Little knowledge , Great challenge ！ This article is participating in “ A programmer must have a little knowledge ” Creative activities .

Once I was a front-end student, I finally took the postgraduate entrance examination , I thought I got rid of it js, result python There is also an iterator generator , Out of line .

iterator

Usually we use for The cycle is smooth and easy, isn't it . such as

What's the principle ？

for  Statement is called on the container object  iter(). This function returns a definition  __next__()  Iterator object for method , This method will access the elements in the container one by one . When the elements are exhausted ,__next__()  Will lead to  StopIteration  Exception to notify termination  for  loop . You can use  next()  Built in function to call  __next__()  Method .

Below we see iterator Yes, it is a The list generates an iterator , Every time you call next Will take one a The elements in .

When you write next When the number of elements is exceeded, there will be StopIteration Abnormal stop .

generator

With a yield The function of is a generator, It's different from ordinary functions , Generate a generator It looks like a function call , But no function code is executed , Until you call next()（ stay for It's called automatically in the loop next()） Just started to execute .

Because I don't learn python Someone who looks directly at the code , I met generator today .

I am here indices There's a whole print(indices), Then call the function in a decent way data_iter(...), The results cannot be printed no matter how they are run or debugged indices, Until I see yield I don't feel good about it ……

Although the execution process still follows the process of the function , But every time it gets to one yield The statement breaks , And return an iteration value , The next execution is from yield Continue with the next statement of . It looks like a function is being yield Several interruptions , Every interrupt passes yield Returns the current iteration value .

yield The benefits are obvious , Rewrite a function to a generator And you get the ability to iterate , Instead of saving the state of a class instance to calculate the next next() Value , Not only is the code concise , And the execution process is extremely clear .

At the end of function execution ,generator Automatically throw out StopIteration abnormal , Indicates that the iteration is complete . stay for In circulation , No need to deal with StopIteration abnormal , The cycle will end normally .

Like a function call , But it's generating a generator ：

yield Action image of return equally , Use next Call generator , Every time the generator executes to yield Throw a return value to stop execution , Until next time next Since the last time yield Continue where you stop , Throw until the iteration object runs out StopIteration Abnormal termination .

Add two small knowledge

1. How to judge whether a function is generator？ If your eyes are OK, you should be able to see yield Of . If you're looking for a way to be tall , That can be used ：

   from inspect import isgeneratorfunction 
   isgeneratorfunction( Function name ) 
    Copy code 

   If so, the generator will return true

2. yield image return, that generator You can have return Do you ？

In a generator function in , without return, The default execution is until the function is finished , If in the course of execution return, And just throw it out StopIteration Termination iteration .