[英]CRC32 in C# fails
I was trying to implement my own CRC32 function in C#. 我试图在C#中实现自己的CRC32函数。 I saw an elegant solution in JS here JavaScript CRC32 So I came up with this:
我在这里的JS JavaScript CRC32中看到了一个优雅的解决方案,因此我想到了:
internal static class Crc32
{
internal static long CalculateCrc32(string str)
{
long[] crcTable = Crc32.MakeCrcTable();
long crc = 0 ^ (-1);
for (int i = 0; i < str.Length; i++)
{
char c = str[i];
crc = (crc >> 8) ^ crcTable[(crc ^ c) & 0xFF];
}
return ~crc; //(crc ^ (-1)) >> 0;
}
internal static long[] MakeCrcTable()
{
long c;
long[] crcTable = new long[256];
for (int n = 0; n < 256; n++)
{
c = n;
for (int k = 0; k < 8; k++)
{
var res = c & 1;
c = (res == 1) ? (0xEDB88320 ^ (c >> 1)) : (c >> 1);
}
crcTable[n] = c;
}
return crcTable;
}
}
The problem is that my solution does not return the same result. 问题是我的解决方案没有返回相同的结果。
Console.WriteLine(Crc32.CalculateCrc32("l"));
results in 1762050814 while the JS function produces 2517025534. The JS result is also the correct one. 结果为1762050814,而JS函数产生2517025534。JS结果也是正确的结果。 What am I doing wrong?
我究竟做错了什么?
The problem here is that you are using the wrong datatype. 这里的问题是您使用了错误的数据类型。 I'm not familiar with the CRC32 algorithm so I googled to find http://sanity-free.org/12/crc32_implementation_in_csharp.html as a reference implementation.
我对CRC32算法不熟悉,因此我在Google上找到了http://sanity-free.org/12/crc32_implementation_in_csharp.html作为参考实现。
the first thing I noticed is that they are using uint
instead of long
. 我注意到的第一件事是他们使用
uint
而不是long
。 This makes sense as I'd assume the 32 in CRC32 means that it would return a 32 bit number. 这很有意义,因为我假设CRC32中的32表示它将返回32位数字。
long
is a 64 bit signed int. long
是一个64位带符号的int。
If you change all the long
for uint
then we nearly get a working program. 如果你改变所有的
long
为uint
那么我们几乎得到了工作程序。 The only line that won't work is c=n
because it can't implicitly convert an int (n) to a uint (c). 唯一行不通的是
c=n
因为它无法隐式将int(n)转换为uint(c)。 However since we know n is always going to be a positive integer we can just change n to be of type uint
as well. 但是,由于我们知道n始终将是一个正整数,因此我们也可以将n更改为
uint
类型。
This leaves me with: 这给我留下了:
internal static uint CalculateCrc32(string str)
{
uint[] crcTable = Crc32.MakeCrcTable();
uint crc = 0xffffffff;
for (int i = 0; i < str.Length; i++)
{
char c = str[i];
crc = (crc >> 8) ^ crcTable[(crc ^ c) & 0xFF];
}
return ~crc; //(crc ^ (-1)) >> 0;
}
internal static uint[] MakeCrcTable()
{
uint c;
uint[] crcTable = new uint[256];
for (uint n = 0; n < 256; n++)
{
c = n;
for (int k = 0; k < 8; k++)
{
var res = c & 1;
c = (res == 1) ? (0xEDB88320 ^ (c >> 1)) : (c >> 1);
}
crcTable[n] = c;
}
return crcTable;
}
With this code Console.WriteLine(Crc32.CalculateCrc32("l"));
使用此代码,
Console.WriteLine(Crc32.CalculateCrc32("l"));
displays 2517025534
as expected. 按预期显示
2517025534
。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.