feat(ExifTool): add capability to overwrite metadata of the PDF generated (#776)

* add new metadata functions to pdf engine interface

* add exiftool module with relevant test cases

* use exiftool to overwrite metadata in libreoffice

* use exiftool to overwrite metadata in chromium

* fix linter issues

* fix more linter issues

* more test cases for better coverage

* remove utils

* minor changes

* remove metadata from pdfoptions

* correct indentation

* read/write metadata one file at a time.
This commit is contained in:
Piyush Srivastava
2024-02-16 18:46:50 +00:00
committed by Julien Neuhart
parent 31e7582216
commit 71911cb1a7
24 changed files with 1433 additions and 39 deletions

View File

@@ -233,6 +233,23 @@ func FormDataChromiumPdfFormats(form *api.FormData) gotenberg.PdfFormats {
}
}
// FormDataMetadata creates metadata object from the form data.
func FormDataMetadata(form *api.FormData) map[string]interface{} {
var metadata map[string]interface{}
form.Custom("metadata", func(value string) error {
metadata = map[string]interface{}{}
if len(value) > 0 {
err := json.Unmarshal([]byte(value), &metadata)
if err != nil {
return err
}
}
return nil
})
return metadata
}
// convertUrlRoute returns an [api.Route] which can convert a URL to PDF.
func convertUrlRoute(chromium Api, engine gotenberg.PdfEngine) api.Route {
return api.Route{
@@ -243,6 +260,7 @@ func convertUrlRoute(chromium Api, engine gotenberg.PdfEngine) api.Route {
ctx := c.Get("context").(*api.Context)
form, options := FormDataChromiumPdfOptions(ctx)
pdfFormats := FormDataChromiumPdfFormats(form)
metadata := FormDataMetadata(form)
var url string
err := form.
@@ -252,7 +270,7 @@ func convertUrlRoute(chromium Api, engine gotenberg.PdfEngine) api.Route {
return fmt.Errorf("validate form data: %w", err)
}
err = convertUrl(ctx, chromium, engine, url, pdfFormats, options)
err = convertUrl(ctx, chromium, engine, url, pdfFormats, options, metadata)
if err != nil {
return fmt.Errorf("convert URL to PDF: %w", err)
}
@@ -302,6 +320,7 @@ func convertHtmlRoute(chromium Api, engine gotenberg.PdfEngine) api.Route {
ctx := c.Get("context").(*api.Context)
form, options := FormDataChromiumPdfOptions(ctx)
pdfFormats := FormDataChromiumPdfFormats(form)
metadata := FormDataMetadata(form)
var inputPath string
err := form.
@@ -312,7 +331,7 @@ func convertHtmlRoute(chromium Api, engine gotenberg.PdfEngine) api.Route {
}
url := fmt.Sprintf("file://%s", inputPath)
err = convertUrl(ctx, chromium, engine, url, pdfFormats, options)
err = convertUrl(ctx, chromium, engine, url, pdfFormats, options, metadata)
if err != nil {
return fmt.Errorf("convert HTML to PDF: %w", err)
}
@@ -363,6 +382,7 @@ func convertMarkdownRoute(chromium Api, engine gotenberg.PdfEngine) api.Route {
ctx := c.Get("context").(*api.Context)
form, options := FormDataChromiumPdfOptions(ctx)
pdfFormats := FormDataChromiumPdfFormats(form)
metadata := FormDataMetadata(form)
var (
inputPath string
@@ -382,7 +402,7 @@ func convertMarkdownRoute(chromium Api, engine gotenberg.PdfEngine) api.Route {
return fmt.Errorf("transform markdown file(s) to HTML: %w", err)
}
err = convertUrl(ctx, chromium, engine, url, pdfFormats, options)
err = convertUrl(ctx, chromium, engine, url, pdfFormats, options, metadata)
if err != nil {
return fmt.Errorf("convert markdown to PDF: %w", err)
}
@@ -506,7 +526,7 @@ func markdownToHtml(ctx *api.Context, inputPath string, markdownPaths []string)
return fmt.Sprintf("file://%s", inputPath), nil
}
func convertUrl(ctx *api.Context, chromium Api, engine gotenberg.PdfEngine, url string, pdfFormats gotenberg.PdfFormats, options PdfOptions) error {
func convertUrl(ctx *api.Context, chromium Api, engine gotenberg.PdfEngine, url string, pdfFormats gotenberg.PdfFormats, options PdfOptions, metadata map[string]interface{}) error {
outputPath := ctx.GeneratePath(".pdf")
err := chromium.Pdf(ctx, ctx.Log(), url, outputPath, options)
@@ -562,6 +582,14 @@ func convertUrl(ctx *api.Context, chromium Api, engine gotenberg.PdfEngine, url
outputPath = convertOutputPath
}
// Writes and potentially overrides metadata entries, if any.
if len(metadata) > 0 {
err = engine.WriteMetadata(ctx, ctx.Log(), outputPath, metadata)
if err != nil {
return fmt.Errorf("write metadata: %w", err)
}
}
err = ctx.AddOutputPaths(outputPath)
if err != nil {
return fmt.Errorf("add output path: %w", err)

View File

@@ -1225,6 +1225,7 @@ func TestConvertUrl(t *testing.T) {
engine gotenberg.PdfEngine
pdfFormats gotenberg.PdfFormats
options PdfOptions
metadata map[string]interface{}
expectError bool
expectHttpError bool
expectHttpStatus int
@@ -1385,10 +1386,45 @@ func TestConvertUrl(t *testing.T) {
expectHttpError: false,
expectOutputPathsCount: 1,
},
{
scenario: "error with metadata write",
ctx: &api.ContextMock{Context: new(api.Context)},
api: &ApiMock{PdfMock: func(ctx context.Context, logger *zap.Logger, url, outputPath string, options PdfOptions) error {
return nil
}},
engine: &gotenberg.PdfEngineMock{WriteMetadataMock: func(ctx context.Context, logger *zap.Logger, inputPath string, newMetadata map[string]interface{}) error {
return fmt.Errorf("error writing metadata to %s: %w", "foo.pdf", errors.New("foo"))
}},
options: DefaultPdfOptions(),
metadata: map[string]interface{}{
"Creator": "foo",
"Producer": "bar",
},
expectError: true,
expectHttpError: false,
},
{
scenario: "success with metadata write",
ctx: &api.ContextMock{Context: new(api.Context)},
api: &ApiMock{PdfMock: func(ctx context.Context, logger *zap.Logger, url, outputPath string, options PdfOptions) error {
return nil
}},
engine: &gotenberg.PdfEngineMock{WriteMetadataMock: func(ctx context.Context, logger *zap.Logger, inputPath string, newMetadata map[string]interface{}) error {
return nil
}},
options: DefaultPdfOptions(),
metadata: map[string]interface{}{
"Creator": "foo",
"Producer": "bar",
},
expectError: false,
expectHttpError: false,
expectOutputPathsCount: 1,
},
} {
t.Run(tc.scenario, func(t *testing.T) {
tc.ctx.SetLogger(zap.NewNop())
err := convertUrl(tc.ctx.Context, tc.api, tc.engine, "", tc.pdfFormats, tc.options)
err := convertUrl(tc.ctx.Context, tc.api, tc.engine, "", tc.pdfFormats, tc.options, tc.metadata)
if tc.expectError && err == nil {
t.Fatal("expected error but got none", err)