G'day everyone,
Well it seems I'm not as intelligent as I once thought. I really need help.
The code below is from the Unicode Consortium and converts UTF32 to UTF16. This is one of 6 similar routines. I figure if can get help with the first, I should be able to wriggle through the rest without losing too much skin.
sourceStart is a pointer to a pointer to an array of unsigned 32 bit values.
sourceEnd is a pointer to the end of the same array.
targetStart is a pointer to a pointer to an array of unsigned 16 bit values.
targetEnd is a pointer to the end of the same array.
flags is an enum, strictConversion or lenientConversion.
This stuff is from the header file and the main source:
typedef unsigned long UTF32; /* at least 32 bits */
typedef unsigned short UTF16; /* at least 16 bits */
typedef unsigned char UTF8; /* typically 8 bits */
typedef unsigned char Boolean; /* 0 or 1 */
/* Some fundamental constants */
#define UNI_REPLACEMENT_CHAR (UTF32)0x0000FFFD
#define UNI_MAX_BMP (UTF32)0x0000FFFF
#define UNI_MAX_UTF16 (UTF32)0x0010FFFF
#define UNI_MAX_UTF32 (UTF32)0x7FFFFFFF
#define UNI_MAX_LEGAL_UTF32 (UTF32)0x0010FFFF
typedef enum {
conversionOK, /* conversion successful */
sourceExhausted, /* partial character in source, but hit end */
targetExhausted, /* insuff. room in target for conversion */
sourceIllegal /* source sequence is illegal/malformed */
} ConversionResult;
typedef enum {
strictConversion = 0,
lenientConversion
} ConversionFlags;
static const int halfShift = 10; /* used for shifting by 10 bits */
static const UTF32 halfBase = 0x0010000UL;
static const UTF32 halfMask = 0x3FFUL;
#define UNI_SUR_HIGH_START (UTF32)0xD800
#define UNI_SUR_HIGH_END (UTF32)0xDBFF
#define UNI_SUR_LOW_START (UTF32)0xDC00
#define UNI_SUR_LOW_END (UTF32)0xDFFF
#define false 0
#define true 1
My biggest hassle at the moment is how to handle the incoming pointer to pointer to array parameters.
Any ideas?
Kind regards,
Bruce.
actual routine
ConversionResult ConvertUTF32toUTF16 (
const UTF32** sourceStart, const UTF32* sourceEnd,
UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) {
ConversionResult result = conversionOK;
const UTF32* source = *sourceStart;
UTF16* target = *targetStart;
while (source < sourceEnd) {
UTF32 ch;
if (target >= targetEnd) {
result = targetExhausted; break;
}
ch = *source++;
if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */
/* UTF-16 surrogate values are illegal in UTF-32; 0xffff or 0xfffe are both reserved values */
if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
if (flags == strictConversion) {
--source; /* return to the illegal value itself */
result = sourceIllegal;
break;
} else {
*target++ = UNI_REPLACEMENT_CHAR;
}
} else {
*target++ = (UTF16)ch; /* normal case */
}
} else if (ch > UNI_MAX_LEGAL_UTF32) {
if (flags == strictConversion) {
result = sourceIllegal;
} else {
*target++ = UNI_REPLACEMENT_CHAR;
}
} else {
/* target is a character in range 0xFFFF - 0x10FFFF. */
if (target + 1 >= targetEnd) {
--source; /* Back up source pointer! */
result = targetExhausted; break;
}
ch -= halfBase;
*target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START);
*target++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START);
}
}
*sourceStart = source;
*targetStart = target;
return result;
}
UTF32** sourceStart
mov eax,sourceStart ; sourceStart
mov ecx,[eax] ; *sourceStart
mov eax,[ecx] ; **sourceStart
You could do it all through eax, but to access the next array pointer all you need to do is:
add ecx,4
mov eax,[ecx]
...and so on for each element of the array.
That simple??!! Wow. So much for that panic-fest.
Thanks.
Bruce.
It's okay, pointers seem to cause people so many problems, but when you get to it low-level they're pretty simple :wink