@voicenter-team/nuxt-llms-generator 0.1.2 → 0.1.3

package/dist/chunks/llms-files-generator.mjs

@@ -615,7 +615,9 @@ function excludeChildrenFromContent(content) {
   return cleanContent;
 }
 function generatePageId(urlItem) {
-  return `${urlItem.TemplateAlias}_${urlItem.nodeID}`;
+  const templateAlias = urlItem.TemplateAlias || "UnknownTemplate";
+  const nodeID = urlItem.nodeID || "UnknownNode";
+  return `${templateAlias}_${nodeID}`;
 }
 function estimateContentTokens(content) {
   try {
@@ -1323,6 +1325,7 @@ class TemplateGenerator {
 class LLMSFilesGenerator {
   config;
   templateGenerator;
+  nhm = new NodeHtmlMarkdown();
   constructor(config) {
     this.config = config;
     this.templateGenerator = new TemplateGenerator(config);
@@ -1439,7 +1442,6 @@ class LLMSFilesGenerator {
 
 `;
     }
-    content += "This document contains all website content in a single file for comprehensive AI analysis.\n\n";
     content += "---\n\n";
     for (const mdFile of mdFiles) {
       const urlItem = umbracoData.urlList.find((item) => item.url === mdFile.url);
@@ -1516,11 +1518,22 @@ class LLMSFilesGenerator {
   }
   extractSiteTitle(umbracoData) {
     const siteData = umbracoData.SiteData;
-    return siteData?.pageTitle || siteData?.mainHeaderBlockTitle || "Website Documentation";
+    const rawTitle = siteData?.pageTitle || siteData?.mainHeaderBlockTitle || "Website Documentation";
+    return this.cleanHtmlContent(rawTitle);
   }
   extractSiteDescription(umbracoData) {
     const siteData = umbracoData.SiteData;
-    return siteData?.pageDescription || siteData?.ogDescription || null;
+    const rawDescription = siteData?.pageDescription || siteData?.ogDescription || null;
+    return rawDescription ? this.cleanHtmlContent(rawDescription) : null;
+  }
+  cleanHtmlContent(content) {
+    if (!this.config.enableHtmlToMarkdown) {
+      return content;
+    }
+    if (/<[^>]+>/.test(content)) {
+      return this.nhm.translate(content).trim();
+    }
+    return content;
   }
   extractPageTitle(umbracoData, urlItem) {
     if (!urlItem)
@@ -1553,7 +1566,14 @@ class LLMSFilesGenerator {
     return names[category] || category.charAt(0).toUpperCase() + category.slice(1);
   }
   sanitizeUrlForFilename(url) {
-    return url.replace(/^\//, "").replace(/\/$/, "").replace(/\//g, "-").replace(/[^a-zA-Z0-9\-_]/g, "").replace(/^$/, "index");
+    let filename = url.replace(/^\//, "").replace(/\/$/, "").replace(/\//g, "-").replace(/[^a-zA-Z0-9\-_.]/g, "").replace(/--+/g, "-").replace(/^-+|-+$/g, "");
+    if (!filename || filename === "") {
+      filename = "index";
+    }
+    if (filename.startsWith("-") || filename.startsWith(".")) {
+      filename = "page-" + filename.replace(/^[-.]/, "");
+    }
+    return filename;
   }
   getRelativeFilePath(fullPath) {
     const filename = fullPath.split("/").pop() || "";

package/dist/module.json

@@ -4,5 +4,5 @@
   "compatibility": {
     "nuxt": "^3.0.0"
   },
-  "version": "0.1.2"
+  "version": "0.1.3"
 }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@voicenter-team/nuxt-llms-generator",
-  "version": "0.1.2",
+  "version": "0.1.3",
   "description": "Nuxt 3 module for automatically generating AI-optimized documentation files (llms.txt, llms-full.txt, and individual .md files) from Umbraco CMS data using Anthropic's Claude API.",
   "repository": "your-org/my-module",
   "license": "MIT",