简介

Generator 函数是 ES6 提供的一种异步编程解决方案,语法行为与传统函数完全不同。Python的协程也是用Generator函数这种方案来实现的。

以前的异步编程方案:

  • 回调函数
  • 事件监听
  • 发布/订阅
  • Promise 对象

Generator 函数有多种理解角度。语法上,首先可以把它理解成,Generator 函数是一个状态机,封装了多个内部状态。

执行 Generator 函数会返回一个遍历器对象,也就是说,Generator 函数除了状态机,还是一个遍历器对象生成函数。返回的遍历器对象,可以依次遍历 Generator 函数内部的每一个状态。

形式上,Generator 函数是一个普通函数,但是有两个特征。一是,function关键字与函数名之间有一个星号;二是,函数体内部使用yield表达式,定义不同的内部状态(yield在英语里的意思就是“产出”)。

1
2
3
4
5
6
7
function* helloWorldGenerator() {
yield 'hello';
yield 'world';
return 'ending';
}

var hw = helloWorldGenerator();

上面代码定义了一个 Generator 函数helloWorldGenerator,它内部有两个yield表达式(helloworld),即该函数有三个状态:helloworldreturn 语句(结束执行)。

然后,Generator 函数的调用方法与普通函数一样,也是在函数名后面加上一对圆括号。不同的是,调用 Generator 函数后,该函数并不执行,返回的也不是函数运行结果,而是一个指向内部状态的指针对象,也就是遍历器对象(Iterator Object)。

下一步,必须调用遍历器对象的next方法,使得指针移向下一个状态。也就是说,每次调用next方法,内部指针就从函数头部或上一次停下来的地方开始执行,直到遇到下一个yield表达式(或return语句)为止。换言之,Generator 函数是分段执行的,yield表达式是暂停执行的标记,而next方法可以恢复执行。

1
2
3
4
5
6
7
8
9
10
11
hw.next()
// { value: 'hello', done: false }

hw.next()
// { value: 'world', done: false }

hw.next()
// { value: 'ending', done: true }

hw.next()
// { value: undefined, done: true }

上面代码一共调用了四次next方法。

第一次调用,Generator 函数开始执行,直到遇到第一个yield表达式为止。next方法返回一个对象,它的value属性就是当前yield表达式的值hellodone属性的值false,表示遍历还没有结束。

第二次调用,Generator 函数从上次yield表达式停下的地方,一直执行到下一个yield表达式。next方法返回的对象的value属性就是当前yield表达式的值worlddone属性的值false,表示遍历还没有结束。

第三次调用,Generator 函数从上次yield表达式停下的地方,一直执行到return语句(如果没有return语句,就执行到函数结束)。next方法返回的对象的value属性,就是紧跟在return语句后面的表达式的值(如果没有return语句,则value属性的值为undefined),done属性的值true,表示遍历已经结束。

第四次调用,此时 Generator 函数已经运行完毕,next方法返回对象的value属性为undefineddone属性为true。以后再调用next方法,返回的都是这个值。

总结一下,调用 Generator 函数,返回一个遍历器对象,代表 Generator 函数的内部指针。以后,每次调用遍历器对象的next方法,就会返回一个有着valuedone两个属性的对象。value属性表示当前的内部状态的值,是yield表达式后面那个表达式的值;done属性是一个布尔值,表示是否遍历结束。

ES6 没有规定,function关键字与函数名之间的星号,写在哪个位置。这导致下面的写法都能通过

1
2
3
4
function * foo(x, y) { ··· }
function *foo(x, y) { ··· }
function* foo(x, y) { ··· }
function*foo(x, y) { ··· }

由于 Generator 函数仍然是普通函数,所以一般的写法是上面的第三种,即星号紧跟在function关键字后面。

yield 表达式

由于 Generator 函数返回的遍历器对象,只有调用next方法才会遍历下一个内部状态,所以其实提供了一种可以暂停执行的函数。yield表达式就是暂停标志。

遍历器对象的next方法的运行逻辑如下。

(1)遇到yield表达式,就暂停执行后面的操作,并将紧跟在yield后面的那个表达式的值,作为返回的对象的value属性值。

(2)下一次调用next方法时,再继续往下执行,直到遇到下一个yield表达式。

(3)如果没有再遇到新的yield表达式,就一直运行到函数结束,直到return语句为止,并将return语句后面的表达式的值,作为返回的对象的value属性值。

(4)如果该函数没有return语句,则返回的对象的value属性值为undefined

需要注意的是,yield表达式后面的表达式,只有当调用next方法、内部指针指向该语句时才会执行,因此等于为 JavaScript 提供了手动的“惰性求值”(Lazy Evaluation)的语法功能。

1
2
3
function* gen() {
yield 123 + 456;
}

上面代码中,yield后面的表达式123 + 456,不会立即求值,只会在next方法将指针移到这一句时,才会求值。

yield表达式与return语句既有相似之处,也有区别。相似之处在于,都能返回紧跟在语句后面的那个表达式的值。区别在于每次遇到yield,函数暂停执行,下一次再从该位置继续向后执行,而return语句不具备位置记忆的功能。一个函数里面,只能执行一次(或者说一个)return语句,但是可以执行多次(或者说多个)yield表达式。正常函数只能返回一个值,因为只能执行一次return;Generator 函数可以返回一系列的值,因为可以有任意多个yield。从另一个角度看,也可以说 Generator 生成了一系列的值,这也就是它的名称的来历(英语中,generator 这个词是“生成器”的意思)。

Generator 函数可以不用yield表达式,这时就变成了一个单纯的暂缓执行函数。

1
2
3
4
5
6
7
8
9
10
11
function* f() {
console.log('Generator函数执行了');
}

var generator = f();
//不会输出‘Generator函数执行了’
function g(){
console.log('函数g()执行了');
}
var gen = g();
//输出‘函数g()执行了’

上面代码中,函数 g 是普通函数,在为变量gen赋值时就会执行。但是,函数 f 是一个 Generator 函数,就变成只有调用next方法时,函数 f 才会执行。

另外需要注意,yield表达式只能用在 Generator 函数里面,用在其他地方都会报错。

另外,yield表达式如果用在另一个表达式之中,必须放在圆括号里面。

1
2
3
4
5
6
7
function* demo() {
console.log('Hello' + yield); // SyntaxError
console.log('Hello' + yield 123); // SyntaxError

console.log('Hello' + (yield)); // OK
console.log('Hello' + (yield 123)); // OK
}

yield表达式用作函数参数或放在赋值表达式的右边,可以不加括号。

1
2
3
4
function* demo() {
foo(yield 'a', yield 'b'); // OK
let input = yield; // OK
}

next 方法的参数

yield表达式本身没有返回值,或者说总是返回undefinednext方法可以带一个参数,该参数就会被当作上一个yield表达式的返回值。

1
2
3
4
5
6
7
8
9
10
11
12
function* f() {
for(var i = 0; true; i++) {
var reset = yield i;
if(reset) { i = -1; }
}
}

var g = f();

g.next() // { value: 0, done: false }
g.next() // { value: 1, done: false }
g.next(true) // { value: 0, done: false }

上面代码先定义了一个可以无限运行的 Generator 函数f,如果next方法没有参数,每次运行到yield表达式,变量reset的值总是undefined。当next方法带一个参数true时,变量reset就被重置为这个参数(即true),因此i会等于-1,下一轮循环就会从-1开始递增。

这个功能有很重要的语法意义。Generator 函数从暂停状态到恢复运行,它的上下文状态(context)是不变的。通过next方法的参数,就有办法在 Generator 函数开始运行之后,继续向函数体内部注入值。也就是说,可以在 Generator 函数运行的不同阶段,从外部向内部注入不同的值,从而调整函数行为。

注意,由于next方法的参数表示上一个yield表达式的返回值,所以在第一次使用next方法时,传递参数是无效的。V8 引擎直接忽略第一次使用next方法时的参数,只有从第二次使用next方法开始,参数才是有效的。从语义上讲,第一个next方法用来启动遍历器对象,所以不用带有参数。

如果想要第一次调用next方法时,就能够输入值,可以在 Generator 函数外面再包一层。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
function wrapper(generatorFunction) {
return function (...args) {
let generatorObject = generatorFunction(...args);
generatorObject.next();
return generatorObject;
};
}

const wrapped = wrapper(function* () {
console.log(`First input: ${yield}`);
return 'DONE';
});

wrapped().next('hello!')
// First input: hello!

Generator.prototype.return()

Generator 函数返回的遍历器对象,还有一个return方法,可以返回给定的值,并且终结遍历 Generator 函数。

1
2
3
4
5
6
7
8
9
10
11
function* gen() {
yield 1;
yield 2;
yield 3;
}

var g = gen();

g.next() // { value: 1, done: false }
g.return('foo') // { value: "foo", done: true }
g.next() // { value: undefined, done: true }

上面代码中,遍历器对象g调用return方法后,返回值的value属性就是return方法的参数foo。并且,Generator 函数的遍历就终止了,返回值的done属性为true,以后再调用next方法,done属性总是返回true

如果return方法调用时,不提供参数,则返回值的value属性为undefined

如果 Generator 函数内部有try...finally代码块,那么return方法会推迟到finally代码块执行完再执行。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
function* numbers () {
yield 1;
try {
yield 2;
yield 3;
} finally {
yield 4;
yield 5;
}
yield 6;
}
var g = numbers();
g.next() // { value: 1, done: false }
g.next() // { value: 2, done: false }
g.return(7) // { value: 4, done: false }
g.next() // { value: 5, done: false }
g.next() // { value: 7, done: true }

上面代码中,调用return方法后,就开始执行finally代码块,然后等到finally代码块执行完,再执行return方法。

Generator.prototype.throw()

Generator 函数返回的遍历器对象,都有一个throw方法,可以在函数体外抛出错误,然后在 Generator 函数体内捕获。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
var g = function* () {
try {
yield;
} catch (e) {
console.log('内部捕获', e);
}
};

var i = g();
i.next();

try {
i.throw('a');
i.throw('b');
} catch (e) {
console.log('外部捕获', e);
}
// 内部捕获 a
// 外部捕获 b

上面代码中,遍历器对象i连续抛出两个错误。第一个错误被 Generator 函数体内的catch语句捕获。i第二次抛出错误,由于 Generator 函数内部的catch语句已经执行过了,不会再捕捉到这个错误了,所以这个错误就被抛出了 Generator 函数体,被函数体外的catch语句捕获。

throw方法可以接受一个参数,该参数会被catch语句接收,建议抛出Error对象的实例。

1
2
3
4
5
6
7
8
9
10
11
12
var g = function* () {
try {
yield;
} catch (e) {
console.log(e);
}
};

var i = g();
i.next();
i.throw(new Error('出错了!'));
// Error: 出错了!(…)

注意,不要混淆遍历器对象的throw方法和全局的throw命令。上面代码的错误,是用遍历器对象的throw方法抛出的,而不是用throw命令抛出的。后者只能被函数体外的catch语句捕获。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
var g = function* () {
while (true) {
try {
yield;
} catch (e) {
if (e != 'a') throw e;
console.log('内部捕获', e);
}
}
};

var i = g();
i.next();

try {
throw new Error('a');
throw new Error('b');
} catch (e) {
console.log('外部捕获', e);
}
// 外部捕获 [Error: a]

上面代码之所以只捕获了a,是因为函数体外的catch语句块,捕获了抛出的a错误以后,就不会再继续try代码块里面剩余的语句了。

如果 Generator 函数内部没有部署try...catch代码块,那么throw方法抛出的错误,将被外部try...catch代码块捕获。

如果 Generator 函数内部和外部,都没有部署try...catch代码块,那么程序将报错,直接中断执行。

throw方法被捕获以后,会附带执行下一条yield表达式。也就是说,会附带执行一次next方法。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
var gen = function* gen(){
try {
yield console.log('a');
} catch (e) {
// ...
}
yield console.log('b');
yield console.log('c');
}

var g = gen();
g.next() // a
g.throw() // b
g.next() // c

上面代码中,g.throw方法被捕获以后,自动执行了一次next方法,所以会打印b。另外,也可以看到,只要 Generator 函数内部部署了try...catch代码块,那么遍历器的throw方法抛出的错误,不影响下一次遍历。

这种函数体内捕获错误的机制,大大方便了对错误的处理。多个yield表达式,可以只用一个try...catch代码块来捕获错误。如果使用回调函数的写法,想要捕获多个错误,就不得不为每个函数内部写一个错误处理语句,现在只在 Generator 函数内部写一次catch语句就可以了。

一旦 Generator 执行过程中抛出错误,且没有被内部捕获,就不会再执行下去了。如果此后还调用next方法,将返回一个value属性等于undefineddone属性等于true的对象,即 JavaScript 引擎认为这个 Generator 已经运行结束了。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
function* g() {
yield 1;
console.log('throwing an exception');
throw new Error('generator broke!');
yield 2;
yield 3;
}

function log(generator) {
var v;
console.log('starting generator');
try {
v = generator.next();
console.log('第一次运行next方法', v);
} catch (err) {
console.log('捕捉错误', v);
}
try {
v = generator.next();
console.log('第二次运行next方法', v);
} catch (err) {
console.log('捕捉错误', v);
}
try {
v = generator.next();
console.log('第三次运行next方法', v);
} catch (err) {
console.log('捕捉错误', v);
}
console.log('caller done');
}

log(g());
// starting generator
// 第一次运行next方法 { value: 1, done: false }
// throwing an exception
// 捕捉错误 { value: 1, done: false }
// 第三次运行next方法 { value: undefined, done: true }
// caller done

上面代码一共三次运行next方法,第二次运行的时候会抛出错误,然后第三次运行的时候,Generator 函数就已经结束了,不再执行下去了。

next()、throw()、return() 的共同点

next()throw()return()这三个方法本质上是同一件事,可以放在一起理解。它们的作用都是让 Generator 函数恢复执行,并且使用不同的语句替换yield表达式。

next()是将yield表达式替换成一个值。

1
2
3
4
5
6
7
8
9
10
11
const g = function* (x, y) {
let result = yield x + y;
return result;
};

const gen = g(1, 2);
gen.next(); // Object {value: 3, done: false}

gen.next(1); // Object {value: 1, done: true}
// 相当于将 let result = yield x + y
// 替换成 let result = 1;

上面代码中,第二个next(1)方法就相当于将yield表达式替换成一个值1。如果next方法没有参数,就相当于替换成undefined

throw()是将yield表达式替换成一个throw语句。

1
2
3
gen.throw(new Error('出错了')); // Uncaught Error: 出错了
// 相当于将 let result = yield x + y
// 替换成 let result = throw(new Error('出错了'));

return()是将yield表达式替换成一个return语句。
gen.return(2); // Object {value: 2, done: true}
// 相当于将 let result = yield x + y
// 替换成 let result = return 2;

一些注意的点

如果Generator函数已经return了,那么Generator函数就不会再执行下去了

1
2
3
4
5
6
7
8
9
function* generator(){
yield 1;
return 2;
yield 3;
}
var gen = generator();
gen.next();//{value: 1, done: false}
gen.next();//{value: 2, done: true}
gen.next();//{value: undefined, done: true}

上述代码,在第二次gen.next()执行之后,Generator函数就已经return,代表结束了,再进行下次gen.next()就不会输出3

如果yield的是一个函数,那这个函数会在value值里面,要调用只能通过value调用

1
2
3
4
5
6
7
function* generator(){
yield function f(){
console.log('f is called');
}
}
var gen = generator();
gen.next().value();//f is called

参考:阮一峰《ECMAScript 6 入门》