Create a PDF from an HTML page in C#

3 June 2015

There are lots of libraries in .NET to create PDFs. We chose because TuesPechkin because it creates PDFs from HTML, and HTML is easy to make and has a flexible layout - so containers expand to fit content on the page.

Add TuesPechkin to your project

Use Nuget to add TuesPechkin and either TuesPechkin.Wkhtmltox.Win64 or 32. Note that adding these unmanaged 64 or 32 bit DLLs mean your application is no longer machine agnostic - you now must deploy to a 64 or 32 bit server. (Ensure you have installed VC runtime on any server you deploy to: https://www.microsoft.com/en-us/download/confirmation.aspx?id=40784)

Make some PDF templates

Add some .html files to your App_Data folder. Use some special string of letters like %% as markers for fields you will replace in code:

<!doctype html>
<html lang='en'>
<head>
    <meta charset='utf-8' />
    <title>%%Title%%</title>
    <meta name='description' content='%%Title%%' />
    <meta name='author' content='ECCR' />
    %%Style%%
</head>
<body>
    %%DraftStamp%%
    %%Header%%
    %%Admission%%
    %%IcdCodes%%
    %%History%%
    %%Investigation%%
    %%Future%%
</body>
</html>

In your stylesheet you can use most modern CSS, such as web fonts. I used tables instead of flexbox and Bootstrap though, the latter didn't seem to convert properly:

    <style>
        @font-face
        {
            font-family: 'Barcode';
            src: url(../Fonts/fre3of9x.ttf);
        }
        @font-face
        {
            font-family: 'RobotoLight';
            src: url(../Fonts/Roboto-Light.ttf);
        }
        .barcode
        {
            font-family: 'Barcode';
            font-size: 36pt;
        }
        html
        {
            font-family: 'RobotoLight';
            font-size: 10pt;
        }
        table
        {
            border-collapse: collapse;
            width: 100%;
        }
        td
        {
            padding: 2px;
        }

Load the root template and replace all placeholders

Open the main html file, e.g. report.html, and then simple do string replacement of all special fields you added. Where a field refers to another html template then load that file and replace its fields, and so on. Don't forget to HtmlEncode() your data so it doesn't break the html tags.

private string GetFuture(PatientVisitConsultation data, string imageFolder, string pdfTemplatesFolder)
{
    var html = File.ReadAllText(pdfTemplatesFolder + "Future.html");
    if (data.Consultation.Administration.FormTypeId == 3)
        html = html.Replace("%%FutureHeader%%", WebUtility.HtmlEncode("POST-MORTEM REQUIREMENTS"));
    else
        html = html.Replace("%%FutureHeader%%", WebUtility.HtmlEncode("FUTURE MANAGEMENT, OUTSTANDING INVESTIGATIONS, AND DESIRED OUTCOME"));
    html = html.Replace("%%FutureText%%", _futureBuilder.GetText(data));
    return html;
}

Save the html to a single file (or two if you want more pages):

File.WriteAllText(pdfFolder + filenameWithoutExtension + "Page2.html", html.ToString());

Warning - don't use opacity

It increased my final PDF file size from 30kb to 350kb.

Convert the html pages to a PDF

Create a singleton instance of TuesPechkin in your IoC container like this:

private static readonly IConverter _pdfConverter =
	new ThreadSafeConverter(
		new RemotingToolset<PdfToolset>(
			new Win64EmbeddedDeployment(
			    new TempFolderDeployment())));

Then convert the html to PDF like this:

private void CreatePdfFromHtmlPage(PatientVisitConsultation data, string pdfFolder, string filenameWithoutExtension)
{
    var document = new HtmlToPdfDocument
    {
        GlobalSettings =
        {
            ProduceOutline = true,
            DocumentTitle = filenameWithoutExtension,
            PaperSize = PaperKind.A4,
            Margins =
            {
                All = 0.1,
                Unit = Unit.Centimeters
            }
        },
        Objects = {}
    };
    if (data.Consultation.Administration.FormTypeId == 2)
    {
        document.Objects.Add(new ObjectSettings { PageUrl = pdfFolder + filenameWithoutExtension + "Page1.html"});
        document.Objects.Add(new ObjectSettings { PageUrl = pdfFolder + filenameWithoutExtension + "Page2.html"});
    }
    else document.Objects.Add(new ObjectSettings { PageUrl = pdfFolder + filenameWithoutExtension + ".html" });
    byte[] result = _pdfConverter.Convert(document);
    File.WriteAllBytes(pdfFolder + filenameWithoutExtension + ".pdf", result);
}

Return the PDF in a web service

If you want to send the PDF to a browser then use this method:

[Route("GetPdfForConsultation")]
[HttpPost]
public IHttpActionResult GetPdfForConsultation(HttpRequestMessage request, long consultationId)
{
    var token = request.Content.ReadAsStringAsync().Result.Substring(6); //"token=243lblahblahblah"
    var hisSignInModel = _tokenManager.ValidateToken(token);
    _userManager.ThrowExceptionIfUserDoesNotHavePermission(token, PermissionName.ViewDischargeSummaries, null);

    var pdfFolder = HostingEnvironment.MapPath(@"~/" + Core.Configuration.DischargePdfFolder);
    var imageFolder = HostingEnvironment.MapPath(@"~/" + Core.Configuration.ImageFolder);
    var pdfTemplatesFolder = HostingEnvironment.MapPath(@"~/" + Core.Configuration.PdfTemplatesFolder);
    _pdfManager.DeleteOldPdfs(pdfFolder);
    var filename = _pdfManager.CreateDischargePdfAndReturnFilename(consultationId, pdfFolder, hisSignInModel, imageFolder, pdfTemplatesFolder);

    //return pdf
    var path = pdfFolder + filename;
    byte[] fileBytes = System.IO.File.ReadAllBytes(path);
    var response = new HttpResponseMessage(HttpStatusCode.OK);
    response.Content = new ByteArrayContent(fileBytes);
    response.Content.Headers.ContentDisposition = new ContentDispositionHeaderValue("inline"); //'attachment' will download instead
    response.Content.Headers.ContentDisposition.FileName = filename;
    response.Content.Headers.ContentType = new MediaTypeHeaderValue("application/pdf");
    return ResponseMessage(response);
}

Richard JE Cooke