void cmark_consolidate_text_nodes(cmark_node *root) {
cmark_iter *iter;
cmark_strbuf buf = CMARK_BUF_INIT(NULL);
cmark_event_type ev_type;
cmark_node *cur, *tmp, *next;
if (root == NULL) {
return;
}
iter = cmark_iter_new(root);
buf.mem = iter->mem;
while ((ev_type = cmark_iter_next(iter)) != CMARK_EVENT_DONE) {
cur = cmark_iter_get_node(iter);
if (ev_type == CMARK_EVENT_ENTER && cur->type == CMARK_NODE_TEXT &&
cur->next && cur->next->type == CMARK_NODE_TEXT) {
cmark_strbuf_clear(&buf);
cmark_strbuf_put(&buf, cur->as.literal.data, cur->as.literal.len);
tmp = cur->next;
while (tmp && tmp->type == CMARK_NODE_TEXT) {
cmark_iter_next(iter); // advance pointer
cmark_strbuf_put(&buf, tmp->as.literal.data, tmp->as.literal.len);
next = tmp->next;
cmark_node_free(tmp);
tmp = next;
}
cmark_chunk_free(iter->mem, &cur->as.literal);
cur->as.literal = cmark_chunk_buf_detach(&buf);
}
}
cmark_strbuf_free(&buf);
cmark_iter_free(iter);
}
char *cmark_render_html_with_mem(cmark_node *root, int options, cmark_llist *extensions, cmark_mem *mem) {
char *result;
cmark_strbuf html = CMARK_BUF_INIT(mem);
cmark_event_type ev_type;
cmark_node *cur;
cmark_html_renderer renderer = {&html, NULL, NULL, 0, 0, NULL};
cmark_iter *iter = cmark_iter_new(root);
for (; extensions; extensions = extensions->next)
if (((cmark_syntax_extension *) extensions->data)->html_filter_func)
renderer.filter_extensions = cmark_llist_append(
mem,
renderer.filter_extensions,
(cmark_syntax_extension *) extensions->data);
while ((ev_type = cmark_iter_next(iter)) != CMARK_EVENT_DONE) {
cur = cmark_iter_get_node(iter);
S_render_node(&renderer, cur, ev_type, options);
}
if (renderer.footnote_ix) {
cmark_strbuf_puts(&html, "</ol>\n</section>\n");
}
result = (char *)cmark_strbuf_detach(&html);
cmark_llist_free(mem, renderer.filter_extensions);
cmark_iter_free(iter);
return result;
}
// Like make_str, but parses entities.
static cmark_node *make_str_with_entities(cmark_mem *mem,
cmark_chunk *content) {
cmark_strbuf unescaped = CMARK_BUF_INIT(mem);
if (houdini_unescape_html(&unescaped, content->data, content->len)) {
return make_str(mem, cmark_chunk_buf_detach(&unescaped));
} else {
return make_str(mem, *content);
}
}
char *cmark_render(cmark_mem *mem, cmark_node *root, int options, int width,
void (*outc)(cmark_renderer *, cmark_node *,
cmark_escaping, int32_t,
unsigned char),
int (*render_node)(cmark_renderer *renderer,
cmark_node *node,
cmark_event_type ev_type, int options)) {
cmark_strbuf pref = CMARK_BUF_INIT(mem);
cmark_strbuf buf = CMARK_BUF_INIT(mem);
cmark_node *cur;
cmark_event_type ev_type;
char *result;
cmark_iter *iter = cmark_iter_new(root);
cmark_renderer renderer = {mem, &buf, &pref, 0, width,
0, 0, true, true, false,
false, outc, S_cr, S_blankline, S_out,
0};
while ((ev_type = cmark_iter_next(iter)) != CMARK_EVENT_DONE) {
cur = cmark_iter_get_node(iter);
if (!render_node(&renderer, cur, ev_type, options)) {
// a false value causes us to skip processing
// the node's contents. this is used for
// autolinks.
cmark_iter_reset(iter, cur, CMARK_EVENT_EXIT);
}
}
// ensure final newline
if (renderer.buffer->ptr[renderer.buffer->size - 1] != '\n') {
cmark_strbuf_putc(renderer.buffer, '\n');
}
result = (char *)cmark_strbuf_detach(renderer.buffer);
cmark_iter_free(iter);
cmark_strbuf_free(renderer.prefix);
cmark_strbuf_free(renderer.buffer);
return result;
}
// Clean a URL: remove surrounding whitespace, and remove \ that escape
// punctuation.
cmark_chunk cmark_clean_url(cmark_mem *mem, cmark_chunk *url) {
cmark_strbuf buf = CMARK_BUF_INIT(mem);
cmark_chunk_trim(url);
if (url->len == 0) {
cmark_chunk result = CMARK_CHUNK_EMPTY;
return result;
}
houdini_unescape_html_f(&buf, url->data, url->len);
cmark_strbuf_unescape(&buf);
return cmark_chunk_buf_detach(&buf);
}
// Parse an entity or a regular "&" string.
// Assumes the subject has an '&' character at the current position.
static cmark_node *handle_entity(subject *subj) {
cmark_strbuf ent = CMARK_BUF_INIT(subj->mem);
bufsize_t len;
advance(subj);
len = houdini_unescape_ent(&ent, subj->input.data + subj->pos,
subj->input.len - subj->pos);
if (len == 0)
return make_str(subj->mem, cmark_chunk_literal("&"));
subj->pos += len;
return make_str(subj->mem, cmark_chunk_buf_detach(&ent));
}
static cmark_chunk cmark_clean_autolink(cmark_mem *mem, cmark_chunk *url,
int is_email) {
cmark_strbuf buf = CMARK_BUF_INIT(mem);
cmark_chunk_trim(url);
if (url->len == 0) {
cmark_chunk result = CMARK_CHUNK_EMPTY;
return result;
}
if (is_email)
cmark_strbuf_puts(&buf, "mailto:");
houdini_unescape_html_f(&buf, url->data, url->len);
return cmark_chunk_buf_detach(&buf);
}
// Parse backtick code section or raw backticks, return an inline.
// Assumes that the subject has a backtick at the current position.
static cmark_node *handle_backticks(subject *subj) {
cmark_chunk openticks = take_while(subj, isbacktick);
bufsize_t startpos = subj->pos;
bufsize_t endpos = scan_to_closing_backticks(subj, openticks.len);
if (endpos == 0) { // not found
subj->pos = startpos; // rewind
return make_str(subj->mem, openticks);
} else {
cmark_strbuf buf = CMARK_BUF_INIT(subj->mem);
cmark_strbuf_set(&buf, subj->input.data + startpos,
endpos - startpos - openticks.len);
cmark_strbuf_trim(&buf);
cmark_strbuf_normalize_whitespace(&buf);
return make_code(subj->mem, cmark_chunk_buf_detach(&buf));
}
}
// Assumes we have a hyphen at the current position.
static cmark_node *handle_hyphen(subject *subj, bool smart) {
cmark_strbuf buf = CMARK_BUF_INIT(NULL);
int startpos = subj->pos;
int en_count = 0;
int em_count = 0;
int numhyphens;
int i;
advance(subj);
if (!smart || peek_char(subj) != '-') {
return make_str(subj->mem, cmark_chunk_literal("-"));
}
while (smart && peek_char(subj) == '-') {
advance(subj);
}
numhyphens = subj->pos - startpos;
buf.mem = subj->mem;
if (numhyphens % 3 == 0) { // if divisible by 3, use all em dashes
em_count = numhyphens / 3;
} else if (numhyphens % 2 == 0) { // if divisible by 2, use all en dashes
en_count = numhyphens / 2;
} else if (numhyphens % 3 == 2) { // use one en dash at end
en_count = 1;
em_count = (numhyphens - 2) / 3;
} else { // use two en dashes at the end
en_count = 2;
em_count = (numhyphens - 4) / 3;
}
for (i = em_count; i > 0; i--) {
cmark_strbuf_puts(&buf, EMDASH);
}
for (i = en_count; i > 0; i--) {
cmark_strbuf_puts(&buf, ENDASH);
}
return make_str(subj->mem, cmark_chunk_buf_detach(&buf));
}
cmark_chunk cmark_clean_title(cmark_mem *mem, cmark_chunk *title) {
cmark_strbuf buf = CMARK_BUF_INIT(mem);
unsigned char first, last;
if (title->len == 0) {
cmark_chunk result = CMARK_CHUNK_EMPTY;
return result;
}
first = title->data[0];
last = title->data[title->len - 1];
// remove surrounding quotes if any:
if ((first == '\'' && last == '\'') || (first == '(' && last == ')') ||
(first == '"' && last == '"')) {
houdini_unescape_html_f(&buf, title->data + 1, title->len - 2);
} else {
houdini_unescape_html_f(&buf, title->data, title->len);
}
cmark_strbuf_unescape(&buf);
return cmark_chunk_buf_detach(&buf);
}
// normalize reference: collapse internal whitespace to single space,
// remove leading/trailing whitespace, case fold
// Return NULL if the reference name is actually empty (i.e. composed
// solely from whitespace)
static unsigned char *normalize_reference(cmark_mem *mem, cmark_chunk *ref) {
cmark_strbuf normalized = CMARK_BUF_INIT(mem);
unsigned char *result;
if (ref == NULL)
return NULL;
if (ref->len == 0)
return NULL;
cmark_utf8proc_case_fold(&normalized, ref->data, ref->len);
cmark_strbuf_trim(&normalized);
cmark_strbuf_normalize_whitespace(&normalized);
result = cmark_strbuf_detach(&normalized);
assert(result);
if (result[0] == '\0') {
free(result);
return NULL;
}
return result;
}