dany/antd-lua-plugin
1
0
Fork 0
mirror of https://github.com/lxsang/antd-lua-plugin synced 2025-02-20 17:52:47 +01:00
Code Issues Projects Releases Wiki Activity

add markdown support

Browse Source
This commit is contained in:
lxsang 2020-09-01 17:21:21 +02:00
parent 22d16d0f9e
commit 2a89d5a441
11 changed files with 9649 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
configure.ac
View File

@ -128,6 +128,7 @@ AC_CONFIG_FILES([
lib/Makefile
lib/core/Makefile
lib/asl/Makefile
lib/md/Makefile
])
if test x"${has_cmake}" == x"yes" ; then

BIN
dist/lua-0.5.2b.tar.gz vendored
View File

Binary file not shown.

2
lib/Makefile.am
View File

@ -1,4 +1,4 @@
SUBDIRS = core asl
SUBDIRS = core asl md
if HAS_CMAKE
SUBDIRS += ann
endif

9
lib/md/Makefile.am Normal file
View File

@ -0,0 +1,9 @@
lib_LTLIBRARIES = md.la
md_la_LDFLAGS = -module -avoid-version -shared
md_la_SOURCES = md.c md4c/entity.c md4c/md4c.c md4c/md4c-html.c
libdir=$(prefix)/lib/lua/
EXTRA_DIST = md4c/*.h

62
lib/md/md.c Normal file
View File

@ -0,0 +1,62 @@
#include "../lualib.h"
#include "md4c/md4c-html.h"
static void md_process_output(const MD_CHAR *buf, MD_SIZE len, void *udata)
{
lua_State *L = (lua_State *)udata;
lua_pushlstring(L, buf, len);
lua_call(L, 1, 0);
lua_pushvalue(L, -1);
}
static int l_md_to_html(lua_State *L)
{
const char *input = luaL_checkstring(L, 1);
if (input == NULL)
{
ERROR("NULL markdown input string");
return 0;
}
if (!lua_isfunction(L, -1))
{
ERROR("Invalid callback function");
return 0;
}
// duplicate top of the stack
lua_pushvalue(L, -1);
if (md_html(input,
strlen(input),
md_process_output,
L,
MD_DIALECT_GITHUB |
MD_HTML_FLAG_VERBATIM_ENTITIES |
MD_FLAG_PERMISSIVEATXHEADERS |
MD_FLAG_NOINDENTEDCODEBLOCKS |
MD_FLAG_NOHTMLBLOCKS |
MD_FLAG_NOHTMLSPANS |
MD_FLAG_NOHTML |
MD_FLAG_COLLAPSEWHITESPACE |
MD_FLAG_PERMISSIVEURLAUTOLINKS |
MD_FLAG_PERMISSIVEWWWAUTOLINKS |
MD_FLAG_PERMISSIVEEMAILAUTOLINKS |
MD_FLAG_PERMISSIVEAUTOLINKS |
MD_FLAG_UNDERLINE,
MD_HTML_FLAG_XHTML) == -1)
{
ERROR("Unable to parse markdown: md_parse() fails");
lua_pop(L,1);
return 0;
}
lua_pop(L,1);
return 1;
}
static const struct luaL_Reg _lib[] = {
{"to_html", l_md_to_html},
{NULL, NULL}};
int luaopen_md(lua_State *L)
{
luaL_newlib(L, _lib);
return 1;
}

2190
lib/md/md4c/entity.c Normal file
View File

File diff suppressed because it is too large Load Diff

42
lib/md/md4c/entity.h Normal file
View File

@ -0,0 +1,42 @@
/*
* MD4C: Markdown parser for C
* (http://github.com/mity/md4c)
*
* Copyright (c) 2016-2019 Martin Mitas
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#ifndef MD4C_ENTITY_H
#define MD4C_ENTITY_H
#include <stdlib.h>
/* Most entities are formed by single Unicode codepoint, few by two codepoints.
* Single-codepoint entities have codepoints[1] set to zero. */
struct entity {
const char* name;
unsigned codepoints[2];
};
const struct entity* entity_lookup(const char* name, size_t name_size);
#endif /* MD4C_ENTITY_H */

573
lib/md/md4c/md4c-html.c Normal file
View File

@ -0,0 +1,573 @@
/*
* MD4C: Markdown parser for C
* (http://github.com/mity/md4c)
*
* Copyright (c) 2016-2019 Martin Mitas
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#include <stdio.h>
#include <string.h>
#include "md4c-html.h"
#include "entity.h"
#if !defined(__STDC_VERSION__) || __STDC_VERSION__ < 199409L
/* C89/90 or old compilers in general may not understand "inline". */
#if defined __GNUC__
#define inline __inline__
#elif defined _MSC_VER
#define inline __inline
#else
#define inline
#endif
#endif
#ifdef _WIN32
#define snprintf _snprintf
#endif
typedef struct MD_HTML_tag MD_HTML;
struct MD_HTML_tag {
void (*process_output)(const MD_CHAR*, MD_SIZE, void*);
void* userdata;
unsigned flags;
int image_nesting_level;
char escape_map[256];
};
#define NEED_HTML_ESC_FLAG 0x1
#define NEED_URL_ESC_FLAG 0x2
/*****************************************
*** HTML rendering helper functions ***
*****************************************/
#define ISDIGIT(ch) ('0' <= (ch) && (ch) <= '9')
#define ISLOWER(ch) ('a' <= (ch) && (ch) <= 'z')
#define ISUPPER(ch) ('A' <= (ch) && (ch) <= 'Z')
#define ISALNUM(ch) (ISLOWER(ch) || ISUPPER(ch) || ISDIGIT(ch))
static inline void
render_verbatim(MD_HTML* r, const MD_CHAR* text, MD_SIZE size)
{
r->process_output(text, size, r->userdata);
}
/* Keep this as a macro. Most compiler should then be smart enough to replace
* the strlen() call with a compile-time constant if the string is a C literal. */
#define RENDER_VERBATIM(r, verbatim) \
render_verbatim((r), (verbatim), (MD_SIZE) (strlen(verbatim)))
static void
render_html_escaped(MD_HTML* r, const MD_CHAR* data, MD_SIZE size)
{
MD_OFFSET beg = 0;
MD_OFFSET off = 0;
/* Some characters need to be escaped in normal HTML text. */
#define NEED_HTML_ESC(ch) (r->escape_map[(unsigned char)(ch)] & NEED_HTML_ESC_FLAG)
while(1) {
/* Optimization: Use some loop unrolling. */
while(off + 3 < size && !NEED_HTML_ESC(data[off+0]) && !NEED_HTML_ESC(data[off+1])
&& !NEED_HTML_ESC(data[off+2]) && !NEED_HTML_ESC(data[off+3]))
off += 4;
while(off < size && !NEED_HTML_ESC(data[off]))
off++;
if(off > beg)
render_verbatim(r, data + beg, off - beg);
if(off < size) {
switch(data[off]) {
case '&': RENDER_VERBATIM(r, "&amp;"); break;
case '<': RENDER_VERBATIM(r, "&lt;"); break;
case '>': RENDER_VERBATIM(r, "&gt;"); break;
case '"': RENDER_VERBATIM(r, "&quot;"); break;
}
off++;
} else {
break;
}
beg = off;
}
}
static void
render_url_escaped(MD_HTML* r, const MD_CHAR* data, MD_SIZE size)
{
static const MD_CHAR hex_chars[] = "0123456789ABCDEF";
MD_OFFSET beg = 0;
MD_OFFSET off = 0;
/* Some characters need to be escaped in URL attributes. */
#define NEED_URL_ESC(ch) (r->escape_map[(unsigned char)(ch)] & NEED_URL_ESC_FLAG)
while(1) {
while(off < size && !NEED_URL_ESC(data[off]))
off++;
if(off > beg)
render_verbatim(r, data + beg, off - beg);
if(off < size) {
char hex[3];
switch(data[off]) {
case '&': RENDER_VERBATIM(r, "&amp;"); break;
default:
hex[0] = '%';
hex[1] = hex_chars[((unsigned)data[off] >> 4) & 0xf];
hex[2] = hex_chars[((unsigned)data[off] >> 0) & 0xf];
render_verbatim(r, hex, 3);
break;
}
off++;
} else {
break;
}
beg = off;
}
}
static unsigned
hex_val(char ch)
{
if('0' <= ch && ch <= '9')
return ch - '0';
if('A' <= ch && ch <= 'Z')
return ch - 'A' + 10;
else
return ch - 'a' + 10;
}
static void
render_utf8_codepoint(MD_HTML* r, unsigned codepoint,
void (*fn_append)(MD_HTML*, const MD_CHAR*, MD_SIZE))
{
static const MD_CHAR utf8_replacement_char[] = { 0xef, 0xbf, 0xbd };
unsigned char utf8[4];
size_t n;
if(codepoint <= 0x7f) {
n = 1;
utf8[0] = codepoint;
} else if(codepoint <= 0x7ff) {
n = 2;
utf8[0] = 0xc0 | ((codepoint >> 6) & 0x1f);
utf8[1] = 0x80 + ((codepoint >> 0) & 0x3f);
} else if(codepoint <= 0xffff) {
n = 3;
utf8[0] = 0xe0 | ((codepoint >> 12) & 0xf);
utf8[1] = 0x80 + ((codepoint >> 6) & 0x3f);
utf8[2] = 0x80 + ((codepoint >> 0) & 0x3f);
} else {
n = 4;
utf8[0] = 0xf0 | ((codepoint >> 18) & 0x7);
utf8[1] = 0x80 + ((codepoint >> 12) & 0x3f);
utf8[2] = 0x80 + ((codepoint >> 6) & 0x3f);
utf8[3] = 0x80 + ((codepoint >> 0) & 0x3f);
}
if(0 < codepoint && codepoint <= 0x10ffff)
fn_append(r, (char*)utf8, n);
else
fn_append(r, utf8_replacement_char, 3);
}
/* Translate entity to its UTF-8 equivalent, or output the verbatim one
* if such entity is unknown (or if the translation is disabled). */
static void
render_entity(MD_HTML* r, const MD_CHAR* text, MD_SIZE size,
void (*fn_append)(MD_HTML*, const MD_CHAR*, MD_SIZE))
{
if(r->flags & MD_HTML_FLAG_VERBATIM_ENTITIES) {
render_verbatim(r, text, size);
return;
}
/* We assume UTF-8 output is what is desired. */
if(size > 3 && text[1] == '#') {
unsigned codepoint = 0;
if(text[2] == 'x' || text[2] == 'X') {
/* Hexadecimal entity (e.g. "&#x1234abcd;")). */
MD_SIZE i;
for(i = 3; i < size-1; i++)
codepoint = 16 * codepoint + hex_val(text[i]);
} else {
/* Decimal entity (e.g. "&1234;") */
MD_SIZE i;
for(i = 2; i < size-1; i++)
codepoint = 10 * codepoint + (text[i] - '0');
}
render_utf8_codepoint(r, codepoint, fn_append);
return;
} else {
/* Named entity (e.g. "&nbsp;"). */
const struct entity* ent;
ent = entity_lookup(text, size);
if(ent != NULL) {
render_utf8_codepoint(r, ent->codepoints[0], fn_append);
if(ent->codepoints[1])
render_utf8_codepoint(r, ent->codepoints[1], fn_append);
return;
}
}
fn_append(r, text, size);
}
static void
render_attribute(MD_HTML* r, const MD_ATTRIBUTE* attr,
void (*fn_append)(MD_HTML*, const MD_CHAR*, MD_SIZE))
{
int i;
for(i = 0; attr->substr_offsets[i] < attr->size; i++) {
MD_TEXTTYPE type = attr->substr_types[i];
MD_OFFSET off = attr->substr_offsets[i];
MD_SIZE size = attr->substr_offsets[i+1] - off;
const MD_CHAR* text = attr->text + off;
switch(type) {
case MD_TEXT_NULLCHAR: render_utf8_codepoint(r, 0x0000, render_verbatim); break;
case MD_TEXT_ENTITY: render_entity(r, text, size, fn_append); break;
default: fn_append(r, text, size); break;
}
}
}
static void
render_open_ol_block(MD_HTML* r, const MD_BLOCK_OL_DETAIL* det)
{
char buf[64];
if(det->start == 1) {
RENDER_VERBATIM(r, "<ol>\n");
return;
}
snprintf(buf, sizeof(buf), "<ol start=\"%u\">\n", det->start);
RENDER_VERBATIM(r, buf);
}
static void
render_open_li_block(MD_HTML* r, const MD_BLOCK_LI_DETAIL* det)
{
if(det->is_task) {
RENDER_VERBATIM(r, "<li class=\"task-list-item\">"
"<input type=\"checkbox\" class=\"task-list-item-checkbox\" disabled");
if(det->task_mark == 'x' || det->task_mark == 'X')
RENDER_VERBATIM(r, " checked");
RENDER_VERBATIM(r, ">");
} else {
RENDER_VERBATIM(r, "<li>");
}
}
static void
render_open_code_block(MD_HTML* r, const MD_BLOCK_CODE_DETAIL* det)
{
RENDER_VERBATIM(r, "<pre><code");
/* If known, output the HTML 5 attribute class="language-LANGNAME". */
if(det->lang.text != NULL) {
RENDER_VERBATIM(r, " class=\"language-");
render_attribute(r, &det->lang, render_html_escaped);
RENDER_VERBATIM(r, "\"");
}
RENDER_VERBATIM(r, ">");
}
static void
render_open_td_block(MD_HTML* r, const MD_CHAR* cell_type, const MD_BLOCK_TD_DETAIL* det)
{
RENDER_VERBATIM(r, "<");
RENDER_VERBATIM(r, cell_type);
switch(det->align) {
case MD_ALIGN_LEFT: RENDER_VERBATIM(r, " align=\"left\">"); break;
case MD_ALIGN_CENTER: RENDER_VERBATIM(r, " align=\"center\">"); break;
case MD_ALIGN_RIGHT: RENDER_VERBATIM(r, " align=\"right\">"); break;
default: RENDER_VERBATIM(r, ">"); break;
}
}
static void
render_open_a_span(MD_HTML* r, const MD_SPAN_A_DETAIL* det)
{
RENDER_VERBATIM(r, "<a href=\"");
render_attribute(r, &det->href, render_url_escaped);
if(det->title.text != NULL) {
RENDER_VERBATIM(r, "\" title=\"");
render_attribute(r, &det->title, render_html_escaped);
}
RENDER_VERBATIM(r, "\">");
}
static void
render_open_img_span(MD_HTML* r, const MD_SPAN_IMG_DETAIL* det)
{
RENDER_VERBATIM(r, "<img src=\"");
render_attribute(r, &det->src, render_url_escaped);
RENDER_VERBATIM(r, "\" alt=\"");
r->image_nesting_level++;
}
static void
render_close_img_span(MD_HTML* r, const MD_SPAN_IMG_DETAIL* det)
{
if(det->title.text != NULL) {
RENDER_VERBATIM(r, "\" title=\"");
render_attribute(r, &det->title, render_html_escaped);
}
RENDER_VERBATIM(r, (r->flags & MD_HTML_FLAG_XHTML) ? "\" />" : "\">");
r->image_nesting_level--;
}
static void
render_open_wikilink_span(MD_HTML* r, const MD_SPAN_WIKILINK_DETAIL* det)
{
RENDER_VERBATIM(r, "<x-wikilink data-target=\"");
render_attribute(r, &det->target, render_html_escaped);
RENDER_VERBATIM(r, "\">");
}
/**************************************
*** HTML renderer implementation ***
**************************************/
static int
enter_block_callback(MD_BLOCKTYPE type, void* detail, void* userdata)
{
static const MD_CHAR* head[6] = { "<h1>", "<h2>", "<h3>", "<h4>", "<h5>", "<h6>" };
MD_HTML* r = (MD_HTML*) userdata;
switch(type) {
case MD_BLOCK_DOC: /* noop */ break;
case MD_BLOCK_QUOTE: RENDER_VERBATIM(r, "<blockquote>\n"); break;
case MD_BLOCK_UL: RENDER_VERBATIM(r, "<ul>\n"); break;
case MD_BLOCK_OL: render_open_ol_block(r, (const MD_BLOCK_OL_DETAIL*)detail); break;
case MD_BLOCK_LI: render_open_li_block(r, (const MD_BLOCK_LI_DETAIL*)detail); break;
case MD_BLOCK_HR: RENDER_VERBATIM(r, (r->flags & MD_HTML_FLAG_XHTML) ? "<hr />\n" : "<hr>\n"); break;
case MD_BLOCK_H: RENDER_VERBATIM(r, head[((MD_BLOCK_H_DETAIL*)detail)->level - 1]); break;
case MD_BLOCK_CODE: render_open_code_block(r, (const MD_BLOCK_CODE_DETAIL*) detail); break;
case MD_BLOCK_HTML: /* noop */ break;
case MD_BLOCK_P: RENDER_VERBATIM(r, "<p>"); break;
case MD_BLOCK_TABLE: RENDER_VERBATIM(r, "<table>\n"); break;
case MD_BLOCK_THEAD: RENDER_VERBATIM(r, "<thead>\n"); break;
case MD_BLOCK_TBODY: RENDER_VERBATIM(r, "<tbody>\n"); break;
case MD_BLOCK_TR: RENDER_VERBATIM(r, "<tr>\n"); break;
case MD_BLOCK_TH: render_open_td_block(r, "th", (MD_BLOCK_TD_DETAIL*)detail); break;
case MD_BLOCK_TD: render_open_td_block(r, "td", (MD_BLOCK_TD_DETAIL*)detail); break;
}
return 0;
}
static int
leave_block_callback(MD_BLOCKTYPE type, void* detail, void* userdata)
{
static const MD_CHAR* head[6] = { "</h1>\n", "</h2>\n", "</h3>\n", "</h4>\n", "</h5>\n", "</h6>\n" };
MD_HTML* r = (MD_HTML*) userdata;
switch(type) {
case MD_BLOCK_DOC: /*noop*/ break;
case MD_BLOCK_QUOTE: RENDER_VERBATIM(r, "</blockquote>\n"); break;
case MD_BLOCK_UL: RENDER_VERBATIM(r, "</ul>\n"); break;
case MD_BLOCK_OL: RENDER_VERBATIM(r, "</ol>\n"); break;
case MD_BLOCK_LI: RENDER_VERBATIM(r, "</li>\n"); break;
case MD_BLOCK_HR: /*noop*/ break;
case MD_BLOCK_H: RENDER_VERBATIM(r, head[((MD_BLOCK_H_DETAIL*)detail)->level - 1]); break;
case MD_BLOCK_CODE: RENDER_VERBATIM(r, "</code></pre>\n"); break;
case MD_BLOCK_HTML: /* noop */ break;
case MD_BLOCK_P: RENDER_VERBATIM(r, "</p>\n"); break;
case MD_BLOCK_TABLE: RENDER_VERBATIM(r, "</table>\n"); break;
case MD_BLOCK_THEAD: RENDER_VERBATIM(r, "</thead>\n"); break;
case MD_BLOCK_TBODY: RENDER_VERBATIM(r, "</tbody>\n"); break;
case MD_BLOCK_TR: RENDER_VERBATIM(r, "</tr>\n"); break;
case MD_BLOCK_TH: RENDER_VERBATIM(r, "</th>\n"); break;
case MD_BLOCK_TD: RENDER_VERBATIM(r, "</td>\n"); break;
}
return 0;
}
static int
enter_span_callback(MD_SPANTYPE type, void* detail, void* userdata)
{
MD_HTML* r = (MD_HTML*) userdata;
if(r->image_nesting_level > 0) {
/* We are inside a Markdown image label. Markdown allows to use any
* emphasis and other rich contents in that context similarly as in
* any link label.
*
* However, unlike in the case of links (where that contents becomes
* contents of the <a>...</a> tag), in the case of images the contents
* is supposed to fall into the attribute alt: <img alt="...">.
*
* In that context we naturally cannot output nested HTML tags. So lets
* suppress them and only output the plain text (i.e. what falls into
* text() callback).
*
* This make-it-a-plain-text approach is the recommended practice by
* CommonMark specification (for HTML output).
*/
return 0;
}
switch(type) {
case MD_SPAN_EM: RENDER_VERBATIM(r, "<em>"); break;
case MD_SPAN_STRONG: RENDER_VERBATIM(r, "<strong>"); break;
case MD_SPAN_U: RENDER_VERBATIM(r, "<u>"); break;
case MD_SPAN_A: render_open_a_span(r, (MD_SPAN_A_DETAIL*) detail); break;
case MD_SPAN_IMG: render_open_img_span(r, (MD_SPAN_IMG_DETAIL*) detail); break;
case MD_SPAN_CODE: RENDER_VERBATIM(r, "<code>"); break;
case MD_SPAN_DEL: RENDER_VERBATIM(r, "<del>"); break;
case MD_SPAN_LATEXMATH: RENDER_VERBATIM(r, "<x-equation>"); break;
case MD_SPAN_LATEXMATH_DISPLAY: RENDER_VERBATIM(r, "<x-equation type=\"display\">"); break;
case MD_SPAN_WIKILINK: render_open_wikilink_span(r, (MD_SPAN_WIKILINK_DETAIL*) detail); break;
}
return 0;
}
static int
leave_span_callback(MD_SPANTYPE type, void* detail, void* userdata)
{
MD_HTML* r = (MD_HTML*) userdata;
if(r->image_nesting_level > 0) {
/* Ditto as in enter_span_callback(), except we have to allow the
* end of the <img> tag. */
if(r->image_nesting_level == 1 && type == MD_SPAN_IMG)
render_close_img_span(r, (MD_SPAN_IMG_DETAIL*) detail);
return 0;
}
switch(type) {
case MD_SPAN_EM: RENDER_VERBATIM(r, "</em>"); break;
case MD_SPAN_STRONG: RENDER_VERBATIM(r, "</strong>"); break;
case MD_SPAN_U: RENDER_VERBATIM(r, "</u>"); break;
case MD_SPAN_A: RENDER_VERBATIM(r, "</a>"); break;
case MD_SPAN_IMG: /*noop, handled above*/ break;
case MD_SPAN_CODE: RENDER_VERBATIM(r, "</code>"); break;
case MD_SPAN_DEL: RENDER_VERBATIM(r, "</del>"); break;
case MD_SPAN_LATEXMATH: /*fall through*/
case MD_SPAN_LATEXMATH_DISPLAY: RENDER_VERBATIM(r, "</x-equation>"); break;
case MD_SPAN_WIKILINK: RENDER_VERBATIM(r, "</x-wikilink>"); break;
}
return 0;
}
static int
text_callback(MD_TEXTTYPE type, const MD_CHAR* text, MD_SIZE size, void* userdata)
{
MD_HTML* r = (MD_HTML*) userdata;
switch(type) {
case MD_TEXT_NULLCHAR: render_utf8_codepoint(r, 0x0000, render_verbatim); break;
case MD_TEXT_BR: RENDER_VERBATIM(r, (r->image_nesting_level == 0
? ((r->flags & MD_HTML_FLAG_XHTML) ? "<br />\n" : "<br>\n")
: " "));
break;
case MD_TEXT_SOFTBR: RENDER_VERBATIM(r, (r->image_nesting_level == 0 ? "\n" : " ")); break;
case MD_TEXT_HTML: render_verbatim(r, text, size); break;
case MD_TEXT_ENTITY: render_entity(r, text, size, render_html_escaped); break;
default: render_html_escaped(r, text, size); break;
}
return 0;
}
static void
debug_log_callback(const char* msg, void* userdata)
{
MD_HTML* r = (MD_HTML*) userdata;
if(r->flags & MD_HTML_FLAG_DEBUG)
fprintf(stderr, "MD4C: %s\n", msg);
}
int
md_html(const MD_CHAR* input, MD_SIZE input_size,
void (*process_output)(const MD_CHAR*, MD_SIZE, void*),
void* userdata, unsigned parser_flags, unsigned renderer_flags)
{
MD_HTML render = { process_output, userdata, renderer_flags, 0, { 0 } };
int i;
MD_PARSER parser = {
0,
parser_flags,
enter_block_callback,
leave_block_callback,
enter_span_callback,
leave_span_callback,
text_callback,
debug_log_callback,
NULL
};
/* Build map of characters which need escaping. */
for(i = 0; i < 256; i++) {
unsigned char ch = (unsigned char) i;
if(strchr("\"&<>", ch) != NULL)
render.escape_map[i] |= NEED_HTML_ESC_FLAG;
if(!ISALNUM(ch) && strchr("-_.+!*(),%#@?=;:/,+$", ch) == NULL)
render.escape_map[i] |= NEED_URL_ESC_FLAG;
}
/* Consider skipping UTF-8 byte order mark (BOM). */
if(renderer_flags & MD_HTML_FLAG_SKIP_UTF8_BOM && sizeof(MD_CHAR) == 1) {
static const MD_CHAR bom[3] = { 0xef, 0xbb, 0xbf };
if(input_size >= sizeof(bom) && memcmp(input, bom, sizeof(bom)) == 0) {
input += sizeof(bom);
input_size -= sizeof(bom);
}
}
return md_parse(input, input_size, &parser, (void*) &render);
}

68
lib/md/md4c/md4c-html.h Normal file
View File

@ -0,0 +1,68 @@
/*
* MD4C: Markdown parser for C
* (http://github.com/mity/md4c)
*
* Copyright (c) 2016-2017 Martin Mitas
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#ifndef MD4C_HTML_H
#define MD4C_HTML_H
#include "md4c.h"
#ifdef __cplusplus
extern "C" {
#endif
/* If set, debug output from md_parse() is sent to stderr. */
#define MD_HTML_FLAG_DEBUG 0x0001
#define MD_HTML_FLAG_VERBATIM_ENTITIES 0x0002
#define MD_HTML_FLAG_SKIP_UTF8_BOM 0x0004
#define MD_HTML_FLAG_XHTML 0x0008
/* Render Markdown into HTML.
*
* Note only contents of <body> tag is generated. Caller must generate
* HTML header/footer manually before/after calling md_html().
*
* Params input and input_size specify the Markdown input.
* Callback process_output() gets called with chunks of HTML output.
* (Typical implementation may just output the bytes to a file or append to
* some buffer).
* Param userdata is just propgated back to process_output() callback.
* Param parser_flags are flags from md4c.h propagated to md_parse().
* Param render_flags is bitmask of MD_HTML_FLAG_xxxx.
*
* Returns -1 on error (if md_parse() fails.)
* Returns 0 on success.
*/
int md_html(const MD_CHAR* input, MD_SIZE input_size,
void (*process_output)(const MD_CHAR*, MD_SIZE, void*),
void* userdata, unsigned parser_flags, unsigned renderer_flags);
#ifdef __cplusplus
} /* extern "C" { */
#endif
#endif /* MD4C_HTML_H */

6306
lib/md/md4c/md4c.c Normal file
View File

File diff suppressed because it is too large Load Diff

397
lib/md/md4c/md4c.h Normal file
View File

@ -0,0 +1,397 @@
/*
* MD4C: Markdown parser for C
* (http://github.com/mity/md4c)
*
* Copyright (c) 2016-2020 Martin Mitas
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#ifndef MD4C_H
#define MD4C_H
#ifdef __cplusplus
extern "C" {
#endif
#if defined MD4C_USE_UTF16
/* Magic to support UTF-16. Note that in order to use it, you have to define
* the macro MD4C_USE_UTF16 both when building MD4C as well as when
* including this header in your code. */
#ifdef _WIN32
#include <windows.h>
typedef WCHAR MD_CHAR;
#else
#error MD4C_USE_UTF16 is only supported on Windows.
#endif
#else
typedef char MD_CHAR;
#endif
typedef unsigned MD_SIZE;
typedef unsigned MD_OFFSET;
/* Block represents a part of document hierarchy structure like a paragraph
* or list item.
*/
typedef enum MD_BLOCKTYPE {
/* <body>...</body> */
MD_BLOCK_DOC = 0,
/* <blockquote>...</blockquote> */
MD_BLOCK_QUOTE,
/* <ul>...</ul>
* Detail: Structure MD_BLOCK_UL_DETAIL. */
MD_BLOCK_UL,
/* <ol>...</ol>
* Detail: Structure MD_BLOCK_OL_DETAIL. */
MD_BLOCK_OL,
/* <li>...</li>
* Detail: Structure MD_BLOCK_LI_DETAIL. */
MD_BLOCK_LI,
/* <hr> */
MD_BLOCK_HR,
/* <h1>...</h1> (for levels up to 6)
* Detail: Structure MD_BLOCK_H_DETAIL. */
MD_BLOCK_H,
/* <pre><code>...</code></pre>
* Note the text lines within code blocks are terminated with '\n'
* instead of explicit MD_TEXT_BR. */
MD_BLOCK_CODE,
/* Raw HTML block. This itself does not correspond to any particular HTML
* tag. The contents of it _is_ raw HTML source intended to be put
* in verbatim form to the HTML output. */
MD_BLOCK_HTML,
/* <p>...</p> */
MD_BLOCK_P,
/* <table>...</table> and its contents.
* Detail: Structure MD_BLOCK_TD_DETAIL (used with MD_BLOCK_TH and MD_BLOCK_TD)
* Note all of these are used only if extension MD_FLAG_TABLES is enabled. */
MD_BLOCK_TABLE,
MD_BLOCK_THEAD,
MD_BLOCK_TBODY,
MD_BLOCK_TR,
MD_BLOCK_TH,
MD_BLOCK_TD
} MD_BLOCKTYPE;
/* Span represents an in-line piece of a document which should be rendered with
* the same font, color and other attributes. A sequence of spans forms a block
* like paragraph or list item. */
typedef enum MD_SPANTYPE {
/* <em>...</em> */
MD_SPAN_EM,
/* <strong>...</strong> */
MD_SPAN_STRONG,
/* <a href="xxx">...</a>
* Detail: Structure MD_SPAN_A_DETAIL. */
MD_SPAN_A,
/* <img src="xxx">...</a>
* Detail: Structure MD_SPAN_IMG_DETAIL.
* Note: Image text can contain nested spans and even nested images.
* If rendered into ALT attribute of HTML <IMG> tag, it's responsibility
* of the parser to deal with it.
*/
MD_SPAN_IMG,
/* <code>...</code> */
MD_SPAN_CODE,
/* <del>...</del>
* Note: Recognized only when MD_FLAG_STRIKETHROUGH is enabled.
*/
MD_SPAN_DEL,
/* For recognizing inline ($) and display ($$) equations
* Note: Recognized only when MD_FLAG_LATEXMATHSPANS is enabled.
*/
MD_SPAN_LATEXMATH,
MD_SPAN_LATEXMATH_DISPLAY,
/* Wiki links
* Note: Recognized only when MD_FLAG_WIKILINKS is enabled.
*/
MD_SPAN_WIKILINK,
/* <u>...</u>
* Note: Recognized only when MD_FLAG_UNDERLINE is enabled. */
MD_SPAN_U
} MD_SPANTYPE;
/* Text is the actual textual contents of span. */
typedef enum MD_TEXTTYPE {
/* Normal text. */
MD_TEXT_NORMAL = 0,
/* NULL character. CommonMark requires replacing NULL character with
* the replacement char U+FFFD, so this allows caller to do that easily. */
MD_TEXT_NULLCHAR,
/* Line breaks.
* Note these are not sent from blocks with verbatim output (MD_BLOCK_CODE
* or MD_BLOCK_HTML). In such cases, '\n' is part of the text itself. */
MD_TEXT_BR, /* <br> (hard break) */
MD_TEXT_SOFTBR, /* '\n' in source text where it is not semantically meaningful (soft break) */
/* Entity.
* (a) Named entity, e.g. &nbsp;
* (Note MD4C does not have a list of known entities.
* Anything matching the regexp /&[A-Za-z][A-Za-z0-9]{1,47};/ is
* treated as a named entity.)
* (b) Numerical entity, e.g. &#1234;
* (c) Hexadecimal entity, e.g. &#x12AB;
*
* As MD4C is mostly encoding agnostic, application gets the verbatim
* entity text into the MD_PARSER::text_callback(). */
MD_TEXT_ENTITY,
/* Text in a code block (inside MD_BLOCK_CODE) or inlined code (`code`).
* If it is inside MD_BLOCK_CODE, it includes spaces for indentation and
* '\n' for new lines. MD_TEXT_BR and MD_TEXT_SOFTBR are not sent for this
* kind of text. */
MD_TEXT_CODE,
/* Text is a raw HTML. If it is contents of a raw HTML block (i.e. not
* an inline raw HTML), then MD_TEXT_BR and MD_TEXT_SOFTBR are not used.
* The text contains verbatim '\n' for the new lines. */
MD_TEXT_HTML,
/* Text is inside an equation. This is processed the same way as inlined code
* spans (`code`). */
MD_TEXT_LATEXMATH
} MD_TEXTTYPE;
/* Alignment enumeration. */
typedef enum MD_ALIGN {
MD_ALIGN_DEFAULT = 0, /* When unspecified. */
MD_ALIGN_LEFT,
MD_ALIGN_CENTER,
MD_ALIGN_RIGHT
} MD_ALIGN;
/* String attribute.
*
* This wraps strings which are outside of a normal text flow and which are
* propagated within various detailed structures, but which still may contain
* string portions of different types like e.g. entities.
*
* So, for example, lets consider this image:
*
* ![image alt text](http://example.org/image.png 'foo &quot; bar')
*
* The image alt text is propagated as a normal text via the MD_PARSER::text()
* callback. However, the image title ('foo &quot; bar') is propagated as
* MD_ATTRIBUTE in MD_SPAN_IMG_DETAIL::title.
*
* Then the attribute MD_SPAN_IMG_DETAIL::title shall provide the following:
* -- [0]: "foo " (substr_types[0] == MD_TEXT_NORMAL; substr_offsets[0] == 0)
* -- [1]: "&quot;" (substr_types[1] == MD_TEXT_ENTITY; substr_offsets[1] == 4)
* -- [2]: " bar" (substr_types[2] == MD_TEXT_NORMAL; substr_offsets[2] == 10)
* -- [3]: (n/a) (n/a ; substr_offsets[3] == 14)
*
* Note that these invariants are always guaranteed:
* -- substr_offsets[0] == 0
* -- substr_offsets[LAST+1] == size
* -- Currently, only MD_TEXT_NORMAL, MD_TEXT_ENTITY, MD_TEXT_NULLCHAR
* substrings can appear. This could change only of the specification
* changes.
*/
typedef struct MD_ATTRIBUTE {
const MD_CHAR* text;
MD_SIZE size;
const MD_TEXTTYPE* substr_types;
const MD_OFFSET* substr_offsets;
} MD_ATTRIBUTE;
/* Detailed info for MD_BLOCK_UL. */
typedef struct MD_BLOCK_UL_DETAIL {
int is_tight; /* Non-zero if tight list, zero if loose. */
MD_CHAR mark; /* Item bullet character in MarkDown source of the list, e.g. '-', '+', '*'. */
} MD_BLOCK_UL_DETAIL;
/* Detailed info for MD_BLOCK_OL. */
typedef struct MD_BLOCK_OL_DETAIL {
unsigned start; /* Start index of the ordered list. */
int is_tight; /* Non-zero if tight list, zero if loose. */
MD_CHAR mark_delimiter; /* Character delimiting the item marks in MarkDown source, e.g. '.' or ')' */
} MD_BLOCK_OL_DETAIL;
/* Detailed info for MD_BLOCK_LI. */
typedef struct MD_BLOCK_LI_DETAIL {
int is_task; /* Can be non-zero only with MD_FLAG_TASKLISTS */
MD_CHAR task_mark; /* If is_task, then one of 'x', 'X' or ' '. Undefined otherwise. */
MD_OFFSET task_mark_offset; /* If is_task, then offset in the input of the char between '[' and ']'. */
} MD_BLOCK_LI_DETAIL;
/* Detailed info for MD_BLOCK_H. */
typedef struct MD_BLOCK_H_DETAIL {
unsigned level; /* Header level (1 - 6) */
} MD_BLOCK_H_DETAIL;
/* Detailed info for MD_BLOCK_CODE. */
typedef struct MD_BLOCK_CODE_DETAIL {
MD_ATTRIBUTE info;
MD_ATTRIBUTE lang;
MD_CHAR fence_char; /* The character used for fenced code block; or zero for indented code block. */
} MD_BLOCK_CODE_DETAIL;
/* Detailed info for MD_BLOCK_TH and MD_BLOCK_TD. */
typedef struct MD_BLOCK_TD_DETAIL {
MD_ALIGN align;
} MD_BLOCK_TD_DETAIL;
/* Detailed info for MD_SPAN_A. */
typedef struct MD_SPAN_A_DETAIL {
MD_ATTRIBUTE href;
MD_ATTRIBUTE title;
} MD_SPAN_A_DETAIL;
/* Detailed info for MD_SPAN_IMG. */
typedef struct MD_SPAN_IMG_DETAIL {
MD_ATTRIBUTE src;
MD_ATTRIBUTE title;
} MD_SPAN_IMG_DETAIL;
/* Detailed info for MD_SPAN_WIKILINK. */
typedef struct MD_SPAN_WIKILINK {
MD_ATTRIBUTE target;
} MD_SPAN_WIKILINK_DETAIL;
/* Flags specifying extensions/deviations from CommonMark specification.
*
* By default (when MD_PARSER::flags == 0), we follow CommonMark specification.
* The following flags may allow some extensions or deviations from it.
*/
#define MD_FLAG_COLLAPSEWHITESPACE 0x0001 /* In MD_TEXT_NORMAL, collapse non-trivial whitespace into single ' ' */
#define MD_FLAG_PERMISSIVEATXHEADERS 0x0002 /* Do not require space in ATX headers ( ###header ) */
#define MD_FLAG_PERMISSIVEURLAUTOLINKS 0x0004 /* Recognize URLs as autolinks even without '<', '>' */
#define MD_FLAG_PERMISSIVEEMAILAUTOLINKS 0x0008 /* Recognize e-mails as autolinks even without '<', '>' and 'mailto:' */
#define MD_FLAG_NOINDENTEDCODEBLOCKS 0x0010 /* Disable indented code blocks. (Only fenced code works.) */
#define MD_FLAG_NOHTMLBLOCKS 0x0020 /* Disable raw HTML blocks. */
#define MD_FLAG_NOHTMLSPANS 0x0040 /* Disable raw HTML (inline). */
#define MD_FLAG_TABLES 0x0100 /* Enable tables extension. */
#define MD_FLAG_STRIKETHROUGH 0x0200 /* Enable strikethrough extension. */
#define MD_FLAG_PERMISSIVEWWWAUTOLINKS 0x0400 /* Enable WWW autolinks (even without any scheme prefix, if they begin with 'www.') */
#define MD_FLAG_TASKLISTS 0x0800 /* Enable task list extension. */
#define MD_FLAG_LATEXMATHSPANS 0x1000 /* Enable $ and $$ containing LaTeX equations. */
#define MD_FLAG_WIKILINKS 0x2000 /* Enable wiki links extension. */
#define MD_FLAG_UNDERLINE 0x4000 /* Enable underline extension (and disables '_' for normal emphasis). */
#define MD_FLAG_PERMISSIVEAUTOLINKS (MD_FLAG_PERMISSIVEEMAILAUTOLINKS | MD_FLAG_PERMISSIVEURLAUTOLINKS | MD_FLAG_PERMISSIVEWWWAUTOLINKS)
#define MD_FLAG_NOHTML (MD_FLAG_NOHTMLBLOCKS | MD_FLAG_NOHTMLSPANS)
/* Convenient sets of flags corresponding to well-known Markdown dialects.
*
* Note we may only support subset of features of the referred dialect.
* The constant just enables those extensions which bring us as close as
* possible given what features we implement.
*
* ABI compatibility note: Meaning of these can change in time as new
* extensions, bringing the dialect closer to the original, are implemented.
*/
#define MD_DIALECT_COMMONMARK 0
#define MD_DIALECT_GITHUB (MD_FLAG_PERMISSIVEAUTOLINKS | MD_FLAG_TABLES | MD_FLAG_STRIKETHROUGH | MD_FLAG_TASKLISTS)
/* Parser structure.
*/
typedef struct MD_PARSER {
/* Reserved. Set to zero.
*/
unsigned abi_version;
/* Dialect options. Bitmask of MD_FLAG_xxxx values.
*/
unsigned flags;
/* Caller-provided rendering callbacks.
*
* For some block/span types, more detailed information is provided in a
* type-specific structure pointed by the argument 'detail'.
*
* The last argument of all callbacks, 'userdata', is just propagated from
* md_parse() and is available for any use by the application.
*
* Note any strings provided to the callbacks as their arguments or as
* members of any detail structure are generally not zero-terminated.
* Application has to take the respective size information into account.
*
* Any rendering callback may abort further parsing of the document by
* returning non-zero.
*/
int (*enter_block)(MD_BLOCKTYPE /*type*/, void* /*detail*/, void* /*userdata*/);
int (*leave_block)(MD_BLOCKTYPE /*type*/, void* /*detail*/, void* /*userdata*/);
int (*enter_span)(MD_SPANTYPE /*type*/, void* /*detail*/, void* /*userdata*/);
int (*leave_span)(MD_SPANTYPE /*type*/, void* /*detail*/, void* /*userdata*/);
int (*text)(MD_TEXTTYPE /*type*/, const MD_CHAR* /*text*/, MD_SIZE /*size*/, void* /*userdata*/);
/* Debug callback. Optional (may be NULL).
*
* If provided and something goes wrong, this function gets called.
* This is intended for debugging and problem diagnosis for developers;
* it is not intended to provide any errors suitable for displaying to an
* end user.
*/
void (*debug_log)(const char* /*msg*/, void* /*userdata*/);
/* Reserved. Set to NULL.
*/
void (*syntax)(void);
} MD_PARSER;
/* For backward compatibility. Do not use in new code.
*/
typedef MD_PARSER MD_RENDERER;
/* Parse the Markdown document stored in the string 'text' of size 'size'.
* The parser provides callbacks to be called during the parsing so the
* caller can render the document on the screen or convert the Markdown
* to another format.
*
* Zero is returned on success. If a runtime error occurs (e.g. a memory
* fails), -1 is returned. If the processing is aborted due any callback
* returning non-zero, the return value of the callback is returned.
*/
int md_parse(const MD_CHAR* text, MD_SIZE size, const MD_PARSER* parser, void* userdata);
#ifdef __cplusplus
} /* extern "C" { */
#endif
#endif /* MD4C_H */