Home | Documentation | Core | CLI | Install |
Home > Documentation > Dependency Manager > Static Resources
- Config
- Controller
- Dependency Manager
- Dynamic Resources
- Properties
- Static Resources
- Model
- Service
- View
Dependency Manager:
Static Resources
Assets
$(':html') $(':css') $(':img') $(':js') | returns an object containing the methods .stream(filename) (under the hood this employs fs.createReadStream()) and .loadURL(fileName, fileExtension). The `fileExtension` argument is optional (use it with images). You can add more static assets to this list via the AssetConfig.js. |
Assets work in conjunction with their respective config. Please the config documentation for more information.
Async
$(':async') | Exposes await/async-friendly .each() and .reduce() methods. |
You will likely be familiar with .forEach
and .reduce
closures. Unfortunately, using await/async
in ecma 8+ is merely syntactic sugar for an underlying promise
structure. Therefore, if you use JavaScript’s traditional .forEach()
and .reduce()
methods on asynchronous calls you probably wont get the results you’re after. The async call resolves this issue as illustrated in the contrived example below.
Suppose you want to create a couple of Sample
objects; assigning each a distinct name:
const _async = $(':async'); // retrieve the async library
const Sample = $('Sample');
await _async.each(
['Foo','Bar'], // the array to iterate over
async (name) => { // use the async library's each method to iterate over array
await new Sample({ name: `My name is: ${name}` }).save();
});
perhaps you’d like to return a count as well, for this we could employ _async.reduce()
. The syntax should look familiar:
const _async = $(':async'); // retrieve the async library
const Sample = $('Sample');
const count = await _async.reduce(
['Foo','Bar'], // the array to iterate over
async (i, name) => { // use the async library's each method to iterate over array
await new Sample({ name: `My name is: ${name}` }).save();
return ++i; // i must be returned at the end of this function.
},
0 // the number to count assigned to 'i' in the function above.
);
$.logger.log(count == 2);
builder
$(':builder') | returns the co-koa-core's builder tool for iterating through files. This is an invaluable tool for handling custom database implementations in your plugins. See the [index.js](https://github.com/jaysaurus/co-koa-mongoose-plugin/blob/master/index.js) of the co-koa-mongoose-plugin to get a feel for how it works. |
The builder takes 2 arguments, the name of the type it is address and a callback for handling each of the files it is being asked to parse. For example:
const _builder = $(':builder');
_builder.build('Model', (model, modelName) => {
if (modelName === 'Foo') {
const file = model.doSomething();
...
}
});
it has been exposed with plugins in mind. for more information on plugins, please visit the plugins pages
Echo
$(':echo') | Exposes the echo-handler NPM module |
The echo call Will load messages based on the folder location specified in config/logger.js
(see the config documentation). See also the echo-handler for more information.
example
You have your default language set as ‘en’ in ./config/config.json
and a json file called en.test.json
in your ./i18n
folder with the following content:
{
"hello": "hello world!"
}
you can load the file and call this message wherever DependencyManager is exposed as below:
const _echo = $(':echo').load('test'); // load ./i18n/en.test.json
_echo.log('hello'); // log with logger.log()
_echo.error('hello'); // log with logger.error()
_echo.raw('hello'); // return the message as a string
_echo.throw('hello'); // throw an error with the supplied message
Using the default ./config/logger.js
configuration; running the call in a test environment, you can supply the log/error calls with an observer:
const _echo = $(':echo').load('test'); // load ./i18n/en.test.json
const observer = [];
_echo.log('hello', observer); // push hello message to observer
// using a test framework like jest, the below would pass:
expect(observer[0]).toBe('hello world!');
Most importantly, the chosen language can be decided at runtime! suppose we also had an es.test.json
as below:
{
"hello": "¡hola todo el mundo!"
}
We can call this as below:
const _echo = $(':echo').load('test', 'es');
_echo.log('hello'); // will log '¡hola todo el mundo!'
Enums
$(':enums') | returns `./api/Enums.js` |
Co.Koa extends mongoose’s core functionality with a numeric enum type courtesy of the ‘mongoose-type-number-enums’ npm package.
Example use:
Let’s suppose you’re coding an application for a book shop. The book shop only sells 3 formats of physical book: paperback, hardback and spiral-bound. We could (arguably) model this logic with an enum. JavaScript doesn’t natively support the Enum type, but we could simulate something similar by adding such an enum to Co.Koa’s ./api/Enums.js
file:
module.exports = {
BookFormats: ['PAPERBACK', 'HARDBACK', 'SPIRAL_BOUND'];
}
mongoose supports an enum type by default, but stores that enum as a string in the database. However, that doesn’t really fall in-line with the “C-like” conventions we might expect. Co.Koa supports mongoose’s default enums, but encourages the definition of numeric integers as below:
module.exports = function Book ($) {
const _enums = $(':enums');
return {
schema: {
format: {
type: 'Enum',
enum: _enums.BookFormats
}
...
}
...
}
}
Subsequently, you can perform requests against that type as below:
const _enums = $(':enums');
await new Book({
format: _enums.BookFormats.indexOf('HARDBACK'),
...
}).save();
The database will store the index number of ‘HARDBACK’ (1) as type “number”
Tree
$(':tree') | returns a stack-tree algorithm that can be used to iterate through object trees. |
This is a powerful tool used by co-koa-core that has been exposed for your convenience. It is used to map mongoose ObjectIds to the place-holder types: “ForeignKey”, “FK” and “ObjectId” defined by clients in their models. You can see it in action on the ModelFactoryHelper
.
suppose you have a complex object you wish to analyse:
const dataset = {
a: {
a1: 'a1',
a2: 'a2',
a3: {
a3_1: {
a3_1_1: {
a3_1_1_1: 'a3_1_1_1'
}
}
}
},
b: {
b1: {
b1_1: 'b1_1',
b1_2: 'b1_2',
b1_3: {
b1_3_1: 'b1_3_1',
b1_3_2: {
b_deep: { deeper: { deepest: 'Deep!' } }
}
},
b1_4: 'b1_4'
}
},
c: 'c'
};
As of co-koa-core
@^1.16.0, The following code easily digests the above:
const keyObserver = [];
const nodeObserver = {};
let outObserver = null;
const _tree = $(':tree');
_tree(dataset)
.process((it) => {
// args:
// it._keyTree: an array of the path to the current node;
// it._out: a pointer to the observed object.
// it.item: the current node's value
// it.key: the current node's key
keyObserver.push(it._keyTree.reduce((list, i) => {
list.push(i);
return list;
}, []));
if (!outObserver) outObserver = it._out
nodeObserver[it.key] = it.item;
});
console.log(keyObserver[0]) // [ 'a' ]
console.log(keyObserver[1]) // [ 'a' ]
console.log(keyObserver[2]) // [ 'a', 'a3', 'a3_1', 'a3_1_1' ]
console.log(keyObserver[3]) // [ 'b', 'b1' ]
console.log(keyObserver[4]) // [ 'b', 'b1' ]
console.log(keyObserver[5]) // [ 'b', 'b1', 'b1_3' ]
console.log(keyObserver[6]) // [ 'b', 'b1', 'b1_3', 'b1_3_2', 'b_deep', 'deeper' ]
console.log(keyObserver[7]) // [ 'b', 'b1' ]
console.log(outObserver === dataset) // true
console.log(nodeObserver) // would render an object as below:
{ a1: 'a1',
a2: 'a2',
a3_1_1_1: 'a3_1_1_1',
b1_1: 'b1_1',
b1_2: 'b1_2',
b1_3_1: 'b1_3_1',
deepest: 'Deep!',
b1_4: 'b1_4',
c: 'c' }