size_t
ups_autolink__www(size_t *rewind_p, struct buf *link, char *data, size_t offset, size_t size)
{
size_t link_end;
int np = 0;
if (offset > 0 && !ispunct(data[-1]) && !isspace(data[-1]))
return 0;
if (size < 4 || memcmp(data, "www.", STRLEN("www.")) != 0)
return 0;
link_end = 0;
while (link_end < size && !isspace(data[link_end])) {
if (data[link_end] == '.')
np++;
link_end++;
}
if (np < 2)
return 0;
link_end = autolink_delim(data, link_end, offset, size);
if (link_end == 0)
return 0;
bufput(link, data, link_end);
*rewind_p = 0;
return (int)link_end;
}
size_t
sd_autolink__www(
size_t *rewind_p,
struct buf *link,
uint8_t *data,
size_t max_rewind,
size_t size,
unsigned int flags)
{
size_t link_end;
if (max_rewind > 0 && !ispunct(data[-1]) && !isspace(data[-1]))
return 0;
if (size < 4 || memcmp(data, "www.", strlen("www.")) != 0)
return 0;
link_end = check_domain(data, size, 0);
if (link_end == 0)
return 0;
while (link_end < size && !isspace(data[link_end]))
link_end++;
link_end = autolink_delim(data, link_end, max_rewind, size);
if (link_end == 0)
return 0;
bufput(link, data, link_end);
*rewind_p = 0;
return (int)link_end;
}
size_t
ups_autolink__url(size_t *rewind_p, struct buf *link, char *data, size_t offset, size_t size)
{
size_t link_end, rewind = 0;
if (size < 4 || data[1] != '/' || data[2] != '/')
return 0;
while (rewind < offset && isalpha(data[-rewind - 1]))
rewind++;
if (!is_safe_link(data - rewind, size + rewind))
return 0;
link_end = 0;
while (link_end < size && !isspace(data[link_end]))
link_end++;
link_end = autolink_delim(data, link_end, offset, size);
if (link_end == 0)
return 0;
bufput(link, data - rewind, link_end + rewind);
*rewind_p = rewind;
return link_end;
}
size_t
sd_autolink__url(size_t *rewind_p, struct buf *link, uint8_t *data, size_t offset, size_t size)
{
size_t link_end, rewind = 0, domain_len;
if (size < 4 || data[1] != '/' || data[2] != '/')
return 0;
while (rewind < offset && isalpha(data[-rewind - 1]))
rewind++;
if (!sd_autolink_issafe(data - rewind, size + rewind))
return 0;
link_end = strlen("://");
domain_len = check_domain(data + link_end, size - link_end);
if (domain_len == 0)
return 0;
link_end += domain_len;
while (link_end < size && !isspace(data[link_end]))
link_end++;
link_end = autolink_delim(data, link_end, offset, size);
if (link_end == 0)
return 0;
bufput(link, data - rewind, link_end + rewind);
*rewind_p = rewind;
return link_end;
}
size_t
rfcdown_autolink__email(
size_t *rewind_p,
rfcdown_buffer *link,
uint8_t *data,
size_t max_rewind,
size_t size,
unsigned int flags)
{
size_t link_end, rewind;
int nb = 0, np = 0;
for (rewind = 0; rewind < max_rewind; ++rewind) {
uint8_t c = data[-1 - rewind];
if (isalnum(c))
continue;
if (strchr(".+-_", c) != NULL)
continue;
break;
}
if (rewind == 0)
return 0;
for (link_end = 0; link_end < size; ++link_end) {
uint8_t c = data[link_end];
if (isalnum(c))
continue;
if (c == '@')
nb++;
else if (c == '.' && link_end < size - 1)
np++;
else if (c != '-' && c != '_')
break;
}
if (link_end < 2 || nb != 1 || np == 0 ||
!isalpha(data[link_end - 1]))
return 0;
link_end = autolink_delim(data, link_end, max_rewind, size);
if (link_end == 0)
return 0;
rfcdown_buffer_put(link, data - rewind, link_end + rewind);
*rewind_p = rewind;
return link_end;
}
size_t
sd_autolink__email(
size_t *rewind_p,
struct buf *link,
uint8_t *data,
size_t offset,
size_t size,
unsigned int flags)
{
size_t link_end, rewind;
int nb = 0, np = 0;
for (rewind = 0; rewind < offset; ++rewind) {
uint8_t c = data[-rewind - 1];
if (isalnum(c))
continue;
if (strchr(".+-_", c) != NULL)
continue;
break;
}
if (rewind == 0)
return 0;
for (link_end = 0; link_end < size; ++link_end) {
uint8_t c = data[link_end];
if (isalnum(c))
continue;
if (c == '@')
nb++;
else if (c == '.' && link_end < size - 1)
np++;
else if (c != '-' && c != '_')
break;
}
if (link_end < 2 || nb != 1 || np == 0)
return 0;
link_end = autolink_delim(data, link_end, offset, size);
if (link_end == 0)
return 0;
bufput(link, data - rewind, link_end + rewind);
*rewind_p = rewind;
return link_end;
}
static cmark_node *www_match(cmark_parser *parser, cmark_node *parent,
cmark_inline_parser *inline_parser) {
cmark_chunk *chunk = cmark_inline_parser_get_chunk(inline_parser);
size_t max_rewind = cmark_inline_parser_get_offset(inline_parser);
uint8_t *data = chunk->data + max_rewind;
size_t size = chunk->len - max_rewind;
int start = cmark_inline_parser_get_column(inline_parser);
size_t link_end;
if (max_rewind > 0 && strchr("*_~(", data[-1]) == NULL &&
!cmark_isspace(data[-1]))
return 0;
if (size < 4 || memcmp(data, "www.", strlen("www.")) != 0)
return 0;
link_end = check_domain(data, size, 0);
if (link_end == 0)
return NULL;
while (link_end < size && !cmark_isspace(data[link_end]))
link_end++;
link_end = autolink_delim(data, link_end);
if (link_end == 0)
return NULL;
cmark_inline_parser_set_offset(inline_parser, (int)(max_rewind + link_end));
cmark_node *node = cmark_node_new_with_mem(CMARK_NODE_LINK, parser->mem);
cmark_strbuf buf;
cmark_strbuf_init(parser->mem, &buf, 10);
cmark_strbuf_puts(&buf, "http://");
cmark_strbuf_put(&buf, data, (bufsize_t)link_end);
node->as.link.url = cmark_chunk_buf_detach(&buf);
cmark_node *text = cmark_node_new_with_mem(CMARK_NODE_TEXT, parser->mem);
text->as.literal =
cmark_chunk_dup(chunk, (bufsize_t)max_rewind, (bufsize_t)link_end);
cmark_node_append_child(node, text);
node->start_line = text->start_line =
node->end_line = text->end_line =
cmark_inline_parser_get_line(inline_parser);
node->start_column = text->start_column = start - 1;
node->end_column = text->end_column = cmark_inline_parser_get_column(inline_parser) - 1;
return node;
}
static cmark_node *url_match(cmark_parser *parser, cmark_node *parent,
cmark_inline_parser *inline_parser) {
size_t link_end, domain_len;
int rewind = 0;
cmark_chunk *chunk = cmark_inline_parser_get_chunk(inline_parser);
int max_rewind = cmark_inline_parser_get_offset(inline_parser);
uint8_t *data = chunk->data + max_rewind;
size_t size = chunk->len - max_rewind;
if (size < 4 || data[1] != '/' || data[2] != '/')
return 0;
while (rewind < max_rewind && cmark_isalpha(data[-rewind - 1]))
rewind++;
if (!sd_autolink_issafe(data - rewind, size + rewind))
return 0;
link_end = strlen("://");
domain_len = check_domain(data + link_end, size - link_end, 1);
if (domain_len == 0)
return 0;
link_end += domain_len;
while (link_end < size && !cmark_isspace(data[link_end]))
link_end++;
link_end = autolink_delim(data, link_end);
if (link_end == 0)
return NULL;
cmark_inline_parser_set_offset(inline_parser, (int)(max_rewind + link_end));
cmark_node_unput(parent, rewind);
cmark_node *node = cmark_node_new_with_mem(CMARK_NODE_LINK, parser->mem);
cmark_chunk url = cmark_chunk_dup(chunk, max_rewind - rewind,
(bufsize_t)(link_end + rewind));
node->as.link.url = url;
cmark_node *text = cmark_node_new_with_mem(CMARK_NODE_TEXT, parser->mem);
text->as.literal = url;
cmark_node_append_child(node, text);
return node;
}
size_t
sd_autolink__url(
size_t *rewind_p,
struct buf *link,
uint8_t *data,
size_t max_rewind,
size_t size,
unsigned int flags)
{
size_t link_end, rewind = 0, domain_len;
if (size < 4 || data[1] != '/' || data[2] != '/')
return 0;
while (rewind < max_rewind && isalpha(data[-rewind - 1]))
rewind++;
if (!sd_autolink_issafe(data - rewind, size + rewind))
return 0;
link_end = strlen("://");
domain_len = check_domain(
data + link_end,
size - link_end,
flags & SD_AUTOLINK_SHORT_DOMAINS);
if (domain_len == 0)
return 0;
link_end += domain_len;
while (link_end < size && !isspace(data[link_end]))
link_end++;
link_end = autolink_delim(data, link_end, max_rewind, size);
if (link_end == 0)
return 0;
bufput(link, data - rewind, link_end + rewind);
*rewind_p = rewind;
return link_end;
}
static void postprocess_text(cmark_parser *parser, cmark_node *text, int offset) {
size_t link_end;
uint8_t *data = text->as.literal.data,
*at;
size_t size = text->as.literal.len;
int rewind, max_rewind,
nb = 0, np = 0, ns = 0;
if (offset < 0 || (size_t)offset >= size)
return;
data += offset;
size -= offset;
at = (uint8_t *)memchr(data, '@', size);
if (!at)
return;
max_rewind = (int)(at - data);
data += max_rewind;
size -= max_rewind;
for (rewind = 0; rewind < max_rewind; ++rewind) {
uint8_t c = data[-rewind - 1];
if (cmark_isalnum(c))
continue;
if (strchr(".+-_", c) != NULL)
continue;
if (c == '/')
ns++;
break;
}
if (rewind == 0 || ns > 0) {
postprocess_text(parser, text, max_rewind + 1 + offset);
return;
}
for (link_end = 0; link_end < size; ++link_end) {
uint8_t c = data[link_end];
if (cmark_isalnum(c))
continue;
if (c == '@')
nb++;
else if (c == '.' && link_end < size - 1 && cmark_isalnum(data[link_end + 1]))
np++;
else if (c != '-' && c != '_')
break;
}
if (link_end < 2 || nb != 1 || np == 0 ||
(!cmark_isalpha(data[link_end - 1]) && data[link_end - 1] != '.')) {
postprocess_text(parser, text, max_rewind + 1 + offset);
return;
}
link_end = autolink_delim(data, link_end);
if (link_end == 0) {
postprocess_text(parser, text, max_rewind + 1 + offset);
return;
}
cmark_chunk_to_cstr(parser->mem, &text->as.literal);
cmark_node *link_node = cmark_node_new_with_mem(CMARK_NODE_LINK, parser->mem);
cmark_strbuf buf;
cmark_strbuf_init(parser->mem, &buf, 10);
cmark_strbuf_puts(&buf, "mailto:");
cmark_strbuf_put(&buf, data - rewind, (bufsize_t)(link_end + rewind));
link_node->as.link.url = cmark_chunk_buf_detach(&buf);
cmark_node *link_text = cmark_node_new_with_mem(CMARK_NODE_TEXT, parser->mem);
cmark_chunk email = cmark_chunk_dup(
&text->as.literal,
offset + max_rewind - rewind,
(bufsize_t)(link_end + rewind));
cmark_chunk_to_cstr(parser->mem, &email);
link_text->as.literal = email;
cmark_node_append_child(link_node, link_text);
cmark_node_insert_after(text, link_node);
cmark_node *post = cmark_node_new_with_mem(CMARK_NODE_TEXT, parser->mem);
post->as.literal = cmark_chunk_dup(&text->as.literal,
(bufsize_t)(offset + max_rewind + link_end),
(bufsize_t)(size - link_end));
cmark_chunk_to_cstr(parser->mem, &post->as.literal);
cmark_node_insert_after(link_node, post);
text->as.literal.len = offset + max_rewind - rewind;
text->as.literal.data[text->as.literal.len] = 0;
postprocess_text(parser, post, 0);
}
