From 7b78b6986e714cbd7653b89287b928974eb5ee4f Mon Sep 17 00:00:00 2001 From: sanine Date: Sat, 1 Jan 2022 13:24:33 -0600 Subject: initial commit --- src/markdown/md4c-html.c | 573 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 573 insertions(+) create mode 100644 src/markdown/md4c-html.c (limited to 'src/markdown/md4c-html.c') diff --git a/src/markdown/md4c-html.c b/src/markdown/md4c-html.c new file mode 100644 index 0000000..d604aec --- /dev/null +++ b/src/markdown/md4c-html.c @@ -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 +#include + +#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, "&"); break; + case '<': RENDER_VERBATIM(r, "<"); break; + case '>': RENDER_VERBATIM(r, ">"); break; + case '"': RENDER_VERBATIM(r, """); 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, "&"); 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, (MD_SIZE)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. "�")). */ + 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. " "). */ + 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, "
    \n"); + return; + } + + snprintf(buf, sizeof(buf), "
      \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, "
    1. " + "task_mark == 'x' || det->task_mark == 'X') + RENDER_VERBATIM(r, " checked"); + RENDER_VERBATIM(r, ">"); + } else { + RENDER_VERBATIM(r, "
    2. "); + } +} + +static void +render_open_code_block(MD_HTML* r, const MD_BLOCK_CODE_DETAIL* det) +{ + RENDER_VERBATIM(r, "
      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, "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, "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, "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] = { "
      ", "
      ", "
      ", "
      ", "
      ", "
      " };
+    MD_HTML* r = (MD_HTML*) userdata;
+
+    switch(type) {
+        case MD_BLOCK_DOC:      /* noop */ break;
+        case MD_BLOCK_QUOTE:    RENDER_VERBATIM(r, "
      \n"); break;
+        case MD_BLOCK_UL:       RENDER_VERBATIM(r, "