feat(pdfengines): add PDF/UA (#714)

This commit is contained in:
Julien Neuhart
2023-11-05 20:14:36 +01:00
committed by GitHub
parent 9c3dfc78df
commit 143b7ce678
39 changed files with 1300 additions and 1013 deletions

View File

@@ -1,3 +1,8 @@
// Package pdftk provides a module which abstracts the CLI tool PDFtk and
// implements the gotenberg.PDFEngine interface.
// Package pdftk provides an implementation of the gotenberg.PdfEngine
// interface using the PDFtk command-line tool. This package allows for the
// merging of PDF files but does not support conversion to specific PDF
// formats. The path to the PDFtk binary must be specified using the
// PDFTK_BIN_PATH environment variable.
//
// See: https://gitlab.com/pdftk-java/pdftk.
package pdftk

View File

@@ -13,26 +13,25 @@ import (
)
func init() {
gotenberg.MustRegisterModule(PDFtk{})
gotenberg.MustRegisterModule(new(PdfTk))
}
// PDFtk abstracts the CLI tool PDFtk and implements the gotenberg.PDFEngine
// PdfTk abstracts the CLI tool PDFtk and implements the [gotenberg.PdfEngine]
// interface.
type PDFtk struct {
type PdfTk struct {
binPath string
}
// Descriptor returns a PDFtk's module descriptor.
func (PDFtk) Descriptor() gotenberg.ModuleDescriptor {
// Descriptor returns a [PdfTk]'s module descriptor.
func (engine *PdfTk) Descriptor() gotenberg.ModuleDescriptor {
return gotenberg.ModuleDescriptor{
ID: "pdftk",
New: func() gotenberg.Module { return new(PDFtk) },
New: func() gotenberg.Module { return new(PdfTk) },
}
}
// Provision sets the modules properties. It returns an error if the
// environment variable PDFTK_BIN_PATH is not set.
func (engine *PDFtk) Provision(_ *gotenberg.Context) error {
// Provision sets the modules properties.
func (engine *PdfTk) Provision(ctx *gotenberg.Context) error {
binPath, ok := os.LookupEnv("PDFTK_BIN_PATH")
if !ok {
return errors.New("PDFTK_BIN_PATH environment variable is not set")
@@ -44,21 +43,22 @@ func (engine *PDFtk) Provision(_ *gotenberg.Context) error {
}
// Validate validates the module properties.
func (engine PDFtk) Validate() error {
func (engine *PdfTk) Validate() error {
_, err := os.Stat(engine.binPath)
if os.IsNotExist(err) {
return fmt.Errorf("PDFtk binary path does not exist: %w", err)
return fmt.Errorf("PdfTk binary path does not exist: %w", err)
}
return nil
}
// Metrics returns the metrics.
func (engine PDFtk) Metrics() ([]gotenberg.Metric, error) {
func (engine *PdfTk) Metrics() ([]gotenberg.Metric, error) {
// TODO: remove deprecated.
return []gotenberg.Metric{
{
Name: "pdftk_active_instances_count",
Description: "Current number of active PDFtk instances.",
Description: "Current number of active PDFtk instances - deprecated.",
Read: func() float64 {
activeInstancesCountMu.RLock()
defer activeInstancesCountMu.RUnlock()
@@ -69,8 +69,8 @@ func (engine PDFtk) Metrics() ([]gotenberg.Metric, error) {
}, nil
}
// Merge merges the given PDFs into a unique PDF.
func (engine PDFtk) Merge(ctx context.Context, logger *zap.Logger, inputPaths []string, outputPath string) error {
// Merge combines multiple PDFs into a single PDF.
func (engine *PdfTk) Merge(ctx context.Context, logger *zap.Logger, inputPaths []string, outputPath string) error {
var args []string
args = append(args, inputPaths...)
args = append(args, "cat", "output", outputPath)
@@ -97,9 +97,9 @@ func (engine PDFtk) Merge(ctx context.Context, logger *zap.Logger, inputPaths []
return fmt.Errorf("merge PDFs with PDFtk: %w", err)
}
// Convert is not available for this PDF engine.
func (engine PDFtk) Convert(_ context.Context, _ *zap.Logger, format, _, _ string) error {
return fmt.Errorf("convert PDF to '%s' with PDFtk: %w", format, gotenberg.ErrPDFEngineMethodNotAvailable)
// Convert is not available in this implementation.
func (engine *PdfTk) Convert(ctx context.Context, logger *zap.Logger, formats gotenberg.PdfFormats, inputPath, outputPath string) error {
return fmt.Errorf("convert PDF to '%+v' with PDFtk: %w", formats, gotenberg.ErrPdfEngineMethodNotSupported)
}
var (
@@ -109,9 +109,9 @@ var (
// Interface guards.
var (
_ gotenberg.Module = (*PDFtk)(nil)
_ gotenberg.Provisioner = (*PDFtk)(nil)
_ gotenberg.Validator = (*PDFtk)(nil)
_ gotenberg.MetricsProvider = (*PDFtk)(nil)
_ gotenberg.PDFEngine = (*PDFtk)(nil)
_ gotenberg.Module = (*PdfTk)(nil)
_ gotenberg.Provisioner = (*PdfTk)(nil)
_ gotenberg.Validator = (*PdfTk)(nil)
_ gotenberg.MetricsProvider = (*PdfTk)(nil)
_ gotenberg.PdfEngine = (*PdfTk)(nil)
)

View File

@@ -12,59 +12,67 @@ import (
"github.com/gotenberg/gotenberg/v7/pkg/gotenberg"
)
func TestPDFtk_Descriptor(t *testing.T) {
descriptor := PDFtk{}.Descriptor()
func TestPdfTk_Descriptor(t *testing.T) {
descriptor := new(PdfTk).Descriptor()
actual := reflect.TypeOf(descriptor.New())
expect := reflect.TypeOf(new(PDFtk))
expect := reflect.TypeOf(new(PdfTk))
if actual != expect {
t.Errorf("expected '%s' but got '%s'", expect, actual)
}
}
func TestPDFtk_Provision(t *testing.T) {
mod := new(PDFtk)
func TestPdfTk_Provision(t *testing.T) {
engine := new(PdfTk)
ctx := gotenberg.NewContext(gotenberg.ParsedFlags{}, nil)
err := mod.Provision(ctx)
err := engine.Provision(ctx)
if err != nil {
t.Errorf("expected no error but got: %v", err)
}
}
func TestPDFtk_Validate(t *testing.T) {
for i, tc := range []struct {
binPath string
expectErr bool
func TestPdfTk_Validate(t *testing.T) {
for _, tc := range []struct {
scenario string
binPath string
expectError bool
}{
{
expectErr: true,
scenario: "empty bin path",
binPath: "",
expectError: true,
},
{
binPath: "/foo",
expectErr: true,
scenario: "bin path does not exist",
binPath: "/foo",
expectError: true,
},
{
binPath: os.Getenv("PDFTK_BIN_PATH"),
scenario: "validate success",
binPath: os.Getenv("PDFTK_BIN_PATH"),
expectError: false,
},
} {
mod := new(PDFtk)
mod.binPath = tc.binPath
err := mod.Validate()
t.Run(tc.scenario, func(t *testing.T) {
engine := new(PdfTk)
engine.binPath = tc.binPath
err := engine.Validate()
if tc.expectErr && err == nil {
t.Errorf("test %d: expected error but got: %v", i, err)
}
if !tc.expectError && err != nil {
t.Fatalf("expected no error but got: %v", err)
}
if !tc.expectErr && err != nil {
t.Errorf("test %d: expected no error but got: %v", i, err)
}
if tc.expectError && err == nil {
t.Fatal("expected error but got none")
}
})
}
}
func TestPDFtk_Metrics(t *testing.T) {
metrics, err := new(PDFtk).Metrics()
func TestPdfTk_Metrics(t *testing.T) {
metrics, err := new(PdfTk).Metrics()
if err != nil {
t.Fatalf("expected no error but got: %v", err)
}
@@ -79,76 +87,80 @@ func TestPDFtk_Metrics(t *testing.T) {
}
}
func TestPDFtk_Merge(t *testing.T) {
for i, tc := range []struct {
ctx context.Context
inputPaths []string
expectErr bool
func TestPdfTk_Merge(t *testing.T) {
for _, tc := range []struct {
scenario string
ctx context.Context
inputPaths []string
expectError bool
}{
{
ctx: context.TODO(),
scenario: "invalid context",
ctx: nil,
expectError: true,
},
{
scenario: "invalid input path",
ctx: context.TODO(),
inputPaths: []string{
"foo",
},
expectError: true,
},
{
scenario: "single file success",
ctx: context.TODO(),
inputPaths: []string{
"/tests/test/testdata/pdfengines/sample1.pdf",
},
},
{
ctx: context.TODO(),
scenario: "many files success",
ctx: context.TODO(),
inputPaths: []string{
"/tests/test/testdata/pdfengines/sample1.pdf",
"/tests/test/testdata/pdfengines/sample2.pdf",
},
},
{
ctx: nil,
expectErr: true,
},
{
ctx: context.TODO(),
inputPaths: []string{
"foo",
},
expectErr: true,
},
} {
func() {
mod := new(PDFtk)
err := mod.Provision(nil)
t.Run(tc.scenario, func(t *testing.T) {
engine := new(PdfTk)
err := engine.Provision(nil)
if err != nil {
t.Fatalf("test %d: expected error but got: %v", i, err)
t.Fatalf("expected error but got: %v", err)
}
fs := gotenberg.NewFileSystem()
outputDir, err := fs.MkdirAll()
if err != nil {
t.Fatalf("test %d: expected error but got: %v", i, err)
t.Fatalf("expected error but got: %v", err)
}
defer func() {
err := os.RemoveAll(fs.WorkingDirPath())
err = os.RemoveAll(fs.WorkingDirPath())
if err != nil {
t.Fatalf("test %d: expected no error while cleaning up but got: %v", i, err)
t.Fatalf("expected no error while cleaning up but got: %v", err)
}
}()
err = mod.Merge(tc.ctx, zap.NewNop(), tc.inputPaths, outputDir+"/foo.pdf")
err = engine.Merge(tc.ctx, zap.NewNop(), tc.inputPaths, outputDir+"/foo.pdf")
if tc.expectErr && err == nil {
t.Errorf("test %d: expected error but got: %v", i, err)
if !tc.expectError && err != nil {
t.Fatalf("expected no error but got: %v", err)
}
if !tc.expectErr && err != nil {
t.Errorf("test %d: expected no error but got: %v", i, err)
if tc.expectError && err == nil {
t.Fatal("expected error but got none")
}
}()
})
}
}
func TestPDFtk_Convert(t *testing.T) {
mod := new(PDFtk)
err := mod.Convert(context.TODO(), zap.NewNop(), "", "", "")
func TestPdfTk_Convert(t *testing.T) {
engine := new(PdfTk)
err := engine.Convert(context.TODO(), zap.NewNop(), gotenberg.PdfFormats{}, "", "")
if !errors.Is(err, gotenberg.ErrPDFEngineMethodNotAvailable) {
t.Errorf("expected error %v, but got: %v", gotenberg.ErrPDFEngineMethodNotAvailable, err)
if !errors.Is(err, gotenberg.ErrPdfEngineMethodNotSupported) {
t.Errorf("expected error %v, but got: %v", gotenberg.ErrPdfEngineMethodNotSupported, err)
}
}