I’m not planning to contribute to this project for a while.
If you are interested in being a contributor, please let me know.
Thanks,
Volkan.
______ ______ _____ ______ __ __
/\ == \ /\ __ \ /\ __-. /\ ___\ /\ "-./ \
\ \ __< \ \ __ \ \ \ \/\ \\ \ __\ \ \ \-./\ \
\ \_____\ \ \_\ \_\ \ \____- \ \_____\ \ \_\ \ \_\
\/_____/ \/_/\/_/ \/____/ \/_____/ \/_/ \/_/
badem
is an abstract component composer that does not require a DOM.
badem
is in its early stages; so anything in its implementation can change.Until it hits
version 1.0
, I’ll be liberally introducing breaking changes, please keep that in mind and fix your versions in your package.json if you depend onbadem
in your apps.Once
badem
hitsversion 1.0
, its API will stabilize, and the changes will be more backwards-compatible, as I will follow the Semantic Versioning 2.0.0 standards.
badem
started as a thought experiment:
“What if you have to create components in an extremely-restricted environment where there is no DOM or
window
?”
In that kind of an environment, you have to create a higher-order mounter, that just creates a tree-like hierarchy and delegates the actual instantiation and initialization of the components “elsewhere” [1].
[1]: “elsewhere” can be anything: A custom TV operating system, an IOC, a native application, a “Façade” to actual real-life components, like a robot’s arm.
Currently, I’m experimenting how to integrate badem
to smartface.io, so that you can render a smartface.io UI just by tracing a JSON descriptor.
Defining how the UI will render and behave declaratively in a JSON file opens on many possibilities.
One of these possibilities is the ability to integrate a custom WYSIWYG designer to the code:
So the designer will generate a JSON instead of a large code-behind autogenerated code.
This approach has several advantages:
- The footprint of the generated JSON is much smaller.
- JSON is much declarative than your standard code-behind IDE code: MUCH easier to read and understand.
- JSON is smaller in size, easier, and faster to generate.
- You can serialize/deserialize a JSON, and it will reflect the same state across the boundaries of your system; you cannot do the same with an IDE-generated code, so parts of it might get “lost in translation” — Code is not serializable; data is.
- The typical designer output is generally some form of an XML or JSON already; so it is much easier to convert it to a
badem
-compatible JSON. - The generation process is be less error-prone (see the “serialization” point above) — There are more edge cases when you generate actual source code: you can create invalid, or incomplete source code, you can have syntax errors, or logic errors that are hard to see)
- Since
badem
JSON is an almost identical representation of the UI, it is a lot easier to spot out bugs and regressions. - Related to the above topic: Testing JSON is much easier than testing code: Once you create a proper schema-validator, you are almost there — no need to stub, or mock anything, because everything is declarative.
This declarative programming is User Interface as Code, which is (much better and) diagonally different from a “user-interface-generated code”.
Install using NPM:
npm install badem --save
There are different badem
distributions for various environments:
- Use
require( 'badem/dist/nodejs' )
in a Node.JS environment. - Use
require( 'badem/dist/js' )
to be used in a JavaScript (ES6+) environment.
I will add more distribution formats later.
Here’s how you mount components with badem
:
const log = console.log;
// Using the Node.JS distribution:
const mount = require( 'badem/dist/nodejs' ).mount;
const store = require( './store' );
const app = require( './app.json' );
mount( store, app )
.then( () => log( 'All should have been done by now!' ) )
.catch( () => log( 'Oh poop!' ) );
See the ./examples
folders for additional details and usage examples.
For development, you’ll need a recent version of Node.JS and NPM.
You will need the current version of Node.JS with all the bells and whistles — You can install it from nodejs.org.
Here are the helper npm scripts that you can run via npm
:
npm test
: Executes the unit tests.npm run lint
: Checks whether the source JavaScript is well-formed.npm run build:node
: Generate a Node.JS distribution inside the./dist
folder.npm run build:js
: Generate a JavaScript (ES6+) distribution inside the./dist
folder.npm run build
: Executes all of thebuild:*
actions.
src
: The source files live here.dist
: Once you run the build script, distribution bundles will be placed here.bin
: Scripts that are mostly used bynpm
.CHANGELOG.md
: A log of what has been done since the last version.CODE_OF_CONDUCT.md
: Tells the collaborators to be nice to each other.README.md
: This very file..babelrc
: Used for development; configuresbabel
..eslintrc
: Used for development; configureseslint
..travis.yml
: Used for CI; configures Travis.webpack.node.config.js
: Helpswebpack
to bundle a Node.JS distribution inside the./dist
folder.
Any help is more than appreciated.
If you want to contribute to the source code, fork this repository and create a pull request.
In lieu of a formal style guide, take care to maintain the existing coding style.
Also, don’t forget to add unit tests for any new or changed functionality.
If you want to report a bug; or share a comment or suggestion, file an issue.
For bug reports and suggestions, please file an issue.
- Project Maintainer: Volkan Özçelik
- Project Website: bytesized.tv
MIT-licensed. — See the license file for details.
We are committed to making participation in this project a harassment-free experience for everyone, regardless of the level of experience, gender, gender identity and expression, sexual orientation, disability, personal appearance, body size, race, ethnicity, age, religion, or nationality.
See the code of conduct for details.
A ByteSized.TV Project
This repository is a part of the Byte-Sized JavaScript VideoCasts.
It is a compilation of short (around ten minutes) screencasts about JavaScript and related technologies.