在.NET中,GetHashCode方法在整个.NET基类库的许多地方都使用。正确执行它对于在集合中或确定相等时快速查找项目尤为重要。
对于如何为自定义类实现GetHashCode,是否有标准算法或最佳实践,以便不会降低性能?
在.NET中,GetHashCode方法在整个.NET基类库的许多地方都使用。正确执行它对于在集合中或确定相等时快速查找项目尤为重要。
对于如何为自定义类实现GetHashCode,是否有标准算法或最佳实践,以便不会降低性能?
当前回答
微软引领了几种哈希方法。。。
//for classes that contain a single int value
return this.value;
//for classes that contain multiple int value
return x ^ y;
//for classes that contain single number bigger than int
return ((int)value ^ (int)(value >> 32));
//for classes that contain class instance fields which inherit from object
return obj1.GetHashCode();
//for classes that contain multiple class instance fields which inherit from object
return obj1.GetHashCode() ^ obj2.GetHashCode() ^ obj3.GetHashCode();
我可以猜测,对于多个大整数,您可以使用这个:
int a=((int)value1 ^ (int)(value1 >> 32));
int b=((int)value2 ^ (int)(value2 >> 32));
int c=((int)value3 ^ (int)(value3 >> 32));
return a ^ b ^ c;
对于多类型也是如此:首先使用GetHashCode()将所有类型转换为int然后int值将被xor'ed,结果是您的哈希值。
对于那些使用哈希作为ID(我的意思是一个唯一的值)的人来说,哈希自然被限制在数字个数,我认为哈希算法是5个字节,至少是MD5。
您可以将多个值转换为哈希值,其中一些值是相同的,因此不要将其用作标识符。(也许有一天我会使用你的组件)
其他回答
与夜编码器的解决方案非常相似,只是如果你想提高素数更容易。
PS:这是你嘴里吐出一点东西的时候之一,因为你知道这可以用9个默认值重构成一个方法,但它会更慢,所以你闭上眼睛,试着忘掉它。
/// <summary>
/// Try not to look at the source code. It works. Just rely on it.
/// </summary>
public static class HashHelper
{
private const int PrimeOne = 17;
private const int PrimeTwo = 23;
public static int GetHashCode<T1, T2, T3, T4, T5, T6, T7, T8, T9, T10>(T1 arg1, T2 arg2, T3 arg3, T4 arg4, T5 arg5, T6 arg6, T7 arg7, T8 arg8, T9 arg9, T10 arg10)
{
unchecked
{
int hash = PrimeOne;
hash = hash * PrimeTwo + arg1.GetHashCode();
hash = hash * PrimeTwo + arg2.GetHashCode();
hash = hash * PrimeTwo + arg3.GetHashCode();
hash = hash * PrimeTwo + arg4.GetHashCode();
hash = hash * PrimeTwo + arg5.GetHashCode();
hash = hash * PrimeTwo + arg6.GetHashCode();
hash = hash * PrimeTwo + arg7.GetHashCode();
hash = hash * PrimeTwo + arg8.GetHashCode();
hash = hash * PrimeTwo + arg9.GetHashCode();
hash = hash * PrimeTwo + arg10.GetHashCode();
return hash;
}
}
public static int GetHashCode<T1, T2, T3, T4, T5, T6, T7, T8, T9>(T1 arg1, T2 arg2, T3 arg3, T4 arg4, T5 arg5, T6 arg6, T7 arg7, T8 arg8, T9 arg9)
{
unchecked
{
int hash = PrimeOne;
hash = hash * PrimeTwo + arg1.GetHashCode();
hash = hash * PrimeTwo + arg2.GetHashCode();
hash = hash * PrimeTwo + arg3.GetHashCode();
hash = hash * PrimeTwo + arg4.GetHashCode();
hash = hash * PrimeTwo + arg5.GetHashCode();
hash = hash * PrimeTwo + arg6.GetHashCode();
hash = hash * PrimeTwo + arg7.GetHashCode();
hash = hash * PrimeTwo + arg8.GetHashCode();
hash = hash * PrimeTwo + arg9.GetHashCode();
return hash;
}
}
public static int GetHashCode<T1, T2, T3, T4, T5, T6, T7, T8>(T1 arg1, T2 arg2, T3 arg3, T4 arg4, T5 arg5, T6 arg6, T7 arg7, T8 arg8)
{
unchecked
{
int hash = PrimeOne;
hash = hash * PrimeTwo + arg1.GetHashCode();
hash = hash * PrimeTwo + arg2.GetHashCode();
hash = hash * PrimeTwo + arg3.GetHashCode();
hash = hash * PrimeTwo + arg4.GetHashCode();
hash = hash * PrimeTwo + arg5.GetHashCode();
hash = hash * PrimeTwo + arg6.GetHashCode();
hash = hash * PrimeTwo + arg7.GetHashCode();
hash = hash * PrimeTwo + arg8.GetHashCode();
return hash;
}
}
public static int GetHashCode<T1, T2, T3, T4, T5, T6, T7>(T1 arg1, T2 arg2, T3 arg3, T4 arg4, T5 arg5, T6 arg6, T7 arg7)
{
unchecked
{
int hash = PrimeOne;
hash = hash * PrimeTwo + arg1.GetHashCode();
hash = hash * PrimeTwo + arg2.GetHashCode();
hash = hash * PrimeTwo + arg3.GetHashCode();
hash = hash * PrimeTwo + arg4.GetHashCode();
hash = hash * PrimeTwo + arg5.GetHashCode();
hash = hash * PrimeTwo + arg6.GetHashCode();
hash = hash * PrimeTwo + arg7.GetHashCode();
return hash;
}
}
public static int GetHashCode<T1, T2, T3, T4, T5, T6>(T1 arg1, T2 arg2, T3 arg3, T4 arg4, T5 arg5, T6 arg6)
{
unchecked
{
int hash = PrimeOne;
hash = hash * PrimeTwo + arg1.GetHashCode();
hash = hash * PrimeTwo + arg2.GetHashCode();
hash = hash * PrimeTwo + arg3.GetHashCode();
hash = hash * PrimeTwo + arg4.GetHashCode();
hash = hash * PrimeTwo + arg5.GetHashCode();
hash = hash * PrimeTwo + arg6.GetHashCode();
return hash;
}
}
public static int GetHashCode<T1, T2, T3, T4, T5>(T1 arg1, T2 arg2, T3 arg3, T4 arg4, T5 arg5)
{
unchecked
{
int hash = PrimeOne;
hash = hash * PrimeTwo + arg1.GetHashCode();
hash = hash * PrimeTwo + arg2.GetHashCode();
hash = hash * PrimeTwo + arg3.GetHashCode();
hash = hash * PrimeTwo + arg4.GetHashCode();
hash = hash * PrimeTwo + arg5.GetHashCode();
return hash;
}
}
public static int GetHashCode<T1, T2, T3, T4>(T1 arg1, T2 arg2, T3 arg3, T4 arg4)
{
unchecked
{
int hash = PrimeOne;
hash = hash * PrimeTwo + arg1.GetHashCode();
hash = hash * PrimeTwo + arg2.GetHashCode();
hash = hash * PrimeTwo + arg3.GetHashCode();
hash = hash * PrimeTwo + arg4.GetHashCode();
return hash;
}
}
public static int GetHashCode<T1, T2, T3>(T1 arg1, T2 arg2, T3 arg3)
{
unchecked
{
int hash = PrimeOne;
hash = hash * PrimeTwo + arg1.GetHashCode();
hash = hash * PrimeTwo + arg2.GetHashCode();
hash = hash * PrimeTwo + arg3.GetHashCode();
return hash;
}
}
public static int GetHashCode<T1, T2>(T1 arg1, T2 arg2)
{
unchecked
{
int hash = PrimeOne;
hash = hash * PrimeTwo + arg1.GetHashCode();
hash = hash * PrimeTwo + arg2.GetHashCode();
return hash;
}
}
}
我在使用上面选择的实现时遇到了浮点和小数的问题。
此测试失败(浮点数;哈希值相同,即使我将2个值切换为负值):
var obj1 = new { A = 100m, B = 100m, C = 100m, D = 100m};
var obj2 = new { A = 100m, B = 100m, C = -100m, D = -100m};
var hash1 = ComputeHash(obj1.A, obj1.B, obj1.C, obj1.D);
var hash2 = ComputeHash(obj2.A, obj2.B, obj2.C, obj2.D);
Assert.IsFalse(hash1 == hash2, string.Format("Hashcode values should be different hash1:{0} hash2:{1}",hash1,hash2));
但是这个测试通过了(int):
var obj1 = new { A = 100m, B = 100m, C = 100, D = 100};
var obj2 = new { A = 100m, B = 100m, C = -100, D = -100};
var hash1 = ComputeHash(obj1.A, obj1.B, obj1.C, obj1.D);
var hash2 = ComputeHash(obj2.A, obj2.B, obj2.C, obj2.D);
Assert.IsFalse(hash1 == hash2, string.Format("Hashcode values should be different hash1:{0} hash2:{1}",hash1,hash2));
我改变了我的实现,不再对原始类型使用GetHashCode,它似乎工作得更好
private static int InternalComputeHash(params object[] obj)
{
unchecked
{
var result = (int)SEED_VALUE_PRIME;
for (uint i = 0; i < obj.Length; i++)
{
var currval = result;
var nextval = DetermineNextValue(obj[i]);
result = (result * MULTIPLIER_VALUE_PRIME) + nextval;
}
return result;
}
}
private static int DetermineNextValue(object value)
{
unchecked
{
int hashCode;
if (value is short
|| value is int
|| value is byte
|| value is sbyte
|| value is uint
|| value is ushort
|| value is ulong
|| value is long
|| value is float
|| value is double
|| value is decimal)
{
return Convert.ToInt32(value);
}
else
{
return value != null ? value.GetHashCode() : 0;
}
}
}
可以尝试采用C++Boost库的方法。类似于:
class HashUtil
{
public static int HashCombine(int seed, int other)
{
unchecked
{
return other + 0x9e3779b9 + (seed << 6) + (seed >> 2);
}
}
}
然后:
class MyClass
{
private string _field1;
private int _field2;
private AnotherClass _field3;
private YetAnotherClass _field4;
public override int GetHashCode()
{
int result = HashUtil.HashCombine(_field1.GetHashCode(), _field2);
result = HashUtil.HashCombine(result, _field3.GetHashCode());
return HashUtil.HashCombine(result, _field4.GetHashCode());
}
}
这是我的简单方法。我使用的是经典的生成器模式。它是类型安全的(无装箱/拆箱),并且与.NET 2.0兼容(无扩展方法等)。
它的用法如下:
public override int GetHashCode()
{
HashBuilder b = new HashBuilder();
b.AddItems(this.member1, this.member2, this.member3);
return b.Result;
}
这里是实际的生成器类:
internal class HashBuilder
{
private const int Prime1 = 17;
private const int Prime2 = 23;
private int result = Prime1;
public HashBuilder()
{
}
public HashBuilder(int startHash)
{
this.result = startHash;
}
public int Result
{
get
{
return this.result;
}
}
public void AddItem<T>(T item)
{
unchecked
{
this.result = this.result * Prime2 + item.GetHashCode();
}
}
public void AddItems<T1, T2>(T1 item1, T2 item2)
{
this.AddItem(item1);
this.AddItem(item2);
}
public void AddItems<T1, T2, T3>(T1 item1, T2 item2, T3 item3)
{
this.AddItem(item1);
this.AddItem(item2);
this.AddItem(item3);
}
public void AddItems<T1, T2, T3, T4>(T1 item1, T2 item2, T3 item3,
T4 item4)
{
this.AddItem(item1);
this.AddItem(item2);
this.AddItem(item3);
this.AddItem(item4);
}
public void AddItems<T1, T2, T3, T4, T5>(T1 item1, T2 item2, T3 item3,
T4 item4, T5 item5)
{
this.AddItem(item1);
this.AddItem(item2);
this.AddItem(item3);
this.AddItem(item4);
this.AddItem(item5);
}
public void AddItems<T>(params T[] items)
{
foreach (T item in items)
{
this.AddItem(item);
}
}
}
如果我们的财产不超过8处(希望如此),这里还有另一种选择。
ValueTuple是一个结构,似乎有一个可靠的GetHashCode实现。
这意味着我们可以简单地这样做:
// Yay, no allocations and no custom implementations!
public override int GetHashCode() => (this.PropA, this.PropB).GetHashCode();
让我们来看看.NETCore当前对ValueTuple的GetHashCode的实现。
这来自ValueTuple:
internal static int CombineHashCodes(int h1, int h2)
{
return HashHelpers.Combine(HashHelpers.Combine(HashHelpers.RandomSeed, h1), h2);
}
internal static int CombineHashCodes(int h1, int h2, int h3)
{
return HashHelpers.Combine(CombineHashCodes(h1, h2), h3);
}
这来自HashHelper:
public static readonly int RandomSeed = Guid.NewGuid().GetHashCode();
public static int Combine(int h1, int h2)
{
unchecked
{
// RyuJIT optimizes this to use the ROL instruction
// Related GitHub pull request: dotnet/coreclr#1830
uint rol5 = ((uint)h1 << 5) | ((uint)h1 >> 27);
return ((int)rol5 + h1) ^ h2;
}
}
英语:
向左旋转(循环移位)h1 5个位置。将结果和h1相加。将结果与h2进行异或运算。首先对{static random seed,h1}执行上述操作。对于每个其他项目,对上一个结果和下一个项目(例如h2)执行操作。
如果能更多地了解这个ROL-5散列代码算法的财产,那就太好了。
遗憾的是,为我们自己的GetHashCode延迟ValueTuple可能不像我们希望的那样快。相关讨论中的这条评论说明了直接调用HashHelpers.Combine更具性能。另一方面,这是内部的,所以我们必须复制代码,牺牲了我们在这里获得的大部分。此外,我们将负责记住首先与随机种子结合。我不知道如果我们跳过这一步会有什么后果。