This time I will bring you a detailed explanation of the use of js data types, what are the precautions when using js data types, the following is a practical case, let’s take a look .
Because I am a wild programmer, I did not pay attention to the basic knowledge of memory when I first started learning programming. As a result, I later mentioned "what is stored in the stack, there is only one ## stored in the stack." #quote" I always look confused when saying this. .
Later, I gradually learned some knowledge about memory, which is still very necessary to understand.Basic data structure
stackStack, a linear list that only allows insertion or deletion operations in a section, is a first-in, last-out data structure.Heap
The heap is a data structure based on a hashing algorithm.Queue
A queue is a first-in-first-out (FIFO) data structure.
Storage of data types in JavaScript
Data types in JavaScript are divided into basic data types and reference data types. One difference between them is the different storage locations.Basic data types
We all know that the basic data types in JavaScript are:##Null
Symbol (Ignore it for now)
Reference data type
Reference data types in JavaScript are:
ArrayIllustration
Now, let’s try declaring a few variables:
var name="axuebin"; var age=25; var job; var arr=[1,2,3]; var obj={age:25};
name
,age, and
job are directly stored in the stack memory, while arr and obj are only stored in the stack memory. Address represents a reference to heap memory.
copy
Basic data types
For basic data types, if copied, the system will automatically allocate a new value in the stack memory for the new variable, which is easy to understand.
Reference data type
For reference data types such as arrays and objects, there will be differences when copying:
The system will also automatically assign a value in stack memory for the new variable, but this value is just an address. In other words, the copied variable has the same address value as the original variable and points to the same object in the heap memory. If shown, after executing var objCopy=obj, obj and objCopy have the same address value and execute the same actual object in the heap memory. What difference does this make? When I modify obj or objCopy, it will cause another variable to change.Why?
Why do basic data types exist on the stack and reference data types exist on the heap?
The heap is larger than the stack, and the stack comparison is faster.The basic data type is relatively stable and takes up relatively little memory.
The reference data type size is dynamic and unlimited.
参考文章
理解js内存分配
原始值和引用值
在ECMAScript中,变量可以存放两种类型的值,即原始值和引用值。
原始值指的就是代表原始数据类型(基本数据类型)的值,即Undefined,Null,Number,String,Boolean类型所表示的值。
引用值指的就是复合数据类型的值,即Object,Function,Array,以及自定义对象,等等
栈和堆
与原始值与引用值对应存在两种结构的内存即栈和堆
栈是一种后进先出的数据结构,在javascript中可以通过Array来模拟栈的行为
原始值是存储在栈中的简单数据,也就是说,他们的值直接存储在变量访问的位置。
堆是基于散列算法的数据结构,在javascript中,引用值是存放在堆中的。
引用值是存储在堆中的对象,也就是说,存储在变量处的值(即指向对象的变量,存储在栈中)是一个指针,指向存储在堆中的实际对象.
例:var obj = new Object(); obj存储在栈中它指向于new Object()这个对象,而new Object()是存放在堆中的。
那为什么引用值要放在堆中,而原始值要放在栈中,不都是在内存中吗,为什么不放在一起呢?那接下来,让我们来探索问题的答案!
首先,我们来看一下代码:
function Person(id,name,age){ this.id = id; this.name = name; this.age = age; } var num = 10; var bol = true; var str = "abc"; var obj = new Object(); var arr = ['a','b','c']; var person = new Person(100,"笨蛋的座右铭",25);
变量num,bol,str为基本数据类型,它们的值,直接存放在栈中,obj,person,arr为复合数据类型,他们的引用变量存储在栈中,指向于存储在堆中的实际对象。
由上图可知,我们无法直接操纵堆中的数据,也就是说我们无法直接操纵对象,但我们可以通过栈中对对象的引用来操作对象,就像我们通过遥控机操作电视机一样,区别在于这个电视机本身并没有控制按钮。
现在让我们来回答为什么引用值要放在堆中,而原始值要放在栈中的问题:
记住一句话:能量是守衡的,无非是时间换空间,空间换时间的问题
堆比栈大,栈比堆的运算速度快,对象是一个复杂的结构,并且可以自由扩展,如:数组可以无限扩充,对象可以自由添加属性。将他们放在堆中是为了不影响栈的效率。而是通过引用的方式查找到堆中的实际对象再进行操作。相对于简单数据类型而言,简单数据类型就比较稳定,并且它只占据很小的内存。不将简单数据类型放在堆是因为通过引用到堆中查找实际对象是要花费时间的,而这个综合成本远大于直接从栈中取得实际值的成本。所以简单数据类型的值直接存放在栈中。
相信看了本文案例你已经掌握了方法,更多精彩请关注php中文网其它相关文章!
推荐阅读:
The above is the detailed content of Detailed explanation of the use of js data types. For more information, please follow other related articles on the PHP Chinese website!