Question

I am stuck trying to debug some code that is designed to convert an web page into a PDF document via a string variable. It uses the iTextSharp c# tool (xmlworker) and is a modification of the example code that comes with the sourceforge source code called html2pdf.csproj. This example code converts an existing html file into a PDF file and saves it in the same directory as the file that was converted. I have a string variable containing html formatted text and I am trying to make it able to be returned as a byte array that will be passed to the client side in a web environment for printing purposes. The problem is that I get an “IOException was unhandled by user code” message that states “The document has no pages.” I’m a little unsure what this is suppose to mean, nor how to go about diagnosing the problem. The example code using the file based system works and I have successfully converted a static version of the html string to PDF. Below is the modified code:

       private byte[] createPDF(string html, string filename) {
            MemoryStream msOutput = new MemoryStream();
            string printPDFCSS = Server.MapPath("/content/printPDF.css");
            Document doc = new Document(PageSize.LETTER);
            doc.SetMargins(doc.LeftMargin, doc.RightMargin, 35, 0);
            PdfWriter pdfWriter = PdfWriter.GetInstance(doc, msOutput);

            doc.Open();
            Dictionary<String, String> substFonts = new Dictionary<String, String>();
            substFonts["Arial Unicode MS"] = "Helvetica";
            CssFilesImpl cssFiles = new CssFilesImpl();
            cssFiles.Add(XMLWorkerHelper.GetCSS(new FileStream(printPDFCSS, FileMode.Open)));
            StyleAttrCSSResolver cssResolver = new StyleAttrCSSResolver(cssFiles);
            HtmlPipelineContext hpc = new HtmlPipelineContext(new CssAppliersImpl(new UnembedFontProvider(XMLWorkerFontProvider.DONTLOOKFORFONTS, substFonts)));
            hpc.SetImageProvider(new ImageProvider(filename));
            hpc.SetAcceptUnknown(true).AutoBookmark(true).SetTagFactory(Tags.GetHtmlTagProcessorFactory());
            HtmlPipeline htmlPipeline = new HtmlPipeline(hpc, new PdfWriterPipeline(doc, pdfWriter));
            IPipeline pipeline = new CssResolverPipeline(cssResolver, htmlPipeline);
            XMLWorker worker = new XMLWorker(pipeline, true);
            XMLParser xmlParse = new XMLParser(true, worker);
            xmlParse.Parse(msOutput);
            doc.Close();
            return msOutput.ToArray();
        }

Answer 1

Bellow is the code I finally found that works, there were a number of issues with the code above, but this seems to work:

private byte[] createPDF(string html, string filename){
        MemoryStream msInput = new MemoryStream(ASCIIEncoding.Default.GetBytes(html));
        MemoryStream msOutput = new MemoryStream();
        string printPDFCSS = Server.MapPath("/content/printPDF.css");
        Document doc = new Document(PageSize.LETTER);
        doc.SetMargins(doc.LeftMargin, doc.RightMargin, 35, 0);
        PdfWriter pdfWriter = PdfWriter.GetInstance(doc, msOutput);
        doc.Open();
        Dictionary<String, String> substFonts = new Dictionary<String, String>();
        substFonts["Arial Unicode MS"] = "Helvetica";
        CssFilesImpl cssFiles = new CssFilesImpl();
        cssFiles.Add(XMLWorkerHelper.GetCSS(new FileStream(printPDFCSS, FileMode.Open)));
        StyleAttrCSSResolver cssResolver = new StyleAttrCSSResolver(cssFiles);
        HtmlPipelineContext hpc = new HtmlPipelineContext(new CssAppliersImpl(new UnembedFontProvider(XMLWorkerFontProvider.DONTLOOKFORFONTS, substFonts)));
        hpc.SetImageProvider(new ImageProvider(filename));
        hpc.SetAcceptUnknown(true).AutoBookmark(true).SetTagFactory(Tags.GetHtmlTagProcessorFactory());
        HtmlPipeline htmlPipeline = new HtmlPipeline(hpc, new PdfWriterPipeline(doc, pdfWriter));
        IPipeline pipeline = new CssResolverPipeline(cssResolver, htmlPipeline);
        XMLWorker worker = new XMLWorker(pipeline, true);
        XMLParser xmlParse = new XMLParser(true, worker);
        xmlParse.Parse(msInput);
        doc.Close();
        return msOutput.ToArray();
    }