adds onCodeBlock callback re #11

example/example.html

@@ -13,16 +13,16 @@ <h2><a id="another" class="anchor" aria-hidden="true" href="#another"></a>Anothe
 <p>line 1
 line 2</p>
 <h2><a id="code-poetry" class="anchor" aria-hidden="true" href="#code-poetry"></a>Code &amp; Poetry</h2>
-<pre><code>You can also indent
-blocks to display
-code or poetry.
+<pre><code>YOU CAN ALSO INDENT
+BLOCKS TO DISPLAY
+CODE OR POETRY.
 
-Indented code/poetry blocks
-can be hard-wrapped.
+INDENTED CODE/POETRY BLOCKS
+CAN BE HARD-WRAPPED.
 </code></pre>
 <p><b>Or, wrap your code in three backticks:</b></p>
-<pre><code class="language-js">function codeBlocks() {
-  return &quot;Can be inserted&quot;
+<pre><code class="language-js">FUNCTION CODEBLOCKS() {
+  RETURN &quot;CAN BE INSERTED&quot;
 }
 </code></pre>
 <h3><a id="block-quotes" class="anchor" aria-hidden="true" href="#block-quotes"></a>Block Quotes</h3>

example/example.js

@@ -1,8 +1,15 @@
 const fs = require("fs")
 const md = require("../dist/markdown.node.js")
+// const md = require("../build/debug/markdown.node.js")
 
 const source = fs.readFileSync(__dirname + "/example.md")
-const outbuf = md.parse(source, { bytes: true })
+const outbuf = md.parse(source, {
+  bytes: true,
+  onCodeBlock(lang, body) {
+    console.log(`onCodeBlock (${lang})`)
+    return html_escape(body.toString().toUpperCase())
+  },
+})
 const outfile = __dirname + "/example.html"
 console.log("write", outfile)
 fs.writeFileSync(outfile, outbuf)
@@ -24,3 +31,9 @@ if (process.argv.includes("-bench")) {
   const timeSpent = Date.now() - timeStart
   console.log(`benchmark end -- avg parse time: ${((timeSpent / ntotal) * 1000).toFixed(1)}us`)
 }
+
+function html_escape(str) {
+  return str.replace(/[&<>'"]/g, tag => ({
+    '&': '&amp;','<': '&lt;','>': '&gt;',"'": '&#39;','"': '&quot;'
+  }[tag]))
+}

markdown.d.ts

@@ -28,10 +28,31 @@ export interface ParseOptions {
    */
   bytes? :boolean
 
+  /**
+   * onCodeBlock is an optional callback which if provided is called for each code block.
+   * langname holds the "language tag", if any, of the block.
+   *
+   * The returned value is inserted into the resulting HTML verbatim, without HTML escaping.
+   * Thus, you should take care of properly escaping any special HTML characters.
+   *
+   * If the function returns null or undefined, or an exception occurs, the body will be
+   * included as-is after going through HTML escaping.
+   *
+   * Note that use of this callback has an adverse impact on performance as it casues
+   * calls and data to be bridged between WASM and JS on every invocation.
+   */
+  onCodeBlock? :(langname :string, body :UTF8Bytes) => Uint8Array|string|null|undefined
+
   /** @depreceated use "bytes" instead (v1.1.1) */
   asMemoryView? :boolean
 }
 
+/** UTF8Bytes is a Uint8Array representing UTF8 text  */
+export interface UTF8Bytes extends Uint8Array {
+  /** toString returns a UTF8 decoded string (lazily decoded and cached) */
+  toString() :string
+}
+
 /** Flags that customize Markdown parsing */
 export enum ParseFlags {
   /** In TEXT, collapse non-trivial whitespace into single ' ' */ COLLAPSE_WHITESPACE,

src/common.h

@@ -54,7 +54,14 @@ typedef int32_t  i32;
 #endif
 #if DEBUG
 #include <stdio.h>
-  #define dlog(...)  printf(__VA_ARGS__)
+  #define dlog(fmt, ...)  printf(fmt "\n", ##__VA_ARGS__)
 #else
   #define dlog(...)
 #endif /* DEBUG > 0 */
+
+#include "wbuf.h"
+
+typedef int(*JSTextFilterFun)(
+  const char*  metaptr, u32 metalen,
+  const char*  inptr,   u32 inlen,
+  const char** outptrp); // return outlen