C#读取纯真IP数据库的代码

 C#读取纯真IP数据库的代码

/// <summary>
/// PHCZIP 的摘要说明。
/// </summary>
///     (转自http://community.csdn.net/Expert/topic/5325/5325875.xml?temp=.3523371)
//辅助类，用于保存IP索引信息
public class CZ_INDEX_INFO
{
public UInt32 IpSet;
public UInt32 IpEnd;
public UInt32 Offset;

public CZ_INDEX_INFO()
{
IpSet = 0;
IpEnd = 0;
Offset = 0;
}
}

//读取纯真IP数据库类
public class PHCZIP
{
protected bool bFilePathInitialized;
protected string FilePath;
protected FileStream FileStrm;
protected UInt32 Index_Set;
protected UInt32 Index_End;
protected UInt32 Index_Count;
protected UInt32 Search_Index_Set;
protected UInt32 Search_Index_End;
protected CZ_INDEX_INFO Search_Set;
protected CZ_INDEX_INFO Search_Mid;
protected CZ_INDEX_INFO Search_End;

public PHCZIP()
{
bFilePathInitialized = false;
}

public PHCZIP(string dbFilePath)
{
bFilePathInitialized = false;
SetDbFilePath(dbFilePath);
}

//使用二分法查找索引区，初始化查找区间
public void Initialize()
{
Search_Index_Set = 0;
Search_Index_End = Index_Count - 1;
}

//关闭文件
public void Dispose()
{
if (bFilePathInitialized)
{
bFilePathInitialized = false;
FileStrm.Close();
//FileStrm.Dispose();
}

}

public bool SetDbFilePath(string dbFilePath)
{
if (dbFilePath == "")
{
return false;
}

try
{
FileStrm = new FileStream(dbFilePath, FileMode.Open, FileAccess.Read, FileShare.Read);
}
catch
{
return false;
}
//检查文件长度
if (FileStrm.Length < 8)
{
FileStrm.Close();
//FileStrm.Dispose();
return false;
}
//得到第一条索引的绝对偏移和最后一条索引的绝对偏移
FileStrm.Seek(0, SeekOrigin.Begin);
Index_Set = GetUInt32();
Index_End = GetUInt32();

//得到总索引条数
Index_Count = (Index_End - Index_Set) / 7 + 1;
bFilePathInitialized = true;

return true;
 
}

public string GetAddressWithIP(string IPValue)
{
if (!bFilePathInitialized)
return "";

Initialize();

UInt32 ip = IPToUInt32(IPValue);

while (true)
{

//首先初始化本轮查找的区间

//区间头
Search_Set = IndexInfoAtPos(Search_Index_Set);
//区间尾
Search_End = IndexInfoAtPos(Search_Index_End);

//判断IP是否在区间头内
if (ip >= Search_Set.IpSet && ip <= Search_Set.IpEnd)
return ReadAddressInfoAtOffset(Search_Set.Offset);

//判断IP是否在区间尾内
if (ip >= Search_End.IpSet && ip <= Search_End.IpEnd)
return ReadAddressInfoAtOffset(Search_End.Offset);

//计算出区间中点
Search_Mid = IndexInfoAtPos((Search_Index_End + Search_Index_Set) / 2);

//判断IP是否在中点
if (ip >= Search_Mid.IpSet && ip <= Search_Mid.IpEnd)
return ReadAddressInfoAtOffset(Search_Mid.Offset);

//本轮没有找到，准备下一轮
if (ip < Search_Mid.IpSet)
//IP比区间中点要小，将区间尾设为现在的中点，将区间缩小1倍。
Search_Index_End = (Search_Index_End + Search_Index_Set) / 2;
else
//IP比区间中点要大，将区间头设为现在的中点，将区间缩小1倍。
Search_Index_Set = (Search_Index_End + Search_Index_Set) / 2;
}
return "";

}

private string ReadAddressInfoAtOffset(UInt32 Offset)
{
string country = "";
string area = "";
UInt32 country_Offset = 0;
byte Tag = 0;
//跳过4字节，因这4个字节是该索引的IP区间上限。
FileStrm.Seek(Offset + 4, SeekOrigin.Begin);

//读取一个字节，得到描述国家信息的“寻址方式”
Tag = GetTag();

if (Tag == 0x01)
{

//模式0x01，表示接下来的3个字节是表示偏移位置
FileStrm.Seek(GetOffset(), SeekOrigin.Begin);

//继续检查“寻址方式”
Tag = GetTag();
if (Tag == 0x02)
{
//模式0x02，表示接下来的3个字节代表国家信息的偏移位置
//先将这个偏移位置保存起来，因为我们要读取它后面的地区信息。
country_Offset = GetOffset();
//读取地区信息（注：按照Luma的说明，好像没有这么多种可能性，但在测试过程中好像有些情况没有考虑到，
//所以写了个ReadArea()来读取。
area = ReadArea();
//读取国家信息
FileStrm.Seek(country_Offset, SeekOrigin.Begin);
country = ReadString();
}
else
{
//这种模式说明接下来就是保存的国家和地区信息了，以'/0'代表结束。
FileStrm.Seek(-1, SeekOrigin.Current);
country = ReadString();
area = ReadArea();

}
}
else if (Tag == 0x02)
{
//模式0x02，说明国家信息是一个偏移位置
country_Offset = GetOffset();
//先读取地区信息
area = ReadArea();
//读取国家信息
FileStrm.Seek(country_Offset, SeekOrigin.Begin);
country = ReadString();
}
else
{
//这种模式最简单了，直接读取国家和地区就OK了
FileStrm.Seek(-1, SeekOrigin.Current);
country = ReadString();
area = ReadArea();

}
string Address = country + " " + area;
return Address;

}

private UInt32 GetOffset()
{
byte[] TempByte4 = new byte[4];
TempByte4[0] = (byte)FileStrm.ReadByte();
TempByte4[1] = (byte)FileStrm.ReadByte();
TempByte4[2] = (byte)FileStrm.ReadByte();
TempByte4[3] = 0;
return BitConverter.ToUInt32(TempByte4, 0);
}

protected string ReadArea()
{
byte Tag = GetTag();

if (Tag == 0x01 || Tag == 0x02)
{
FileStrm.Seek(GetOffset(), SeekOrigin.Begin);
return ReadString();
}
else
{
FileStrm.Seek(-1, SeekOrigin.Current);
return ReadString();
}
}

protected string ReadString()
{
UInt32 Offset = 0;
byte[] TempByteArray = new byte[256];
TempByteArray[Offset] = (byte)FileStrm.ReadByte();
while (TempByteArray[Offset] != 0x00)
{
Offset += 1;
TempByteArray[Offset] = (byte)FileStrm.ReadByte();
}
return System.Text.Encoding.Default.GetString(TempByteArray).TrimEnd('/0');
}

protected byte GetTag()
{
return (byte)FileStrm.ReadByte();
}

protected CZ_INDEX_INFO IndexInfoAtPos(UInt32 Index_Pos)
{
CZ_INDEX_INFO Index_Info = new CZ_INDEX_INFO();
//根据索引编号计算出在文件中在偏移位置
FileStrm.Seek(Index_Set + 7 * Index_Pos, SeekOrigin.Begin);
Index_Info.IpSet = GetUInt32();
Index_Info.Offset = GetOffset();
FileStrm.Seek(Index_Info.Offset, SeekOrigin.Begin);
Index_Info.IpEnd = GetUInt32();

return Index_Info;
}

public UInt32 IPToUInt32(string IpValue)
{
string[] IpByte = IpValue.Split('.');
Int32 nUpperBound = IpByte.GetUpperBound(0);
if (nUpperBound != 3)
{
IpByte = new string[4];
for (Int32 i = 1; i <= 3 - nUpperBound; i++)
IpByte[nUpperBound + i] = "0";
}

byte[] TempByte4 = new byte[4];
for (Int32 i = 0; i <= 3; i++)
{
//'如果是.Net 2.0可以支持TryParse。
//'If Not (Byte.TryParse(IpByte(i), TempByte4(3 - i))) Then
//'    TempByte4(3 - i) = &H0
//'End If
if (IsNumeric(IpByte[i]))
TempByte4[3 - i] = (byte)(Convert.ToInt32(IpByte[i]) & 0xff);
}

return BitConverter.ToUInt32(TempByte4, 0);
}

protected bool IsNumeric(string str)
{
if (str != null && System.Text.RegularExpressions.Regex.IsMatch(str, @"^-?/d+$"))
return true;
else
return false;
}

protected UInt32 GetUInt32()
{
byte[] TempByte4 = new byte[4];
FileStrm.Read(TempByte4, 0, 4);
return BitConverter.ToUInt32(TempByte4, 0);
}
}