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
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
|
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
namespace SocketHttpListener.Net
{
// we use this static class as a helper class to encode/decode HTTP headers.
// what we need is a 1-1 correspondence between a char in the range U+0000-U+00FF
// and a byte in the range 0x00-0xFF (which is the range that can hit the network).
// The Latin-1 encoding (ISO-88591-1) (GetEncoding(28591)) works for byte[] to string, but is a little slow.
// It doesn't work for string -> byte[] because of best-fit-mapping problems.
internal static class WebHeaderEncoding
{
// We don't want '?' replacement characters, just fail.
private static readonly Encoding s_utf8Decoder = Encoding.GetEncoding("utf-8", EncoderFallback.ExceptionFallback, DecoderFallback.ExceptionFallback);
internal static unsafe string GetString(byte[] bytes, int byteIndex, int byteCount)
{
fixed (byte* pBytes = bytes)
return GetString(pBytes + byteIndex, byteCount);
}
internal static unsafe string GetString(byte* pBytes, int byteCount)
{
if (byteCount < 1)
return "";
string s = new string('\0', byteCount);
fixed (char* pStr = s)
{
char* pString = pStr;
while (byteCount >= 8)
{
pString[0] = (char)pBytes[0];
pString[1] = (char)pBytes[1];
pString[2] = (char)pBytes[2];
pString[3] = (char)pBytes[3];
pString[4] = (char)pBytes[4];
pString[5] = (char)pBytes[5];
pString[6] = (char)pBytes[6];
pString[7] = (char)pBytes[7];
pString += 8;
pBytes += 8;
byteCount -= 8;
}
for (int i = 0; i < byteCount; i++)
{
pString[i] = (char)pBytes[i];
}
}
return s;
}
internal static int GetByteCount(string myString)
{
return myString.Length;
}
internal static unsafe void GetBytes(string myString, int charIndex, int charCount, byte[] bytes, int byteIndex)
{
if (myString.Length == 0)
{
return;
}
fixed (byte* bufferPointer = bytes)
{
byte* newBufferPointer = bufferPointer + byteIndex;
int finalIndex = charIndex + charCount;
while (charIndex < finalIndex)
{
*newBufferPointer++ = (byte)myString[charIndex++];
}
}
}
internal static unsafe byte[] GetBytes(string myString)
{
byte[] bytes = new byte[myString.Length];
if (myString.Length != 0)
{
GetBytes(myString, 0, myString.Length, bytes, 0);
}
return bytes;
}
// The normal client header parser just casts bytes to chars (see GetString).
// Check if those bytes were actually utf-8 instead of ASCII.
// If not, just return the input value.
internal static string DecodeUtf8FromString(string input)
{
if (string.IsNullOrWhiteSpace(input))
{
return input;
}
bool possibleUtf8 = false;
for (int i = 0; i < input.Length; i++)
{
if (input[i] > (char)255)
{
return input; // This couldn't have come from the wire, someone assigned it directly.
}
else if (input[i] > (char)127)
{
possibleUtf8 = true;
break;
}
}
if (possibleUtf8)
{
byte[] rawBytes = new byte[input.Length];
for (int i = 0; i < input.Length; i++)
{
if (input[i] > (char)255)
{
return input; // This couldn't have come from the wire, someone assigned it directly.
}
rawBytes[i] = (byte)input[i];
}
try
{
return s_utf8Decoder.GetString(rawBytes);
}
catch (ArgumentException) { } // Not actually Utf-8
}
return input;
}
}
}
|
