/**
* Calculate the length (in bytes) of a UTF-8 character
*
* \param s Pointer to start of character
* \return Length of character, in bytes
*/
size_t utf8_char_byte_length(const char *s)
{
size_t len;
parserutils_error perror;
perror = parserutils_charset_utf8_char_byte_length((const uint8_t *) s,
&len);
assert(perror == PARSERUTILS_OK);
return len;
}
/** Convert the given string to lowercase
*
* \param source
* \param ascii_only Whether to only convert [a-z] to [A-Z]
* \param lower Result pointer for lowercase string. Caller owns ref
*
* \return DOM_NO_ERR on success.
*
* \note Right now, will return DOM_NOT_SUPPORTED_ERR if ascii_only is false.
*/
dom_exception
dom_string_tolower(dom_string *source, bool ascii_only, dom_string **lower)
{
const uint8_t *orig_s = (const uint8_t *) dom_string_data(source);
const size_t nbytes = dom_string_byte_length(source);
uint8_t *copy_s;
size_t index = 0, clen;
parserutils_error err;
dom_exception exc;
if (ascii_only == false)
return DOM_NOT_SUPPORTED_ERR;
copy_s = malloc(nbytes);
if (copy_s == NULL)
return DOM_NO_MEM_ERR;
memcpy(copy_s, orig_s, nbytes);
while (index < nbytes) {
err = parserutils_charset_utf8_char_byte_length(orig_s + index,
&clen);
if (err != PARSERUTILS_OK) {
free(copy_s);
/** \todo Find a better exception */
return DOM_NO_MEM_ERR;
}
if (clen == 1) {
if (orig_s[index] >= 'A' &&
orig_s[index] <= 'Z')
copy_s[index] += 'a' - 'A';
}
index += clen;
}
if (((dom_string_internal*)source)->type == DOM_STRING_CDATA) {
exc = dom_string_create(copy_s, nbytes, lower);
} else {
exc = dom_string_create_interned(copy_s, nbytes, lower);
}
free(copy_s);
return exc;
}
/**
* Get the UCS4 character at position index
*
* \param index The position of the charater
* \param ch The UCS4 character
* \return DOM_NO_ERR on success, appropriate dom_exception on failure.
*/
dom_exception dom_string_at(dom_string *str, uint32_t index,
uint32_t *ch)
{
const uint8_t *s;
size_t clen, slen;
uint32_t c, i;
parserutils_error err;
s = (const uint8_t *) dom_string_data(str);
slen = dom_string_byte_length(str);
i = 0;
while (slen > 0) {
err = parserutils_charset_utf8_char_byte_length(s, &clen);
if (err != PARSERUTILS_OK) {
return (uint32_t) -1;
}
i++;
if (i == index + 1)
break;
s += clen;
slen -= clen;
}
if (i == index + 1) {
err = parserutils_charset_utf8_to_ucs4(s, slen, &c, &clen);
if (err != PARSERUTILS_OK) {
return (uint32_t) -1;
}
*ch = c;
return DOM_NO_ERR;
} else {
return DOM_DOMSTRING_SIZE_ERR;
}
}
