Identify Blank PDF generated by go-wkhtmltopdf

[jagdevsingh9709]

Is there any way to identify that the go-wkhtmltopdf library has created/generated a blank pdf? We are using dynamic html page url to create a PDF. The dynamic html page may or may not have content. If it has content then a PDF will be generated which is working fine but if the html content is empty then a blank pdf is being generated(with our headers and footers which we had set in the page instance). We need to check if the generated PDF has blank/empty body content then create a kafka message and send it to retry topic.

// Create a new input page from an URL
src := "<any html page url>"
// Create new PDF generator
var pdfg *wkhtmltopdf.PDFGenerator
pdfg, err := wkhtmltopdf.NewPDFGenerator()
if err != nil {
    logFunc.Error(err)
    pending.abort(job.ID)
    return nil, err
}
//Margins
pdfg.MarginLeft.Set(0)
pdfg.MarginRight.Set(0)

page := wkhtmltopdf.NewPage(src)

// Add headers and footers
if siteConf.HeaderHTML != "" {
    page.HeaderHTML.Set(siteConf.HeaderHTML)
    page.HeaderSpacing.Set(3.0)
}
if siteConf.FooterHTML != "" {
    page.FooterHTML.Set(siteConf.FooterHTML)
    page.FooterSpacing.Set(3.0)
}

// Add to document
pdfg.AddPage(page)

// Create PDF document in internal buffer
err = pdfg.Create()
if err != nil {
    logFunc.Error(err)
    pending.abort(job.ID)
    return nil, err
}

// Write buffer contents to file on disk
logFunc.Info("Writing PDF... ")
err = pdfg.WriteFile("./test.pdf")
if err != nil {
    log.Fatal(err)
}

out.bytes = pdfg.Bytes()
return out, nil

[SebastiaanKlippert]

Do you actually need to know if the PDF has content or do you just want to know if the source URL has content?

You could check the URL first:

srcResp, err := http.Get("https://example.com")
if err != nil {
    logFunc.Error(err)
    pending.abort(job.ID)
    return nil, err
}
defer srcResp.Body.Close()

// you could check the content length (if available)
if srcResp.ContentLength == 0 {
        err := errors.New("no content") 
    logFunc.Error(err)
    pending.abort(job.ID)
    return nil, err
}

srcBody, err := io.ReadAll(srcResp.Body)
if err != nil {
    logFunc.Error(err)
    pending.abort(job.ID)
    return nil, err
}

// or the body
if len(srcBody) == 0 {
        err := errors.New("no body") 
    logFunc.Error(err)
    pending.abort(job.ID)
    return nil, err
}

log.Println(string(srcBody)) //etc

pdfg, err := NewPDFGenerator()
if err != nil {
    log.Fatal(err)
}

// pass the body as reader
pdfg.AddPage(NewPageReader(bytes.NewReader(srcBody)))

err = pdfg.Create()
if err != nil {
    log.Fatal(err)
}

Or if you actually want to check the PDF it is now supported here, but you could use a PDF parser like https://github.com/dslipak/pdf