Created
March 31, 2023 14:41
-
-
Save tslater2006/1e8a304976ae3267f7a72ebd221ff43a to your computer and use it in GitHub Desktop.
Converts from Unicode codepoint to UTF 8 and vice versa. Courtesy of GPT-4
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
using System.Text; | |
static byte[] UnicodeCodepointToUtf8(int codepoint) | |
{ | |
if (codepoint >= 0 && codepoint <= 127) | |
{ | |
return new byte[] { (byte)codepoint }; | |
} | |
else if (codepoint >= 128 && codepoint <= 2047) | |
{ | |
byte firstByte = (byte)(192 + (codepoint / 64)); | |
byte secondByte = (byte)(128 + (codepoint % 64)); | |
return new byte[] { firstByte, secondByte }; | |
} | |
else if (codepoint >= 2048 && codepoint <= 65535) | |
{ | |
byte firstByte = (byte)(224 + (codepoint / 4096)); | |
int remainder = codepoint % 4096; | |
byte secondByte = (byte)(128 + (remainder / 64)); | |
byte thirdByte = (byte)(128 + (remainder % 64)); | |
return new byte[] { firstByte, secondByte, thirdByte }; | |
} | |
else if (codepoint >= 65536 && codepoint <= 1114111) | |
{ | |
byte firstByte = (byte)(240 + (codepoint / 262144)); | |
int remainder = codepoint % 262144; | |
byte secondByte = (byte)(128 + (remainder / 4096)); | |
remainder = remainder % 4096; | |
byte thirdByte = (byte)(128 + (remainder / 64)); | |
byte fourthByte = (byte)(128 + (remainder % 64)); | |
return new byte[] { firstByte, secondByte, thirdByte, fourthByte }; | |
} | |
else | |
{ | |
throw new ArgumentOutOfRangeException("Invalid Unicode codepoint"); | |
} | |
} | |
static uint Utf8BytesToCodepoint(MemoryStream ms) | |
{ | |
int firstByte = ms.ReadByte(); | |
if (firstByte < 0) return 0; // End of stream | |
uint codepoint = 0; | |
int additionalBytes = 0; | |
if (firstByte < 128) | |
{ | |
codepoint = (uint)firstByte; | |
} | |
else if (firstByte >= 192 && firstByte < 224) | |
{ | |
codepoint = (uint)(firstByte - 192); | |
additionalBytes = 1; | |
} | |
else if (firstByte >= 224 && firstByte < 240) | |
{ | |
codepoint = (uint)(firstByte - 224); | |
additionalBytes = 2; | |
} | |
else if (firstByte >= 240 && firstByte < 248) | |
{ | |
codepoint = (uint)(firstByte - 240); | |
additionalBytes = 3; | |
} | |
else | |
{ | |
throw new ArgumentException("Invalid UTF-8 byte sequence."); | |
} | |
for (int i = 0; i < additionalBytes; i++) | |
{ | |
int nextByte = ms.ReadByte(); | |
if (nextByte < 128 || nextByte >= 192) | |
{ | |
throw new ArgumentException("Invalid UTF-8 byte sequence."); | |
} | |
codepoint = codepoint * 64 + (uint)(nextByte - 128); | |
} | |
return codepoint; | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment