@learnpack/learnpack 5.0.334 → 5.0.339

package/src/utils/export/epub.ts

@@ -1,553 +1,553 @@
-import * as path from "path";
-import * as fs from "fs";
-import * as mkdirp from "mkdirp";
-import * as rimraf from "rimraf";
-import { v4 as uuidv4 } from "uuid";
-import { spawn } from "child_process";
-import { ExportOptions, CourseMetadata, EpubMetadata } from "./types";
-import { downloadS3Folder, getCourseMetadata } from "./shared";
-
-const getLocalizedText = (language: string) => {
-  const translations = {
-    en: "Write your answer here",
-    es: "Escribe tu respuesta aquí",
-    fr: "Écrivez votre réponse ici",
-    de: "Schreiben Sie Ihre Antwort hier",
-    it: "Scrivi la tua risposta qui",
-    pt: "Escreva sua resposta aqui",
-    ja: "ここに答えを書いてください",
-    ko: "여기에 답을 작성하세요",
-    zh: "在这里写下你的答案",
-    ar: "اكتب إجابتك هنا",
-    ru: "Напишите свой ответ здесь",
-  };
-
-  return translations[language as keyof typeof translations] || translations.en;
-};
-
-const replaceQuestionsFromMarkdown = (
-  markdown: string,
-  language: string = "en"
-) => {
-  // Search for all code blocks like this (No breaking lines sensitive)
-  // ```question some parameters
-  // some text
-  // ```
-  const questionRegex = /```question([\s\S]*?)```/g;
-  const localizedText = getLocalizedText(language);
-  return markdown.replace(
-    questionRegex,
-    `<div type="text" class="open-question">${localizedText}</div>`
-  );
-};
-
-const replaceMermaidFromMarkdown = (markdown: string) => {
-  // Search for all code blocks like this (No breaking lines sensitive)
-  // ```mermaid
-  // some graphs in mermaid syntax
-  // ```
-  const mermaidRegex = /```mermaid([\s\S]*?)```/g;
-  return markdown.replace(mermaidRegex, "");
-};
-
-const cleanQuizOptions = (markdown: string) => {
-  // Remove [x] from quiz options, keeping only [ ] for all options
-  // This handles both - [x] and - [x] patterns
-  const quizRegex = /^(\s*-\s*)\[x\](\s+.*)$/gm;
-  return markdown.replace(quizRegex, "$1[ ]$2");
-};
-
-// Download cover image from bucket
-async function downloadCoverImage(
-  bucket: any,
-  courseSlug: string,
-  outputDir: string
-): Promise<string | undefined> {
-  try {
-    const coverImagePath = path.join(outputDir, "cover.png");
-    const [coverBuf] = await bucket
-      .file(`courses/${courseSlug}/preview.png`)
-      .download();
-    fs.writeFileSync(coverImagePath, coverBuf);
-    console.log("Cover image downloaded:", coverImagePath);
-    return coverImagePath;
-  } catch (error) {
-    console.log("No cover image found, skipping...");
-    return undefined;
-  }
-}
-
-// Generate EPUB using pandoc directly from markdown files
-async function generateEpub(
-  markdownFiles: string[],
-  outputPath: string,
-  metadata: CourseMetadata,
-  epubMetadata: EpubMetadata,
-  coverImagePath?: string
-): Promise<void> {
-  return new Promise((resolve, reject) => {
-    let stdout = "";
-    let stderr = "";
-
-    const yamlPath = path.join(path.dirname(outputPath), "metadata.yaml");
-
-    const escapeYaml = (str: string) => {
-      if (!str) return "";
-      return str.replace(/"/g, '\\"').replace(/\n/g, "\\n");
-    };
-
-    const yamlContent = `---
-title: "${epubMetadata.title}"
-creator: "${epubMetadata.creator}"
-publisher: "${epubMetadata.publisher}"
-rights: "${epubMetadata.rights}"
-lang: "${epubMetadata.lang}"
-description: "${escapeYaml(metadata.description || "")}"
-subject: "${metadata.technologies?.join(", ") || "Programming"}"
----`;
-
-    fs.writeFileSync(yamlPath, yamlContent, "utf-8");
-
-    // Include all markdown files in the EPUB
-    const pandocArgs = [
-      ...markdownFiles, // All markdown files
-      "-o",
-      outputPath,
-      "--from",
-      "markdown",
-      "--to",
-      "epub",
-      "--metadata-file",
-      yamlPath,
-      "--css",
-      path.join(__dirname, "../templates/epub/epub.css"),
-    ];
-
-    // Add cover image if available - use absolute path
-    if (coverImagePath) {
-      pandocArgs.push("--epub-cover-image", coverImagePath);
-    }
-
-    console.log("🔧 Executing pandoc EPUB command:");
-    console.log("Command:", "pandoc", pandocArgs.join(" "));
-
-    const pandoc = spawn("pandoc", pandocArgs);
-
-    pandoc.stdout.on("data", (data) => {
-      const output = data.toString();
-      stdout += output;
-      console.log("📤 Pandoc STDOUT:", output);
-    });
-
-    pandoc.stderr.on("data", (data) => {
-      const output = data.toString();
-      stderr += output;
-      console.log("⚠️ Pandoc STDERR:", output);
-    });
-
-    pandoc.on("close", (code) => {
-      console.log(`🏁 Pandoc process finished with exit code: ${code}`);
-      if (code === 0) {
-        console.log("✅ EPUB generation successful!");
-        resolve();
-      } else {
-        const errorMessage = `Pandoc process exited with code ${code}\n\nSTDOUT:\n${stdout}\n\nSTDERR:\n${stderr}`;
-        console.error("❌ Pandoc Error Details:", errorMessage);
-        reject(new Error(errorMessage));
-      }
-    });
-
-    pandoc.on("error", (err) => {
-      console.error("❌ Pandoc spawn error:", err);
-      reject(err);
-    });
-  });
-}
-
-// Process exercises and collect markdown files for direct EPUB conversion
-async function processExercises(
-  exercisesDir: string,
-  outputDir: string,
-  language: string = "en",
-  learnAssetsDir?: string
-): Promise<string[]> {
-  const processedMarkdownFiles: string[] = [];
-
-  console.log("Processing exercises directory:", exercisesDir);
-  console.log("Output directory:", outputDir);
-  console.log("Language:", language);
-  console.log("Learn assets directory:", learnAssetsDir);
-
-  if (!fs.existsSync(exercisesDir)) {
-    console.log("Exercises directory does not exist:", exercisesDir);
-    return processedMarkdownFiles;
-  }
-
-  console.log("Exercises directory exists, scanning for files...");
-
-  const processDirectory = async (dir: string, relativePath: string = "") => {
-    console.log(`🔍 Processing directory: ${dir} (relative: ${relativePath})`);
-    const entries = fs.readdirSync(dir, { withFileTypes: true });
-    console.log(
-      `📁 Found ${entries.length} entries in ${dir}:`,
-      entries.map((e) => e.name)
-    );
-
-    // Sort entries to prioritize README files first
-    entries.sort((a, b) => {
-      const aName = a.name.toLowerCase();
-      const bName = b.name.toLowerCase();
-
-      // Prioritize README files
-      if (aName === "readme.md" && bName !== "readme.md") return -1;
-      if (bName === "readme.md" && aName !== "readme.md") return 1;
-
-      return aName.localeCompare(bName);
-    });
-
-    for (const entry of entries) {
-      const fullPath = path.join(dir, entry.name);
-      const relativeFilePath = path.join(relativePath, entry.name);
-
-      if (entry.isDirectory()) {
-        console.log("Processing subdirectory:", relativeFilePath);
-        // esling-disable-next-line
-        await processDirectory(fullPath, relativeFilePath);
-      } else if (
-        entry.name.endsWith(".md") ||
-        entry.name.endsWith(".markdown")
-      ) {
-        // Apply language filtering for all markdown files
-        const baseName = entry.name.replace(/\.(md|markdown)$/, "");
-
-        if (language === "en") {
-          // For English: only include README.md and files without language suffixes
-          if (entry.name.toLowerCase() === "readme.md") {
-            console.log(`✅ Including English README: ${relativeFilePath}`);
-          } else if (
-            baseName.includes(".") &&
-            /\.(es|fr|de|it|pt|ja|ko|zh|ar|ru)$/i.test(baseName)
-          ) {
-            console.log(
-              `❌ Skipping language-specific file for English: ${relativeFilePath}`
-            );
-            continue;
-          } else {
-            console.log(`✅ Including English file: ${relativeFilePath}`);
-          }
-        } else {
-          // For other languages: only include files with the correct language suffix
-          if (entry.name.toLowerCase() === "readme.md") {
-            console.log(
-              `❌ Skipping English README for language ${language}: ${relativeFilePath}`
-            );
-            continue;
-          } else if (!baseName.endsWith(`.${language}`)) {
-            console.log(
-              `❌ Skipping file with different language: ${relativeFilePath}`
-            );
-            continue;
-          } else {
-            console.log(`✅ Including ${language} file: ${relativeFilePath}`);
-          }
-        }
-
-        // Process markdown content and save to output directory
-        console.log("Processing markdown file:", relativeFilePath);
-        const processedContent = processMarkdownContent(
-          fullPath,
-          learnAssetsDir,
-          language
-        );
-
-        const processedMarkdownPath = path.join(
-          outputDir,
-          relativeFilePath.replace(/\.(md|markdown)$/, ".processed.md")
-        );
-
-        console.log("Processed markdown output path:", processedMarkdownPath);
-        mkdirp.sync(path.dirname(processedMarkdownPath));
-        fs.writeFileSync(processedMarkdownPath, processedContent, "utf-8");
-
-        processedMarkdownFiles.push(processedMarkdownPath);
-        console.log(
-          "Successfully processed:",
-          relativeFilePath,
-          "->",
-          processedMarkdownPath
-        );
-      } else if (entry.name.endsWith(".html")) {
-        // Convert HTML files to markdown for consistency
-        console.log("Converting HTML file to markdown:", relativeFilePath);
-        const htmlContent = fs.readFileSync(fullPath, "utf-8");
-        const processedMarkdownPath = path.join(
-          outputDir,
-          relativeFilePath.replace(/\.html$/, ".processed.md")
-        );
-
-        mkdirp.sync(path.dirname(processedMarkdownPath));
-        fs.writeFileSync(processedMarkdownPath, htmlContent, "utf-8");
-        processedMarkdownFiles.push(processedMarkdownPath);
-        console.log(
-          "Successfully converted HTML:",
-          relativeFilePath,
-          "->",
-          processedMarkdownPath
-        );
-      } else {
-        console.log(
-          "Skipping file:",
-          relativeFilePath,
-          "(not markdown or HTML)"
-        );
-      }
-    }
-  };
-
-  await processDirectory(exercisesDir);
-  return processedMarkdownFiles;
-}
-
-// Process markdown content for EPUB conversion
-function processMarkdownContent(
-  markdownPath: string,
-  learnAssetsDir?: string,
-  language: string = "en"
-): string {
-  let markdownContent = fs.readFileSync(markdownPath, "utf-8");
-
-  if (learnAssetsDir && fs.existsSync(learnAssetsDir)) {
-    // Find all image references in markdown
-    const imageRegex = /!\[([^\]]*)\]\(([^)]+)\)/g;
-    let match;
-
-    while ((match = imageRegex.exec(markdownContent)) !== null) {
-      const [fullMatch, altText, imagePath] = match;
-
-      // Check if it's a relative path starting with /learn/assets/
-      if (imagePath.startsWith("/.learn/assets/")) {
-        const imageName = path.basename(imagePath);
-        const absoluteImagePath = path.join(learnAssetsDir, imageName);
-
-        console.log(`📸 Looking for image: ${imageName}`);
-        console.log(`📁 Learn assets dir: ${learnAssetsDir}`);
-        console.log(`📁 Absolute path: ${absoluteImagePath}`);
-        console.log(`📁 Path exists: ${fs.existsSync(absoluteImagePath)}`);
-
-        if (fs.existsSync(absoluteImagePath)) {
-          // Use absolute path for the image
-          console.log(
-            `✅ Found image, replacing path: ${imagePath} -> ${absoluteImagePath}`
-          );
-          markdownContent = markdownContent.replace(
-            fullMatch,
-            `![${altText}](${absoluteImagePath})`
-          );
-        } else {
-          console.log(`❌ Image not found: ${absoluteImagePath}`);
-          // List files in the assets directory for debugging
-          if (fs.existsSync(learnAssetsDir)) {
-            const files = fs.readdirSync(learnAssetsDir);
-            console.log(`📁 Files in assets directory: ${files.join(", ")}`);
-          } else {
-            console.log(
-              `❌ Assets directory does not exist: ${learnAssetsDir}`
-            );
-          }
-        }
-      }
-    }
-  }
-
-  markdownContent = replaceQuestionsFromMarkdown(markdownContent, language);
-  markdownContent = replaceMermaidFromMarkdown(markdownContent);
-  markdownContent = cleanQuizOptions(markdownContent);
-
-  return markdownContent;
-}
-
-const titlelify = (title: string) => {
-  return title.replace(/[-_]/g, " ").replace(/\b\w/g, (l) => l.toUpperCase());
-};
-
-// Read sidebar.json to get exercise titles
-function readSidebarData(learnDir: string): any {
-  try {
-    const sidebarPath = path.join(learnDir, "sidebar.json");
-    if (fs.existsSync(sidebarPath)) {
-      const sidebarContent = fs.readFileSync(sidebarPath, "utf-8");
-      return JSON.parse(sidebarContent);
-    }
-  } catch (error) {
-    console.log("⚠️ Could not read sidebar.json:", error);
-  }
-  return null;
-}
-
-export async function exportToEpub(
-  options: ExportOptions,
-  epubMetadata: EpubMetadata
-): Promise<string> {
-  const { courseSlug, bucket, outDir, language = "en" } = options;
-
-  console.log("🚀 Starting EPUB export for course:", courseSlug);
-  console.log("📁 Output directory:", outDir);
-  console.log("🌍 Language:", language);
-  console.log("📋 EPUB Metadata:", epubMetadata);
-
-  // 1. Create temporary folder
-  const tmpName = uuidv4();
-  const epubOutDir = path.join(outDir, tmpName);
-  rimraf.sync(epubOutDir);
-  mkdirp.sync(epubOutDir);
-
-  console.log("Created temporary directory:", epubOutDir);
-
-  // 2. Create EPUB template directory
-  const epubTemplateDir = path.join(epubOutDir, "template");
-  mkdirp.sync(epubTemplateDir);
-
-  // 3. Download exercises, .learn, and cover image from bucket
-  console.log("📥 Downloading exercises...");
-  await downloadS3Folder(
-    bucket,
-    `courses/${courseSlug}/exercises/`,
-    path.join(epubOutDir, "exercises")
-  );
-  console.log("✅ Exercises downloaded successfully");
-
-  console.log("📥 Downloading .learn files...");
-  await downloadS3Folder(
-    bucket,
-    `courses/${courseSlug}/.learn/`,
-    path.join(epubOutDir, "learn")
-  );
-  console.log("✅ .learn files downloaded successfully");
-
-  console.log("📥 Downloading cover image...");
-  const coverImagePath = await downloadCoverImage(
-    bucket,
-    courseSlug,
-    epubOutDir
-  );
-  console.log(
-    "✅ Cover image download completed:",
-    coverImagePath ? "Found" : "Not found"
-  );
-
-  // 4. Read learn.json for course info
-  console.log("Reading course metadata...");
-  const learnJson = await getCourseMetadata(bucket, courseSlug);
-
-  const metadata: CourseMetadata = {
-    title: learnJson.title?.[epubMetadata.lang] || courseSlug,
-    description:
-      learnJson.description?.[epubMetadata.lang] || learnJson.description,
-    language: epubMetadata.lang,
-    technologies: learnJson.technologies || [],
-    difficulty: learnJson.difficulty || "beginner",
-  };
-
-  console.log("Course metadata:", metadata);
-
-  // 5. Process exercises and prepare markdown files
-  console.log("🔄 Processing exercises and preparing markdown files...");
-
-  // First, let's see what files were downloaded
-  const exercisesPath = path.join(epubOutDir, "exercises");
-  console.log("📂 Exercises directory contents:");
-  if (fs.existsSync(exercisesPath)) {
-    const exerciseDirs = fs.readdirSync(exercisesPath, { withFileTypes: true });
-    console.log(`Found ${exerciseDirs.length} exercise directories:`);
-    exerciseDirs.forEach((dir) => {
-      if (dir.isDirectory()) {
-        const dirPath = path.join(exercisesPath, dir.name);
-        const files = fs.readdirSync(dirPath);
-        console.log(
-          `  📁 ${dir.name}: ${files.length} files - ${files.join(", ")}`
-        );
-      }
-    });
-  } else {
-    console.log("❌ Exercises directory does not exist!");
-  }
-
-  const processedMarkdownFiles = await processExercises(
-    path.join(epubOutDir, "exercises"),
-    path.join(epubOutDir, "processed"),
-    language,
-    path.join(epubOutDir, "learn", "assets")
-  );
-
-  // 6. Read sidebar data and create table of contents
-  const sidebarData = readSidebarData(path.join(epubOutDir, "learn"));
-  console.log("📋 Sidebar data loaded:", sidebarData ? "Success" : "Not found");
-
-  // 7. Create main content file as markdown
-  const mainContent = `# ${metadata.title}
-
-${metadata.description || ""}
-
-**Technologies:** ${metadata.technologies?.join(", ") || "Programming"}
-
-**Difficulty:** ${metadata.difficulty}
-
-**Language:** ${language}
-
----
-
-## Table of Contents
-
-${processedMarkdownFiles
-  .map((file, index) => {
-    const fileName = path.basename(file, ".processed.md");
-    const parentDir = path.basename(path.dirname(file));
-
-    // Get the title from sidebar.json using the parent directory name
-    let displayName = fileName;
-    if (sidebarData && sidebarData[parentDir]) {
-      displayName =
-        sidebarData[parentDir][language] ||
-        sidebarData[parentDir]["en"] ||
-        titlelify(parentDir);
-    } else {
-      displayName = titlelify(parentDir);
-    }
-
-    return `${index + 1}. ${displayName}`;
-  })
-  .join("\n")}
-
----
-
-`;
-
-  const mainContentPath = path.join(epubOutDir, "processed", "main.md");
-  mkdirp.sync(path.dirname(mainContentPath));
-  fs.writeFileSync(mainContentPath, mainContent, "utf-8");
-
-  // 8. Prepare all markdown files for EPUB generation
-  const allMarkdownFiles = [mainContentPath, ...processedMarkdownFiles];
-
-  // 9. Generate EPUB using pandoc
-  console.log("📖 Generating EPUB file...");
-  const epubPath = path.join(epubOutDir, `${courseSlug}.epub`);
-  await generateEpub(
-    allMarkdownFiles,
-    epubPath,
-    metadata,
-    epubMetadata,
-    coverImagePath
-  );
-
-  console.log("EPUB generated successfully:", epubPath);
-
-  // Clean up temporary files
-  console.log("Cleaning up temporary files...");
-  rimraf.sync(path.join(epubOutDir, "processed"));
-  rimraf.sync(path.join(epubOutDir, "exercises"));
-  rimraf.sync(path.join(epubOutDir, "learn"));
-
-  console.log("EPUB export completed successfully");
-  return epubPath;
-}
+import * as path from "path";
+import * as fs from "fs";
+import * as mkdirp from "mkdirp";
+import * as rimraf from "rimraf";
+import { v4 as uuidv4 } from "uuid";
+import { spawn } from "child_process";
+import { ExportOptions, CourseMetadata, EpubMetadata } from "./types";
+import { downloadS3Folder, getCourseMetadata } from "./shared";
+
+const getLocalizedText = (language: string) => {
+  const translations = {
+    en: "Write your answer here",
+    es: "Escribe tu respuesta aquí",
+    fr: "Écrivez votre réponse ici",
+    de: "Schreiben Sie Ihre Antwort hier",
+    it: "Scrivi la tua risposta qui",
+    pt: "Escreva sua resposta aqui",
+    ja: "ここに答えを書いてください",
+    ko: "여기에 답을 작성하세요",
+    zh: "在这里写下你的答案",
+    ar: "اكتب إجابتك هنا",
+    ru: "Напишите свой ответ здесь",
+  };
+
+  return translations[language as keyof typeof translations] || translations.en;
+};
+
+const replaceQuestionsFromMarkdown = (
+  markdown: string,
+  language: string = "en"
+) => {
+  // Search for all code blocks like this (No breaking lines sensitive)
+  // ```question some parameters
+  // some text
+  // ```
+  const questionRegex = /```question([\s\S]*?)```/g;
+  const localizedText = getLocalizedText(language);
+  return markdown.replace(
+    questionRegex,
+    `<div type="text" class="open-question">${localizedText}</div>`
+  );
+};
+
+const replaceMermaidFromMarkdown = (markdown: string) => {
+  // Search for all code blocks like this (No breaking lines sensitive)
+  // ```mermaid
+  // some graphs in mermaid syntax
+  // ```
+  const mermaidRegex = /```mermaid([\s\S]*?)```/g;
+  return markdown.replace(mermaidRegex, "");
+};
+
+const cleanQuizOptions = (markdown: string) => {
+  // Remove [x] from quiz options, keeping only [ ] for all options
+  // This handles both - [x] and - [x] patterns
+  const quizRegex = /^(\s*-\s*)\[x\](\s+.*)$/gm;
+  return markdown.replace(quizRegex, "$1[ ]$2");
+};
+
+// Download cover image from bucket
+async function downloadCoverImage(
+  bucket: any,
+  courseSlug: string,
+  outputDir: string
+): Promise<string | undefined> {
+  try {
+    const coverImagePath = path.join(outputDir, "cover.png");
+    const [coverBuf] = await bucket
+      .file(`courses/${courseSlug}/preview.png`)
+      .download();
+    fs.writeFileSync(coverImagePath, coverBuf);
+    console.log("Cover image downloaded:", coverImagePath);
+    return coverImagePath;
+  } catch (error) {
+    console.log("No cover image found, skipping...");
+    return undefined;
+  }
+}
+
+// Generate EPUB using pandoc directly from markdown files
+async function generateEpub(
+  markdownFiles: string[],
+  outputPath: string,
+  metadata: CourseMetadata,
+  epubMetadata: EpubMetadata,
+  coverImagePath?: string
+): Promise<void> {
+  return new Promise((resolve, reject) => {
+    let stdout = "";
+    let stderr = "";
+
+    const yamlPath = path.join(path.dirname(outputPath), "metadata.yaml");
+
+    const escapeYaml = (str: string) => {
+      if (!str) return "";
+      return str.replace(/"/g, '\\"').replace(/\n/g, "\\n");
+    };
+
+    const yamlContent = `---
+title: "${epubMetadata.title}"
+creator: "${epubMetadata.creator}"
+publisher: "${epubMetadata.publisher}"
+rights: "${epubMetadata.rights}"
+lang: "${epubMetadata.lang}"
+description: "${escapeYaml(metadata.description || "")}"
+subject: "${metadata.technologies?.join(", ") || "Programming"}"
+---`;
+
+    fs.writeFileSync(yamlPath, yamlContent, "utf-8");
+
+    // Include all markdown files in the EPUB
+    const pandocArgs = [
+      ...markdownFiles, // All markdown files
+      "-o",
+      outputPath,
+      "--from",
+      "markdown",
+      "--to",
+      "epub",
+      "--metadata-file",
+      yamlPath,
+      "--css",
+      path.join(__dirname, "../templates/epub/epub.css"),
+    ];
+
+    // Add cover image if available - use absolute path
+    if (coverImagePath) {
+      pandocArgs.push("--epub-cover-image", coverImagePath);
+    }
+
+    console.log("🔧 Executing pandoc EPUB command:");
+    console.log("Command:", "pandoc", pandocArgs.join(" "));
+
+    const pandoc = spawn("pandoc", pandocArgs);
+
+    pandoc.stdout.on("data", (data) => {
+      const output = data.toString();
+      stdout += output;
+      console.log("📤 Pandoc STDOUT:", output);
+    });
+
+    pandoc.stderr.on("data", (data) => {
+      const output = data.toString();
+      stderr += output;
+      console.log("⚠️ Pandoc STDERR:", output);
+    });
+
+    pandoc.on("close", (code) => {
+      console.log(`🏁 Pandoc process finished with exit code: ${code}`);
+      if (code === 0) {
+        console.log("✅ EPUB generation successful!");
+        resolve();
+      } else {
+        const errorMessage = `Pandoc process exited with code ${code}\n\nSTDOUT:\n${stdout}\n\nSTDERR:\n${stderr}`;
+        console.error("❌ Pandoc Error Details:", errorMessage);
+        reject(new Error(errorMessage));
+      }
+    });
+
+    pandoc.on("error", (err) => {
+      console.error("❌ Pandoc spawn error:", err);
+      reject(err);
+    });
+  });
+}
+
+// Process exercises and collect markdown files for direct EPUB conversion
+async function processExercises(
+  exercisesDir: string,
+  outputDir: string,
+  language: string = "en",
+  learnAssetsDir?: string
+): Promise<string[]> {
+  const processedMarkdownFiles: string[] = [];
+
+  console.log("Processing exercises directory:", exercisesDir);
+  console.log("Output directory:", outputDir);
+  console.log("Language:", language);
+  console.log("Learn assets directory:", learnAssetsDir);
+
+  if (!fs.existsSync(exercisesDir)) {
+    console.log("Exercises directory does not exist:", exercisesDir);
+    return processedMarkdownFiles;
+  }
+
+  console.log("Exercises directory exists, scanning for files...");
+
+  const processDirectory = async (dir: string, relativePath: string = "") => {
+    console.log(`🔍 Processing directory: ${dir} (relative: ${relativePath})`);
+    const entries = fs.readdirSync(dir, { withFileTypes: true });
+    console.log(
+      `📁 Found ${entries.length} entries in ${dir}:`,
+      entries.map((e) => e.name)
+    );
+
+    // Sort entries to prioritize README files first
+    entries.sort((a, b) => {
+      const aName = a.name.toLowerCase();
+      const bName = b.name.toLowerCase();
+
+      // Prioritize README files
+      if (aName === "readme.md" && bName !== "readme.md") return -1;
+      if (bName === "readme.md" && aName !== "readme.md") return 1;
+
+      return aName.localeCompare(bName);
+    });
+
+    for (const entry of entries) {
+      const fullPath = path.join(dir, entry.name);
+      const relativeFilePath = path.join(relativePath, entry.name);
+
+      if (entry.isDirectory()) {
+        console.log("Processing subdirectory:", relativeFilePath);
+        // esling-disable-next-line
+        await processDirectory(fullPath, relativeFilePath);
+      } else if (
+        entry.name.endsWith(".md") ||
+        entry.name.endsWith(".markdown")
+      ) {
+        // Apply language filtering for all markdown files
+        const baseName = entry.name.replace(/\.(md|markdown)$/, "");
+
+        if (language === "en") {
+          // For English: only include README.md and files without language suffixes
+          if (entry.name.toLowerCase() === "readme.md") {
+            console.log(`✅ Including English README: ${relativeFilePath}`);
+          } else if (
+            baseName.includes(".") &&
+            /\.(es|fr|de|it|pt|ja|ko|zh|ar|ru)$/i.test(baseName)
+          ) {
+            console.log(
+              `❌ Skipping language-specific file for English: ${relativeFilePath}`
+            );
+            continue;
+          } else {
+            console.log(`✅ Including English file: ${relativeFilePath}`);
+          }
+        } else {
+          // For other languages: only include files with the correct language suffix
+          if (entry.name.toLowerCase() === "readme.md") {
+            console.log(
+              `❌ Skipping English README for language ${language}: ${relativeFilePath}`
+            );
+            continue;
+          } else if (!baseName.endsWith(`.${language}`)) {
+            console.log(
+              `❌ Skipping file with different language: ${relativeFilePath}`
+            );
+            continue;
+          } else {
+            console.log(`✅ Including ${language} file: ${relativeFilePath}`);
+          }
+        }
+
+        // Process markdown content and save to output directory
+        console.log("Processing markdown file:", relativeFilePath);
+        const processedContent = processMarkdownContent(
+          fullPath,
+          learnAssetsDir,
+          language
+        );
+
+        const processedMarkdownPath = path.join(
+          outputDir,
+          relativeFilePath.replace(/\.(md|markdown)$/, ".processed.md")
+        );
+
+        console.log("Processed markdown output path:", processedMarkdownPath);
+        mkdirp.sync(path.dirname(processedMarkdownPath));
+        fs.writeFileSync(processedMarkdownPath, processedContent, "utf-8");
+
+        processedMarkdownFiles.push(processedMarkdownPath);
+        console.log(
+          "Successfully processed:",
+          relativeFilePath,
+          "->",
+          processedMarkdownPath
+        );
+      } else if (entry.name.endsWith(".html")) {
+        // Convert HTML files to markdown for consistency
+        console.log("Converting HTML file to markdown:", relativeFilePath);
+        const htmlContent = fs.readFileSync(fullPath, "utf-8");
+        const processedMarkdownPath = path.join(
+          outputDir,
+          relativeFilePath.replace(/\.html$/, ".processed.md")
+        );
+
+        mkdirp.sync(path.dirname(processedMarkdownPath));
+        fs.writeFileSync(processedMarkdownPath, htmlContent, "utf-8");
+        processedMarkdownFiles.push(processedMarkdownPath);
+        console.log(
+          "Successfully converted HTML:",
+          relativeFilePath,
+          "->",
+          processedMarkdownPath
+        );
+      } else {
+        console.log(
+          "Skipping file:",
+          relativeFilePath,
+          "(not markdown or HTML)"
+        );
+      }
+    }
+  };
+
+  await processDirectory(exercisesDir);
+  return processedMarkdownFiles;
+}
+
+// Process markdown content for EPUB conversion
+function processMarkdownContent(
+  markdownPath: string,
+  learnAssetsDir?: string,
+  language: string = "en"
+): string {
+  let markdownContent = fs.readFileSync(markdownPath, "utf-8");
+
+  if (learnAssetsDir && fs.existsSync(learnAssetsDir)) {
+    // Find all image references in markdown
+    const imageRegex = /!\[([^\]]*)\]\(([^)]+)\)/g;
+    let match;
+
+    while ((match = imageRegex.exec(markdownContent)) !== null) {
+      const [fullMatch, altText, imagePath] = match;
+
+      // Check if it's a relative path starting with /learn/assets/
+      if (imagePath.startsWith("/.learn/assets/")) {
+        const imageName = path.basename(imagePath);
+        const absoluteImagePath = path.join(learnAssetsDir, imageName);
+
+        console.log(`📸 Looking for image: ${imageName}`);
+        console.log(`📁 Learn assets dir: ${learnAssetsDir}`);
+        console.log(`📁 Absolute path: ${absoluteImagePath}`);
+        console.log(`📁 Path exists: ${fs.existsSync(absoluteImagePath)}`);
+
+        if (fs.existsSync(absoluteImagePath)) {
+          // Use absolute path for the image
+          console.log(
+            `✅ Found image, replacing path: ${imagePath} -> ${absoluteImagePath}`
+          );
+          markdownContent = markdownContent.replace(
+            fullMatch,
+            `![${altText}](${absoluteImagePath})`
+          );
+        } else {
+          console.log(`❌ Image not found: ${absoluteImagePath}`);
+          // List files in the assets directory for debugging
+          if (fs.existsSync(learnAssetsDir)) {
+            const files = fs.readdirSync(learnAssetsDir);
+            console.log(`📁 Files in assets directory: ${files.join(", ")}`);
+          } else {
+            console.log(
+              `❌ Assets directory does not exist: ${learnAssetsDir}`
+            );
+          }
+        }
+      }
+    }
+  }
+
+  markdownContent = replaceQuestionsFromMarkdown(markdownContent, language);
+  markdownContent = replaceMermaidFromMarkdown(markdownContent);
+  markdownContent = cleanQuizOptions(markdownContent);
+
+  return markdownContent;
+}
+
+const titlelify = (title: string) => {
+  return title.replace(/[-_]/g, " ").replace(/\b\w/g, (l) => l.toUpperCase());
+};
+
+// Read sidebar.json to get exercise titles
+function readSidebarData(learnDir: string): any {
+  try {
+    const sidebarPath = path.join(learnDir, "sidebar.json");
+    if (fs.existsSync(sidebarPath)) {
+      const sidebarContent = fs.readFileSync(sidebarPath, "utf-8");
+      return JSON.parse(sidebarContent);
+    }
+  } catch (error) {
+    console.log("⚠️ Could not read sidebar.json:", error);
+  }
+  return null;
+}
+
+export async function exportToEpub(
+  options: ExportOptions,
+  epubMetadata: EpubMetadata
+): Promise<string> {
+  const { courseSlug, bucket, outDir, language = "en" } = options;
+
+  console.log("🚀 Starting EPUB export for course:", courseSlug);
+  console.log("📁 Output directory:", outDir);
+  console.log("🌍 Language:", language);
+  console.log("📋 EPUB Metadata:", epubMetadata);
+
+  // 1. Create temporary folder
+  const tmpName = uuidv4();
+  const epubOutDir = path.join(outDir, tmpName);
+  rimraf.sync(epubOutDir);
+  mkdirp.sync(epubOutDir);
+
+  console.log("Created temporary directory:", epubOutDir);
+
+  // 2. Create EPUB template directory
+  const epubTemplateDir = path.join(epubOutDir, "template");
+  mkdirp.sync(epubTemplateDir);
+
+  // 3. Download exercises, .learn, and cover image from bucket
+  console.log("📥 Downloading exercises...");
+  await downloadS3Folder(
+    bucket,
+    `courses/${courseSlug}/exercises/`,
+    path.join(epubOutDir, "exercises")
+  );
+  console.log("✅ Exercises downloaded successfully");
+
+  console.log("📥 Downloading .learn files...");
+  await downloadS3Folder(
+    bucket,
+    `courses/${courseSlug}/.learn/`,
+    path.join(epubOutDir, "learn")
+  );
+  console.log("✅ .learn files downloaded successfully");
+
+  console.log("📥 Downloading cover image...");
+  const coverImagePath = await downloadCoverImage(
+    bucket,
+    courseSlug,
+    epubOutDir
+  );
+  console.log(
+    "✅ Cover image download completed:",
+    coverImagePath ? "Found" : "Not found"
+  );
+
+  // 4. Read learn.json for course info
+  console.log("Reading course metadata...");
+  const learnJson = await getCourseMetadata(bucket, courseSlug);
+
+  const metadata: CourseMetadata = {
+    title: learnJson.title?.[epubMetadata.lang] || courseSlug,
+    description:
+      learnJson.description?.[epubMetadata.lang] || learnJson.description,
+    language: epubMetadata.lang,
+    technologies: learnJson.technologies || [],
+    difficulty: learnJson.difficulty || "beginner",
+  };
+
+  console.log("Course metadata:", metadata);
+
+  // 5. Process exercises and prepare markdown files
+  console.log("🔄 Processing exercises and preparing markdown files...");
+
+  // First, let's see what files were downloaded
+  const exercisesPath = path.join(epubOutDir, "exercises");
+  console.log("📂 Exercises directory contents:");
+  if (fs.existsSync(exercisesPath)) {
+    const exerciseDirs = fs.readdirSync(exercisesPath, { withFileTypes: true });
+    console.log(`Found ${exerciseDirs.length} exercise directories:`);
+    exerciseDirs.forEach((dir) => {
+      if (dir.isDirectory()) {
+        const dirPath = path.join(exercisesPath, dir.name);
+        const files = fs.readdirSync(dirPath);
+        console.log(
+          `  📁 ${dir.name}: ${files.length} files - ${files.join(", ")}`
+        );
+      }
+    });
+  } else {
+    console.log("❌ Exercises directory does not exist!");
+  }
+
+  const processedMarkdownFiles = await processExercises(
+    path.join(epubOutDir, "exercises"),
+    path.join(epubOutDir, "processed"),
+    language,
+    path.join(epubOutDir, "learn", "assets")
+  );
+
+  // 6. Read sidebar data and create table of contents
+  const sidebarData = readSidebarData(path.join(epubOutDir, "learn"));
+  console.log("📋 Sidebar data loaded:", sidebarData ? "Success" : "Not found");
+
+  // 7. Create main content file as markdown
+  const mainContent = `# ${metadata.title}
+
+${metadata.description || ""}
+
+**Technologies:** ${metadata.technologies?.join(", ") || "Programming"}
+
+**Difficulty:** ${metadata.difficulty}
+
+**Language:** ${language}
+
+---
+
+## Table of Contents
+
+${processedMarkdownFiles
+  .map((file, index) => {
+    const fileName = path.basename(file, ".processed.md");
+    const parentDir = path.basename(path.dirname(file));
+
+    // Get the title from sidebar.json using the parent directory name
+    let displayName = fileName;
+    if (sidebarData && sidebarData[parentDir]) {
+      displayName =
+        sidebarData[parentDir][language] ||
+        sidebarData[parentDir]["en"] ||
+        titlelify(parentDir);
+    } else {
+      displayName = titlelify(parentDir);
+    }
+
+    return `${index + 1}. ${displayName}`;
+  })
+  .join("\n")}
+
+---
+
+`;
+
+  const mainContentPath = path.join(epubOutDir, "processed", "main.md");
+  mkdirp.sync(path.dirname(mainContentPath));
+  fs.writeFileSync(mainContentPath, mainContent, "utf-8");
+
+  // 8. Prepare all markdown files for EPUB generation
+  const allMarkdownFiles = [mainContentPath, ...processedMarkdownFiles];
+
+  // 9. Generate EPUB using pandoc
+  console.log("📖 Generating EPUB file...");
+  const epubPath = path.join(epubOutDir, `${courseSlug}.epub`);
+  await generateEpub(
+    allMarkdownFiles,
+    epubPath,
+    metadata,
+    epubMetadata,
+    coverImagePath
+  );
+
+  console.log("EPUB generated successfully:", epubPath);
+
+  // Clean up temporary files
+  console.log("Cleaning up temporary files...");
+  rimraf.sync(path.join(epubOutDir, "processed"));
+  rimraf.sync(path.join(epubOutDir, "exercises"));
+  rimraf.sync(path.join(epubOutDir, "learn"));
+
+  console.log("EPUB export completed successfully");
+  return epubPath;
+}