Likeanonymous methods,
iterators in C# are very complex syntactic sugar.
You could do it all yourself (after all, you did have to do
it all yourself in earlier versions of C#),
but the compiler transformation makes for much greater convenience.

The idea behind iterators is that they take a function withyield return
statements
(and possible some yield break statements)
and convert it into a state machine.
When you yield return, the state of the function is
recorded, and execution resumes from that state the next time the
iterator is called upon to produce another object.

Here’s the basic idea:
All the local variables of the iterator (treating iterator parameters
as pre-initialized local variables, including the hidden this
parameter)
become member variables of a helper class.
The helper class also has an internal state member that keeps
track of where execution left off and an internal current
member that holds the object most recently enumerated.

class MyClass {

 int limit = ;

 public MyClass(int limit) { this.limit = limit; }

 public IEnumerable<int> CountFrom(int start)

 {

  for (int i = start; i <= limit; i++) {

   yield return i;

  }

 }

}

The CountFrom method produces an integer
enumerator that spits out the integers starting at start
and continuing up to and including limit.
The compiler internally converts this enumerator into
something like this:

class MyClass_Enumerator : IEnumerable<int> {

  int state$ = ;// internal member

  int current$;  // internal member

  MyClass this$; // implicit parameter to CountFrom

  int start;      // explicit parameter to CountFrom

  int i;          // local variable of CountFrom

  public int Current {

   get { return current$; }

  }

  public bool MoveNext()

  {

   switch (state$) {

   case : goto resume$;

   case : goto resume$;

   case : return false;

   }

 resume$:;

   for (i = start; i <= this$.limit; i++) {

    current$ = i;

    state$ = ;

    return true;

 resume$:;

   }

   state$ = ;

   return false;

  }

  … other bookkeeping, not important here …

 }

 public IEnumerable<int> CountFrom(int start)

 {

  MyClass_Enumerator e = new MyClass_Enumerator();

  e.this$ = this;

  e.start = start;

  return e;

 }

用dnSpy反编译上面的代码,同时在配置中

得到如下代码,是一个状态机

    // Token: 0x02000005 RID: 5
internal class MyClass
{
// Token: 0x06000006 RID: 6 RVA: 0x000020C9 File Offset: 0x000002C9
public MyClass(int limit)
{
this.limit = limit;
} // Token: 0x06000007 RID: 7 RVA: 0x000020E1 File Offset: 0x000002E1
public IEnumerable<int> CountFrom(int start)
{
MyClass.<CountFrom>d__2 <CountFrom>d__ = new MyClass.<CountFrom>d__2(-);
<CountFrom>d__.<>4__this = this;
<CountFrom>d__.<>3__start = start;
return <CountFrom>d__;
} // Token: 0x04000001 RID: 1
private int limit = ; // Token: 0x02000006 RID: 6
[CompilerGenerated]
private sealed class <CountFrom>d__2 : IEnumerable<int>, IEnumerable, IEnumerator<int>, IDisposable, IEnumerator
{
// Token: 0x06000008 RID: 8 RVA: 0x000020F8 File Offset: 0x000002F8
[DebuggerHidden]
public <CountFrom>d__2(int <>1__state)
{
this.<>1__state = <>1__state;
this.<>l__initialThreadId = Environment.CurrentManagedThreadId;
} // Token: 0x06000009 RID: 9 RVA: 0x00002113 File Offset: 0x00000313
[DebuggerHidden]
void IDisposable.Dispose()
{
} // Token: 0x0600000A RID: 10 RVA: 0x00002118 File Offset: 0x00000318
bool IEnumerator.MoveNext()
{
int num = this.<>1__state;
if (num != )
{
if (num != )
{
return false;
}
this.<>1__state = -;
int num2 = this.<i>5__1;
this.<i>5__1 = num2 + ;
}
else
{
this.<>1__state = -;
this.<i>5__1 = this.start;
}
if (this.<i>5__1 > this.<>4__this.limit)
{
return false;
}
this.<>2__current = this.<i>5__1;
this.<>1__state = ;
return true;
} // Token: 0x17000001 RID: 1
// (get) Token: 0x0600000B RID: 11 RVA: 0x0000219C File Offset: 0x0000039C
int IEnumerator<int>.Current
{
[DebuggerHidden]
get
{
return this.<>2__current;
}
} // Token: 0x0600000C RID: 12 RVA: 0x000021A4 File Offset: 0x000003A4
[DebuggerHidden]
void IEnumerator.Reset()
{
throw new NotSupportedException();
} // Token: 0x17000002 RID: 2
// (get) Token: 0x0600000D RID: 13 RVA: 0x000021AB File Offset: 0x000003AB
object IEnumerator.Current
{
[DebuggerHidden]
get
{
return this.<>2__current;
}
} // Token: 0x0600000E RID: 14 RVA: 0x000021B8 File Offset: 0x000003B8
[DebuggerHidden]
IEnumerator<int> IEnumerable<int>.GetEnumerator()
{
MyClass.<CountFrom>d__2 <CountFrom>d__;
if (this.<>1__state == - && this.<>l__initialThreadId == Environment.CurrentManagedThreadId)
{
this.<>1__state = ;
<CountFrom>d__ = this;
}
else
{
<CountFrom>d__ = new MyClass.<CountFrom>d__2();
<CountFrom>d__.<>4__this = this.<>4__this;
}
<CountFrom>d__.start = this.<>3__start;
return <CountFrom>d__;
} // Token: 0x0600000F RID: 15 RVA: 0x00002207 File Offset: 0x00000407
[DebuggerHidden]
IEnumerator IEnumerable.GetEnumerator()
{
return this.System.Collections.Generic.IEnumerable<System.Int32>.GetEnumerator();
} // Token: 0x04000002 RID: 2
private int <>1__state; // Token: 0x04000003 RID: 3
private int <>2__current; // Token: 0x04000004 RID: 4
private int <>l__initialThreadId; // Token: 0x04000005 RID: 5
private int start; // Token: 0x04000006 RID: 6
public int <>3__start; // Token: 0x04000007 RID: 7
public MyClass <>4__this; // Token: 0x04000008 RID: 8
private int <i>5__1;
}
}

The enumerator class is auto-generated by the compiler
and, as promised, it contains two internal members for the
state and current object,
plus a member for each parameter
(including the hidden this parameter),
plus a member for each local variable.
The Current property merely returns the current object.
All the real work happens in MoveNext.

To generate the MoveNext method, the compiler
takes the code you write and performs a few transformations.
First, all the references to variables and parameters need to
be adjusted since the code moved to a helper class.

Notice that this transformation is quite different fromthe enumeration model we built based on coroutines and fibers.
The C# method is far more efficient in terms of memory usage
since it doesn’t consume an entire stack (typically a megabyte in size)
like the fiber approach does.
Instead it just borrows the stack of the caller,
and anything that it needs to save across calls to MoveNext
are stored in a helper object (which goes on the heap rather than the stack).
This fake-out is normally quite effective—most
people don’t even realize that it’s happening—but there are places
where the difference is significant, and we’ll see that shortly.

The implementation of iterators in C# and its consequences (part 1) Raymond Chen的更多相关文章

  1. What is the yield keyword used for in C#?

    What is the yield keyword used for in C#? https://stackoverflow.com/a/39496/3782855 The yield keywor ...

  2. 一次C#和C++的实际应用性能比较(C++允许我们使用任何手段来提高效率,只要愿意做出足够的努力)

    05年时,在微软的Rico Mariani做了一次实际应用的C#和C++的性能比较.事情起源于微软著名的元老Raymond Chen(在下敬仰的超级牛人)用C++写了一个英汉词典程序,来描述讲解优化C ...

  3. cvpr2015papers

    @http://www-cs-faculty.stanford.edu/people/karpathy/cvpr2015papers/ CVPR 2015 papers (in nicer forma ...

  4. Python 的上下文管理器是怎么设计的?

    花下猫语:最近,我在看 Python 3.10 版本的更新内容时,发现有一个关于上下文管理器的小更新,然后,突然发现上下文管理器的设计 PEP 竟然还没人翻译过!于是,我断断续续花了两周时间,终于把这 ...

  5. Implementation with Java

    Implementation with Java From:http://jcsc.sourceforge.net In general, follow the Sun coding conventi ...

  6. Python标准模块--Iterators和Generators

    1 模块简介 当你开始使用Python编程时,你或许已经使用了iterators(迭代器)和generators(生成器),你当时可能并没有意识到.在本篇博文中,我们将会学习迭代器和生成器是什么.当然 ...

  7. Design and Implementation of the Sun Network File System

    Introduction The network file system(NFS) is a client/service application that provides shared file ...

  8. [转]Objective-c中@interface、@implementation、@protocal

    原处:http://blog.csdn.net/l271640625/article/details/8393531 以下Objective-c简称OC 从事java开发的程序员们都知道,在java中 ...

  9. Implementation Model Editor of AVEVA in OpenSceneGraph

    Implementation Model Editor of AVEVA in OpenSceneGraph eryar@163.com 摘要Abstract:本文主要对工厂和海工设计软件AVEVA的 ...

随机推荐

  1. dubbo和mq的使用场景

    MQ:消息队列.生产者消费者模式,可用于对消息实时性要求不高的场景.多进程之间间接调用关系 Dubbo:RPC实现.多进程之间直接调用关系 dubbo 1,rpc的分布式集群支持:负载均衡是对外提供一 ...

  2. c# 值传递

  3. 牛客练习赛6 C 手铐

    手铐 思路: 缩环+树形dp 代码: #pragma GCC optimize(2) #pragma GCC optimize(3) #pragma GCC optimize(4) #include& ...

  4. Maven dependencies說明

    一.今天搭建了一个maven项目,在pom.xml文件引入依赖后,发现项目结构中没有Maven Dependencies文件(下图),但是在java Build Path中又可以找到Maven Dep ...

  5. 多继承以及MRO顺序

    class A: def test(self): print("A --- test方法") def demo(self): print("A --- demo方法&qu ...

  6. tsql 通过row_number() over() 产生行号

    先按userIP分组,再按时间排序,最后编号. select row_number() over (partition by UserIp order by insertTime),* from us ...

  7. SQL Server查询性能

    sql server常用语句总结 http://ace105.blog.51cto.com/639741/792519 SQL Server 性能调优(一)--从等待状态判断系统资源瓶颈        ...

  8. 为什么深度(Ubuntu)Linux挂载NTFS分区只读不可写?

      如前所述,经扩展,Dell Vostro 1520笔记本电脑拥有了两块硬盘. 本着旧物利用的心思,在其中一块256GiB固态盘上安装了深度linux,同时挂载另一块2TiB机械盘作为存储盘. 这块 ...

  9. 爬虫与seo优化

    爬虫及爬行方式 爬虫有很多名字,比如web机器人.spider等,它是一种可以在无需人类干预的情况下自动进行一系列web事务处理的软件程序.web爬虫是一种机器人,它们会递归地对各种信息性的web站点 ...

  10. fopen 的mode

    转自 http://blog.csdn.net/todd911/article/details/8976543 r 打开只读文件,该文件必须存在. r+具有读写属性,从文件头开始写,保留原文件中没有被 ...