Base/core/src-c/runes.c

#include <librr/runes.h>

isz_t rr_distance_to_last_utf8_rune_start(const char *string, isz_t offset)
{
    usz_t                       bytes_walked                    = 0;
    while((offset - bytes_walked) >= 0)
    {
        if(bytes_walked > 4)
            return -1;
        
        if((string[offset - bytes_walked] >> 6) != 0b10)
        {
            return bytes_walked;
        }
        ++bytes_walked;
    }
    return -2;
}

isz_t rr_identify_utf8_rune_length(const char *string, usz_t offset)
{
    char                        head_byte                       = string[offset];
    
    // If this is ASCII
    if((head_byte & (1 << 7)) == 0)
        return  1;
    
    // UTF-8 - only

    usz_t                       length                          = 0;
    while(length < 5)
    {
        head_byte                     <<= 1;
        if((head_byte & (1 << 7)) == 0)
            break;
        ++length;
    }
    
    if(length > 4)
        return -1;
    
    if(length < 2)
        return -2;
    
    return length;
}

rune_t rr_postprocess_utf8_head_byte(char byte, usz_t num_bytes)
{
    switch(num_bytes)
    {
        case 1: return byte;
        case 2: return byte & 0b11100000;
        case 3: return byte & 0b11110000;
        case 4: return byte & 0b11111000;
    }
    return 0;
}

rune_t rr_postprocess_utf8_bytes(const char *bytes, usz_t num_bytes)
{
    rune_t                      result                          = rr_postprocess_utf8_head_byte(bytes[0], num_bytes);

    usz_t                       byte_index                      = 1;
    while(byte_index < num_bytes)
    {
        result                    <<= 6;
        result                     |= bytes[byte_index] & 0b00111111;
        ++byte_index;
    }
    return result;
}

rune_t rr_extract_utf8(const char *string, usz_t offset, usz_t *increase)
{
    if(string[offset] == 0x00)
        return ZERO;
    
    isz_t                       offset_into_rune                = rr_distance_to_last_utf8_rune_start(string, offset);
    if(offset_into_rune < 0)
        return ZERO;
    offset                         -= offset_into_rune;

    isz_t                       rune_length                     = rr_identify_utf8_rune_length(string, offset);
    if(rune_length < 0)
        return ZERO;
    
    *increase                      += rune_length - offset_into_rune;
    return rr_postprocess_utf8_bytes(&string[offset], rune_length);
}

bool_t rr_check_newline(const char *string, usz_t offset, usz_t *next)
{
    rune_t                      subject                         = rr_extract_utf8(string, offset, &offset);
    if(subject == '\r')
    {
        usz_t                       offset_backup                   = offset;
        if(rr_extract_utf8(string, offset, &offset) != '\n')
            (*next)                         = offset_backup;
        return TRUE;
    }
    if(subject == '\n')
    {
        (*next)                         = offset;
        return TRUE;
    }
    return FALSE;
}

bool_t rr_rune_is_lower(rune_t rune)
{
    if(rune < 'a') return FALSE;
    if(rune > 'z') return FALSE;
    return TRUE;
}

bool_t rr_rune_is_upper(rune_t rune)
{
    if(rune < 'A') return FALSE;
    if(rune > 'Z') return FALSE;
    return TRUE;
}

bool_t rr_rune_is_letter(rune_t rune)
{
    if(rr_rune_is_lower(rune)) return TRUE;
    if(rr_rune_is_upper(rune)) return TRUE;
    return FALSE;
}

bool_t rr_rune_is_digit(rune_t rune)
{
    if(rune < '0') return FALSE;
    if(rune > '9') return FALSE;
    return TRUE;
}

bool_t rr_rune_is_in_ascii_special_block_1(rune_t rune)
{
    if(rune < '!') return FALSE;
    if(rune > '/') return FALSE;
    return TRUE;
}

bool_t rr_rune_is_in_ascii_special_block_2(rune_t rune)
{
    if(rune < ':') return FALSE;
    if(rune > '@') return FALSE;
    return TRUE;
}

bool_t rr_rune_is_in_ascii_special_block_3(rune_t rune)
{
    if(rune < '[') return FALSE;
    if(rune > '`') return FALSE;
    return TRUE;
}

bool_t rr_rune_is_in_ascii_special_block_4(rune_t rune)
{
    if(rune < '{') return FALSE;
    if(rune > '~') return FALSE;
    return TRUE;
}

bool_t rr_rune_is_ascii_special(rune_t rune)
{
    if(rr_rune_is_in_ascii_special_block_1(rune)) return TRUE;
    if(rr_rune_is_in_ascii_special_block_2(rune)) return TRUE;
    if(rr_rune_is_in_ascii_special_block_3(rune)) return TRUE;
    if(rr_rune_is_in_ascii_special_block_4(rune)) return TRUE;
    return FALSE;
}

rr_ascii_sign_e rr_rune_to_ascii_sign(rune_t rune)
{
    switch(rune)
    {
        case '!':  return RR_ASCII_EXCLAMATION_MARK;
        case '"':  return RR_ASCII_DOUBLE_QUOTATION_MARK;
        case '#':  return RR_ASCII_HASH_SIGN;
        case '$':  return RR_ASCII_DOLLAR_SIGN;
        case '%':  return RR_ASCII_PERCENT_SIGN;
        case '&':  return RR_ASCII_AMPERSAND;
        case '\'': return RR_ASCII_SINGLE_QUOTATION_MARK;
        case '(':  return RR_ASCII_OPENING_PARENTHESIS;
        case ')':  return RR_ASCII_CLOSING_PARENTHESIS;
        case '*':  return RR_ASCII_ASTERISK;
        case '+':  return RR_ASCII_PLUS;
        case ',':  return RR_ASCII_COMMA;
        case '-':  return RR_ASCII_MINUS;
        case '.':  return RR_ASCII_POINT;
        case '/':  return RR_ASCII_SLASH;
        case ':':  return RR_ASCII_COLON;
        case ';':  return RR_ASCII_SEMICOLON;
        case '<':  return RR_ASCII_SMALLER_THAN;
        case '=':  return RR_ASCII_EQUALS_SIGN;
        case '>':  return RR_ASCII_BIGGER_THAN;
        case '?':  return RR_ASCII_QUESTION_MARK;
        case '@':  return RR_ASCII_AT_SIGN;
        case '[':  return RR_ASCII_OPENING_SQUARE_BRACKET;
        case '\\': return RR_ASCII_BACKSLASH;
        case ']':  return RR_ASCII_CLOSING_SQUARE_BRACKET;
        case '^':  return RR_ASCII_CIRCUMFLEX;
        case '_':  return RR_ASCII_UNDERSCORE;
        case '`':  return RR_ASCII_TICK;
        case '{':  return RR_ASCII_OPENING_CURLY_BRACE;
        case '|':  return RR_ASCII_VERTICAL_BAR;
        case '}':  return RR_ASCII_CLOSING_CURLY_BRACE;
        case '~':  return RR_ASCII_TILDE;
    }
    return RR_ASCII_NOT_A_SIGN;
}