char *cmark_render_xml(cmark_node *root, long options)
{
char *result;
cmark_strbuf xml = GH_BUF_INIT;
cmark_event_type ev_type;
cmark_node *cur;
struct render_state state = { &xml, 0 };
if (options & CMARK_OPT_NORMALIZE) {
cmark_consolidate_text_nodes(root);
}
cmark_iter *iter = cmark_iter_new(root);
cmark_strbuf_puts(state.xml,
"<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n");
cmark_strbuf_puts(state.xml,
"<!DOCTYPE CommonMark SYSTEM \"CommonMark.dtd\">\n");
while ((ev_type = cmark_iter_next(iter)) != CMARK_EVENT_DONE) {
cur = cmark_iter_get_node(iter);
S_render_node(cur, ev_type, &state, options);
}
result = (char *)cmark_strbuf_detach(&xml);
cmark_iter_free(iter);
cmark_strbuf_free(&xml);
return result;
}
void cmark_consolidate_text_nodes(cmark_node *root)
{
cmark_iter *iter = cmark_iter_new(root);
cmark_strbuf buf = GH_BUF_INIT;
cmark_event_type ev_type;
cmark_node *cur, *tmp, *next;
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_puts(&buf, cmark_node_get_literal(cur));
tmp = cur->next;
while (tmp && tmp->type == CMARK_NODE_TEXT) {
cmark_iter_next(iter); // advance pointer
cmark_strbuf_puts(&buf, cmark_node_get_literal(tmp));
next = tmp->next;
cmark_node_free(tmp);
tmp = next;
}
cmark_node_set_literal(cur, (char *)cmark_strbuf_detach(&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;
}
static unsigned char *cmark_clean_autolink(cmark_chunk *url, int is_email)
{
cmark_strbuf buf = GH_BUF_INIT;
cmark_chunk_trim(url);
if (url->len == 0)
return NULL;
if (is_email)
cmark_strbuf_puts(&buf, "mailto:");
houdini_unescape_html_f(&buf, url->data, url->len);
return cmark_strbuf_detach(&buf);
}
char*
cmark_render(cmark_node *root,
int options,
int width,
void (*outc)(cmark_renderer*,
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 = GH_BUF_INIT;
cmark_strbuf buf = GH_BUF_INIT;
cmark_node *cur;
cmark_event_type ev_type;
char *result;
cmark_iter *iter = cmark_iter_new(root);
cmark_renderer renderer = { &buf, &pref, 0, width,
0, 0, true, false, false,
outc, S_cr, S_blankline, S_out
};
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 surrounding <>,
// and remove \ that escape punctuation.
unsigned char *cmark_clean_url(cmark_chunk *url)
{
cmark_strbuf buf = GH_BUF_INIT;
cmark_chunk_trim(url);
if (url->len == 0)
return NULL;
if (url->data[0] == '<' && url->data[url->len - 1] == '>') {
houdini_unescape_html_f(&buf, url->data + 1, url->len - 2);
} else {
houdini_unescape_html_f(&buf, url->data, url->len);
}
cmark_strbuf_unescape(&buf);
return cmark_strbuf_detach(&buf);
}
char *cmark_render_html(cmark_node *root, int options)
{
char *result;
cmark_strbuf html = GH_BUF_INIT;
cmark_event_type ev_type;
cmark_node *cur;
struct render_state state = { &html, NULL };
cmark_iter *iter = cmark_iter_new(root);
while ((ev_type = cmark_iter_next(iter)) != CMARK_EVENT_DONE) {
cur = cmark_iter_get_node(iter);
S_render_node(cur, ev_type, &state, options);
}
result = (char *)cmark_strbuf_detach(&html);
cmark_iter_free(iter);
return result;
}
char *cmark_render_man(cmark_node *root, long options)
{
char *result;
cmark_strbuf man = GH_BUF_INIT;
struct render_state state = { &man, NULL };
cmark_node *cur;
cmark_event_type ev_type;
cmark_iter *iter = cmark_iter_new(root);
if (options == 0) options = 0; // avoid warning about unused parameters
while ((ev_type = cmark_iter_next(iter)) != CMARK_EVENT_DONE) {
cur = cmark_iter_get_node(iter);
S_render_node(cur, ev_type, &state);
}
result = (char *)cmark_strbuf_detach(&man);
cmark_iter_free(iter);
cmark_strbuf_free(&man);
return result;
}
unsigned char *cmark_clean_title(cmark_chunk *title)
{
cmark_strbuf buf = GH_BUF_INIT;
unsigned char first, last;
if (title->len == 0)
return NULL;
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_strbuf_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;
}
