char*
CFCPerlPod_md_to_pod(const char *md, CFCClass *klass, int header_level) {
int options = CMARK_OPT_SMART
| CMARK_OPT_VALIDATE_UTF8;
cmark_node *doc = cmark_parse_document(md, strlen(md), options);
char *pod = S_node_to_pod(doc, klass, header_level);
cmark_node_free(doc);
return pod;
}
char*
CFCPerlPod_md_to_pod(CFCPerlPod *self, CFCClass *klass, const char *md) {
(void)self; // unused
cmark_node *doc = cmark_parse_document(md, strlen(md));
char *pod = S_nodes_to_pod(klass, doc);
cmark_node_free(doc);
return pod;
}
char *cmark_markdown_to_html(const char *text, size_t len, int options) {
cmark_node *doc;
char *result;
doc = cmark_parse_document(text, len, options);
result = cmark_render_html(doc, options, NULL);
cmark_node_free(doc);
return result;
}
char *cmark_markdown_to_html(const char *text, int len)
{
cmark_node *doc;
char *result;
doc = cmark_parse_document(text, len, CMARK_OPT_DEFAULT);
result = cmark_render_html(doc, CMARK_OPT_DEFAULT);
cmark_node_free(doc);
return result;
}
static void
test_content(test_batch_runner *runner, cmark_node_type type,
int allowed_content)
{
cmark_node *node = cmark_node_new(type);
for (int i = 0; i < num_node_types; ++i) {
cmark_node_type child_type = node_types[i];
cmark_node *child = cmark_node_new(child_type);
int got = cmark_node_append_child(node, child);
int expected = (allowed_content >> child_type) & 1;
INT_EQ(runner, got, expected,
"add %d as child of %d", child_type, type);
cmark_node_free(child);
}
cmark_node_free(node);
}
static void
iterator_delete(test_batch_runner *runner) {
static const char md[] =
"a *b* c\n"
"\n"
"* item1\n"
"* item2\n"
"\n"
"a `b` c\n"
"\n"
"* item1\n"
"* item2\n";
cmark_node *doc = cmark_parse_document(md, sizeof(md) - 1,
CMARK_OPT_DEFAULT);
cmark_iter *iter = cmark_iter_new(doc);
cmark_event_type ev_type;
while ((ev_type = cmark_iter_next(iter)) != CMARK_EVENT_DONE) {
cmark_node *node = cmark_iter_get_node(iter);
// Delete list, emph, and code nodes.
if ((ev_type == CMARK_EVENT_EXIT &&
node->type == CMARK_NODE_LIST) ||
(ev_type == CMARK_EVENT_EXIT &&
node->type == CMARK_NODE_EMPH) ||
(ev_type == CMARK_EVENT_ENTER &&
node->type == CMARK_NODE_CODE)) {
cmark_node_free(node);
}
}
char *html = cmark_render_html(doc, CMARK_OPT_DEFAULT);
static const char expected[] =
"<p>a c</p>\n"
"<p>a c</p>\n";
STR_EQ(runner, html, expected, "iterate and delete nodes");
free(html);
cmark_iter_free(iter);
cmark_node_free(doc);
}
static void
node_check(test_batch_runner *runner) {
// Construct an incomplete tree.
cmark_node *doc = cmark_node_new(CMARK_NODE_DOCUMENT);
cmark_node *p1 = cmark_node_new(CMARK_NODE_PARAGRAPH);
cmark_node *p2 = cmark_node_new(CMARK_NODE_PARAGRAPH);
doc->first_child = p1;
p1->next = p2;
INT_EQ(runner, cmark_node_check(doc, NULL), 4, "node_check works");
INT_EQ(runner, cmark_node_check(doc, NULL), 0,
"node_check fixes tree");
cmark_node_free(doc);
}
static void md_escape_inline(cmark_iter *const iter) {
for(;;) {
cmark_event_type const event = cmark_iter_next(iter);
if(CMARK_EVENT_DONE == event) break;
if(CMARK_EVENT_ENTER != event) continue;
cmark_node *const node = cmark_iter_get_node(iter);
if(CMARK_NODE_INLINE_HTML != cmark_node_get_type(node)) continue;
char const *const str = cmark_node_get_literal(node);
cmark_node *text = cmark_node_new(CMARK_NODE_TEXT);
cmark_node_set_literal(text, str);
cmark_node_insert_before(node, text);
cmark_node_free(node);
}
}
static void
iterator(test_batch_runner *runner) {
cmark_node *doc = cmark_parse_document("> a *b*\n\nc", 10, CMARK_OPT_DEFAULT);
int parnodes = 0;
cmark_event_type ev_type;
cmark_iter *iter = cmark_iter_new(doc);
cmark_node *cur;
while ((ev_type = cmark_iter_next(iter)) != CMARK_EVENT_DONE) {
cur = cmark_iter_get_node(iter);
if (cur->type == CMARK_NODE_PARAGRAPH &&
ev_type == CMARK_EVENT_ENTER) {
parnodes += 1;
}
}
INT_EQ(runner, parnodes, 2, "iterate correctly counts paragraphs");
cmark_iter_free(iter);
cmark_node_free(doc);
}
char*
CFCPerlPod_md_doc_to_pod(const char *module, const char *md) {
int options = CMARK_OPT_SMART
| CMARK_OPT_VALIDATE_UTF8;
cmark_node *doc = cmark_parse_document(md, strlen(md), options);
cmark_node *maybe_header = cmark_node_first_child(doc);
char *name;
char *desc;
if (maybe_header
&& cmark_node_get_type(maybe_header) == CMARK_NODE_HEADER
) {
cmark_node *header_child = cmark_node_first_child(maybe_header);
char *short_desc = S_nodes_to_pod(header_child, NULL, 1);
name = CFCUtil_sprintf("%s - %s", module, short_desc);
FREEMEM(short_desc);
cmark_node *remaining = cmark_node_next(maybe_header);
desc = S_nodes_to_pod(remaining, NULL, 1);
}
else {
// No header found.
name = CFCUtil_strdup(module);
desc = S_node_to_pod(doc, NULL, 1);
}
const char *pattern =
"=head1 NAME\n"
"\n"
"%s\n"
"\n"
"=head1 DESCRIPTION\n"
"\n"
"%s";
char *retval = CFCUtil_sprintf(pattern, name, desc);
FREEMEM(name);
FREEMEM(desc);
cmark_node_free(doc);
return retval;
}
static delimiter *insert(cmark_syntax_extension *self, cmark_parser *parser,
cmark_inline_parser *inline_parser, delimiter *opener,
delimiter *closer) {
cmark_node *strikethrough;
cmark_node *tmp, *next;
delimiter *delim, *tmp_delim;
delimiter *res = closer->next;
strikethrough = opener->inl_text;
if (!cmark_node_set_type(strikethrough, CMARK_NODE_STRIKETHROUGH))
goto done;
cmark_node_set_syntax_extension(strikethrough, self);
cmark_node_set_string_content(strikethrough, "~");
tmp = cmark_node_next(opener->inl_text);
while (tmp) {
if (tmp == closer->inl_text)
break;
next = cmark_node_next(tmp);
cmark_node_append_child(strikethrough, tmp);
tmp = next;
}
strikethrough->end_column = closer->inl_text->start_column + closer->inl_text->as.literal.len - 1;
cmark_node_free(closer->inl_text);
delim = closer;
while (delim != NULL && delim != opener) {
tmp_delim = delim->previous;
cmark_inline_parser_remove_delimiter(inline_parser, delim);
delim = tmp_delim;
}
cmark_inline_parser_remove_delimiter(inline_parser, opener);
done:
return res;
}
static void
constructor(test_batch_runner *runner)
{
for (int i = 0; i < num_node_types; ++i) {
cmark_node_type type = node_types[i];
cmark_node *node = cmark_node_new(type);
OK(runner, node != NULL, "new type %d", type);
INT_EQ(runner, cmark_node_get_type(node), type,
"get_type %d", type);
switch (node->type) {
case CMARK_NODE_HEADER:
INT_EQ(runner, cmark_node_get_header_level(node), 1,
"default header level is 1");
node->as.header.level = 1;
break;
case CMARK_NODE_LIST:
INT_EQ(runner, cmark_node_get_list_type(node),
CMARK_BULLET_LIST,
"default is list type is bullet");
INT_EQ(runner, cmark_node_get_list_delim(node),
CMARK_NO_DELIM,
"default is list delim is NO_DELIM");
INT_EQ(runner, cmark_node_get_list_start(node), 1,
"default is list start is 1");
INT_EQ(runner, cmark_node_get_list_tight(node), 0,
"default is list is loose");
break;
default:
break;
}
cmark_node_free(node);
}
}
int main(int argc, char *argv[])
{
int i, numfps = 0;
int *files;
int ok;
char buffer[4096];
cmark_parser *parser;
size_t bytes;
cmark_node *document;
char *outfile = NULL;
int options = CMARK_OPT_DEFAULT | CMARK_OPT_SAFE | CMARK_OPT_NORMALIZE;
#if defined(_WIN32) && !defined(__CYGWIN__)
_setmode(_fileno(stdout), _O_BINARY);
#endif
files = (int *)malloc(argc * sizeof(*files));
for (i = 1; i < argc; i++) {
if (strcmp(argv[i], "--version") == 0) {
printf("cmark %s", CMARK_VERSION_STRING);
printf(" - CommonMark converter\n(C) 2014, 2015 John MacFarlane\n");
exit(0);
} else if (strcmp(argv[i], "--sourcepos") == 0) {
options |= CMARK_OPT_SOURCEPOS;
} else if (strcmp(argv[i], "--hardbreaks") == 0) {
options |= CMARK_OPT_HARDBREAKS;
} else if (strcmp(argv[i], "--smart") == 0) {
options |= CMARK_OPT_SMART;
} else if (strcmp(argv[i], "--validate-utf8") == 0) {
options |= CMARK_OPT_VALIDATE_UTF8;
} else if ((strcmp(argv[i], "--help") == 0) ||
(strcmp(argv[i], "-h") == 0)) {
print_usage();
exit(0);
} else if ((strcmp(argv[i], "-o") == 0) ||
(strcmp(argv[i], "--output") == 0)) {
i += 1;
if (i < argc) {
outfile = argv[i];
} else {
fprintf(stderr, "No argument provided for %s\n",
argv[i - 1]);
exit(1);
}
} else if (*argv[i] == '-') {
print_usage();
exit(1);
} else { // treat as file argument
files[numfps++] = i;
}
}
if (!outfile) {
fprintf(stderr, "Specify an output file with -o/--output\n");
exit(1);
}
parser = cmark_parser_new(options);
for (i = 0; i < numfps; i++) {
FILE *fp = fopen(argv[files[i]], "r");
if (fp == NULL) {
fprintf(stderr, "Error opening file %s: %s\n",
argv[files[i]], strerror(errno));
exit(1);
}
while ((bytes = fread(buffer, 1, sizeof(buffer), fp)) > 0) {
cmark_parser_feed(parser, buffer, bytes);
if (bytes < sizeof(buffer)) {
break;
}
}
fclose(fp);
}
if (numfps == 0) {
while ((bytes = fread(buffer, 1, sizeof(buffer), stdin)) > 0) {
cmark_parser_feed(parser, buffer, bytes);
if (bytes < sizeof(buffer)) {
break;
}
}
}
document = cmark_parser_finish(parser);
cmark_parser_free(parser);
ok = cmark_render_pdf(document, options, outfile);
free(files);
cmark_node_free(document);
return ok ? 0 : 1;
}
static cmark_node *fixup_nodes(cmark_syntax_extension *self,
cmark_parser *parser,
cmark_inline_parser *inline_parser,
cmark_node *parent,
int start_offset,
int size)
{
int node_text_len;
cmark_node *prev = NULL;
cmark_node *tmp;
int name_size = size;
cmark_strbuf *name;
NamedLink *named_link;
for (prev = cmark_node_last_child(parent); prev; prev = cmark_node_previous(prev)) {
if (cmark_node_get_type(prev) == CMARK_NODE_TEXT) {
const char *text = cmark_node_get_literal(prev);
node_text_len = strlen(text);
size -= node_text_len;
if (size <= 0) {
if (size < 0) {
char *split_text = my_strndup(text, size * -1);
cmark_node *split = cmark_node_new(CMARK_NODE_TEXT);
cmark_node_set_literal(split, split_text);
free(split_text);
split_text = my_strndup(text + (size * - 1), node_text_len - size);
cmark_node_set_literal(prev, split_text);
free(split_text);
cmark_node_insert_before(prev, split);
}
break;
}
} else {
return NULL;
}
}
name = cmark_strbuf_new(name_size + 1);
tmp = prev;
while (tmp) {
cmark_node *next = cmark_node_next(tmp);
cmark_strbuf_puts(name, cmark_node_get_literal(tmp));
if (tmp != prev)
cmark_node_free(tmp);
tmp = next;
}
named_link = PRIV(self)->link_resolve_func(cmark_strbuf_get(name));
if (!named_link || !named_link->ref) {
int actual_line, actual_col;
translate_sourcepos(get_first_parent_block(parent),
start_offset, &actual_line, &actual_col);
cmark_strbuf *message = cmark_strbuf_new(0);
cmark_strbuf_puts(message, "Trying to link to non-existing symbol ‘");
cmark_strbuf_puts(message, cmark_strbuf_get(name));
cmark_strbuf_puts(message, "’");
diagnose("gtk-doc-bad-link", cmark_strbuf_get(message), actual_line - 1,
actual_col - 1);
cmark_strbuf_free(message);
cmark_node_set_literal(prev, cmark_strbuf_get(name));
cmark_strbuf_free(name);
return prev;
}
free_named_link(named_link);
cmark_node_set_type(prev, CMARK_NODE_LINK);
cmark_node_set_url(prev, cmark_strbuf_get(name));
cmark_strbuf_free(name);
return prev;
}
// Return a link, an image, or a literal close bracket.
static cmark_node *handle_close_bracket(subject *subj) {
bufsize_t initial_pos, after_link_text_pos;
bufsize_t endurl, starttitle, endtitle, endall;
bufsize_t sps, n;
cmark_reference *ref = NULL;
cmark_chunk url_chunk, title_chunk;
cmark_chunk url, title;
bracket *opener;
cmark_node *inl;
cmark_chunk raw_label;
int found_label;
cmark_node *tmp, *tmpnext;
bool is_image;
advance(subj); // advance past ]
initial_pos = subj->pos;
// get last [ or ![
opener = subj->last_bracket;
if (opener == NULL) {
return make_str(subj->mem, cmark_chunk_literal("]"));
}
if (!opener->active) {
// take delimiter off stack
pop_bracket(subj);
return make_str(subj->mem, cmark_chunk_literal("]"));
}
// If we got here, we matched a potential link/image text.
// Now we check to see if it's a link/image.
is_image = opener->image;
after_link_text_pos = subj->pos;
// First, look for an inline link.
if (peek_char(subj) == '(' &&
((sps = scan_spacechars(&subj->input, subj->pos + 1)) > -1) &&
((n = manual_scan_link_url(&subj->input, subj->pos + 1 + sps, &url_chunk)) > -1)) {
// try to parse an explicit link:
endurl = subj->pos + 1 + sps + n;
starttitle = endurl + scan_spacechars(&subj->input, endurl);
// ensure there are spaces btw url and title
endtitle = (starttitle == endurl)
? starttitle
: starttitle + scan_link_title(&subj->input, starttitle);
endall = endtitle + scan_spacechars(&subj->input, endtitle);
if (peek_at(subj, endall) == ')') {
subj->pos = endall + 1;
title_chunk =
cmark_chunk_dup(&subj->input, starttitle, endtitle - starttitle);
url = cmark_clean_url(subj->mem, &url_chunk);
title = cmark_clean_title(subj->mem, &title_chunk);
cmark_chunk_free(subj->mem, &url_chunk);
cmark_chunk_free(subj->mem, &title_chunk);
goto match;
} else {
// it could still be a shortcut reference link
subj->pos = after_link_text_pos;
}
}
// Next, look for a following [link label] that matches in refmap.
// skip spaces
raw_label = cmark_chunk_literal("");
found_label = link_label(subj, &raw_label);
if (!found_label) {
// If we have a shortcut reference link, back up
// to before the spacse we skipped.
subj->pos = initial_pos;
}
if ((!found_label || raw_label.len == 0) && !opener->bracket_after) {
cmark_chunk_free(subj->mem, &raw_label);
raw_label = cmark_chunk_dup(&subj->input, opener->position,
initial_pos - opener->position - 1);
found_label = true;
}
if (found_label) {
ref = cmark_reference_lookup(subj->refmap, &raw_label);
cmark_chunk_free(subj->mem, &raw_label);
}
if (ref != NULL) { // found
url = chunk_clone(subj->mem, &ref->url);
title = chunk_clone(subj->mem, &ref->title);
goto match;
} else {
goto noMatch;
}
noMatch:
// If we fall through to here, it means we didn't match a link:
pop_bracket(subj); // remove this opener from delimiter list
subj->pos = initial_pos;
return make_str(subj->mem, cmark_chunk_literal("]"));
match:
inl = make_simple(subj->mem, is_image ? CMARK_NODE_IMAGE : CMARK_NODE_LINK);
inl->as.link.url = url;
inl->as.link.title = title;
cmark_node_insert_before(opener->inl_text, inl);
// Add link text:
tmp = opener->inl_text->next;
while (tmp) {
tmpnext = tmp->next;
cmark_node_append_child(inl, tmp);
tmp = tmpnext;
}
// Free the bracket [:
cmark_node_free(opener->inl_text);
process_emphasis(subj, opener->previous_delimiter);
pop_bracket(subj);
// Now, if we have a link, we also want to deactivate earlier link
// delimiters. (This code can be removed if we decide to allow links
// inside links.)
if (!is_image) {
opener = subj->last_bracket;
while (opener != NULL) {
if (!opener->image) {
if (!opener->active) {
break;
} else {
opener->active = false;
}
}
opener = opener->previous;
}
}
return NULL;
}
static void
create_tree(test_batch_runner *runner)
{
char *html;
cmark_node *doc = cmark_node_new(CMARK_NODE_DOCUMENT);
cmark_node *p = cmark_node_new(CMARK_NODE_PARAGRAPH);
OK(runner, !cmark_node_insert_before(doc, p),
"insert before root fails");
OK(runner, !cmark_node_insert_after(doc, p),
"insert after root fails");
OK(runner, cmark_node_append_child(doc, p), "append1");
INT_EQ(runner, cmark_node_check(doc, NULL), 0, "append1 consistent");
OK(runner, cmark_node_parent(p) == doc, "node_parent");
cmark_node *emph = cmark_node_new(CMARK_NODE_EMPH);
OK(runner, cmark_node_prepend_child(p, emph), "prepend1");
INT_EQ(runner, cmark_node_check(doc, NULL), 0, "prepend1 consistent");
cmark_node *str1 = cmark_node_new(CMARK_NODE_TEXT);
cmark_node_set_literal(str1, "Hello, ");
OK(runner, cmark_node_prepend_child(p, str1), "prepend2");
INT_EQ(runner, cmark_node_check(doc, NULL), 0, "prepend2 consistent");
cmark_node *str3 = cmark_node_new(CMARK_NODE_TEXT);
cmark_node_set_literal(str3, "!");
OK(runner, cmark_node_append_child(p, str3), "append2");
INT_EQ(runner, cmark_node_check(doc, NULL), 0, "append2 consistent");
cmark_node *str2 = cmark_node_new(CMARK_NODE_TEXT);
cmark_node_set_literal(str2, "world");
OK(runner, cmark_node_append_child(emph, str2), "append3");
INT_EQ(runner, cmark_node_check(doc, NULL), 0, "append3 consistent");
html = cmark_render_html(doc, CMARK_OPT_DEFAULT);
STR_EQ(runner, html, "<p>Hello, <em>world</em>!</p>\n",
"render_html");
free(html);
OK(runner, cmark_node_insert_before(str1, str3), "ins before1");
INT_EQ(runner, cmark_node_check(doc, NULL), 0,
"ins before1 consistent");
// 31e
OK(runner, cmark_node_first_child(p) == str3, "ins before1 works");
OK(runner, cmark_node_insert_before(str1, emph), "ins before2");
INT_EQ(runner, cmark_node_check(doc, NULL), 0,
"ins before2 consistent");
// 3e1
OK(runner, cmark_node_last_child(p) == str1, "ins before2 works");
OK(runner, cmark_node_insert_after(str1, str3), "ins after1");
INT_EQ(runner, cmark_node_check(doc, NULL), 0,
"ins after1 consistent");
// e13
OK(runner, cmark_node_next(str1) == str3, "ins after1 works");
OK(runner, cmark_node_insert_after(str1, emph), "ins after2");
INT_EQ(runner, cmark_node_check(doc, NULL), 0,
"ins after2 consistent");
// 1e3
OK(runner, cmark_node_previous(emph) == str1, "ins after2 works");
cmark_node_unlink(emph);
html = cmark_render_html(doc, CMARK_OPT_DEFAULT);
STR_EQ(runner, html, "<p>Hello, !</p>\n",
"render_html after shuffling");
free(html);
cmark_node_free(doc);
// TODO: Test that the contents of an unlinked inline are valid
// after the parent block was destroyed. This doesn't work so far.
cmark_node_free(emph);
}
static cmark_node*
finalize(cmark_parser *parser, cmark_node* b)
{
int pos;
cmark_node* item;
cmark_node* subitem;
cmark_node* parent;
parent = b->parent;
assert(b->open); // shouldn't call finalize on closed blocks
b->open = false;
if (parser->curline->size == 0) {
// end of input - line number has not been incremented
b->end_line = parser->line_number;
b->end_column = parser->last_line_length;
} else if (b->type == NODE_DOCUMENT ||
(b->type == NODE_CODE_BLOCK && b->as.code.fenced) ||
(b->type == NODE_HEADER && b->as.header.setext)) {
b->end_line = parser->line_number;
b->end_column = parser->curline->size;
if (b->end_column && parser->curline->ptr[b->end_column - 1] == '\n')
b->end_column--;
if (b->end_column && parser->curline->ptr[b->end_column - 1] == '\r')
b->end_column--;
} else {
b->end_line = parser->line_number - 1;
b->end_column = parser->last_line_length;
}
switch (b->type) {
case NODE_PARAGRAPH:
while (cmark_strbuf_at(&b->string_content, 0) == '[' &&
(pos = cmark_parse_reference_inline(&b->string_content, parser->refmap))) {
cmark_strbuf_drop(&b->string_content, pos);
}
if (is_blank(&b->string_content, 0)) {
// remove blank node (former reference def)
cmark_node_free(b);
}
break;
case NODE_CODE_BLOCK:
if (!b->as.code.fenced) { // indented code
remove_trailing_blank_lines(&b->string_content);
cmark_strbuf_putc(&b->string_content, '\n');
} else {
// first line of contents becomes info
for (pos = 0; pos < b->string_content.size; ++pos) {
if (b->string_content.ptr[pos] == '\r' ||
b->string_content.ptr[pos] == '\n')
break;
}
assert(pos < b->string_content.size);
cmark_strbuf tmp = GH_BUF_INIT;
houdini_unescape_html_f(
&tmp,
b->string_content.ptr,
pos
);
cmark_strbuf_trim(&tmp);
cmark_strbuf_unescape(&tmp);
b->as.code.info = cmark_chunk_buf_detach(&tmp);
if (b->string_content.ptr[pos] == '\r')
pos += 1;
if (b->string_content.ptr[pos] == '\n')
pos += 1;
cmark_strbuf_drop(&b->string_content, pos);
}
b->as.code.literal = cmark_chunk_buf_detach(&b->string_content);
break;
case NODE_HTML:
b->as.literal = cmark_chunk_buf_detach(&b->string_content);
break;
case NODE_LIST: // determine tight/loose status
b->as.list.tight = true; // tight by default
item = b->first_child;
while (item) {
// check for non-final non-empty list item ending with blank line:
if (item->last_line_blank && item->next) {
b->as.list.tight = false;
break;
}
// recurse into children of list item, to see if there are
// spaces between them:
subitem = item->first_child;
while (subitem) {
if (ends_with_blank_line(subitem) &&
(item->next || subitem->next)) {
b->as.list.tight = false;
break;
}
subitem = subitem->next;
}
if (!(b->as.list.tight)) {
break;
}
item = item->next;
}
break;
default:
break;
}
return parent;
}
static void
accessors(test_batch_runner *runner)
{
static const char markdown[] =
"## Header\n"
"\n"
"* Item 1\n"
"* Item 2\n"
"\n"
"2. Item 1\n"
"\n"
"3. Item 2\n"
"\n"
"\n"
" code\n"
"\n"
"``` lang\n"
"fenced\n"
"```\n"
"\n"
"<div>html</div>\n"
"\n"
"[link](url 'title')\n";
cmark_node *doc = cmark_parse_document(markdown, sizeof(markdown) - 1, CMARK_OPT_DEFAULT);
// Getters
cmark_node *header = cmark_node_first_child(doc);
INT_EQ(runner, cmark_node_get_header_level(header), 2,
"get_header_level");
cmark_node *bullet_list = cmark_node_next(header);
INT_EQ(runner, cmark_node_get_list_type(bullet_list),
CMARK_BULLET_LIST, "get_list_type bullet");
INT_EQ(runner, cmark_node_get_list_tight(bullet_list), 1,
"get_list_tight tight");
cmark_node *ordered_list = cmark_node_next(bullet_list);
INT_EQ(runner, cmark_node_get_list_type(ordered_list),
CMARK_ORDERED_LIST, "get_list_type ordered");
INT_EQ(runner, cmark_node_get_list_delim(ordered_list),
CMARK_PERIOD_DELIM, "get_list_delim ordered");
INT_EQ(runner, cmark_node_get_list_start(ordered_list), 2,
"get_list_start");
INT_EQ(runner, cmark_node_get_list_tight(ordered_list), 0,
"get_list_tight loose");
cmark_node *code = cmark_node_next(ordered_list);
STR_EQ(runner, cmark_node_get_literal(code), "code\n",
"get_literal indented code");
cmark_node *fenced = cmark_node_next(code);
STR_EQ(runner, cmark_node_get_literal(fenced), "fenced\n",
"get_literal fenced code");
STR_EQ(runner, cmark_node_get_fence_info(fenced), "lang",
"get_fence_info");
cmark_node *html = cmark_node_next(fenced);
STR_EQ(runner, cmark_node_get_literal(html),
"<div>html</div>\n", "get_literal html");
cmark_node *paragraph = cmark_node_next(html);
INT_EQ(runner, cmark_node_get_start_line(paragraph), 19,
"get_start_line");
INT_EQ(runner, cmark_node_get_start_column(paragraph), 1,
"get_start_column");
INT_EQ(runner, cmark_node_get_end_line(paragraph), 19,
"get_end_line");
cmark_node *link = cmark_node_first_child(paragraph);
STR_EQ(runner, cmark_node_get_url(link), "url",
"get_url");
STR_EQ(runner, cmark_node_get_title(link), "title",
"get_title");
cmark_node *string = cmark_node_first_child(link);
STR_EQ(runner, cmark_node_get_literal(string), "link",
"get_literal string");
// Setters
OK(runner, cmark_node_set_header_level(header, 3),
"set_header_level");
OK(runner, cmark_node_set_list_type(bullet_list, CMARK_ORDERED_LIST),
"set_list_type ordered");
OK(runner, cmark_node_set_list_delim(bullet_list, CMARK_PAREN_DELIM),
"set_list_delim paren");
OK(runner, cmark_node_set_list_start(bullet_list, 3),
"set_list_start");
OK(runner, cmark_node_set_list_tight(bullet_list, 0),
"set_list_tight loose");
OK(runner, cmark_node_set_list_type(ordered_list, CMARK_BULLET_LIST),
"set_list_type bullet");
OK(runner, cmark_node_set_list_tight(ordered_list, 1),
"set_list_tight tight");
OK(runner, cmark_node_set_literal(code, "CODE\n"),
"set_literal indented code");
OK(runner, cmark_node_set_literal(fenced, "FENCED\n"),
"set_literal fenced code");
OK(runner, cmark_node_set_fence_info(fenced, "LANG"),
"set_fence_info");
OK(runner, cmark_node_set_literal(html, "<div>HTML</div>\n"),
"set_literal html");
OK(runner, cmark_node_set_url(link, "URL"),
"set_url");
OK(runner, cmark_node_set_title(link, "TITLE"),
"set_title");
OK(runner, cmark_node_set_literal(string, "LINK"),
"set_literal string");
char *rendered_html = cmark_render_html(doc, CMARK_OPT_DEFAULT);
static const char expected_html[] =
"<h3>Header</h3>\n"
"<ol start=\"3\">\n"
"<li>\n"
"<p>Item 1</p>\n"
"</li>\n"
"<li>\n"
"<p>Item 2</p>\n"
"</li>\n"
"</ol>\n"
"<ul>\n"
"<li>Item 1</li>\n"
"<li>Item 2</li>\n"
"</ul>\n"
"<pre><code>CODE\n"
"</code></pre>\n"
"<pre><code class=\"language-LANG\">FENCED\n"
"</code></pre>\n"
"<div>HTML</div>\n"
"<p><a href=\"URL\" title=\"TITLE\">LINK</a></p>\n";
STR_EQ(runner, rendered_html, expected_html, "setters work");
free(rendered_html);
// Getter errors
INT_EQ(runner, cmark_node_get_header_level(bullet_list), 0,
"get_header_level error");
INT_EQ(runner, cmark_node_get_list_type(header), CMARK_NO_LIST,
"get_list_type error");
INT_EQ(runner, cmark_node_get_list_start(code), 0,
"get_list_start error");
INT_EQ(runner, cmark_node_get_list_tight(fenced), 0,
"get_list_tight error");
OK(runner, cmark_node_get_literal(ordered_list) == NULL,
"get_literal error");
OK(runner, cmark_node_get_fence_info(paragraph) == NULL,
"get_fence_info error");
OK(runner, cmark_node_get_url(html) == NULL,
"get_url error");
OK(runner, cmark_node_get_title(header) == NULL,
"get_title error");
// Setter errors
OK(runner, !cmark_node_set_header_level(bullet_list, 3),
"set_header_level error");
OK(runner, !cmark_node_set_list_type(header, CMARK_ORDERED_LIST),
"set_list_type error");
OK(runner, !cmark_node_set_list_start(code, 3),
"set_list_start error");
OK(runner, !cmark_node_set_list_tight(fenced, 0),
"set_list_tight error");
OK(runner, !cmark_node_set_literal(ordered_list, "content\n"),
"set_literal error");
OK(runner, !cmark_node_set_fence_info(paragraph, "lang"),
"set_fence_info error");
OK(runner, !cmark_node_set_url(html, "url"),
"set_url error");
OK(runner, !cmark_node_set_title(header, "title"),
"set_title error");
OK(runner, !cmark_node_set_header_level(header, 0),
"set_header_level too small");
OK(runner, !cmark_node_set_header_level(header, 7),
"set_header_level too large");
OK(runner, !cmark_node_set_list_type(bullet_list, CMARK_NO_LIST),
"set_list_type invalid");
OK(runner, !cmark_node_set_list_start(bullet_list, -1),
"set_list_start negative");
cmark_node_free(doc);
}
static cmark_node*
finalize(cmark_parser *parser, cmark_node* b, int line_number)
{
int firstlinelen;
int pos;
cmark_node* item;
cmark_node* subitem;
cmark_node* parent;
parent = b->parent;
// don't do anything if the cmark_node is already closed
if (!b->open)
return parent;
b->open = false;
if (line_number > b->start_line) {
b->end_line = line_number - 1;
} else {
b->end_line = line_number;
}
switch (b->type) {
case NODE_PARAGRAPH:
while (cmark_strbuf_at(&b->string_content, 0) == '[' &&
(pos = cmark_parse_reference_inline(&b->string_content, parser->refmap))) {
cmark_strbuf_drop(&b->string_content, pos);
}
if (is_blank(&b->string_content, 0)) {
// remove blank node (former reference def)
cmark_node_free(b);
}
break;
case NODE_CODE_BLOCK:
if (!b->as.code.fenced) { // indented code
remove_trailing_blank_lines(&b->string_content);
cmark_strbuf_putc(&b->string_content, '\n');
} else {
// first line of contents becomes info
firstlinelen = cmark_strbuf_strchr(&b->string_content, '\n', 0);
cmark_strbuf tmp = GH_BUF_INIT;
houdini_unescape_html_f(
&tmp,
b->string_content.ptr,
firstlinelen
);
cmark_strbuf_trim(&tmp);
cmark_strbuf_unescape(&tmp);
b->as.code.info = cmark_chunk_buf_detach(&tmp);
cmark_strbuf_drop(&b->string_content, firstlinelen + 1);
}
b->as.code.literal = cmark_chunk_buf_detach(&b->string_content);
break;
case NODE_HTML:
b->as.literal = cmark_chunk_buf_detach(&b->string_content);
break;
case NODE_LIST: // determine tight/loose status
b->as.list.tight = true; // tight by default
item = b->first_child;
while (item) {
// check for non-final non-empty list item ending with blank line:
if (item->last_line_blank && item->next) {
b->as.list.tight = false;
break;
}
// recurse into children of list item, to see if there are
// spaces between them:
subitem = item->first_child;
while (subitem) {
if (ends_with_blank_line(subitem) &&
(item->next || subitem->next)) {
b->as.list.tight = false;
break;
}
subitem = subitem->next;
}
if (!(b->as.list.tight)) {
break;
}
item = item->next;
}
break;
default:
break;
}
return parent;
}
void
hierarchy(test_batch_runner *runner)
{
cmark_node *bquote1 = cmark_node_new(CMARK_NODE_BLOCK_QUOTE);
cmark_node *bquote2 = cmark_node_new(CMARK_NODE_BLOCK_QUOTE);
cmark_node *bquote3 = cmark_node_new(CMARK_NODE_BLOCK_QUOTE);
OK(runner, cmark_node_append_child(bquote1, bquote2),
"append bquote2");
OK(runner, cmark_node_append_child(bquote2, bquote3),
"append bquote3");
OK(runner, !cmark_node_append_child(bquote3, bquote3),
"adding a node as child of itself fails");
OK(runner, !cmark_node_append_child(bquote3, bquote1),
"adding a parent as child fails");
cmark_node_free(bquote1);
int max_node_type = CMARK_NODE_LAST_BLOCK > CMARK_NODE_LAST_INLINE
? CMARK_NODE_LAST_BLOCK : CMARK_NODE_LAST_INLINE;
OK(runner, max_node_type < 32, "all node types < 32");
int list_item_flag = 1 << CMARK_NODE_ITEM;
int top_level_blocks =
(1 << CMARK_NODE_BLOCK_QUOTE) |
(1 << CMARK_NODE_LIST) |
(1 << CMARK_NODE_CODE_BLOCK) |
(1 << CMARK_NODE_HTML) |
(1 << CMARK_NODE_PARAGRAPH) |
(1 << CMARK_NODE_HEADER) |
(1 << CMARK_NODE_HRULE);
int all_inlines =
(1 << CMARK_NODE_TEXT) |
(1 << CMARK_NODE_SOFTBREAK) |
(1 << CMARK_NODE_LINEBREAK) |
(1 << CMARK_NODE_CODE) |
(1 << CMARK_NODE_INLINE_HTML) |
(1 << CMARK_NODE_EMPH) |
(1 << CMARK_NODE_STRONG) |
(1 << CMARK_NODE_LINK) |
(1 << CMARK_NODE_IMAGE);
test_content(runner, CMARK_NODE_DOCUMENT, top_level_blocks);
test_content(runner, CMARK_NODE_BLOCK_QUOTE, top_level_blocks);
test_content(runner, CMARK_NODE_LIST, list_item_flag);
test_content(runner, CMARK_NODE_ITEM, top_level_blocks);
test_content(runner, CMARK_NODE_CODE_BLOCK , 0);
test_content(runner, CMARK_NODE_HTML, 0);
test_content(runner, CMARK_NODE_PARAGRAPH, all_inlines);
test_content(runner, CMARK_NODE_HEADER, all_inlines);
test_content(runner, CMARK_NODE_HRULE, 0);
test_content(runner, CMARK_NODE_TEXT, 0);
test_content(runner, CMARK_NODE_SOFTBREAK, 0);
test_content(runner, CMARK_NODE_LINEBREAK, 0);
test_content(runner, CMARK_NODE_CODE, 0);
test_content(runner, CMARK_NODE_INLINE_HTML, 0);
test_content(runner, CMARK_NODE_EMPH, all_inlines);
test_content(runner, CMARK_NODE_STRONG, all_inlines);
test_content(runner, CMARK_NODE_LINK, all_inlines);
test_content(runner, CMARK_NODE_IMAGE, all_inlines);
}
static delimiter*
S_insert_emph(subject *subj, delimiter *opener, delimiter *closer)
{
delimiter *delim, *tmp_delim;
int use_delims;
cmark_node *opener_inl = opener->inl_text;
cmark_node *closer_inl = closer->inl_text;
int opener_num_chars = opener_inl->as.literal.len;
int closer_num_chars = closer_inl->as.literal.len;
cmark_node *tmp, *emph, *first_child, *last_child;
// calculate the actual number of characters used from this closer
if (closer_num_chars < 3 || opener_num_chars < 3) {
use_delims = closer_num_chars <= opener_num_chars ?
closer_num_chars : opener_num_chars;
} else { // closer and opener both have >= 3 characters
use_delims = closer_num_chars % 2 == 0 ? 2 : 1;
}
// remove used characters from associated inlines.
opener_num_chars -= use_delims;
closer_num_chars -= use_delims;
opener_inl->as.literal.len = opener_num_chars;
closer_inl->as.literal.len = closer_num_chars;
// free delimiters between opener and closer
delim = closer->previous;
while (delim != NULL && delim != opener) {
tmp_delim = delim->previous;
remove_delimiter(subj, delim);
delim = tmp_delim;
}
first_child = opener_inl->next;
last_child = closer_inl->prev;
// if opener has 0 characters, remove it and its associated inline
if (opener_num_chars == 0) {
// replace empty opener inline with emph
cmark_chunk_free(&(opener_inl->as.literal));
emph = opener_inl;
emph->type = use_delims == 1 ? NODE_EMPH : NODE_STRONG;
// remove opener from list
remove_delimiter(subj, opener);
} else {
// create new emph or strong, and splice it in to our inlines
// between the opener and closer
emph = use_delims == 1 ? make_emph() : make_strong();
emph->parent = opener_inl->parent;
emph->prev = opener_inl;
opener_inl->next = emph;
}
// push children below emph
emph->next = closer_inl;
closer_inl->prev = emph;
emph->first_child = first_child;
emph->last_child = last_child;
// fix children pointers
first_child->prev = NULL;
last_child->next = NULL;
for (tmp = first_child; tmp != NULL; tmp = tmp->next) {
tmp->parent = emph;
}
// if closer has 0 characters, remove it and its associated inline
if (closer_num_chars == 0) {
// remove empty closer inline
cmark_node_free(closer_inl);
// remove closer from list
tmp_delim = closer->next;
remove_delimiter(subj, closer);
closer = tmp_delim;
}
return closer;
}
static delimiter *S_insert_emph(subject *subj, delimiter *opener,
delimiter *closer) {
delimiter *delim, *tmp_delim;
bufsize_t use_delims;
cmark_node *opener_inl = opener->inl_text;
cmark_node *closer_inl = closer->inl_text;
bufsize_t opener_num_chars = opener_inl->as.literal.len;
bufsize_t closer_num_chars = closer_inl->as.literal.len;
cmark_node *tmp, *tmpnext, *emph;
// calculate the actual number of characters used from this closer
if (closer_num_chars < 3 || opener_num_chars < 3) {
use_delims = closer_num_chars <= opener_num_chars ? closer_num_chars
: opener_num_chars;
} else { // closer and opener both have >= 3 characters
use_delims = closer_num_chars % 2 == 0 ? 2 : 1;
}
// remove used characters from associated inlines.
opener_num_chars -= use_delims;
closer_num_chars -= use_delims;
opener_inl->as.literal.len = opener_num_chars;
closer_inl->as.literal.len = closer_num_chars;
// free delimiters between opener and closer
delim = closer->previous;
while (delim != NULL && delim != opener) {
tmp_delim = delim->previous;
remove_delimiter(subj, delim);
delim = tmp_delim;
}
// create new emph or strong, and splice it in to our inlines
// between the opener and closer
emph = use_delims == 1 ? make_emph(subj->mem) : make_strong(subj->mem);
tmp = opener_inl->next;
while (tmp && tmp != closer_inl) {
tmpnext = tmp->next;
cmark_node_append_child(emph, tmp);
tmp = tmpnext;
}
cmark_node_insert_after(opener_inl, emph);
// if opener has 0 characters, remove it and its associated inline
if (opener_num_chars == 0) {
cmark_node_free(opener_inl);
remove_delimiter(subj, opener);
}
// if closer has 0 characters, remove it and its associated inline
if (closer_num_chars == 0) {
// remove empty closer inline
cmark_node_free(closer_inl);
// remove closer from list
tmp_delim = closer->next;
remove_delimiter(subj, closer);
closer = tmp_delim;
}
return closer;
}
int main(int argc, char *argv[]) {
int i, numfps = 0;
int *files;
char buffer[4096];
cmark_parser *parser;
size_t bytes;
cmark_node *document;
int width = 0;
char *unparsed;
writer_format writer = FORMAT_HTML;
int options = CMARK_OPT_DEFAULT;
#if defined(_WIN32) && !defined(__CYGWIN__)
_setmode(_fileno(stdout), _O_BINARY);
#endif
files = (int *)malloc(argc * sizeof(*files));
for (i = 1; i < argc; i++) {
if (strcmp(argv[i], "--version") == 0) {
printf("cmark %s", CMARK_VERSION_STRING);
printf(" - CommonMark converter\n(C) 2014, 2015 John MacFarlane\n");
exit(0);
} else if (strcmp(argv[i], "--sourcepos") == 0) {
options |= CMARK_OPT_SOURCEPOS;
} else if (strcmp(argv[i], "--hardbreaks") == 0) {
options |= CMARK_OPT_HARDBREAKS;
} else if (strcmp(argv[i], "--smart") == 0) {
options |= CMARK_OPT_SMART;
} else if (strcmp(argv[i], "--safe") == 0) {
options |= CMARK_OPT_SAFE;
} else if (strcmp(argv[i], "--normalize") == 0) {
options |= CMARK_OPT_NORMALIZE;
} else if (strcmp(argv[i], "--validate-utf8") == 0) {
options |= CMARK_OPT_VALIDATE_UTF8;
} else if ((strcmp(argv[i], "--help") == 0) ||
(strcmp(argv[i], "-h") == 0)) {
print_usage();
exit(0);
} else if (strcmp(argv[i], "--width") == 0) {
i += 1;
if (i < argc) {
width = (int)strtol(argv[i], &unparsed, 10);
if (unparsed && strlen(unparsed) > 0) {
fprintf(stderr, "failed parsing width '%s' at '%s'\n", argv[i],
unparsed);
exit(1);
}
} else {
fprintf(stderr, "--width requires an argument\n");
exit(1);
}
} else if ((strcmp(argv[i], "-t") == 0) || (strcmp(argv[i], "--to") == 0)) {
i += 1;
if (i < argc) {
if (strcmp(argv[i], "man") == 0) {
writer = FORMAT_MAN;
} else if (strcmp(argv[i], "html") == 0) {
writer = FORMAT_HTML;
} else if (strcmp(argv[i], "xml") == 0) {
writer = FORMAT_XML;
} else if (strcmp(argv[i], "commonmark") == 0) {
writer = FORMAT_COMMONMARK;
} else if (strcmp(argv[i], "latex") == 0) {
writer = FORMAT_LATEX;
} else {
fprintf(stderr, "Unknown format %s\n", argv[i]);
exit(1);
}
} else {
fprintf(stderr, "No argument provided for %s\n", argv[i - 1]);
exit(1);
}
} else if (*argv[i] == '-') {
print_usage();
exit(1);
} else { // treat as file argument
files[numfps++] = i;
}
}
parser = cmark_parser_new(options);
for (i = 0; i < numfps; i++) {
FILE *fp = fopen(argv[files[i]], "rb");
if (fp == NULL) {
fprintf(stderr, "Error opening file %s: %s\n", argv[files[i]],
strerror(errno));
exit(1);
}
while ((bytes = fread(buffer, 1, sizeof(buffer), fp)) > 0) {
cmark_parser_feed(parser, buffer, bytes);
if (bytes < sizeof(buffer)) {
break;
}
}
fclose(fp);
}
if (numfps == 0) {
while ((bytes = fread(buffer, 1, sizeof(buffer), stdin)) > 0) {
cmark_parser_feed(parser, buffer, bytes);
if (bytes < sizeof(buffer)) {
break;
}
}
}
document = cmark_parser_finish(parser);
cmark_parser_free(parser);
print_document(document, writer, options, width);
cmark_node_free(document);
free(files);
return 0;
}
int main(int argc, char *argv[])
{
int i, numfps = 0;
bool ast = false;
int *files;
char buffer[4096];
cmark_parser *parser;
size_t bytes;
cmark_node *document;
parser = cmark_parser_new();
files = (int *)malloc(argc * sizeof(*files));
for (i = 1; i < argc; i++) {
if (strcmp(argv[i], "--version") == 0) {
printf("cmark %s", CMARK_VERSION);
printf(" - CommonMark converter (c) 2014 John MacFarlane\n");
exit(0);
} else if ((strcmp(argv[i], "--help") == 0) ||
(strcmp(argv[i], "-h") == 0)) {
print_usage();
exit(0);
} else if (strcmp(argv[i], "--ast") == 0) {
ast = true;
} else if (*argv[i] == '-') {
print_usage();
exit(1);
} else { // treat as file argument
files[numfps++] = i;
}
}
for (i = 0; i < numfps; i++) {
FILE *fp = fopen(argv[files[i]], "r");
if (fp == NULL) {
fprintf(stderr, "Error opening file %s: %s\n",
argv[files[i]], strerror(errno));
exit(1);
}
start_timer();
while ((bytes = fread(buffer, 1, sizeof(buffer), fp)) > 0) {
cmark_parser_feed(parser, buffer, bytes);
}
end_timer("processing lines");
fclose(fp);
}
if (numfps == 0) {
/*
document = cmark_parse_file(stdin);
print_document(document, ast);
exit(0);
*/
while ((bytes = fread(buffer, 1, sizeof(buffer), stdin)) > 0) {
cmark_parser_feed(parser, buffer, bytes);
}
}
start_timer();
document = cmark_parser_finish(parser);
end_timer("finishing document");
cmark_parser_free(parser);
start_timer();
print_document(document, ast);
end_timer("print_document");
start_timer();
cmark_node_free(document);
end_timer("free_blocks");
free(files);
return 0;
}
