Add EPUB and PDF cover extraction with metadata support

- Add extractEPUBCover() to extract embedded covers from EPUB files - Parse OPF manifest for cover-image properties - Support meta name="cover" tags - Fall back to common cover paths (cover.jpg, images/cover.jpg) - Add findSidecarCover() for sidecar cover detection - Check cover.jpg, cover.png, cover.webp - Check folder.jpg, folder.png - Check {basename}.jpg (same name as media file) - Add extractPDFCover() to extract first page images from PDFs - Use pdfcpu API to extract images from page 1 - Save largest image as cover - Update extractPDFMetadata() to use pdfcpu API - Extract Title, Author, Subject, Creator, Producer - Call extractPDFCover for embedded covers - Fall back to sidecar covers - Update extractMetadata() for EPUB to call extractEPUBCover - Try embedded cover first, then sidecar
2026-02-25 20:53:18 -05:00
parent 570048c96c
commit 213e7b9a9d
1 changed files with 406 additions and 8 deletions
@@ -31,6 +31,7 @@ import (
 	"github.com/jackc/pgx/v5"
 	"github.com/jackc/pgx/v5/pgtype"
 	"github.com/nwaples/rardecode"
+	pdfcpuapi "github.com/pdfcpu/pdfcpu/pkg/api"
 )

 // MediaMetadata contains extracted metadata for media files (ebooks, comics, manga)
@@ -555,7 +556,25 @@ func (s *MediaScanner) extractMetadata(path string) (*MediaMetadata, error) {

 	switch ext {
 	case ".epub":
-		return s.extractEPUBMetadata(path)
+		metadata, err := s.extractEPUBMetadata(path)
+		if err != nil {
+			return metadata, err
+		}
+		// Try to extract embedded cover
+		coverPath, err := s.extractEPUBCover(path)
+		if err != nil {
+			fmt.Printf("Warning: failed to extract EPUB cover from %s: %v\n", path, err)
+		} else if coverPath != "" {
+			metadata.CoverPath = coverPath
+		}
+		// If no embedded cover, try sidecar
+		if metadata.CoverPath == "" {
+			sidecarCover := findSidecarCover(path)
+			if sidecarCover != "" {
+				metadata.CoverPath = sidecarCover
+			}
+		}
+		return metadata, nil
 	case ".pdf":
 		return s.extractPDFMetadata(path)
 	default:
@@ -652,14 +671,393 @@ func (s *MediaScanner) extractEPUBMetadata(path string) (*MediaMetadata, error)
 	return metadata, nil
 }

-func (s *MediaScanner) extractPDFMetadata(path string) (*MediaMetadata, error) {
-	// For now, return basic metadata since PDF extraction requires additional libraries
-	// In a future enhancement, you could use libraries like github.com/ledongthuc/pdf
-	filename := strings.TrimSuffix(filepath.Base(path), ".pdf")
+// extractEPUBCover extracts the cover image from an EPUB file.
+// It looks for:
+// 1. An item with properties="cover-image" in the manifest
+// 2. A meta tag with name="cover" pointing to an image
+// 3. Common cover image paths like cover.jpg, cover.jpeg, cover.png
+// Returns the path to the saved cover image, or empty string if no cover found.
+func (s *MediaScanner) extractEPUBCover(epubPath string) (string, error) {
+	// Open the EPUB as a zip file to extract images
+	r, err := zip.OpenReader(epubPath)
+	if err != nil {
+		return "", fmt.Errorf("failed to open EPUB as zip: %v", err)
+	}
+	defer r.Close()

-	return &MediaMetadata{
-		Title: filename,
-	}, nil
+	// Try to find cover image from OPF metadata
+	coverImageName := ""
+
+	// Attempt to read the OPF file to find cover reference
+	// First, find container.xml to locate the OPF
+	var opfPath string
+	for _, f := range r.File {
+		if f.Name == "META-INF/container.xml" {
+			rc, err := f.Open()
+			if err != nil {
+				continue
+			}
+			content, err := io.ReadAll(rc)
+			rc.Close()
+			if err != nil {
+				continue
+			}
+			// Parse container.xml to find OPF path
+			// Simple string search since we just need the path
+			opfStart := bytes.Index(content, []byte("<rootfile "))
+			if opfStart == -1 {
+				continue
+			}
+			opfStartAttr := bytes.Index(content[opfStart:], []byte("full-path="))
+			if opfStartAttr == -1 {
+				continue
+			}
+			opfStartAttr += len("full-path=")
+			quote := content[opfStart+opfStartAttr]
+			opfStartQuote := opfStart + opfStartAttr + 1
+			opfEndQuote := bytes.Index(content[opfStartQuote:], []byte{byte(quote)})
+			if opfEndQuote == -1 {
+				continue
+			}
+			opfPath = string(content[opfStartQuote : opfStartQuote+opfEndQuote])
+			break
+		}
+	}
+
+	if opfPath == "" {
+		// No OPF found, try common cover image paths
+		coverImageName = findCoverImageInZip(r.File)
+	} else {
+		// Read OPF to find cover reference
+		opfContent, err := readFileFromZip(r.File, opfPath)
+		if err != nil {
+			// Fall back to common paths
+			coverImageName = findCoverImageInZip(r.File)
+		} else {
+			coverImageName = findCoverInOPF(opfContent, r.File, opfPath)
+		}
+	}
+
+	if coverImageName == "" {
+		return "", nil // No cover found
+	}
+
+	// Extract the cover image
+	coverImage, err := extractImageFromZip(r.File, coverImageName, opfPath)
+	if err != nil {
+		return "", fmt.Errorf("failed to extract cover image: %v", err)
+	}
+	if len(coverImage) == 0 {
+		return "", nil
+	}
+
+	// Save cover to disk (same pattern as comics: {book_path}.cover.jpg)
+	coverPath := epubPath + ".cover.jpg"
+	if err := os.WriteFile(coverPath, coverImage, 0644); err != nil {
+		return "", fmt.Errorf("failed to write cover file: %v", err)
+	}
+
+	return coverPath, nil
+}
+
+// findCoverImageInZip searches for common cover image filenames in the zip
+func findCoverImageInZip(files []*zip.File) string {
+	coverNames := []string{"cover.jpg", "cover.jpeg", "cover.png", "cover.webp",
+		"Cover.jpg", "Cover.jpeg", "Cover.png", "Cover.webp",
+		"images/cover.jpg", "Images/cover.jpg", "OEBPS/images/cover.jpg"}
+
+	for _, name := range coverNames {
+		for _, f := range files {
+			if strings.ToLower(f.Name) == strings.ToLower(name) {
+				return f.Name
+			}
+		}
+	}
+	return ""
+}
+
+// findCoverInOPF parses OPF content to find cover image reference
+func findCoverInOPF(opfContent []byte, files []*zip.File, opfDir string) string {
+	contentStr := string(opfContent)
+
+	// Look for item with properties="cover-image"
+	coverImageRE := regexp.MustCompile(`<item[^>]*properties="[^"]*cover-image[^"]*"[^>]*id="([^"]+)"`)
+	matches := coverImageRE.FindStringSubmatch(contentStr)
+	if len(matches) > 1 {
+		coverID := matches[1]
+		// Find the href for this ID
+		hrefRE := regexp.MustCompile(fmt.Sprintf(`<item[^>]*id="%s"[^>]*href="([^"]+)"`, coverID))
+		hrefMatches := hrefRE.FindStringSubmatch(contentStr)
+		if len(hrefMatches) > 1 {
+			return resolveOPFPath(opfDir, hrefMatches[1])
+		}
+	}
+
+	// Look for meta name="cover"
+	metaCoverRE := regexp.MustCompile(`<meta[^>]*name="cover"[^>]*content="([^"]+)"`)
+	metaMatches := metaCoverRE.FindStringSubmatch(contentStr)
+	if len(metaMatches) > 1 {
+		coverContent := metaMatches[1]
+		// Could be "image-id" format
+		if strings.HasPrefix(coverContent, "image-") {
+			coverID := strings.TrimPrefix(coverContent, "image-")
+			hrefRE := regexp.MustCompile(fmt.Sprintf(`<item[^>]*id="%s"[^>]*href="([^"]+)"`, coverID))
+			hrefMatches := hrefRE.FindStringSubmatch(contentStr)
+			if len(hrefMatches) > 1 {
+				return resolveOPFPath(opfDir, hrefMatches[1])
+			}
+		}
+	}
+
+	// Fall back to searching common paths
+	return findCoverImageInZip(files)
+}
+
+// resolveOPFPath resolves a relative path against the OPF directory
+func resolveOPFPath(opfDir, href string) string {
+	if opfDir == "" {
+		return href
+	}
+	// Handle ../ in href
+	if strings.HasPrefix(href, "../") {
+		// Simple case: just use the href as-is for now
+		return href
+	}
+	// Join the directory with the href
+	return filepath.Join(filepath.Dir(opfDir), href)
+}
+
+// readFileFromZip reads a file from the zip by name
+func readFileFromZip(files []*zip.File, name string) ([]byte, error) {
+	// Normalize the name for comparison
+	name = filepath.ToSlash(name)
+	for _, f := range files {
+		fName := filepath.ToSlash(f.Name)
+		if fName == name {
+			rc, err := f.Open()
+			if err != nil {
+				return nil, err
+			}
+			defer rc.Close()
+			return io.ReadAll(rc)
+		}
+	}
+	return nil, fmt.Errorf("file not found: %s", name)
+}
+
+// extractImageFromZip extracts an image file and returns its contents
+func extractImageFromZip(files []*zip.File, imagePath, opfDir string) ([]byte, error) {
+	// Try direct match first
+	for _, f := range files {
+		if strings.ToLower(f.Name) == strings.ToLower(imagePath) {
+			rc, err := f.Open()
+			if err != nil {
+				return nil, err
+			}
+			defer rc.Close()
+			return io.ReadAll(rc)
+		}
+	}
+
+	// Try resolved path
+	resolvedPath := resolveOPFPath(opfDir, imagePath)
+	for _, f := range files {
+		if strings.ToLower(f.Name) == strings.ToLower(resolvedPath) {
+			rc, err := f.Open()
+			if err != nil {
+				return nil, err
+			}
+			defer rc.Close()
+			return io.ReadAll(rc)
+		}
+	}
+
+	return nil, fmt.Errorf("image not found: %s", imagePath)
+}
+
+// findSidecarCover looks for cover images in the same directory as the media file.
+// It checks for common cover filename patterns in priority order:
+// 1. cover.jpg, cover.jpeg, cover.png, cover.webp
+// 2. folder.jpg, folder.jpeg, folder.png, folder.webp
+// 3. {basename}.jpg, {basename}.jpeg, etc. (same name as media file)
+// 4. .folder.jpg (hidden file)
+// Returns the full path to the cover file, or empty string if not found.
+func findSidecarCover(mediaPath string) string {
+	dir := filepath.Dir(mediaPath)
+	baseName := strings.TrimSuffix(filepath.Base(mediaPath), filepath.Ext(mediaPath))
+
+	// Priority order for cover filenames
+	coverPatterns := []string{
+		"cover.jpg",
+		"cover.jpeg",
+		"cover.png",
+		"cover.webp",
+		"folder.jpg",
+		"folder.jpeg",
+		"folder.png",
+		"folder.webp",
+		".folder.jpg",
+		".folder.jpeg",
+		".folder.png",
+	}
+
+	// First, check exact match cover/folder names
+	for _, coverName := range coverPatterns {
+		coverPath := filepath.Join(dir, coverName)
+		if _, err := os.Stat(coverPath); err == nil {
+			return coverPath
+		}
+	}
+
+	// Second, check for {basename}.{ext} pattern
+	extensions := []string{".jpg", ".jpeg", ".png", ".webp"}
+	for _, ext := range extensions {
+		coverPath := filepath.Join(dir, baseName+ext)
+		if _, err := os.Stat(coverPath); err == nil {
+			return coverPath
+		}
+		// Also check uppercase extension
+		coverPathUpper := filepath.Join(dir, baseName+strings.ToUpper(ext))
+		if _, err := os.Stat(coverPathUpper); err == nil {
+			return coverPathUpper
+		}
+	}
+
+	return ""
+}
+
+func (s *MediaScanner) extractPDFMetadata(path string) (*MediaMetadata, error) {
+	metadata := &MediaMetadata{}
+
+	// Open PDF file for reading metadata
+	f, err := os.Open(path)
+	if err != nil {
+		fmt.Printf("Warning: failed to open PDF file %s: %v\n", path, err)
+		metadata.Title = strings.TrimSuffix(filepath.Base(path), ".pdf")
+		return metadata, nil
+	}
+	defer f.Close()
+
+	// Use pdfcpu API to read PDF metadata
+	// Configuration: nil = default (lenient mode)
+	pdfInfo, err := pdfcpuapi.PDFInfo(f, filepath.Base(path), nil, nil)
+	if err != nil {
+		fmt.Printf("Warning: failed to read PDF info from %s: %v\n", path, err)
+		// Fall back to filename as title
+		metadata.Title = strings.TrimSuffix(filepath.Base(path), ".pdf")
+		return metadata, nil
+	}
+
+	// Extract title
+	if pdfInfo.Title != "" {
+		metadata.Title = pdfInfo.Title
+	} else {
+		metadata.Title = strings.TrimSuffix(filepath.Base(path), ".pdf")
+	}
+
+	// Extract author
+	if pdfInfo.Author != "" {
+		metadata.Author = pdfInfo.Author
+	}
+
+	// Extract subject (use as description)
+	if pdfInfo.Subject != "" {
+		metadata.Description = pdfInfo.Subject
+	}
+
+	// Extract creator (use as author fallback)
+	if pdfInfo.Creator != "" && metadata.Author == "" {
+		metadata.Author = pdfInfo.Creator
+	}
+
+	// Extract producer (use as publisher)
+	if pdfInfo.Producer != "" {
+		metadata.Publisher = pdfInfo.Producer
+	}
+
+	// Try to extract cover image
+	coverPath, err := s.extractPDFCover(path)
+	if err != nil {
+		fmt.Printf("Warning: failed to extract PDF cover from %s: %v\n", path, err)
+	} else if coverPath != "" {
+		metadata.CoverPath = coverPath
+	}
+
+	// If no embedded cover, try sidecar
+	if metadata.CoverPath == "" {
+		sidecarCover := findSidecarCover(path)
+		if sidecarCover != "" {
+			metadata.CoverPath = sidecarCover
+		}
+	}
+
+	return metadata, nil
+}
+
+// extractPDFCover extracts a cover image from a PDF file.
+// It uses pdfcpu API to extract images from the first page.
+// Returns the path to the saved cover, or empty string if no cover found.
+func (s *MediaScanner) extractPDFCover(pdfPath string) (string, error) {
+	// Create a temporary directory for extracted images
+	tmpDir, err := os.MkdirTemp("", "pdf-cover-")
+	if err != nil {
+		return "", fmt.Errorf("failed to create temp dir: %v", err)
+	}
+	defer os.RemoveAll(tmpDir)
+
+	// Use pdfcpu API to extract images from first page
+	// ExtractImagesFile(inFile, outDir string, selectedPages []string, conf *model.Configuration) error
+	err = pdfcpuapi.ExtractImagesFile(pdfPath, tmpDir, []string{"1"}, nil)
+	if err != nil {
+		// No images found or extraction failed - this is OK, just return empty
+		return "", nil
+	}
+
+	// Check for extracted images in the temp directory
+	entries, err := os.ReadDir(tmpDir)
+	if err != nil || len(entries) == 0 {
+		return "", nil
+	}
+
+	// Find the largest image (likely the cover)
+	var largestImage string
+	var largestSize int64
+
+	for _, entry := range entries {
+		if entry.IsDir() {
+			continue
+		}
+		info, err := entry.Info()
+		if err != nil {
+			continue
+		}
+		// Skip very small files (likely thumbnails or icons)
+		if info.Size() < 1000 {
+			continue
+		}
+		if info.Size() > largestSize {
+			largestImage = filepath.Join(tmpDir, entry.Name())
+			largestSize = info.Size()
+		}
+	}
+
+	if largestImage == "" {
+		return "", nil
+	}
+
+	// Read the image
+	imageData, err := os.ReadFile(largestImage)
+	if err != nil || len(imageData) == 0 {
+		return "", nil
+	}
+
+	// Save cover to disk (same pattern as comics: {pdf_path}.cover.jpg)
+	coverPath := pdfPath + ".cover.jpg"
+	if err := os.WriteFile(coverPath, imageData, 0644); err != nil {
+		return "", fmt.Errorf("failed to write cover file: %v", err)
+	}
+
+	return coverPath, nil
 }

 // ComicInfo represents metadata from ComicInfo.xml