From d2d8c4d74da37354ba0c69c3b8875cee090dd38c Mon Sep 17 00:00:00 2001
From: bbedward <bbedward@gmail.com>
Date: Sat, 26 Jul 2025 16:37:45 -0400
Subject: [PATCH] cleanup markdown better,  don't process code block content

---
 Common/markdown2html.js | 64 +++++++++++++++++++++++++++++++++++------
 1 file changed, 56 insertions(+), 8 deletions(-)

diff --git a/Common/markdown2html.js b/Common/markdown2html.js
index db1a4085..18e831bd 100644
--- a/Common/markdown2html.js
+++ b/Common/markdown2html.js
@@ -3,9 +3,36 @@
 function markdownToHtml(text) {
     if (!text) return "";
     
-    let html = text;
+    // Store code blocks and inline code to protect them from further processing
+    const codeBlocks = [];
+    const inlineCode = [];
+    let blockIndex = 0;
+    let inlineIndex = 0;
     
-    // Escape HTML entities first
+    // First, extract and replace code blocks with placeholders
+    let html = text.replace(/```([\s\S]*?)```/g, (match, code) => {
+        // Trim leading and trailing blank lines only
+        const trimmedCode = code.replace(/^\n+|\n+$/g, '');
+        // Escape HTML entities in code
+        const escapedCode = trimmedCode.replace(/&/g, '&amp;')
+                                       .replace(/</g, '&lt;')
+                                       .replace(/>/g, '&gt;');
+        codeBlocks.push(`<pre><code>${escapedCode}</code></pre>`);
+        return `\x00CODEBLOCK${blockIndex++}\x00`;
+    });
+    
+    // Extract and replace inline code
+    html = html.replace(/`([^`]+)`/g, (match, code) => {
+        // Escape HTML entities in code
+        const escapedCode = code.replace(/&/g, '&amp;')
+                               .replace(/</g, '&lt;')
+                               .replace(/>/g, '&gt;');
+        inlineCode.push(`<code>${escapedCode}</code>`);
+        return `\x00INLINECODE${inlineIndex++}\x00`;
+    });
+    
+    // Now process everything else
+    // Escape HTML entities (but not in code blocks)
     html = html.replace(/&/g, '&amp;')
                 .replace(/</g, '&lt;')
                 .replace(/>/g, '&gt;');
@@ -15,7 +42,7 @@ function markdownToHtml(text) {
     html = html.replace(/^## (.*?)$/gm, '<h2>$1</h2>');
     html = html.replace(/^# (.*?)$/gm, '<h1>$1</h1>');
     
-    // Bold and italic
+    // Bold and italic (order matters!)
     html = html.replace(/\*\*\*(.*?)\*\*\*/g, '<b><i>$1</i></b>');
     html = html.replace(/\*\*(.*?)\*\*/g, '<b>$1</b>');
     html = html.replace(/\*(.*?)\*/g, '<i>$1</i>');
@@ -23,10 +50,6 @@ function markdownToHtml(text) {
     html = html.replace(/__(.*?)__/g, '<b>$1</b>');
     html = html.replace(/_(.*?)_/g, '<i>$1</i>');
     
-    // Code blocks
-    html = html.replace(/```([\s\S]*?)```/g, '<pre><code>$1</code></pre>');
-    html = html.replace(/`(.*?)`/g, '<code>$1</code>');
-    
     // Links
     html = html.replace(/\[([^\]]+)\]\(([^)]+)\)/g, '<a href="$2">$1</a>');
     
@@ -43,8 +66,16 @@ function markdownToHtml(text) {
     // Detect plain URLs and wrap them in anchor tags (but not inside existing <a> or markdown links)
     html = html.replace(/(^|[^"'>])((https?|file):\/\/[^\s<]+)/g, '$1<a href="$2">$2</a>');
 
+    // Restore code blocks and inline code BEFORE line break processing
+    html = html.replace(/\x00CODEBLOCK(\d+)\x00/g, (match, index) => {
+        return codeBlocks[parseInt(index)];
+    });
     
-    // Line breaks
+    html = html.replace(/\x00INLINECODE(\d+)\x00/g, (match, index) => {
+        return inlineCode[parseInt(index)];
+    });
+    
+    // Line breaks (after code blocks are restored)
     html = html.replace(/\n\n/g, '</p><p>');
     html = html.replace(/\n/g, '<br/>');
     
@@ -53,5 +84,22 @@ function markdownToHtml(text) {
         html = '<p>' + html + '</p>';
     }
     
+    // Clean up the final HTML
+    // Remove <br/> tags immediately before block elements
+    html = html.replace(/<br\/>\s*<pre>/g, '<pre>');
+    html = html.replace(/<br\/>\s*<ul>/g, '<ul>');
+    html = html.replace(/<br\/>\s*<h[1-6]>/g, '<h$1>');
+    
+    // Remove empty paragraphs
+    html = html.replace(/<p>\s*<\/p>/g, '');
+    html = html.replace(/<p>\s*<br\/>\s*<\/p>/g, '');
+    
+    // Remove excessive line breaks
+    html = html.replace(/(<br\/>){3,}/g, '<br/><br/>'); // Max 2 consecutive line breaks
+    html = html.replace(/(<\/p>)\s*(<p>)/g, '$1$2'); // Remove whitespace between paragraphs
+    
+    // Remove leading/trailing whitespace
+    html = html.trim();
+    
     return html;
 }
\ No newline at end of file