feat(libreoffice): add generateTaggedPdf form field

Tags the PDF without declaring PDF/UA conformance. Today tagging is only
available through pdfua, which also writes pdfuaid:part=1 into the XMP
metadata. Mirrors the Chromium routes, where generateTaggedPdf and pdfua
are separate fields. Defaults to false, so existing requests are
unchanged.

Closes #1676

test/integration/scenario/scenario.go

@@ -1795,6 +1795,7 @@ func InitializeScenario(ctx *godog.ScenarioContext) {
 	ctx.Then(`^the (response|webhook request) PDF\(s\) should declare Factur-X XMP with conformance level "([^"]*)"$`, s.thePdfsShouldDeclareFacturXConformanceLevel)
 	ctx.Then(`^the "([^"]*)" PDF should have (\d+) page\(s\)$`, s.thePdfShouldHavePages)
 	ctx.Then(`^the "([^"]*)" PDF (should|should NOT) be set to landscape orientation$`, s.thePdfShouldBeSetToLandscapeOrientation)
+	ctx.Then(`^the "([^"]*)" PDF (should|should NOT) be tagged$`, s.thePdfShouldBeTagged)
 	ctx.Then(`^the "([^"]*)" PDF (should|should NOT) have a document outline$`, s.thePdfShouldHaveADocumentOutline)
 	ctx.Then(`^the "([^"]*)" PDF (should|should NOT) have the following content at page (\d+):$`, s.thePdfShouldHaveTheFollowingContentAtPage)
 	ctx.Then(`^the "([^"]*)" PDF (should|should NOT) have content matching "([^"]*)" at page (\d+)$`, s.thePdfShouldHaveContentMatchingAtPage)

test/integration/scenario/tagged.go

@@ -0,0 +1,44 @@
+package scenario
+
+import (
+	"context"
+	"fmt"
+	"os"
+	"path/filepath"
+	"strings"
+)
+
+func (s *scenario) thePdfShouldBeTagged(ctx context.Context, name, kind string) error {
+	path := filepath.Join(s.workdir, s.resp.Header().Get("Gotenberg-Trace"), name)
+	if _, err := os.Stat(path); err != nil {
+		return fmt.Errorf("stat PDF %q: %w", path, err)
+	}
+
+	cmd := []string{"pdfinfo", filepath.Base(path)}
+	output, err := execCommandInIntegrationToolsContainer(ctx, cmd, path)
+	if err != nil {
+		return fmt.Errorf("exec %q: %w", cmd, err)
+	}
+
+	return checkTagged(output, kind != "should NOT")
+}
+
+// checkTagged reads the "Tagged" line of pdfinfo, which reflects the Marked
+// entry of the document catalog's MarkInfo dictionary.
+func checkTagged(output string, want bool) error {
+	for line := range strings.SplitSeq(output, "\n") {
+		value, found := strings.CutPrefix(strings.TrimSpace(line), "Tagged:")
+		if !found {
+			continue
+		}
+
+		tagged := strings.TrimSpace(value) == "yes"
+		if tagged != want {
+			return fmt.Errorf("expected tagged PDF to be %t, but actual is %t", want, tagged)
+		}
+
+		return nil
+	}
+
+	return fmt.Errorf("invalid pdfinfo output, no Tagged line: %q", output)
+}

test/integration/scenario/tagged_test.go

@@ -0,0 +1,67 @@
+package scenario
+
+import (
+	"strings"
+	"testing"
+)
+
+func TestCheckTagged(t *testing.T) {
+	t.Parallel()
+
+	pdfinfo := func(tagged string) string {
+		return "Producer:        LibreOffice 26.8\nTagged:          " + tagged + "\nPages:           1\n"
+	}
+
+	tests := []struct {
+		name    string
+		output  string
+		want    bool
+		wantErr string
+	}{
+		{
+			name:   "tagged PDF expected and found",
+			output: pdfinfo("yes"),
+			want:   true,
+		},
+		{
+			name:   "untagged PDF expected and found",
+			output: pdfinfo("no"),
+			want:   false,
+		},
+		{
+			name:    "tagged PDF expected but untagged",
+			output:  pdfinfo("no"),
+			want:    true,
+			wantErr: "expected tagged PDF to be true, but actual is false",
+		},
+		{
+			name:    "untagged PDF expected but tagged",
+			output:  pdfinfo("yes"),
+			want:    false,
+			wantErr: "expected tagged PDF to be false, but actual is true",
+		},
+		{
+			name:    "no Tagged line",
+			output:  "Pages:           1\n",
+			want:    true,
+			wantErr: "invalid pdfinfo output, no Tagged line:",
+		},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			t.Parallel()
+
+			err := checkTagged(tc.output, tc.want)
+			if tc.wantErr == "" {
+				if err != nil {
+					t.Fatalf("expected no error, got: %v", err)
+				}
+				return
+			}
+			if err == nil || !strings.Contains(err.Error(), tc.wantErr) {
+				t.Fatalf("expected error containing %q, got: %v", tc.wantErr, err)
+			}
+		})
+	}
+}