#include "utf8.h"


#define UTF8_ONE_BYTE_MASK 0b10000000
#define UTF8_ONE_BYTE_COUNT 0

#define UTF8_TWO_BYTE_MASK 0b11100000
#define UTF8_TWO_BYTE_COUNT 0b11000000

#define UTF8_THREE_BYTE_MASK 0b11110000
#define UTF8_THREE_BYTE_COUNT 0b11100000

#define UTF8_FOUR_BYTE_MASK 0b11111000
#define UTF8_FOUR_BYTE_COUNT 0b11110000

// This one could use a better name, I just don't know a better one (yet?)
#define UTF8_OTHER_MASK 0b00111111

size_t utf8_codepoint_size(uint8_t text)
{
	if((text & UTF8_ONE_BYTE_MASK) == UTF8_ONE_BYTE_COUNT) {
		return 1;
	}

	if((text & UTF8_TWO_BYTE_MASK) == UTF8_TWO_BYTE_COUNT) {
		return 2;
	}

	if((text & UTF8_THREE_BYTE_MASK) == UTF8_THREE_BYTE_COUNT) {
        	return 3;
	}
	return 4;
}

/*
	Compute the size (number of visible chars) of an UTF-8 string.
	For each character, the check function is called.
	If it returns false, utf8_scan end and return -1.

	Normally, returns the number of characters in the string.

	It also validate that UTF-8 character of size 2, 3 and 4 contains confirming content.
	It returns -1 if not confirming.

	So it performs basic validation of an UTF-8 string plus custom validation of ASCII chars.
*/
int utf8_scan (std::string_view s, std::function<bool(int size, const char firstcar)> check)
{
	int ret = 0;
	size_t s_size = s.size();
	auto pt = s.begin();
	for (unsigned i=0; i<s_size; i++, pt++){
		char c = *pt;
		size_t size = utf8_codepoint_size(c);
		if (i+size <= s_size){
			if (!check(size,c)){
				ret = -1;
				break;
			}else{
				ret++;
				if (size > 1){
					size--;
					i += size;
					pt++;
					for (unsigned j=0; j<size; j++, pt++){
						if (*pt < 0x80 || *pt >= 0xc0){
							ret = -1;
							break;
						}
					}
					if (ret == -1) break;
					pt--;
				}
			}
		}else{
			// Incomplete UTF-8 character
			ret = -1;
			break;
		}
	}
	return ret;
}

#ifdef TEST

#include <stdio.h>
#include <stdlib.h>
#include <ctype.h>
#include <string.h>
#include <tlmplib.h>
#include <string>

using namespace std;

<mod>
int main (int argc, char *argv[])
{
	glocal int ret = -1;
	glocal.ret = <call tlmpprogram>(argc,argv);
	<f init>
		setproginfo ("","0.0","...");
	</f>
	<f main>
		int ret = -1;
		for (int i=0; i<argc; i++){
			int len = utf8_scan(argv[i],[](int size, const char car){
				return true;
			});
			printf ("just count len = %d\n",len);
			len = utf8_scan(argv[i],[](int size, const char car){
				bool ret = false;
				if (size == 1 && isalpha(car)){
					ret = true;
				}else if (size > 1){
					ret = true;
				}
				return ret;
			});
			printf ("accept only alpha + multi character UTF-8len = %d\n",len);
		}
		return ret;
	</f>
	</call>
	return glocal.ret;
}
</mod>

#endif