static bufsize_t manual_scan_link_url_2(cmark_chunk *input, bufsize_t offset, cmark_chunk *output) {
bufsize_t i = offset;
size_t nb_p = 0;
while (i < input->len) {
if (input->data[i] == '\\' &&
i + 1 < input-> len &&
cmark_ispunct(input->data[i+1]))
i += 2;
else if (input->data[i] == '(') {
++nb_p;
++i;
} else if (input->data[i] == ')') {
if (nb_p == 0)
break;
--nb_p;
++i;
} else if (cmark_isspace(input->data[i]))
break;
else
++i;
}
if (i >= input->len)
return -1;
{
cmark_chunk result = {input->data + offset, i - offset, 0};
*output = result;
}
return i - offset;
}
// Destructively unescape a string: remove backslashes before punctuation chars.
extern void cmark_strbuf_unescape(cmark_strbuf *buf) {
bufsize_t r, w;
for (r = 0, w = 0; r < buf->size; ++r) {
if (buf->ptr[r] == '\\' && cmark_ispunct(buf->ptr[r + 1]))
r++;
buf->ptr[w++] = buf->ptr[r];
}
cmark_strbuf_truncate(buf, w);
}
// Parse backslash-escape or just a backslash, returning an inline.
static cmark_node *handle_backslash(subject *subj) {
advance(subj);
unsigned char nextchar = peek_char(subj);
if (cmark_ispunct(
nextchar)) { // only ascii symbols and newline can be escaped
advance(subj);
return make_str(subj->mem, cmark_chunk_dup(&subj->input, subj->pos - 1, 1));
} else if (!is_eof(subj) && skip_line_end(subj)) {
return make_linebreak(subj->mem);
} else {
return make_str(subj->mem, cmark_chunk_literal("\\"));
}
}
static CMARK_INLINE void outc(cmark_renderer *renderer, cmark_escaping escape,
int32_t c, unsigned char nextc) {
bool needs_escaping = false;
bool follows_digit =
renderer->buffer->size > 0 &&
cmark_isdigit(renderer->buffer->ptr[renderer->buffer->size - 1]);
char encoded[ENCODED_SIZE];
needs_escaping =
c < 0x80 && escape != LITERAL &&
((escape == NORMAL &&
(c < 0x20 ||
c == '*' || c == '_' || c == '[' || c == ']' || c == '#' || c == '<' ||
c == '>' || c == '\\' || c == '`' || c == '!' ||
(c == '&' && cmark_isalpha(nextc)) || (c == '!' && nextc == '[') ||
(renderer->begin_content && (c == '-' || c == '+' || c == '=') &&
// begin_content doesn't get set to false til we've passed digits
// at the beginning of line, so...
!follows_digit) ||
(renderer->begin_content && (c == '.' || c == ')') && follows_digit &&
(nextc == 0 || cmark_isspace(nextc))))) ||
(escape == URL &&
(c == '`' || c == '<' || c == '>' || cmark_isspace(c) || c == '\\' ||
c == ')' || c == '(')) ||
(escape == TITLE &&
(c == '`' || c == '<' || c == '>' || c == '"' || c == '\\')));
if (needs_escaping) {
if (escape == URL && cmark_isspace(c)) {
// use percent encoding for spaces
snprintf(encoded, ENCODED_SIZE, "%%%2X", c);
cmark_strbuf_puts(renderer->buffer, encoded);
renderer->column += 3;
} else if (cmark_ispunct(c)) {
cmark_render_ascii(renderer, "\\");
cmark_render_code_point(renderer, c);
} else { // render as entity
snprintf(encoded, ENCODED_SIZE, "&#%d;", c);
cmark_strbuf_puts(renderer->buffer, encoded);
renderer->column += strlen(encoded);
}
} else {
cmark_render_code_point(renderer, c);
}
}
// Parse a link label. Returns 1 if successful.
// Note: unescaped brackets are not allowed in labels.
// The label begins with `[` and ends with the first `]` character
// encountered. Backticks in labels do not start code spans.
static int link_label(subject* subj, cmark_chunk *raw_label)
{
bufsize_t startpos = subj->pos;
int length = 0;
unsigned char c;
// advance past [
if (peek_char(subj) == '[') {
advance(subj);
} else {
return 0;
}
while ((c = peek_char(subj)) && c != '[' && c != ']') {
if (c == '\\') {
advance(subj);
length++;
if (cmark_ispunct(peek_char(subj))) {
advance(subj);
length++;
}
} else {
advance(subj);
length++;
}
if (length > MAX_LINK_LABEL_LENGTH) {
goto noMatch;
}
}
if (c == ']') { // match found
*raw_label = cmark_chunk_dup(&subj->input, startpos + 1, subj->pos - (startpos + 1));
cmark_chunk_trim(raw_label);
advance(subj); // advance past ]
return 1;
}
noMatch:
subj->pos = startpos; // rewind
return 0;
}
