Found this at MDC but how if I'd wanted to add a private variable to the
var dataset = {
tables:{
customers:{
cols:[ /*here*/ ],
rows:[ /*here*/ ]
},
orders:{
cols:[ /*here*/ ],
rows:[ /*here*/ ]
}
},
relations:{
0:{
parent:'customers',
child:'orders',
keyparent:'custid',
keychild:'orderid',
onetomany:true
}
}
}
The way I understand OOP in Javascript, I'd have access to dataset.tables.customers.cols[0] if such an item exists.
But if I wanted to place a private variable into customers, what would that look like?
Adding var index = 0;
results in a runtime error.
Alternatively, we may also use the “this” keyword to make method (function) calls to stick to the main method itself which thus makes the variables private. The main idea for using the “this” keyword is just to make things directly visible that is making methods directly accessible.
You also can add closures as members, so create private fields which are method private rather than object private. dataset = { secretCounter: ( function () { var c = 0; return function () { return ++c; } })(), ... So dataset. secretCounter() has a varable c which is private to that function only.
In its current state, there is no “direct” way to create a private variable in JavaScript.
You can't have "private" variables without a function involved. Functions are the only way to introduce a new scope in javascript.
But never fear, you can add functions in the right place to gain this sort of functionality with your object
var dataset = {
tables: {
customers:(function(){
var privateVar = 'foo';
return {
cols:[ /*here*/ ],
rows:[ /*here*/ ]
}
}()),
orders:{
cols:[ /*here*/ ],
rows:[ /*here*/ ]
}
},
relations: [{
parent:'customers',
child:'orders',
keyparent:'custid',
keychild:'orderid',
onetomany:true
}]
};
But this doesn't gain us much. This is still mostly just a bunch of literal objects. These types of "Private" variables have zero meaning since there are no methods - nothing that would actually read or otherwise use the variables in the scope we created by adding a function (a closure).
But if we had a method, that might actually start to become useful.
var dataset = {
tables: {
customers:(function(){
var privateVar = 'foo';
return {
cols:[ /*here*/ ],
rows:[ /*here*/ ],
getPrivateVar: function()
{
return privateVar;
}
};
}()),
orders:{
cols:[ /*here*/ ],
rows:[ /*here*/ ]
}
},
relations: [{
parent:'customers',
child:'orders',
keyparent:'custid',
keychild:'orderid',
onetomany:true
}]
};
alert( dataset.tables.customers.getPrivateVar() );
JavaScript lacks the sort of access controls you get in more rigid languages. You can simulate private access for objects' data using closures, but your example is an object literal - a simple data structure - rather than a constructed object.
It rather depends what you want to do with the object - the normal technique for 'private' members means they are accessible only by member functions, and requires that you use a constructor to create the object. The literal syntax is used for data structures or light weight objects with public data and functions.
The problem with using the private closure pattern is that the fields within a literal are in public scope, but the privacy given by the closure is because the variable is defined in a function, so is scoped locally. You could either create a function which clones the literal and adds private fields, or add a public field which has private data. You also can add closures as members, so create private fields which are method private rather than object private.
dataset = {
secretCounter: (
function () {
var c = 0;
return function () { return ++c; }
})(),
...
So dataset.secretCounter()
has a varable c
which is private to that function only.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With