Build Your Kit page and full Micromelon website

Complete website build including: - Build Your Kit store page with cart system, sectioned layout (Hardware, Software, Attachments, Spare Parts), inline quote request form, and sticky sidebar summary - 16+ pages: Education, Platform, Resources, News, About Us, Download, Contact, Rover, Code Editor, Robot Simulator, etc. - 89+ MDX resource articles and 18 news posts - Store product images scraped from micromelon.com.au - Quote request API route with Airtable integration - Dynamic back links and cover photos on resource pages - Redesigned downloads page - Fixed corrupted MDX code blocks
2026-02-28 19:00:42 +10:00
parent 5233233662
commit 707c49dd3f
320 changed files with 22333 additions and 107 deletions
--- a/scripts/cleanup-mdx.mjs
+++ b/scripts/cleanup-mdx.mjs
@@ -0,0 +1,45 @@
+import fs from 'fs';
+import path from 'path';
+import { fileURLToPath } from 'url';
+
+const __dirname = path.dirname(fileURLToPath(import.meta.url));
+const dir = path.join(__dirname, '..', 'content', 'resources');
+
+let fixed = 0;
+for (const f of fs.readdirSync(dir).filter(f => f.endsWith('.mdx'))) {
+  const filepath = path.join(dir, f);
+  const original = fs.readFileSync(filepath, 'utf8');
+  let content = original;
+
+  // Clean excerpts that contain HTML/CSS artifacts
+  const excerptMatch = content.match(/^excerpt: "(.*)"$/m);
+  if (excerptMatch) {
+    let excerpt = excerptMatch[1];
+    excerpt = excerpt.replace(/#block-[^\s]+ \{[^}]*\}/g, '');
+    excerpt = excerpt.replace(/!\[[^\]]*\]\([^)]*\)/g, '');
+    excerpt = excerpt.replace(/\[([^\]]*)\]\([^)]*\)/g, '$1');
+    excerpt = excerpt.replace(/\*\*/g, '');
+    excerpt = excerpt.replace(/\s+/g, ' ').trim().substring(0, 250);
+    if (excerpt !== excerptMatch[1]) {
+      content = content.replace(excerptMatch[0], `excerpt: "${excerpt}"`);
+    }
+  }
+
+  // Remove BACK TO POSTS lines
+  content = content.replace(/\[?\s*\*?\*?BACK TO POSTS\s*\*?\*?\s*\]?\s*\([^)]*\)\s*(\*?\*?\s*\|?\s*\*?\*?\s*\[?\*?\*?OPEN PAGE AS PDF\*?\*?\]?\([^)]*\))?/gi, '');
+
+  // Remove --> artifacts
+  content = content.replace(/^-->\s*$/gm, '');
+
+  // Remove CSS block definitions
+  content = content.replace(/#block-[a-zA-Z0-9_-]+ \{[^}]*\}/g, '');
+
+  // Clean excessive blank lines
+  content = content.replace(/\n{3,}/g, '\n\n');
+
+  if (content !== original) {
+    fs.writeFileSync(filepath, content);
+    fixed++;
+  }
+}
+console.log(`Cleaned ${fixed} files`);
--- a/scripts/download-assets.sh
+++ b/scripts/download-assets.sh
@@ -0,0 +1,56 @@
+#!/bin/bash
+# Download assets from current Squarespace site
+BASE="https://images.squarespace-cdn.com/content/v1/60a43bf842d7b601064a8828"
+OUT="public/images"
+
+# Founders
+curl -sL "$BASE/207a7c93-62e4-4cce-960e-3aac9d0f326d/adam.png" -o "$OUT/founders/adam.png"
+curl -sL "$BASE/1a6645b4-9e27-4b49-b8ff-c3644c358f4c/tim.png" -o "$OUT/founders/tim.png"
+
+# Partner logos
+curl -sL "$BASE/1633901579287-HDYLZ442KFVAZWZCPSXI/codingathome.png" -o "$OUT/partners/codingathome.png"
+curl -sL "$BASE/1633901627921-UKKWATKWSZF65M8PILDY/worldsciencefair.png" -o "$OUT/partners/worldsciencefair.png"
+curl -sL "$BASE/1633901657858-W25D5TI13OTSG76POUDS/cisco.png" -o "$OUT/partners/cisco.png"
+curl -sL "$BASE/1633901706899-JBER9BFFNRKAW6ZF0KCH/uq-logo-lockup-purple.png" -o "$OUT/partners/uq.png"
+curl -sL "$BASE/1633901726291-HK73RNB7LCNZ17IXQYMC/shapelabs.png" -o "$OUT/partners/shapelabs.png"
+curl -sL "$BASE/1633901745735-MR5B6UE4A89BLW9H4PLM/roborave.png" -o "$OUT/partners/roborave.png"
+curl -sL "$BASE/1634585888937-VZJBKXYBI5V28XGXN8T9/robo-gals-new.png" -o "$OUT/partners/robogals.png"
+curl -sL "$BASE/961ac034-e08c-4585-aeee-90cff217a3cc/Screen-Shot-2022-01-19-at-2.41.17-pm.png" -o "$OUT/partners/somerset.png"
+curl -sL "$BASE/4a1c8c44-deba-49fe-bed9-100bced46984/MARS-logo.png" -o "$OUT/partners/mars.png"
+curl -sL "$BASE/1633912027308-Z2PUHHVZR07R4A55Z1NJ/queensland+robotics.png" -o "$OUT/partners/queensland-robotics.png"
+curl -sL "$BASE/24345bac-45df-4625-bbe9-4c2b9ddc2bbd/001.jpg" -o "$OUT/partners/nambour.jpg"
+curl -sL "$BASE/2b8da32b-f29f-4fbc-a69f-62b674ff6455/HFSDesign_Logo1-ColourBlack.png" -o "$OUT/partners/hfs-design.png"
+curl -sL "$BASE/42bb721d-a1de-48a5-b595-1a6ee0809d31/reddirtrobotics%2Blogo.jpg" -o "$OUT/partners/reddirtrobotics.jpg"
+curl -sL "$BASE/102b1693-2145-4aea-8012-4b41f4b101f7/195909638_151774876989903_7462155829817761256_n.jpg" -o "$OUT/partners/stpeters.jpg"
+curl -sL "$BASE/74593547-155c-41d6-a3ec-c9f2870043e0/1631322817046.jpeg" -o "$OUT/partners/bgs.jpeg"
+curl -sL "$BASE/298bbc5f-49ee-40c1-aa0c-255bca5d6f12/GISP_black_logo.png" -o "$OUT/partners/gisp.png"
+
+# Award badges
+curl -sL "$BASE/f2922da5-e964-497d-8e45-bfde14f9a16f/body-1-good-design-award_gold-winner_logo-jpg-ximg_-l_8_m-smart_-e1599799902729.jpg" -o "$OUT/awards/good-design.jpg"
+curl -sL "$BASE/1633901132092-1CSSNZ4JPWFXQG9XDGP4/advanced-queensland.png" -o "$OUT/awards/advance-queensland.png"
+curl -sL "$BASE/1633901205930-0TMN1CR2WD3UFNKMOT15/youngstarterslogo.png" -o "$OUT/awards/young-starters.png"
+curl -sL "$BASE/58a43758-897b-4821-92f8-16f21b3a7bab/Australian-Made-Owned-full-colour-logo.jpg" -o "$OUT/awards/australian-made.jpg"
+curl -sL "$BASE/1633901223722-ULL99PJTYXHRA7W8VQQ7/icralogo.png" -o "$OUT/awards/icra.png"
+curl -sL "$BASE/1633901251644-ZCCMHUZF7QIK8NXWNKZI/blueboxlogo.png" -o "$OUT/awards/bluebox.png"
+curl -sL "$BASE/b205cc03-1282-44ef-8e94-9a99037dbca5/PL.jpg" -o "$OUT/awards/pl.jpg"
+
+# Product/Platform images
+curl -sL "$BASE/d7e06c90-5ae8-4a25-aa51-695b63542b76/platform.png" -o "$OUT/hero/platform.png"
+curl -sL "$BASE/15c6cb9e-36ed-4b8f-96e6-4485ec587d77/rover-candid.png" -o "$OUT/products/rover-candid.png"
+curl -sL "$BASE/a738f6d7-a969-445f-982c-71d9f80943b1/rover-big.png" -o "$OUT/products/rover-big.png"
+curl -sL "$BASE/afa456f1-1da2-4e8f-9a13-44503985bf1d/driving-school-duckie-town.png" -o "$OUT/hero/driving-school.png"
+curl -sL "$BASE/e3de22cd-11b8-4bf4-87c1-155d84607e84/simulator-big.png" -o "$OUT/products/simulator-big.png"
+curl -sL "$BASE/1d1b50df-b357-4eb2-8e83-4a7ea36152ae/code-editor-big.png" -o "$OUT/products/code-editor-big.png"
+curl -sL "$BASE/1633409192147-YZB8VCH6IQVIX6JJ00R7/class-management.png" -o "$OUT/products/class-management.png"
+curl -sL "$BASE/b8018b7c-dd04-4b0e-a77f-5d648be55b85/amo-2021-logo-rgb-full-colour-clear.png" -o "$OUT/awards/australian-made-owned.png"
+curl -sL "$BASE/04fae17e-865c-4f22-af38-fe8c023406a4/KeyShot+Renders.70.jpg" -o "$OUT/products/rover-render.jpg"
+curl -sL "$BASE/1633397571502-7P7WRMB5N3X0WDVSV1K4/rover.png" -o "$OUT/products/rover.png"
+curl -sL "$BASE/1633398402343-BCFYMKQICJQC5M34D296/rover-digger-full.png" -o "$OUT/products/rover-digger.png"
+curl -sL "$BASE/1633398257202-RKGBJDDOEVTIYVGZWDG3/cords.png" -o "$OUT/products/cords.png"
+curl -sL "$BASE/1633398231053-6EF4N4ZGLIG92FXJONTI/case.png" -o "$OUT/products/case.png"
+curl -sL "$BASE/1633400391317-34X5T9TL6I1ASWIZYH0X/image-asset.png" -o "$OUT/products/rover-specs.png"
+curl -sL "$BASE/1633988093034-A27ZE2HMSHGGDK7VDSM9/header+video+with+laptop.gif" -o "$OUT/hero/code-editor-demo.gif"
+curl -sL "$BASE/514f66e4-0c76-4ca8-8451-b38d0c4b39e6/2021-12-16_16-01-19_1639630906.jpg" -o "$OUT/hero/3d-printing.jpg"
+curl -sL "$BASE/1633402428790-4PHP6JHFUYICKH614XQW/from-rover-to-software.gif" -o "$OUT/hero/rover-to-software.gif"
+
+echo "Done downloading assets"
--- a/scripts/fix-all-mdx.mjs
+++ b/scripts/fix-all-mdx.mjs
@@ -0,0 +1,111 @@
+import fs from 'fs';
+import path from 'path';
+import { fileURLToPath } from 'url';
+
+const __dirname = path.dirname(fileURLToPath(import.meta.url));
+const dir = path.join(__dirname, '..', 'content', 'resources');
+
+let fixed = 0;
+for (const f of fs.readdirSync(dir).filter(f => f.endsWith('.mdx'))) {
+  const filepath = path.join(dir, f);
+  const original = fs.readFileSync(filepath, 'utf8');
+
+  // Split frontmatter from content
+  const secondDash = original.indexOf('---', 4);
+  if (secondDash === -1) continue;
+  let frontmatter = original.substring(0, secondDash + 3);
+  let content = original.substring(secondDash + 3);
+
+  // === Fix body content ===
+
+  // Remove BACK TO POSTS and OPEN PAGE AS PDF
+  content = content.replace(/\[\s*\*?\*?\s*BACK TO POSTS\s*\*?\*?\s*\]\([^)]*\)/gi, '');
+  content = content.replace(/\*?\*?\s*BACK TO POSTS\s*\*?\*?/gi, '');
+  content = content.replace(/\|?\s*\[?\s*\*?\*?\s*OPEN PAGE AS PDF\s*\*?\*?\s*\]?\s*\([^)]*\)/gi, '');
+
+  // Remove CSS block definitions
+  content = content.replace(/#block-[a-zA-Z0-9_-]+ \{[^}]*?\}/g, '');
+
+  // Remove --> artifacts
+  content = content.replace(/^-->\s*$/gm, '');
+
+  // Escape ALL < that aren't part of:
+  // 1. Markdown images: ![...](...) - these don't have <
+  // 2. iframe tags: <iframe ... > or </iframe>
+  // 3. Already escaped: \<
+  // Strategy: protect safe tags, escape everything else, restore safe tags
+
+  // Temporarily replace safe tags with placeholders
+  const safeTags = [];
+  content = content.replace(/<(\/?)iframe([^>]*)>/gi, (match) => {
+    safeTags.push(match);
+    return `__SAFE_TAG_${safeTags.length - 1}__`;
+  });
+  content = content.replace(/<(\/?)video([^>]*)>/gi, (match) => {
+    safeTags.push(match);
+    return `__SAFE_TAG_${safeTags.length - 1}__`;
+  });
+  content = content.replace(/<source([^>]*)>/gi, (match) => {
+    safeTags.push(match);
+    return `__SAFE_TAG_${safeTags.length - 1}__`;
+  });
+
+  // Now escape all remaining < and > that aren't in image markdown
+  // Split by lines to handle image markdown
+  const lines = content.split('\n');
+  for (let i = 0; i < lines.length; i++) {
+    const line = lines[i];
+    // Skip lines that are image markdown
+    if (line.trim().startsWith('![')) continue;
+    // Skip lines that only contain safe tag placeholders
+    if (line.trim().match(/^__SAFE_TAG_\d+__$/)) continue;
+
+    // Escape < and > that aren't already escaped or part of placeholders
+    lines[i] = line.replace(/<(?!_SAFE_TAG)/g, (match, offset) => {
+      // Check if already escaped
+      if (offset > 0 && line[offset - 1] === '\\') return match;
+      return '&lt;';
+    });
+    lines[i] = lines[i].replace(/>(?!_)/g, (match, offset) => {
+      const before = lines[i].substring(0, offset);
+      // Don't escape if it's after a safe tag placeholder
+      if (before.match(/__SAFE_TAG_\d+$/)) return match;
+      // Don't escape if already escaped
+      if (offset > 0 && lines[i][offset - 1] === '\\') return match;
+      // Don't escape markdown blockquotes at start of line
+      if (before.match(/^\s*$/)) return match;
+      return '&gt;';
+    });
+  }
+  content = lines.join('\n');
+
+  // Restore safe tags
+  for (let i = 0; i < safeTags.length; i++) {
+    content = content.replace(`__SAFE_TAG_${i}__`, safeTags[i]);
+  }
+
+  // Clean excessive blank lines
+  content = content.replace(/\n{3,}/g, '\n\n');
+
+  // === Fix frontmatter excerpt ===
+  const excerptMatch = frontmatter.match(/^excerpt: "(.*)"$/m);
+  if (excerptMatch) {
+    let excerpt = excerptMatch[1];
+    excerpt = excerpt.replace(/\*?\*?BACK TO POSTS\*?\*?/gi, '');
+    excerpt = excerpt.replace(/!\[[^\]]*\]\([^)]*\)/g, '');
+    excerpt = excerpt.replace(/\[([^\]]*)\]\([^)]*\)/g, '$1');
+    excerpt = excerpt.replace(/\*\*/g, '');
+    excerpt = excerpt.replace(/#block-[^\s]+ \{[^}]*?\}/g, '');
+    excerpt = excerpt.replace(/\s*\|\s*/g, ' ');
+    excerpt = excerpt.replace(/\s+/g, ' ').trim();
+    excerpt = excerpt.substring(0, 250);
+    frontmatter = frontmatter.replace(excerptMatch[0], `excerpt: "${excerpt}"`);
+  }
+
+  const result = frontmatter + content;
+  if (result !== original) {
+    fs.writeFileSync(filepath, result);
+    fixed++;
+  }
+}
+console.log(`Fixed ${fixed} files`);
--- a/scripts/fix-bullets.mjs
+++ b/scripts/fix-bullets.mjs
@@ -0,0 +1,37 @@
+import fs from 'fs';
+import path from 'path';
+import { fileURLToPath } from 'url';
+
+const __dirname = path.dirname(fileURLToPath(import.meta.url));
+const dir = path.join(__dirname, '..', 'content', 'resources');
+
+let fixed = 0;
+for (const f of fs.readdirSync(dir).filter(f => f.endsWith('.mdx'))) {
+  const filepath = path.join(dir, f);
+  const original = fs.readFileSync(filepath, 'utf8');
+
+  // Split frontmatter from content
+  const secondDash = original.indexOf('---', 4);
+  if (secondDash === -1) continue;
+  const frontmatter = original.substring(0, secondDash + 3);
+  let content = original.substring(secondDash + 3);
+
+  // Fix broken bullet lists where "-" is on its own line followed by blank line then content
+  // Pattern: line with just "-", blank line, then content text
+  // Should become: "- content text"
+  content = content.replace(/^- *\n\n(.+)/gm, '- $1');
+
+  // Also fix numbered lists with same pattern: "1." on own line, blank, content
+  content = content.replace(/^(\d+)\. *\n\n(.+)/gm, '$1. $2');
+
+  // Clean up any resulting triple+ blank lines
+  content = content.replace(/\n{3,}/g, '\n\n');
+
+  const result = frontmatter + content;
+  if (result !== original) {
+    fs.writeFileSync(filepath, result);
+    fixed++;;
+    console.log(`  Fixed: ${f}`);
+  }
+}
+console.log(`Fixed bullet lists in ${fixed} files`);
--- a/scripts/fix-code-blocks.mjs
+++ b/scripts/fix-code-blocks.mjs
@@ -0,0 +1,59 @@
+import fs from "fs";
+import path from "path";
+import { fileURLToPath } from "url";
+
+const __dirname = path.dirname(fileURLToPath(import.meta.url));
+const contentDir = path.join(__dirname, "..", "content", "resources");
+
+const files = fs.readdirSync(contentDir).filter((f) => f.endsWith(".mdx"));
+
+let totalFixed = 0;
+
+for (const file of files) {
+  const filePath = path.join(contentDir, file);
+  const raw = fs.readFileSync(filePath, "utf-8");
+
+  // Split frontmatter from content
+  const fmMatch = raw.match(/^---\n([\s\S]*?)\n---\n([\s\S]*)$/);
+  if (!fmMatch) continue;
+
+  const frontmatter = fmMatch[1];
+  const content = fmMatch[2];
+
+  const lines = content.split("\n");
+  let insideFencedBlock = false;
+  let changed = false;
+  const result = [];
+
+  for (let line of lines) {
+    // Toggle fenced code block state
+    if (/^```/.test(line.trimStart())) {
+      insideFencedBlock = !insideFencedBlock;
+      result.push(line);
+      continue;
+    }
+
+    if (insideFencedBlock) {
+      // Inside fenced code block: unescape HTML entities
+      const newLine = line.replace(/&lt;/g, "<").replace(/&gt;/g, ">");
+      if (newLine !== line) changed = true;
+      result.push(newLine);
+    } else {
+      // Outside fenced block: fix inline code spans
+      const newLine = line.replace(/`[^`]+`/g, (match) => {
+        return match.replace(/&lt;/g, "<").replace(/&gt;/g, ">");
+      });
+      if (newLine !== line) changed = true;
+      result.push(newLine);
+    }
+  }
+
+  if (changed) {
+    const output = `---\n${frontmatter}\n---\n${result.join("\n")}`;
+    fs.writeFileSync(filePath, output, "utf-8");
+    totalFixed++;
+    console.log(`Fixed: ${file}`);
+  }
+}
+
+console.log(`\nDone. Fixed ${totalFixed} files.`);
--- a/scripts/fix-mdx-jsx.mjs
+++ b/scripts/fix-mdx-jsx.mjs
@@ -0,0 +1,52 @@
+import fs from 'fs';
+import path from 'path';
+import { fileURLToPath } from 'url';
+
+const __dirname = path.dirname(fileURLToPath(import.meta.url));
+const dir = path.join(__dirname, '..', 'content', 'resources');
+
+// Known safe HTML/JSX tags that MDX should handle
+const safeTags = new Set([
+  'iframe', 'video', 'source', 'br', 'hr', 'img',
+  'div', 'span', 'p', 'a',
+  'h1', 'h2', 'h3', 'h4', 'h5', 'h6',
+  'ul', 'ol', 'li',
+  'strong', 'em', 'b', 'i', 'u',
+  'code', 'pre',
+  'blockquote',
+  'table', 'tr', 'td', 'th', 'thead', 'tbody',
+]);
+
+let fixed = 0;
+for (const f of fs.readdirSync(dir).filter(f => f.endsWith('.mdx'))) {
+  const filepath = path.join(dir, f);
+  const original = fs.readFileSync(filepath, 'utf8');
+
+  // Split into frontmatter and content
+  const fmEnd = original.indexOf('---', 4);
+  if (fmEnd === -1) continue;
+  const frontmatter = original.substring(0, fmEnd + 3);
+  let content = original.substring(fmEnd + 3);
+
+  // Escape angle brackets that look like HTML tags but aren't valid JSX
+  // Match < followed by something that's not a known safe tag or /
+  content = content.replace(/<(?!\/?(?:iframe|video|source|br|hr|img)\b)([^>]*?)>/g, (match, inner) => {
+    // Keep markdown image syntax ![...](...) - these don't match since they start with !
+    // Keep valid looking JSX/HTML
+    const tagName = inner.split(/[\s/]/)[0].toLowerCase();
+    if (safeTags.has(tagName) || tagName.startsWith('!--')) return match;
+    // It's a stray angle bracket - escape it
+    return `\\<${inner}\\>`;
+  });
+
+  // Also escape bare < that aren't part of tags (like math: x < y)
+  // But be careful not to double-escape
+  content = content.replace(/([^\\])<([^/!a-zA-Z\\])/g, '$1\\<$2');
+
+  const result = frontmatter + content;
+  if (result !== original) {
+    fs.writeFileSync(filepath, result);
+    fixed++;
+  }
+}
+console.log(`Fixed JSX issues in ${fixed} files`);
--- a/scripts/scrape-resources.mjs
+++ b/scripts/scrape-resources.mjs
@@ -0,0 +1,280 @@
+#!/usr/bin/env node
+/**
+ * Scrape all resources from the Squarespace site using the JSON API.
+ * Usage: node scripts/scrape-resources.mjs
+ */
+
+import fs from 'fs';
+import path from 'path';
+import { fileURLToPath } from 'url';
+
+const __dirname = path.dirname(fileURLToPath(import.meta.url));
+const CONTENT_DIR = path.join(__dirname, '..', 'content', 'resources');
+const IMAGE_DIR = path.join(__dirname, '..', 'public', 'images', 'resources');
+
+fs.mkdirSync(CONTENT_DIR, { recursive: true });
+fs.mkdirSync(IMAGE_DIR, { recursive: true });
+
+function parseResourceUrls(xml) {
+  const urls = [];
+  const regex = /<loc>(https?:\/\/micromelon\.com\.au\/resources\/[^<]+)<\/loc>/g;
+  let match;
+  while ((match = regex.exec(xml)) !== null) {
+    const url = match[1];
+    if (url.includes('/category/') || url.includes('/tag/')) continue;
+    if (url === 'https://micromelon.com.au/resources') continue;
+    urls.push(url);
+  }
+  return urls;
+}
+
+async function fetchPageJson(url) {
+  const res = await fetch(`${url}?format=json-pretty`, {
+    headers: { 'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7)' },
+  });
+  if (!res.ok) return null;
+  const text = await res.text();
+  if (!text.trim()) return null;
+  try { return JSON.parse(text); } catch { return null; }
+}
+
+function sqsBodyToMarkdown(html) {
+  if (!html) return '';
+
+  // Step 1: Extract meaningful content from Squarespace block structure
+  // The body is wrapped in sqs-layout divs. We need to find the actual content blocks.
+  let md = html;
+
+  // Remove Squarespace layout/grid wrapper divs but keep sqs-block-content
+  // First, extract all image URLs from data-src attributes (Squarespace lazy loading)
+  const imageSrcs = [];
+  const imgDataSrcRegex = /data-src="([^"]+)"/g;
+  let imgMatch;
+  while ((imgMatch = imgDataSrcRegex.exec(html)) !== null) {
+    const src = imgMatch[1].startsWith('//') ? `https:${imgMatch[1]}` : imgMatch[1];
+    imageSrcs.push(src);
+  }
+
+  // Extract video embeds (YouTube, Vimeo, etc.)
+  const videos = [];
+  const videoRegex = /data-html="([^"]*)"/g;
+  let videoMatch;
+  while ((videoMatch = videoRegex.exec(html)) !== null) {
+    const decoded = videoMatch[1]
+      .replace(/&lt;/g, '<').replace(/&gt;/g, '>')
+      .replace(/&amp;/g, '&').replace(/&quot;/g, '"');
+    const srcMatch = decoded.match(/src=["']([^"']+)["']/);
+    if (srcMatch) videos.push(srcMatch[1]);
+  }
+
+  // Also find direct iframe embeds
+  const iframeRegex = /<iframe[^>]*src=["']([^"']+)["'][^>]*>/gi;
+  let iframeMatch;
+  while ((iframeMatch = iframeRegex.exec(html)) !== null) {
+    if (!videos.includes(iframeMatch[1])) videos.push(iframeMatch[1]);
+  }
+
+  // Now do the HTML-to-markdown conversion
+  // Remove all Squarespace-specific wrapper divs, keeping inner content
+  // Remove style blocks
+  md = md.replace(/<style[^>]*>[\s\S]*?<\/style>/gi, '');
+
+  // Remove script blocks
+  md = md.replace(/<script[^>]*>[\s\S]*?<\/script>/gi, '');
+
+  // Handle Squarespace image blocks - replace with clean markdown images
+  // Match the noscript img tags which have the actual src
+  md = md.replace(/<noscript><img[^>]*src="([^"]+)"[^>]*\/?><\/noscript>/gi, (m, src) => {
+    const cleanSrc = src.startsWith('//') ? `https:${src}` : src;
+    return `\n\n![](${cleanSrc})\n\n`;
+  });
+
+  // Handle remaining img tags with data-src (lazy loaded)
+  md = md.replace(/<img[^>]*data-src="([^"]+)"[^>]*>/gi, (m, src) => {
+    const cleanSrc = src.startsWith('//') ? `https:${src}` : src;
+    return `\n\n![](${cleanSrc})\n\n`;
+  });
+
+  // Handle regular img tags
+  md = md.replace(/<img[^>]*src="([^"]+)"[^>]*>/gi, (m, src) => {
+    // Skip tracking pixels and tiny images
+    if (src.includes('static1.squarespace.com/static') && !src.includes('content/')) return '';
+    const cleanSrc = src.startsWith('//') ? `https:${src}` : src;
+    return `\n\n![](${cleanSrc})\n\n`;
+  });
+
+  // Handle video embeds
+  md = md.replace(/<div[^>]*class="[^"]*sqs-video-wrapper[^"]*"[^>]*data-html="([^"]*)"[^>]*>[\s\S]*?<\/div>/gi, (m, encoded) => {
+    const decoded = encoded.replace(/&lt;/g, '<').replace(/&gt;/g, '>').replace(/&amp;/g, '&').replace(/&quot;/g, '"');
+    const srcMatch = decoded.match(/src=["']([^"']+)["']/);
+    if (srcMatch) {
+      return `\n\n<iframe width="560" height="315" src="${srcMatch[1]}" frameBorder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowFullScreen></iframe>\n\n`;
+    }
+    return '';
+  });
+
+  // Handle iframes
+  md = md.replace(/<iframe[^>]*src="([^"]+)"[^>]*>[\s\S]*?<\/iframe>/gi, (m, src) => {
+    return `\n\n<iframe width="560" height="315" src="${src}" frameBorder="0" allowFullScreen></iframe>\n\n`;
+  });
+
+  // Headers
+  md = md.replace(/<h1[^>]*>([\s\S]*?)<\/h1>/gi, '\n\n# $1\n\n');
+  md = md.replace(/<h2[^>]*>([\s\S]*?)<\/h2>/gi, '\n\n## $1\n\n');
+  md = md.replace(/<h3[^>]*>([\s\S]*?)<\/h3>/gi, '\n\n### $1\n\n');
+  md = md.replace(/<h4[^>]*>([\s\S]*?)<\/h4>/gi, '\n\n#### $1\n\n');
+
+  // Links
+  md = md.replace(/<a[^>]*href="([^"]+)"[^>]*>([\s\S]*?)<\/a>/gi, '[$2]($1)');
+
+  // Bold/italic
+  md = md.replace(/<strong[^>]*>([\s\S]*?)<\/strong>/gi, '**$1**');
+  md = md.replace(/<b[^>]*>([\s\S]*?)<\/b>/gi, '**$1**');
+  md = md.replace(/<em[^>]*>([\s\S]*?)<\/em>/gi, '*$1*');
+
+  // Code
+  md = md.replace(/<code[^>]*>([\s\S]*?)<\/code>/gi, '`$1`');
+  md = md.replace(/<pre[^>]*>([\s\S]*?)<\/pre>/gi, '\n\n```\n$1\n```\n\n');
+
+  // Lists
+  md = md.replace(/<\/li>\s*<li[^>]*>/gi, '</li>\n<li>');
+  md = md.replace(/<li[^>]*>([\s\S]*?)<\/li>/gi, '- $1');
+  md = md.replace(/<\/?[uo]l[^>]*>/gi, '\n');
+
+  // Paragraphs
+  md = md.replace(/<p[^>]*>([\s\S]*?)<\/p>/gi, '\n\n$1\n\n');
+  md = md.replace(/<br\s*\/?>/gi, '\n');
+  md = md.replace(/<hr\s*\/?>/gi, '\n\n---\n\n');
+
+  // Blockquotes
+  md = md.replace(/<blockquote[^>]*>([\s\S]*?)<\/blockquote>/gi, '\n\n> $1\n\n');
+
+  // Remove all remaining HTML tags
+  md = md.replace(/<[^>]+>/g, '');
+
+  // Decode entities
+  md = md.replace(/&amp;/g, '&');
+  md = md.replace(/&lt;/g, '<');
+  md = md.replace(/&gt;/g, '>');
+  md = md.replace(/&quot;/g, '"');
+  md = md.replace(/&#39;/g, "'");
+  md = md.replace(/&nbsp;/g, ' ');
+  md = md.replace(/&#x27;/g, "'");
+  md = md.replace(/&#x2F;/g, '/');
+
+  // Remove "BACK TO POSTS" link that Squarespace adds
+  md = md.replace(/\[?\s*\*?\*?BACK TO POSTS\*?\*?\s*\]?\s*\([^)]*\)/gi, '');
+
+  // Clean up excessive whitespace
+  md = md.replace(/[ \t]+$/gm, ''); // trailing spaces
+  md = md.replace(/^\s+$/gm, ''); // whitespace-only lines -> empty
+  md = md.replace(/\n{3,}/g, '\n\n'); // max 2 newlines
+  md = md.trim();
+
+  return md;
+}
+
+function getCleanExcerpt(item) {
+  // Use Squarespace excerpt if available
+  if (item.excerpt) {
+    return item.excerpt.replace(/<[^>]+>/g, '').replace(/\s+/g, ' ').trim().substring(0, 250);
+  }
+  // Fallback: extract from body
+  const text = (item.body || '').replace(/<[^>]+>/g, '').replace(/\s+/g, ' ').trim();
+  return text.substring(0, 250);
+}
+
+async function downloadImage(url, filename) {
+  try {
+    const filepath = path.join(IMAGE_DIR, filename);
+    if (fs.existsSync(filepath)) return `/images/resources/${filename}`;
+
+    const res = await fetch(url, { headers: { 'User-Agent': 'Mozilla/5.0' } });
+    if (!res.ok) return null;
+
+    const buffer = Buffer.from(await res.arrayBuffer());
+    if (buffer.length < 100) return null; // skip tiny/empty files
+    fs.writeFileSync(filepath, buffer);
+    return `/images/resources/${filename}`;
+  } catch { return null; }
+}
+
+async function main() {
+  console.log('Fetching sitemap...');
+  const sitemapRes = await fetch('https://www.micromelon.com.au/sitemap.xml', {
+    headers: { 'User-Agent': 'Mozilla/5.0' },
+  });
+  const sitemapXml = await sitemapRes.text();
+  const urls = parseResourceUrls(sitemapXml);
+  console.log(`Found ${urls.length} resource posts.\n`);
+
+  // Clean out old files
+  for (const f of fs.readdirSync(CONTENT_DIR)) {
+    if (f.endsWith('.mdx')) fs.unlinkSync(path.join(CONTENT_DIR, f));
+  }
+
+  let count = 0;
+  let skipped = 0;
+
+  for (const url of urls) {
+    const slug = url.split('/resources/')[1];
+    if (!slug) continue;
+
+    count++;
+    process.stdout.write(`[${count}/${urls.length}] ${slug}...`);
+
+    const data = await fetchPageJson(url);
+    if (!data || !data.item) {
+      console.log(' SKIP (no JSON)');
+      skipped++;
+      continue;
+    }
+
+    const item = data.item;
+    const title = (item.title || slug).replace(/"/g, "'");
+    const date = item.publishOn
+      ? new Date(item.publishOn).toISOString().split('T')[0]
+      : new Date(item.addedOn || Date.now()).toISOString().split('T')[0];
+    const categories = item.categories || [];
+    const tags = item.tags || [];
+    const excerpt = getCleanExcerpt(item).replace(/"/g, "'");
+
+    // Featured image
+    let featuredImage = '';
+    if (item.assetUrl) {
+      const imgUrl = item.assetUrl.startsWith('//') ? `https:${item.assetUrl}` : item.assetUrl;
+      const ext = path.extname(imgUrl.split('?')[0]) || '.jpg';
+      const localPath = await downloadImage(imgUrl, `${slug}${ext}`);
+      if (localPath) featuredImage = localPath;
+    }
+
+    // Convert body
+    const content = sqsBodyToMarkdown(item.body || '');
+
+    // Build frontmatter
+    const fm = [
+      '---',
+      `title: "${title}"`,
+      `date: "${date}"`,
+      `categories: [${categories.map(c => `"${c}"`).join(', ')}]`,
+      `tags: [${tags.map(t => `"${t}"`).join(', ')}]`,
+      `excerpt: "${excerpt}"`,
+    ];
+    if (featuredImage) fm.push(`featuredImage: "${featuredImage}"`);
+    fm.push('---');
+
+    fs.writeFileSync(
+      path.join(CONTENT_DIR, `${slug}.mdx`),
+      fm.join('\n') + '\n\n' + content + '\n'
+    );
+
+    console.log(` OK (${content.length} chars)`);
+    await new Promise(r => setTimeout(r, 300));
+  }
+
+  const files = fs.readdirSync(CONTENT_DIR).filter(f => f.endsWith('.mdx'));
+  const images = fs.readdirSync(IMAGE_DIR);
+  console.log(`\nDone! ${files.length} MDX files, ${images.length} images. Skipped: ${skipped}`);
+}
+
+main().catch(console.error);