Usermode - Moved UTF-8 handling into a library

[tpg/acess2.git] / Usermode / Libraries / libunicode.so_src / utf-8.c

diff --git a/Usermode/Libraries/libunicode.so_src/utf-8.c b/Usermode/Libraries/libunicode.so_src/utf-8.c
new file mode 100644 (file)
index 0000000..3aa9d1a
--- /dev/null
+++ b/Usermode/Libraries/libunicode.so_src/utf-8.c
@@ -0,0 +1,153 @@
+/*
+ * Acess2 "libunicode" UTF Parser
+ * - By John Hodge (thePowersGang)
+ *
+ * utf-8.c
+ * - UTF-8 Parsing code
+ */
+#include <stdint.h>
+#include <unicode.h>
+
+/**
+ * \brief Read a UTF-8 character from a string
+ * \param Input        Source UTF-8 encoded string
+ * \param Val  Destination for read codepoint
+ * \return Number of bytes read/used
+ */
+int ReadUTF8(const char *Input, uint32_t *Val)
+{
+       const uint8_t   *str = (const uint8_t *)Input;
+       *Val = 0xFFFD;  // Assume invalid character
+       
+       // ASCII
+       if( !(*str & 0x80) ) {
+               *Val = *str;
+               return 1;
+       }
+       
+       // Middle of a sequence
+       if( (*str & 0xC0) == 0x80 ) {
+               return 1;
+       }
+       
+       // Two Byte
+       if( (*str & 0xE0) == 0xC0 ) {
+               *Val = (*str & 0x1F) << 6;      // Upper 6 Bits
+               str ++;
+               if( (*str & 0xC0) != 0x80)      return -1;      // Validity check
+               *Val |= (*str & 0x3F);  // Lower 6 Bits
+               return 2;
+       }
+       
+       // Three Byte
+       if( (*str & 0xF0) == 0xE0 ) {
+               *Val = (*str & 0x0F) << 12;     // Upper 4 Bits
+               str ++;
+               if( (*str & 0xC0) != 0x80)      return -1;      // Validity check
+               *Val |= (*str & 0x3F) << 6;     // Middle 6 Bits
+               str ++;
+               if( (*str & 0xC0) != 0x80)      return -1;      // Validity check
+               *Val |= (*str & 0x3F);  // Lower 6 Bits
+               return 3;
+       }
+       
+       // Four Byte
+       if( (*str & 0xF8) == 0xF0 ) {
+               *Val = (*str & 0x07) << 18;     // Upper 3 Bits
+               str ++;
+               if( (*str & 0xC0) != 0x80)      return -1;      // Validity check
+               *Val |= (*str & 0x3F) << 12;    // Middle-upper 6 Bits
+               str ++;
+               if( (*str & 0xC0) != 0x80)      return -1;      // Validity check
+               *Val |= (*str & 0x3F) << 6;     // Middle-lower 6 Bits
+               str ++;
+               if( (*str & 0xC0) != 0x80)      return -1;      // Validity check
+               *Val |= (*str & 0x3F);  // Lower 6 Bits
+               return 4;
+       }
+       
+       // UTF-8 Doesn't support more than four bytes
+       return 4;
+}
+
+/**
+ * \brief Get the UTF-8 character before the 
+ * \
+ */
+int ReadUTF8Rev(const char *Base, int Offset, uint32_t *Val)
+{
+        int    len = 0;
+       
+       // Scan backwards for the beginning of the character
+       while( Offset > 0 && (Base[Offset--] & 0xC0) == 0x80 )
+               len ++;
+       // Invalid string (no beginning)
+       if(Offset == 0 && (Base[Offset] & 0xC0) == 0x80 )
+               return len;
+       
+       len ++; // First character
+       if( ReadUTF8(Base+Offset, Val) != len ) {
+               *Val = 0xFFFD;
+       }
+       return len;
+}
+
+/**
+ * \brief Write a UTF-8 character sequence to a string
+ * \param buf  Destination buffer (must have at least 4 bytes available)
+ * \param Val  Unicode codepoint to write
+ * \return Number of bytes written
+ * \note Does not NULL terminate the string in \a buf
+ */
+int WriteUTF8(char *buf, uint32_t Val)
+{
+       uint8_t *str = (void*)buf;
+       
+       // ASCII
+       if( Val < 128 ) {
+               if(str) {
+                       *str = Val;
+               }
+               return 1;
+       }
+       
+       // Two Byte
+       if( Val < 0x8000 ) {
+               if(str) {
+                       *str = 0xC0 | (Val >> 6);
+                       str ++;
+                       *str = 0x80 | (Val & 0x3F);
+               }
+               return 2;
+       }
+       
+       // Three Byte
+       if( Val < 0x10000 ) {
+               if(str) {
+                       *str = 0xE0 | (Val >> 12);
+                       str ++;
+                       *str = 0x80 | ((Val >> 6) & 0x3F);
+                       str ++;
+                       *str = 0x80 | (Val & 0x3F);
+               }
+               return 3;
+       }
+       
+       // Four Byte
+       if( Val < 0x110000 ) {
+               if(str) {
+                       *str = 0xF0 | (Val >> 18);
+                       str ++;
+                       *str = 0x80 | ((Val >> 12) & 0x3F);
+                       str ++;
+                       *str = 0x80 | ((Val >> 6) & 0x3F);
+                       str ++;
+                       *str = 0x80 | (Val & 0x3F);
+               }
+               return 4;
+       }
+       
+       // UTF-8 Doesn't support more than four bytes
+       return 0;
+}
+