Function chr_encode

kernel/generic/src/lib/str.c:217–271

Encode a single character to string representation. * * Encode a single character to string representation (i.e. UTF-8) and store * it into a buffer at @a offset. Encoding starts at @a offset and this offset * is moved to the position where the next character can be written to. * * @param ch Input character. * @param str Output buffer. * @param offset Byte offset where to start writ

 *         code was invalid.
 */
errno_t chr_encode(const char32_t ch, char *str, size_t *offset, size_t size)
{
	if (*offset >= size)
		return EOVERFLOW;

	if (!chr_check(ch))
		return EINVAL;

	/*
	 * Unsigned version of ch (bit operations should only be done
	 * on unsigned types).
	 */
	uint32_t cc = (uint32_t) ch;

	/* Determine how many continuation bytes are needed */

	unsigned int b0_bits;  /* Data bits in first byte */
	unsigned int cbytes;   /* Number of continuation bytes */

	if ((cc & ~LO_MASK_32(7)) == 0) {
		b0_bits = 7;
		cbytes = 0;
	} else if ((cc & ~LO_MASK_32(11)) == 0) {
		b0_bits = 5;
		cbytes = 1;
	} else if ((cc & ~LO_MASK_32(16)) == 0) {
		b0_bits = 4;
		cbytes = 2;
	} else if ((cc & ~LO_MASK_32(21)) == 0) {
		b0_bits = 3;
		cbytes = 3;
	} else {
		/* Codes longer than 21 bits are not supported */
		return EINVAL;
	}

	/* Check for available space in buffer */
	if (*offset + cbytes >= size)
		return EOVERFLOW;

	/* Encode continuation bytes */
	unsigned int i;
	for (i = cbytes; i > 0; i--) {
		str[*offset + i] = 0x80 | (cc & LO_MASK_32(CONT_BITS));
		cc = cc >> CONT_BITS;
	}

	/* Encode first byte */
	str[*offset] = (cc & LO_MASK_32(b0_bits)) | HI_MASK_8(8 - b0_bits - 1);

	/* Advance offset */
	*offset += cbytes + 1;

	return EOK;
}

/** Get size of string.
 *