Parsee/src/Unistr.c

#include <Unistring.h>

#include <Cytoplasm/Memory.h>
#include <Cytoplasm/Str.h>
#include <Cytoplasm/Log.h>
#include <Cytoplasm/Log.h>

#include <string.h>
#include <stdarg.h>

struct Unistr {
    size_t length;
    uint32_t *codepoints;
};

void 
UnistrAddch(Unistr *unistr, uint32_t u)
{
    if (!unistr || !u)
    {
        return;
    }
    unistr->length++;
    unistr->codepoints = Realloc(
        unistr->codepoints, 
        unistr->length * sizeof(*unistr->codepoints)
    );

    unistr->codepoints[unistr->length - 1] = u;
}

static bool
UTFIsN(char *off, size_t available, size_t n, uint8_t pc)
{
    size_t i;
    uint8_t *offu = (uint8_t *) off;
    if (((available < n) || ((*offu >> (8-n-1)) != pc)) && (n >= 1))
    {
        return false;
    }

    for (i = 0; i < n - 1; i++)
    {
        if ((offu[i+1] >> 6) != 0x2)
        {
            return false;
        }
    }
    return true;
}

Unistr *
UnistrCreate(char *src)
{
    size_t len, i;
    Unistr *str;
    if (!src)
    {
        return NULL;
    }

    len = strlen(src);
    str = Malloc(sizeof(*str));
    str->length = 0;
    str->codepoints = NULL;

    /* We can't just set the length to {len}. */
    for (i = 0; i < len; i++)
    {
        char byte = src[i];
        size_t available = len - i;
        if ((byte & 0x80) == 0)
        {
            /* This is a regular codepoint */
            UnistrAddch(str, byte & 0x7F);
            continue;
        }
        else if (UTFIsN(&src[i], available, 2, 0x06))
        {
            char a = src[i+0] & 0x1F;
            char b = src[i+1] & 0x3F;
            uint32_t u = (a << (6 * 1)) | b;

            /* Overlongs are errors. */
            if (u < 0x0080 || u > 0x07FF)
            {
                UnistrFree(str);
                return NULL;
            }

            UnistrAddch(str, u);
            i += 2 - 1;
            continue;
        }
        else if (UTFIsN(&src[i], available, 3, 0x0E))
        {
            char a = src[i+0] & 0x0F;
            char b = src[i+1] & 0x3F;
            char c = src[i+2] & 0x3F;
            uint32_t u = 
                (a << (6 * 2)) | 
                (b << (6 * 1)) | 
                (c << (6 * 0)) ;

            /* Overlongs are errors. */
            if (u < 0x0800 || u > 0xFFFF)
            {
                UnistrFree(str);
                return NULL;
            }

            UnistrAddch(str, u);
            i += 3 - 1;
            continue;
        }
        else if (UTFIsN(&src[i], available, 4, 0x1E))
        {
            char a = src[i+0] & 0x07;
            char b = src[i+1] & 0x3F;
            char c = src[i+2] & 0x3F;
            char d = src[i+3] & 0x3F;
            uint32_t u = 
                (a << (6 * 3)) | 
                (b << (6 * 2)) | 
                (c << (6 * 1)) |
                (d << (6 * 0)) ;

            /* Overlongs are errors. */
            if (u < 0x10000 || u > 0x10FFFF)
            {
                UnistrFree(str);
                return NULL;
            }

            UnistrAddch(str, u);
            i += 4 - 1;
            continue;
            
        }
    }

    return str;
}
void 
UnistrFree(Unistr *unistr)
{
    if (!unistr)
    {
        return;
    }

    Free(unistr->codepoints);
    Free(unistr);
}
char * 
UnistrC(Unistr *unistr)
{
    char *ret, *tmp, *utf;
    size_t i;
    if (!unistr)
    {
        return NULL;
    }

    ret = NULL;
    for (i = 0; i < unistr->length; i++)
    {
        uint32_t code = unistr->codepoints[i];
        utf = StrUtf8Encode(code);

        tmp = ret;
        ret = StrConcat(2, ret, utf);
        Free(tmp);
        Free(utf);
    }

    return ret;
}
size_t
UnistrSize(Unistr *unistr)
{
    return unistr ? unistr->length : 0;
}
uint32_t 
UnistrGetch(Unistr *unistr, size_t i)
{
    if (!unistr)
    {
        return 0;
    }

    return i < unistr->length ? unistr->codepoints[i] : 0;
}
bool
UnistrIsASCII(uint32_t u)
{
    if (u == 0)
    {
        return NULL;
    }

    return u < 0x7F;
}
bool
UnistrIsBMP(uint32_t u)
{
    if (u == 0)
    {
        return NULL;
    }

    return u <= 0xFFFF;
}
Unistr *
UnistrFilter(Unistr *str, UnistrFilterFunc filter)
{
    Unistr *unistr;
    size_t i;
    if (!str || !filter)
    {
        return NULL;
    }

    unistr = UnistrCreate("");
    for (i = 0; i < UnistrSize(str); i++)
    {
        uint32_t code = UnistrGetch(str, i);
        if (!filter(code))
        {
            continue;
        }
        UnistrAddch(unistr, code);
    }

    return unistr;
}

Unistr *
UnistrConcat(size_t n, ...)
{
    va_list list;
    size_t i;
    Unistr *ret = UnistrCreate("");

    va_start(list, n);
    for (i = 0; i < n; i++)
    {
        Unistr *to_concat = va_arg(list, Unistr *);
        size_t j;
        for (j = 0; j < UnistrSize(to_concat); j++)
        {
            UnistrAddch(ret, UnistrGetch(to_concat, j));
        }
    }

    va_end(list);
    return ret;
}
size_t 
UnistrGetOffset(Unistr *str, uint32_t sep)
{
    size_t i;
    uint32_t prev = 0x0A;
    if (!str || !sep)
    {
        return 0;
    }

    for (i = 0; i < str->length; i++)
    {
        uint32_t curr = str->codepoints[i];
        if (prev == 0x0A && curr != sep)
        {
            return i;
        }
        prev = curr;
    }
    return 0;
}