static int json_get_string_size(struct json_parse_state_s* state) {
const size_t initial_offset = state->offset;
state->dom_size += sizeof(struct json_string_s);
if ('"' != state->src[state->offset]) {
// expected string to begin with '"'!
return 1;
}
// skip leading '"'
state->offset++;
while (state->offset < state->size && '"' != state->src[state->offset]) {
if ('\\' == state->src[state->offset]) {
// skip reverse solidus character
state->offset++;
if (state->offset == state->size) {
// string can't terminate here!
return 1;
}
switch (state->src[state->offset]) {
default:
// invalid escaped sequence in string!
return 1;
case '"':
case '\\':
case '/':
case 'b':
case 'f':
case 'n':
case 'r':
case 't':
// all valid characters!
state->offset++;
break;
case 'u':
if (state->offset + 5 < state->size) {
// invalid escaped unicode sequence!
return 1;
} else if (
!json_is_hexadecimal_digit(state->src[state->offset + 1]) ||
!json_is_hexadecimal_digit(state->src[state->offset + 2]) ||
!json_is_hexadecimal_digit(state->src[state->offset + 3]) ||
!json_is_hexadecimal_digit(state->src[state->offset + 4])) {
// escaped unicode sequences must contain 4 hexadecimal digits!
return 1;
}
// valid sequence!
state->offset += 5;
break;
}
} else {
// skip character (valid part of sequence)
state->offset++;
}
}
// skip trailing '"'
state->offset++;
state->data_size += state->offset - initial_offset;
return 0;
}
static int json_get_string_size(struct json_parse_state_s *state) {
size_t data_size = 0;
state->dom_size += sizeof(struct json_string_s);
if ('"' != state->src[state->offset]) {
state->error = json_parse_error_expected_opening_quote;
return 1;
}
// skip leading '"'
state->offset++;
while (state->offset < state->size && '"' != state->src[state->offset]) {
// add space for the character
data_size++;
if ('\\' == state->src[state->offset]) {
// skip reverse solidus character
state->offset++;
if (state->offset == state->size) {
// string can't terminate here!
return 1;
}
switch (state->src[state->offset]) {
default:
state->error = json_parse_error_invalid_string_escape_sequence;
return 1;
case '"':
case '\\':
case '/':
case 'b':
case 'f':
case 'n':
case 'r':
case 't':
// all valid characters!
state->offset++;
// if we have to preserve reverse solidus...
if (!(json_parse_flags_allow_string_simplification & state->flags_bitset)) {
// ... allocate enough space for it
data_size++;
}
break;
case 'u':
if (state->offset + 5 < state->size) {
// invalid escaped unicode sequence!
state->error = json_parse_error_invalid_string_escape_sequence;
return 1;
} else if (!json_is_hexadecimal_digit(state->src[state->offset + 1]) ||
!json_is_hexadecimal_digit(state->src[state->offset + 2]) ||
!json_is_hexadecimal_digit(state->src[state->offset + 3]) ||
!json_is_hexadecimal_digit(state->src[state->offset + 4])) {
// escaped unicode sequences must contain 4 hexadecimal digits!
state->error = json_parse_error_invalid_string_escape_sequence;
return 1;
}
// valid sequence!
state->offset += 5;
// add space for the 5 character sequence too
data_size += 5;
break;
}
} else {
// skip character (valid part of sequence)
state->offset++;
}
}
// skip trailing '"'
state->offset++;
// add enough space to store the string
state->data_size += data_size;
// one more byte for null terminator ending the string!
state->data_size++;
return 0;
}
