chore(deps): update pdfcpu to v0.15.0 (#1628)

Makefile

@@ -91,7 +91,7 @@ PDFENGINES_WATERMARK_ENGINES=pdfcpu,pdftk
 PDFENGINES_STAMP_ENGINES=pdfcpu,pdftk
 PDFENGINES_ENCRYPT_ENGINES=qpdf,pdfcpu,pdftk
 PDFENGINES_ROTATE_ENGINES=pdfcpu,pdftk
-PDFENGINES_EMBED_ENGINES=qpdf,pdfcpu
+PDFENGINES_EMBED_ENGINES=pdfcpu,qpdf
 PDFENGINES_EMBED_METADATA_ENGINES=qpdf
 PDFENGINES_FACTUR_X_ENGINES=qpdf
 PROMETHEUS_NAMESPACE=gotenberg

build/Dockerfile

@@ -11,7 +11,7 @@ ARG GOLANG_VERSION=1.26.5
 FROM golang:$GOLANG_VERSION AS pdfcpu-binary-stage
 
 # See https://github.com/pdfcpu/pdfcpu/releases.
-ARG PDFCPU_VERSION=v0.13.0
+ARG PDFCPU_VERSION=v0.15.0
 ENV CGO_ENABLED=0
 
 # Define the working directory outside of $GOPATH (we're using go modules).

pkg/modules/exiftool/exiftool.go

@@ -115,18 +115,29 @@ func buildExifToolWriteArgs(metadata map[string]any) ([]string, error) {
 			return nil, fmt.Errorf("write PDF metadata with ExifTool: invalid metadata key %q: %w", key, gotenberg.ErrPdfEngineMetadataValueNotSupported)
 		}
 
+		tag := key
+		if key == "Trapped" {
+			// ExifTool writes the document info /Trapped entry as a malformed
+			// name-in-a-string, e.g. "(/Unknown)". pdfcpu's stricter validation
+			// (as of v0.15) rejects it, which breaks later pdfcpu operations on
+			// the file such as embedding. Writing Trapped to XMP keeps the value
+			// readable without the invalid document info entry.
+			// See https://github.com/gotenberg/gotenberg/issues/1628.
+			tag = "XMP-pdf:Trapped"
+		}
+
 		switch val := value.(type) {
 		case string:
 			if err := validateMetadataValue(key, val); err != nil {
 				return nil, err
 			}
-			args = append(args, fmt.Sprintf("-%s=%s", key, val))
+			args = append(args, fmt.Sprintf("-%s=%s", tag, val))
 		case []string:
 			for _, s := range val {
 				if err := validateMetadataValue(key, s); err != nil {
 					return nil, err
 				}
-				args = append(args, fmt.Sprintf("-%s=%s", key, s))
+				args = append(args, fmt.Sprintf("-%s=%s", tag, s))
 			}
 		case []any:
 			// See https://github.com/gotenberg/gotenberg/issues/1048.
@@ -138,18 +149,18 @@ func buildExifToolWriteArgs(metadata map[string]any) ([]string, error) {
 				if err := validateMetadataValue(key, s); err != nil {
 					return nil, err
 				}
-				args = append(args, fmt.Sprintf("-%s=%s", key, s))
+				args = append(args, fmt.Sprintf("-%s=%s", tag, s))
 			}
 		case bool:
-			args = append(args, fmt.Sprintf("-%s=%t", key, val))
+			args = append(args, fmt.Sprintf("-%s=%t", tag, val))
 		case int:
-			args = append(args, fmt.Sprintf("-%s=%d", key, val))
+			args = append(args, fmt.Sprintf("-%s=%d", tag, val))
 		case int64:
-			args = append(args, fmt.Sprintf("-%s=%d", key, val))
+			args = append(args, fmt.Sprintf("-%s=%d", tag, val))
 		case float32:
-			args = append(args, fmt.Sprintf("-%s=%g", key, val))
+			args = append(args, fmt.Sprintf("-%s=%g", tag, val))
 		case float64:
-			args = append(args, fmt.Sprintf("-%s=%g", key, val))
+			args = append(args, fmt.Sprintf("-%s=%g", tag, val))
 		default:
 			return nil, fmt.Errorf("write PDF metadata with ExifTool: unsupported type %T for key %q: %w", value, key, gotenberg.ErrPdfEngineMetadataValueNotSupported)
 		}

pkg/modules/pdfengines/pdfengines.go

@@ -62,7 +62,7 @@ func (mod *PdfEngines) Descriptor() gotenberg.ModuleDescriptor {
 			fs.StringSlice("pdfengines-read-metadata-engines", []string{"exiftool"}, "Set the PDF engines and their order for the read metadata feature - empty means all")
 			fs.StringSlice("pdfengines-write-metadata-engines", []string{"exiftool"}, "Set the PDF engines and their order for the write metadata feature - empty means all")
 			fs.StringSlice("pdfengines-encrypt-engines", []string{"qpdf", "pdftk", "pdfcpu"}, "Set the PDF engines and their order for the password protection feature - empty means all")
-			fs.StringSlice("pdfengines-embed-engines", []string{"pdfcpu"}, "Set the PDF engines and their order for the file embedding feature - empty means all")
+			fs.StringSlice("pdfengines-embed-engines", []string{"pdfcpu", "qpdf"}, "Set the PDF engines and their order for the file embedding feature - empty means all")
 			fs.StringSlice("pdfengines-embed-metadata-engines", []string{"qpdf"}, "Set the PDF engines and their order for the embed metadata feature - empty means all")
 			fs.StringSlice("pdfengines-read-bookmarks-engines", []string{"pdfcpu"}, "Set the PDF engines and their order for the read bookmarks feature - empty means all")
 			fs.StringSlice("pdfengines-write-bookmarks-engines", []string{"pdfcpu"}, "Set the PDF engines and their order for the write bookmarks feature - empty means all")

pkg/modules/qpdf/qpdf.go

@@ -419,17 +419,57 @@ func (engine *QPdf) Encrypt(ctx context.Context, logger *slog.Logger, inputPath
 }
 
 // EmbedFiles is not available in this implementation.
+// EmbedFiles embeds the given files into a PDF as attachments using qpdf's
+// --add-attachment. It is a fallback for pdfcpu: qpdf reads PDFs whose document
+// info pdfcpu's stricter validation rejects (for example a non-conformant
+// /Trapped value written by ExifTool). See
+// https://github.com/gotenberg/gotenberg/issues/1628.
 func (engine *QPdf) EmbedFiles(ctx context.Context, logger *slog.Logger, filePaths []string, inputPath string) error {
-	_, span := gotenberg.Tracer().Start(ctx, "qpdf.EmbedFiles",
+	ctx, span := gotenberg.Tracer().Start(ctx, "qpdf.EmbedFiles",
 		trace.WithSpanKind(trace.SpanKindClient),
 		trace.WithAttributes(engine.spanAttrs()...),
 	)
 	defer span.End()
 
-	err := fmt.Errorf("embed files with QPDF: %w", gotenberg.ErrPdfEngineMethodNotSupported)
-	span.RecordError(err)
-	span.SetStatus(codes.Error, err.Error())
-	return err
+	if len(filePaths) == 0 {
+		span.SetStatus(codes.Ok, "")
+		return nil
+	}
+
+	fail := func(err error) error {
+		span.RecordError(err)
+		span.SetStatus(codes.Error, err.Error())
+		return err
+	}
+
+	// Each --add-attachment group must be terminated by "--", so qpdf attaches
+	// one file per invocation. Chain them, writing to a temporary file and
+	// moving it back over the input each time.
+	outputPath := inputPath + "~embed"
+	for _, filePath := range filePaths {
+		args := make([]string, 0, 6+len(engine.globalArgs))
+		args = append(args, engine.globalArgs...)
+		args = append(args, "--add-attachment", filePath, "--key="+filepath.Base(filePath))
+		args = append(args, "--", inputPath, outputPath)
+
+		cmd, err := gotenberg.CommandContext(ctx, logger, engine.binPath, args...)
+		if err != nil {
+			return fail(fmt.Errorf("create command: %w", err))
+		}
+
+		_, err = cmd.Exec()
+		if err != nil {
+			return fail(fmt.Errorf("embed files with QPDF: %w", err))
+		}
+
+		err = os.Rename(outputPath, inputPath)
+		if err != nil {
+			return fail(fmt.Errorf("embed files with QPDF: replace input: %w", err))
+		}
+	}
+
+	span.SetStatus(codes.Ok, "")
+	return nil
 }
 
 // EmbedFilesMetadata sets metadata on already-embedded files in a PDF using