Experimental building blocks for rich type metadata.
Combinatorics combinations (alias: combos) Cartesian product across nested arrays; stops early if any sub-array is empty.
Async coordination call_multi, call_multiple_callback_functions, Fns, Publisher Mix callback ergonomics with promise-friendly utilities and readiness helpers.

Full source documentation lives inline within lang-mini.js; search for function names to see implementation notes and historical commentary.

Usage patterns & examples

Iterating with each

const { each } = require('lang-mini');

const labelled = each({ a: 1, b: 2 }, (value, key) => `${key}:${value}`);
// labelled → { a: 'a:1', b: 'b:2' }

const doubleUp = each([1, 2, 3], (value, index, stop) => {
	if (value === 3) stop();
	return value * 2;
});
// doubleUp → [2, 4]

Polymorphism with mfp

const { mfp } = require('lang-mini');

const sum = mfp({ name: 'sum' }, {
	'n,n': (a, b) => a + b,
	'[n]': (arr) => arr.reduce((acc, n) => acc + n, 0),
	'default': (_, signature) => {
		throw new TypeError(`sum has no handler for signature ${signature}`);
	}
});

sum(2, 3);        // → 5
sum([1, 2, 3, 4]); // → 10

Creating validated, reactive models

const lang = require('lang-mini');
const person = lang.eventify({});

const IntegerType = new lang.Functional_Data_Type({
	name: 'integer',
	validate: Number.isInteger,
	parse_string: (value) => {
		const parsed = Number(value);
		return Number.isInteger(parsed) ? parsed : undefined;
	}
});

lang.field(person, 'name', (value) => value.trim());
lang.field(person, 'age', IntegerType, 18);

person.on('change', ({ name, old, value }) => {
	console.log(`${name} changed from ${old} → ${value}`);
});

person.name = ' Ada ';
person.age = '21';
// Console:
// name changed from undefined → Ada
// age changed from 18 → 21

Working with combinations

const { combinations } = require('lang-mini');

const colourways = combinations([
	['red', 'green'],
	['S', 'M', 'L'],
	['cotton', 'linen']
]);

console.log(colourways.length); // 12

Grammar-powered inference

const { Grammar } = require('lang-mini');

const geoGrammar = new Grammar({
	name: 'Geo',
	def: {
		coordinate: { def: ['number', 'number'], plural: 'coordinates' },
		route: { def: ['coordinate', 'coordinate'], plural: 'routes' }
	}
});

geoGrammar.tof([51.5, -0.1]);                 // → 'coordinate'
geoGrammar.sig([[51.5, -0.1], [40.7, -74]]);   // → 'route'
geoGrammar.sig([ [ [0,0], [1,1] ], [ [2,2], [3,3] ] ]); // → 'routes'

Tip: Some advanced grammar APIs remain marked NYI; consult inline comments before relying on them in production.

Asynchronous coordination

Lang-mini bundles a tiny but expressive concurrency helper stack for orchestrating callback-heavy workflows without pulling in a promise library. The primitives mirror the asynchronous utilities used throughout the historical jsgui projects and are now fully documented in docs/Async_Coordination.md.

call_multi & call_multiple_callback_functions

  • Accepts an array of work items where each item can be a bare function, [fn, params], [context, fn], or [fn, params, perTaskCallback].
  • Optional flags support:
    • Parallelism: pass lang.call_multi(tasks, parallelism, done) to limit concurrency.
    • Delays: pass both parallelism and delay to stagger invocations when throttling external systems.
    • Parameter echoing: lang.call_multi(tasks, done, true) returns [[params], result] tuples so you can rebuild association tables after asynchronous fan-outs.
  • Errors short-circuit the execution and bubble through the final callback, making it easy to surface failures while still allowing per-task callbacks to run.
const lang = require('lang-mini');

const work = [
        (cb) => setTimeout(() => cb(null, 'A'), 5),
        [({ message }, cb) => cb(null, message.toUpperCase()), [{ message: 'b' }]],
        [
                (value, cb) => cb(null, value * 2),
                [21],
                (err, doubled) => console.log('per-task:', doubled)
        ]
];

lang.call_multi(work, 2, 1, (err, results) => {
        if (err) throw err;
        console.log(results); // → ['A', { message: 'B' }, 42]
});

Fns.go

Fns(arr) wraps an array of functions and exposes a .go() helper that forwards to call_multi, preserving the same argument permutations. This is particularly useful when dynamically building pipelines:

const steps = lang.Fns([
        (cb) => readFile('./input.json', cb),
        (contents, cb) => cb(null, JSON.parse(contents)),
        (data, cb) => cb(null, transform(data))
]);

steps.go((err, [buffer, parsed, transformed]) => {
        if (err) throw err;
        console.log(transformed);
});

Publisher

Publisher extends Evented_Class and adds a when_ready getter that returns a promise resolving once the instance raises ready. The helper caches readiness so subsequent calls resolve synchronously, letting you bridge older callback code with modern async/await:

const publisher = new lang.Publisher();

async function ensureReady() {
        await publisher.when_ready; // waits until publisher.raise('ready')
        console.log('ready to emit');
}

ensureReady();
setTimeout(() => publisher.raise('ready'), 20);

Project layout

lang-mini/
├─ lang-mini.js           # Single-file implementation (~2800 lines)
├─ lib-lang-mini.js       # CommonJS re-export for consumers
├─ examples/              # Runnable usage samples (Node scripts)
├─ docs/                  # Additional design notes (e.g., Control_Dom)
├─ tests/                 # Jest + legacy test suites
│  ├─ *.test.js           # Modern Jest suites
│  ├─ all-test.js         # Legacy runner
│  └─ new-tests.js        # Legacy runner extension
├─ AI-NOTES.md            # Development notes and open questions
├─ TEST-SUMMARY.md        # Snapshot of recent test runs
└─ package.json           # Metadata, scripts, dev dependencies (jest, fnl)

Testing & development

Run the full suite (Jest + legacy tests):

npm test

Run only the Jest suites:

npm run test:jest

Run legacy suites (useful when iterating on old tests):

npm run test:legacy

Contribution tips

  • Keep lang-mini.js changes focused; it's a shared dependency across multiple projects.
  • Document behavioral observations or known issues in AI-NOTES.md before making large refactors.
  • Add or update tests under tests/ and ensure npm test passes before submitting PRs.

Status & roadmap notes

  • Version 0.0.40 (see package.json).
  • No production dependencies; fnl and jest are dev-only.
  • Several Grammar and advanced type-system methods are intentionally marked NYI; contributions welcome but coordinate via AI-NOTES.md first.
  • Roadmap items such as richer type representations and additional documentation live in AI-README.md and roadmap.md.
  • Keep up with the current focus areas in docs/Whats_Next.md, which consolidates the remaining gaps and suggested next steps for lang-mini.

License

MIT © James Vickers / Metabench

About

No description, website, or topics provided.

Resources

Readme
MIT license
Activity

Stars

0 stars

Watchers

1 watching

Forks

0 forks
Report repository

Releases

Packages

Used by

Contributors

Languages