// Scan ***, **, or * and return number scanned, or 0.
// Advances position.
static int
scan_delims(subject* subj, unsigned char c, bool * can_open, bool * can_close)
{
int numdelims = 0;
int before_char_pos;
int32_t after_char = 0;
int32_t before_char = 0;
int len;
bool left_flanking, right_flanking;
if (subj->pos == 0) {
before_char = 10;
} else {
before_char_pos = subj->pos - 1;
// walk back to the beginning of the UTF_8 sequence:
while (peek_at(subj, before_char_pos) >> 6 == 2 &&
before_char_pos > 0) {
before_char_pos -= 1;
}
len = utf8proc_iterate(subj->input.data + before_char_pos,
subj->pos - before_char_pos, &before_char);
if (len == -1) {
before_char = 10;
}
}
while (peek_char(subj) == c) {
numdelims++;
advance(subj);
}
len = utf8proc_iterate(subj->input.data + subj->pos,
subj->input.len - subj->pos, &after_char);
if (len == -1) {
after_char = 10;
}
left_flanking = numdelims > 0 && !utf8proc_is_space(after_char) &&
!(utf8proc_is_punctuation(after_char) &&
!utf8proc_is_space(before_char) &&
!utf8proc_is_punctuation(before_char));
right_flanking = numdelims > 0 && !utf8proc_is_space(before_char) &&
!(utf8proc_is_punctuation(before_char) &&
!utf8proc_is_space(after_char) &&
!utf8proc_is_punctuation(after_char));
if (c == '_') {
*can_open = left_flanking && !right_flanking;
*can_close = right_flanking && !left_flanking;
} else {
*can_open = left_flanking;
*can_close = right_flanking;
}
return numdelims;
}
size_t utf8proc_find_space(const uint8_t *str, size_t pos, size_t size)
{
while (pos < size) {
const size_t last = pos;
int32_t uc = utf8proc_next(str, &pos);
if (utf8proc_is_space(uc))
return last;
}
return size;
}
