首页 > 其他 > 详细

FCL研究-集合- System.Collections 接口和对象集合

时间:2014-03-30 23:49:13      阅读:1159      评论:0      收藏:0      [点我收藏+]

【目录】

  发现自己已经有很长一段时间写代码没什么进步了,随便读读FCL的源码,看看之前一直用的方法是如何实现的,也顺便提高下自己。FCL很是庞大,很难下口,于是用最笨的办法,先看常见的命名空间,逐个展开。战五渣的水平,必定有很多理解上的错误,欢迎斧正,不胜感激。

 System.Collections namespace contains interfaces and classes that define various collections of objects, such as lists, queues, bit arrays, hash tables and dictionaries.‘ data-guid="f227f10a151d6eb7f5562d3ea1206dbe"> System.Collections  命名空间中的集合包含(如列表、队列、位数组、哈希表和字典)的集合。

本篇目录:

 IEnumerable,ICollection,IList,IDictionary

 ArrayList

 Queue

 Stack

 SortedList

 集合的接口

在刚开始写程序的时候的时候经常会写一些接口,哪怕这个接口只被用到了一次,也要抽象一个接口出来,这样显得牛X一些。到后来,接口几乎从日常的代码中消失,能简单就简单。看了FCL的源码,发现,这些接口很有必要,而且抽象的恰到好处,不经拍案叫绝。

很难想象,如果没有这些接口,庞大的FCL将如何构建,如何约束那些类。每个集合的操作方法类似,名称各不相同,对于使用者来说,也绝对是件很苦逼的事情。接口 是一种规范,实现了某一个接口,便具备了改接口的功能。所以了解某一个集合的性质和功能,首先需要了解它实现了哪些接口。

集合中常见的接口有IEnumerable,IEnumerator,ICollection,IComparer,IDictionary,IDictionaryEnumerator,ListDictionaryInternal,IEnumerator,IHashCodeProvider,IList,IStructuralComparable等。彻底晕菜了!新建 Class Diagram ,将几个主要接口拖入后,结构便很清晰了。

bubuko.com,布布扣

所有的集合都是继承了IEnumerable,逐个分析每个接口的实现。IEnumerable的源码,其中 PureAttribute来表示自己是很纯的,协定的东西。DispId 属性被用来指定一个OLE 的自动化 DISPID,COM交互时会使用。查了下使用的地方在 ComAwareEventInfo.cs 中,请自行查阅。 IEnumerator GetEnumerator(); 使用了组合模式,正是因为这个方法,所有的集合才可以使用Foreach方法。在后面研究集合的时候会详细的看下IEnumerator的实现。

IEnumerable源码

IEnumerator源码

接口ICollection 继承了IEnumerable,定义了集合基本的元素,大小(count)枚举器(继承自IEnumerable 的GetEnumerator),同步方法(使用 IsSynchronized,SyncRoot),这里涉及一个锁的问题,如果对这个集合元素锁定后不可读与写,那么锁定这个集合的本身,如果锁定这个集合,不可写,可以读,那么锁定这个集合的SyncRoot。为什么使用Synchronized 方法返回的类是线程安全的呢,来看下具体的实现方式吧,以ArrayList为例

1
var al=  ArrayList.Synchronized(new ArrayList());

 

1
2
3
4
5
6
7
public static ArrayList Synchronized(ArrayList list) {
           if (list==null)
               throw new ArgumentNullException("list");
           Contract.Ensures(Contract.Result<ArrayList>() != null);
           Contract.EndContractBlock();
           return new SyncArrayList(list);
       }

 具体的实现在SyncArrayList中下面是部分源码:

 实现的方式也很简单,所有的数据操作全部上锁,所以就线程安全了

 学习数据结构的时候,线性表有几种操作,初始化,清空,获取某一个位置元素,判断元素是否存在,插入,删除,获取长度。很多元素是可以标准化的,IList就是干这个的。

IList的具体实现:

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
public interface IList : ICollection
  {
     //索引器
      Object this[int index] {
          get;
          set;
      }
    //插入
      int Add(Object value);
    //判断是否包含
      bool Contains(Object value);
    //清空
      void Clear();
 
    //判断是否为只读 只读集合在创建之后不允许添加、移除或修改元素。
      bool IsReadOnly
      { get; }
     //是否是固定大小
      bool IsFixedSize
      {
          get;
      }
     //取元素索引值
      int IndexOf(Object value);
    //指定索引位置插入元素
      void Insert(int index, Object value);
    //删除某个元素
      void Remove(Object value);
     //删除指定索引的元素
      void RemoveAt(int index);
  }

 字典集合的抽象接口为IDictionary,字典类型的操作有,获取keys,获取values,添加,删除,清空等。实现源码如下:

 ArrayList

ArrayList动态容量的实现

ArrayList 方法底层实现探究

 

ArrayList动态容量的实现

 

ArrayList  为动态数组,动态的添加和减少线性表的长度,不用担心长度不够而抛异常。首先我们来探究下,这个动态的长度是如何实现的。查看arraylist.cs文件。

1
2
3
4
5
6
7
public virtual int Add(Object value) {
    Contract.Ensures(Contract.Result<int>() >= 0);
    if (_size == _items.Length) EnsureCapacity(_size + 1);
    _items[_size] = value;
    _version++;
    return _size++;
}

 ArrayList添加元素方法, Contract.Ensures(Contract.Result<int>() >= 0);这句可以忽略,契约式编程,可以自己搜索。顺腾摸瓜,进入EnsureCapacity函数:

1
2
3
4
5
6
7
8
9
10
private void EnsureCapacity(int min) {
            if (_items.Length < min) {
                int newCapacity = _items.Length == 0? _defaultCapacity: _items.Length * 2;//定义一个新的容量,如果当然容量是0,就用默认的,否则就当前容量*2
                // Allow the list to grow to maximum possible capacity (~2G elements) before encountering overflow.
                // Note that this check works even when _items.Length overflowed thanks to the (uint) cast
                if ((uint)newCapacity > Array.MaxArrayLength) newCapacity = Array.MaxArrayLength;//如果新的容量比Array数组的最大值还大,那么就赋最大的值
                if (newCapacity < min) newCapacity = min;//如果新的容量比传入的最小值要小,那么赋最小值
                Capacity = newCapacity;//容量等于新的容量
            }
        }

 学过数据结构我们都知道,线性表增加容量,肯定要移动元素的。这个地方没看到,那么在找Capacity的Set方法。如下:

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
public virtual int Capacity {
           get {
               Contract.Ensures(Contract.Result<int>() >= Count);
               return _items.Length;
           }
           set {
               if (value < _size) {
                   throw new ArgumentOutOfRangeException("value", Environment.GetResourceString("ArgumentOutOfRange_SmallCapacity"));
               }
               Contract.Ensures(Capacity >= 0);
               Contract.EndContractBlock();
               // We don‘t want to update the version number when we change the capacity.
               // Some existing applications have dependency on this.
               if (value != _items.Length) {
                   if (value > 0) {
                       Object[] newItems = new Object[value];
                       if (_size > 0) {
                         Array.Copy(_items, 0, newItems, 0, _size);//ArrayList的长度发生改变时,就要来一次迁移
                       }
                       _items = newItems;
                   }
                   else {
                       _items = new Object[_defaultCapacity];
                   }
               }           
           }
       }

  private const int _defaultCapacity = 4;默认的容量是4.不管容量是变大还是变小,都要移动元素,性能肯定是会有到影响的。

ArrayList 方法底层实现探究

继续跟踪Array.Copy 来看下具体是怎么实现的,最后跟踪的代码:

1
2
3
4
5
[System.Security.SecurityCritical]  // auto-generated
[ReliabilityContract(Consistency.MayCorruptInstance, Cer.MayFail)]
[ResourceExposure(ResourceScope.None)]
[MethodImplAttribute(MethodImplOptions.InternalCall)]
internal static extern void Copy(Array sourceArray, int sourceIndex, Array destinationArray, int destinationIndex, int length, bool reliable);

      [MethodImplAttribute(MethodImplOptions.InternalCall)]由这个属性猜出来,这个是CLR内部实现的,没法看。好奇心有强怎么办,好吧,拿出sscli(.net 2.0 的clr源码),在ecall.cpp里面看到这个

1
FCFuncElement("Copy", SystemNative::ArrayCopy)//

 仔细看,能看出个大概。经过阅读FCL源码会发现,几乎所有的集合Copy,CopyTo 方法,最终都是调用Array.Copy,Array.Copy最终调用的是下面这个CLR中的方法

CLR中的源码如下:

 IndexOf 的具体实现如下,为了更容易阅读,我把里面的前置判断去掉了。可以看到一个简单的函数,但是极为严谨,再回想自己写的代码,弱爆了。加了一点注释。

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
40
41
public static int IndexOf(Array array, Object value, int startIndex, int count) {
          int lb = array.GetLowerBound(0);
          // Try calling a quick native method to handle primitive types.
          int retVal;
          bool r = TrySZIndexOf(array, startIndex, count, value, out retVal);
          if (r)
              return retVal;
 
          Object[] objArray = array as Object[]; //转换为object 数组
          int endIndex = startIndex + count;
          if (objArray != null) { //转换之后不为null
              if (value == null) { //如果传入的值为null,则查找objArray里面为null的,返回改值的位置
                  for (int i = startIndex; i < endIndex; i++) {
                      if (objArray[i] == null) return i;
                  }
              }
              else {//如果传入的值不为null,则逐个查找,找到后返回该值的位置
                  for (int i = startIndex; i < endIndex; i++) {
                      Object obj = objArray[i];  //这个地方为什么要单独出来呢,直接这样  if (objArray[i]!= null && objArray[i].Equals(value)) ,之前自己的代码全是这么写的。
                                                 // 仔细推敲下,在if()里面只要取一次就好,性能应该可以稍微快一点。          
                      if (obj != null && obj.Equals(value)) return i;
                  }
              }
          }
          else {
              for (int i = startIndex; i < endIndex; i++) {
                  Object obj = array.GetValue(i);//这个地方应该和上面类似
                  if( obj == null) {
                      if(value == null) return i;
                  }
                  else {
                      if( obj.Equals(value)) return i;
                  }
              }
          }
          // Return one less than the lower bound of the array.  This way,
          // for arrays with a lower bound of -1 we will not return -1 when the
          // item was not found.  And for SZArrays (the vast majority), -1 still
          // works for them.
          return lb-1;
      }

 里面的调用,可以在sscli中找到具体的实现

1
bool r = TrySZIndexOf(array, startIndex, count, value, out retVal);
1
 

 最终的c++代码,很简单不是么

1
2
3
4
5
6
7
8
9
static int IndexOf(KIND array[], UINT32 index, UINT32 count, KIND value) {
      LEAF_CONTRACT;
 
      _ASSERTE(array != NULL && index >= 0 && count >= 0);
      for(UINT32 i=index; i<index+count; i++)
          if (array[i] == value)
              return i;
      return -1;
  }

 LastIndexOf方法和IndexOf方法类似,不详细看了。看下Add和Insert方法的实现:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
public virtual int Add(Object value) {
     if (_size == _items.Length) EnsureCapacity(_size + 1);//首先确保容量要够,不够会自动成倍添加,上面说过
           _items[_size] = value;
           _version++;
           return _size++;
       }
 
public virtual void Insert(int index, Object value) {
 
          if (index < 0 || index > _size) throw new ArgumentOutOfRangeException("index", Environment.GetResourceString("ArgumentOutOfRange_ArrayListInsert"));
 
           if (_size == _items.Length) EnsureCapacity(_size + 1);
           if (index < _size) {//如果插入的index比当前的长度要小,那么index之后的元素要后移
               Array.Copy(_items, index, _items, index + 1, _size - index); //调用内部方法,上面已给出
           }
           _items[index] = value;
           _size++;
           _version++;
       }

 有源码可以看出,Add方法是直接添加到线性表的表尾,Insert方法是直接插入到指定位置,制定位置之后的元素要依次后移。显然Add方法的效率要高一些。用于添加的方法还有  AddRange和InsertRange,顾名思义就是插入一个范围数据即插入集合。那这两个方法有什么异同呢?先看AddRange吧

1
2
3
public virtual void AddRange(ICollection c) {
           InsertRange(_size, c);
       }

 AddRange的方法是直接调用的InsertRange,从末尾插入一个集合。InsertRange源码如下:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
public virtual void InsertRange(int index, ICollection c) {
           int count = c.Count;
           if (count > 0) {
               EnsureCapacity(_size + count);
               // shift existing items
               if (index < _size) {//依次向后移位,腾出位置
                   Array.Copy(_items, index, _items, index + count, _size - index);
               }
 
               Object[] itemsToInsert = new Object[count];//新建一个object数组
               c.CopyTo(itemsToInsert, 0);//将新加的集合拷贝到新建的object数组中
               itemsToInsert.CopyTo(_items, index);//再讲这个新建的数组拷贝到源列表中
               _size += count;
               _version++;
           }
       }

 

 下面看下ArrayList的排序方法的实现Sort:

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
public virtual void Sort()
        {
            Sort(0, Count, Comparer.Default);
        }
//fcl里的最终的实现
 public static void Sort(Array keys, Array items, int index, int length, IComparer comparer) {
         
            if (length > 1) {
                //如果是默认的,那么调用内部方法,下面会详细给出
                if (comparer == Comparer.Default || comparer == null) {
                    bool r = TrySZSort(keys, items, index, index + length - 1);
                    if (r)
                        return;
                }
  
                Object[] objKeys = keys as Object[];
                Object[] objItems = null;
                if (objKeys != null)
                    objItems = items as Object[];
                if (objKeys != null && (items==null || objItems != null)) {
                    SorterObjectArray sorter = new SorterObjectArray(objKeys, objItems, comparer);
                    sorter.QuickSort(index, index + length - 1);
                }
                else {
                    SorterGenericArray sorter = new SorterGenericArray(keys, items, comparer);
                    sorter.QuickSort(index, index + length - 1);
                }
            }
        }

 可以自己实现一个比较器IComparer,也可以使用默认的比较器。如果使用的是默认的比较器,那么将会调用clr底层的快速排序方法,下面是从sscli中查到的C++源码:

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
static void QuickSort(KIND keys[], KIND items[], int left, int right) {//KIND 在头部给出了定义 template <class KIND>
      WRAPPER_CONTRACT;
 
      // Make sure left != right in your own code.
      _ASSERTE(keys != NULL && left < right);
      do {
          int i = left;
          int j = right;
          KIND x = keys[i + ((j - i) >> 1)];
          do {
              while (keys[i] < x) i++;
              while (x < keys[j]) j--;
              _ASSERTE(i>=left && j<=right);
              if (i > j) break;
              if (i < j) {
                  KIND key = keys[i];
                  keys[i] = keys[j];
                  keys[j] = key;
                  if (items != NULL) {
                      KIND item = items[i];
                      items[i] = items[j];
                      items[j] = item;
                  }
              }
              i++;
              j--;
          } while (i <= j);
          if (j - left <= right - i) {
              if (left < j) QuickSort(keys, items, left, j);
              left = i;
          }
          else {
              if (i < right) QuickSort(keys, items, i, right);
              right = j;
          }
      } while (left < right);
  }

 和Sort方法类似,BinarySearch(二分查找)方法也可以使用自定义的比较器看下BinarySearch的具体实现:

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
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
public static int BinarySearch(Array array, int index, int length, Object value, IComparer comparer) {
          //去掉一些前置判断
           if (comparer == null) comparer = Comparer.Default;
           if (comparer == Comparer.Default) {
               int retval;
               bool r = TrySZBinarySearch(array, index, length, value, out retval);
               if (r)
                   return retval;
           }
 
           int lo = index;
           int hi = index + length - 1;
           Object[] objArray = array as Object[];
           if(objArray != null) {
               while (lo <= hi) {
                   // i might overflow if lo and hi are both large positive numbers.
                   int i = GetMedian(lo, hi);//取中位数
 
                   int c;
                   try {
                       c = comparer.Compare(objArray[i], value);//比较这个中间值是否是要查找的值,c=0找到 c为负数在右边,c为正数在左边
                   }
                   catch (Exception e) {
                       throw new InvalidOperationException(Environment.GetResourceString("InvalidOperation_IComparerFailed"), e);
                   }
                   if (c == 0) return i; //找到,返回下标
                   if (c < 0) {
                       lo = i + 1;
                   }
                   else {
                       hi = i - 1;
                   }
               }
           }
           else {
               while (lo <= hi) {
                   int i = GetMedian(lo, hi);
 
                   int c;
                   try {
                       c = comparer.Compare(array.GetValue(i), value);
                   }
                   catch (Exception e) {
                       throw new InvalidOperationException(Environment.GetResourceString("InvalidOperation_IComparerFailed"), e);
                   }
                   if (c == 0) return i;
                   if (c < 0) {
                       lo = i + 1;
                   }
                   else {
                       hi = i - 1;
                   }
               }
           }
           return ~lo;
       }

 看下这个TrySZBinarySearch在clr中的具体实现吧,同样很易懂。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
static int BinarySearchBitwiseEquals(KIND array[], int index, int length, KIND value) {
      WRAPPER_CONTRACT;
 
      _ASSERTE(array != NULL && length >= 0 && index >= 0);
      int lo = index;
      int hi = index + length - 1;
      // Note: if length == 0, hi will be Int32.MinValue, and our comparison
      // here between 0 & -1 will prevent us from breaking anything.
      while (lo <= hi) {
          int i = lo + ((hi - lo) >> 1);
          if (array[i] < value) {
              lo = i + 1;
          }
          else if (array[i] > value){
              hi = i - 1;
          }
          else {
              return i;
          }
      }
      return ~lo;
  }

 ArrayList的其他方法,也极为易懂和类似,不在罗列。

 Queue

队列是特殊的线性表,先进先出的结构。从源码中可以看出,FCL中的Queue是一种循环队列。先看Queue的属性

1
2
3
4
5
6
7
8
9
10
11
private Object[] _array; //存储的数据
private int _head;       // 对一个有效元素
private int _tail;       // 最后一个有效元素
private int _size;       // 元素数量
private int _growFactor; // 增长因素 100 == 1.0, 130 == 1.3, 200 == 2.0,取值范围 1.0到10.0之间
private int _version;
[NonSerialized]
private Object _syncRoot;
 
private const int _MinimumGrow = 4; //最小增长量
private const int _ShrinkThreshold = 32;//这个地方极为扯淡,定义了没用,下面直接写死32

 初始化:

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
public Queue()
          : this(32, (float)2.0) {
      }
      // Creates a queue with room for capacity objects. The default grow factor
      // is used.
      //
      public Queue(int capacity)
          : this(capacity, (float)2.0) {
      }
       
      // Creates a queue with room for capacity objects. When full, the new
      // capacity is set to the old capacity * growFactor.
      //
      public Queue(int capacity, float growFactor) {
          if (capacity < 0)
              throw new ArgumentOutOfRangeException("capacity", Environment.GetResourceString("ArgumentOutOfRange_NeedNonNegNum"));
          if (!(growFactor >= 1.0 && growFactor <= 10.0))
              throw new ArgumentOutOfRangeException("growFactor", Environment.GetResourceString("ArgumentOutOfRange_QueueGrowFactor", 1, 10));
          Contract.EndContractBlock();
   
          _array = new Object[capacity];
          _head = 0;
          _tail = 0;
          _size = 0;
          _growFactor = (int)(growFactor * 100);
      }
  

 入队操作:

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
public virtual void Enqueue(Object obj) {
           if (_size == _array.Length) {//如果队满,则重新分配容量
               int newcapacity = (int)((long)_array.Length * (long)_growFactor / 100);
               if (newcapacity < _array.Length + _MinimumGrow) {//如果新分配的容量小于当前容量加上最小增长量,那么把当前容量加最小增长量分配给新分配的容量
                   newcapacity = _array.Length + _MinimumGrow;
               }
               SetCapacity(newcapacity);//重新设置容量
           }
    
           _array[_tail] = obj;
           _tail = (_tail + 1) % _array.Length;//如果_taill+1<_array.Length 那么 _tail=_tail+1;否则,_tail=0;表示队列已满。
                                               //循环队列的写法
           _size++;
           _version++;
       }
   <br>private void SetCapacity(int capacity) {
           Object[] newarray = new Object[capacity];
           if (_size > 0) {
               if (_head < _tail) {//环形队列头部项索引在尾部索引前面
                   Array.Copy(_array, _head, newarray, 0, _size);
               } else {//环形队列头部项索引在尾部索引后面
                   Array.Copy(_array, _head, newarray, 0, _array.Length - _head);//copy _head 到 Length部分
 
                   Array.Copy(_array, 0, newarray, _array.Length - _head, _tail);//copy 0 到_tail 部分
               }
           }
    
           _array = newarray;
           _head = 0;
           _tail = (_size == capacity) ? 0 : _size;           
           _version++;
       }

 出队操作Dequeue(),出队,并从队列中删除;Peek()方法,取队列的第一位元素,不从队列中删除。

1
2
3
4
5
6
7
8
9
10
11
12
public virtual Object Dequeue() {
            if (Count == 0)
                throw new InvalidOperationException(Environment.GetResourceString("InvalidOperation_EmptyQueue"));
            Contract.EndContractBlock();
 
            Object removed = _array[_head]; //取元素
            _array[_head] = null//删除元素
            _head = (_head + 1) % _array.Length;//循环队列,头部索引移到下一位
            _size--;
            _version++;
            return removed;
        }

 

1
2
3
4
5
6
7
public virtual Object Peek() {
        if (Count == 0)
            throw new InvalidOperationException(Environment.GetResourceString("InvalidOperation_EmptyQueue"));
        Contract.EndContractBlock();
 
        return _array[_head];
    }

 可以看出队列的入队和出队的时间复杂度是O(1);但是入列的时候如果需要重置容量,那么时间复杂度会变为O(n)

 队列中的其他操作:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
//是否包含       
public virtual bool Contains(Object obj) {
            int index = _head;
            int count = _size;
     
            while (count-- > 0) {
                if (obj == null) {
                    if (_array[index] == null)
                        return true;
                } else if (_array[index] != null && _array[index].Equals(obj)) {
                    return true;
                }
                index = (index + 1) % _array.Length;//index向后移
            }
     
            return false;
        }
//取某个元素
internal Object GetElement(int i){
      return _array[(_head + i) % _array.Length]
       }

 和ArrayList类似,Queue也有线程安全的实现,Queue.Synchronized(),返回一个 同步Queue。实现的方式就是在队列的操作上加锁。

 Stack

 栈,先进后出,像弹夹。属性定义:

1
2
3
4
5
private Object[] _array;     // Storage for stack elements
private int _size;           // Number of items in the stack.
private int _version;        // Used to keep enumerator in [....] w/ collection.
private Object _syncRoot;
private const int _defaultCapacity = 10;

 可以看出栈的结构属性更为简单,默认的容量是10;初始化操作:

  入栈操作:

1
2
3
4
5
6
7
8
9
10
public virtual void Push(Object obj) {
          //Contract.Ensures(Count == Contract.OldValue(Count) + 1);
          if (_size == _array.Length) {//如果容量满了,那么成2倍增加
              Object[] newArray = new Object[2*_array.Length];
              Array.Copy(_array, 0, newArray, 0, _size);
              _array = newArray;
          }
          _array[_size++] = obj;
          _version++;
      }

 出栈操作:

1
2
3
4
5
6
public virtual Object Pop() {
          _version++;
          Object obj = _array[--_size];//取元素并将长度减1
          _array[_size] = null;     // 删除元素
          return obj;
      }

 和Queue类似,Stack也有Peek操作,实现方式类似,同样也有Synchronized方法,实现方式也是类似的。

 SortedList

 属性:

1
2
3
4
5
6
7
8
9
10
11
private Object[] keys;//键数组
private Object[] values;//值数组
private int _size;
private int version;
private IComparer comparer;
private KeyList keyList;//建集合 继承IList 内部使用 SortList 与keys关联
private ValueList valueList;//值集合 继承IList 内部使用 SortList 与 values关联
[NonSerialized]
private Object _syncRoot;
private const int _defaultCapacity = 16;//默认容量 16
private static Object[] emptyArray = EmptyArray<Object>.Value;//空数组 等效 New Object[0];

 SortedList的容量也是动态的

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
  private void EnsureCapacity(int min) {
            int newCapacity = keys.Length == 0? 16: keys.Length * 2;//2倍增上
            // Allow the list to grow to maximum possible capacity (~2G elements) before encountering overflow.
            // Note that this check works even when _items.Length overflowed thanks to the (uint) cast
            if ((uint)newCapacity > Array.MaxArrayLength) newCapacity = Array.MaxArrayLength;
            if (newCapacity < min) newCapacity = min;
            Capacity = newCapacity;
        }
 
public virtual int Capacity {
            get {
                return keys.Length;
            }
            set {
                if (value < Count) {
                    throw new ArgumentOutOfRangeException("value", Environment.GetResourceString("ArgumentOutOfRange_SmallCapacity"));
                }
                Contract.EndContractBlock();
 
                if (value != keys.Length) {
                    if (value > 0) {//如果容量发生改变,新建键数组和值指数组,并将源键值数组复制进去。如果新的容量不大于0则将键数组和值数组置空
                        Object[] newKeys = new Object[value];
                        Object[] newValues = new Object[value];
                        if (_size > 0) {
                            Array.Copy(keys, 0, newKeys, 0, _size);
                            Array.Copy(values, 0, newValues, 0, _size);
                        }
                        keys = newKeys;
                        values = newValues;
                    }
                    else {
                        // size can only be zero here.
                        Contract.Assert( _size == 0, "Size is not zero");
                        keys = emptyArray;
                        values = emptyArray;                     
                    }
                }
            }
        }

 SortedList 的IndexOfKey方法,可以发现内部使用二分法查找,具体C++代码上面ArrayList中已经给出。

1
2
3
4
5
6
7
public virtual int IndexOfKey(Object key) {
         if (key == null)
             throw new ArgumentNullException("key", Environment.GetResourceString("ArgumentNull_Key"));
         Contract.EndContractBlock();
         int ret = Array.BinarySearch(keys, 0, _size, key, comparer);
         return ret >=0 ? ret : -1;
     }

 SortedList的IndexOfValue方法内部也是调用的IndexOfKey这个方法,如下:

1
2
3
public virtual int IndexOfValue(Object value) {
          return Array.IndexOf(values, value, 0, _size);
      }

 插入操作如下,很易懂。可以看出,Add一个元素的时候

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
private void Insert(int index, Object key, Object value) {
            if (_size == keys.Length) EnsureCapacity(_size + 1);
            if (index < _size) {
                Array.Copy(keys, index, keys, index + 1, _size - index);
                Array.Copy(values, index, values, index + 1, _size - index);
            }
            keys[index] = key;
            values[index] = value;
            _size++;
            version++;
        }
 public virtual void Add(Object key, Object value) {
         if (key == null) throw new ArgumentNullException("key", Environment.GetResourceString("ArgumentNull_Key"));
           Contract.EndContractBlock();
           int i = Array.BinarySearch(keys, 0, _size, key, comparer);//这个地方是二分法查找。comparer是排序器,可以自己实现。默认的是按照key来排序。每次新添加元素都会重新排序。
            if (i >= 0)
             throw new ArgumentException(Environment.GetResourceString("Argument_AddingDuplicate__", GetKey(i), key));
           Insert(~i, key, value);//这个地方比较绕,Array.BinarySearch 如果找不到,返回的是index取反,结果为-1,在对这个-1取反,结果为0.在0这个位置插入<br>        }
  

 删除操作,同样也很易懂,代码如下:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
public virtual void RemoveAt(int index) {
         if (index < 0 || index >= Count) throw new ArgumentOutOfRangeException("index", Environment.GetResourceString("ArgumentOutOfRange_Index"));
         Contract.EndContractBlock();
         _size--;
         if (index < _size) {
             Array.Copy(keys, index + 1, keys, index, _size - index);
             Array.Copy(values, index + 1, values, index, _size - index);
         }
         keys[_size] = null;
         values[_size] = null;
         version++;
     }
 
public virtual void Remove(Object key) {
         int i = IndexOfKey(key);
         if (i >= 0)
         RemoveAt(i);
     }

 同样的,SortedList也有线程同步的方法 SortedList.Synchronized() 实现方式和ArrayList,Queue,Stack 并无二致。

 至此,告一段落。几个常用的集合,自己有了更为深刻的理解。阅读优秀的代码是一种享受,阅读渣渣的代码,是虐心!

FCL研究-集合- System.Collections 接口和对象集合,布布扣,bubuko.com

FCL研究-集合- System.Collections 接口和对象集合

原文:http://www.cnblogs.com/BangQ/p/3634214.html

(0)
(0)
   
举报
评论 一句话评论(0
关于我们 - 联系我们 - 留言反馈 - 联系我们:wmxa8@hotmail.com
© 2014 bubuko.com 版权所有
打开技术之扣,分享程序人生!