iTextSharp PDF/A-3 来自 C# 中的 HTML

Question

我一直在尝试使用 iTextSharp 库从 HTML 字符串创建 PDF 文件，它可以创建普通的 PDF，但是当我尝试从相同的 HTML 创建 PDF/A-3 时，它总是返回错误。

消息： 文档没有页面。

Stack_Trace： 在 iTextSharp.text.pdf.PdfPages.WritePageTree() 在 iTextSharp.text.pdf.PdfWriter.Close() 在 iTextSharp.text.pdf.PdfAWriter.Close() 在 iTextSharp.text.pdf.PdfDocument。 Close() 在 iTextSharp.text.Document.Close() 在 Common.PDFHelper.GeneratePDF(String html, String fileName, Boolean isNormalPDF, String detailsJSON)

这是我的代码：

using iTextSharp.text.pdf;
using iTextSharp.tool.xml;
using System;
using System.IO;
using System.Text;
using System.Threading;
using System.Web;

namespace Common
{
    public class PDFHelper
    {
        public void GeneratePDF(string html, string fileName = "", bool isNormalPDF = false, string detailsJSON = "")
        {
            try
            {
                if (string.IsNullOrWhiteSpace(fileName))
                    fileName = Helpers.GenerateRandomString(8);

                fileName += Guid.NewGuid().ToString();

                byte[] bytes;
                using (MemoryStream memoryStream = new MemoryStream())
                {
                    XMLWorkerFontProvider fontProvider = new XMLWorkerFontProvider(Environment.GetFolderPath(Environment.SpecialFolder.Fonts));
                    if (isNormalPDF) //Create normal PDF
                    {
                        Document pdfDoc = new Document(PageSize.A4);
                        PdfWriter writer = PdfWriter.GetInstance(pdfDoc, memoryStream);

                        pdfDoc.Open();

                        var xmlWorker = XMLWorkerHelper.GetInstance();
                        MemoryStream htmlContent = new MemoryStream(Encoding.UTF8.GetBytes(html));
                        xmlWorker.ParseXHtml(writer, pdfDoc, htmlContent, null, Encoding.UTF8, fontProvider);

                        pdfDoc.Close();
                    }
                    else //Create PDF/A-3
                    {
                        Document pdfDoc = new Document(PageSize.A4);
                        PdfAWriter writer = PdfAWriter.GetInstance(pdfDoc, memoryStream, PdfAConformanceLevel.PDF_A_3A);

                        pdfDoc.Open();
                        writer.CreateXmpMetadata();

                        string sRGBCSprofile = HttpContext.Current.Server.MapPath("~/Resources/Color/sRGB_CS_profile.icm");
                        FileStream sRGBCSproFileStream = new FileStream(sRGBCSprofile, FileMode.Open, FileAccess.Read);
                        var sRGBCSproFileByte = new byte[sRGBCSproFileStream.Length];
                        sRGBCSproFileStream.Read(sRGBCSproFileByte, 0, sRGBCSproFileByte.Length);

                        ICC_Profile iccProfile = ICC_Profile.GetInstance(sRGBCSproFileByte);
                        writer.SetOutputIntents("Custom", "", "http://www.color.org", "sRGB IEC61966-2.1", iccProfile);

                        PdfDictionary markInfo = new PdfDictionary(PdfName.MARKINFO);
                        markInfo.Put(PdfName.MARKED, new PdfBoolean("true"));
                        writer.ExtraCatalog.Put(PdfName.MARKINFO, markInfo);

                        var xmlWorker = XMLWorkerHelper.GetInstance();
                        MemoryStream htmlContent = new MemoryStream(Encoding.UTF8.GetBytes(html));
                        xmlWorker.ParseXHtml(writer, pdfDoc, htmlContent, null, Encoding.UTF8, fontProvider);

                        pdfDoc.Close();
                    }

                    bytes = memoryStream.ToArray();
                    memoryStream.Close();
                }

                // Clears all content output from the buffer stream
                HttpContext.Current.Response.Clear();
                HttpContext.Current.Response.ContentType = "application/pdf";
                HttpContext.Current.Response.AddHeader("content-disposition", "attachment;filename=" + fileName + ".pdf");
                HttpContext.Current.Response.Cache.SetCacheability(HttpCacheability.NoCache);
                HttpContext.Current.Response.BinaryWrite(bytes);

                HttpContext.Current.Response.End();
            }
            catch (ThreadAbortException)
            {
            }
            catch (Exception ex)
            {

            }
        }
    }
}

Answer 1

所以在看了 iText7 文档之后

https://kb.itextpdf.com/home/it7kb/ebooks/itext-7-converting-html-to-pdf-with-pdfhtml/chapter-4-creating-reports-using-pdfhtml

我终于想出了如何去做，这是我的代码，仅供需要帮助的人使用

using iText.Kernel.Pdf;
using iText.Kernel.Pdf.Filespec;
using System;
using System.IO;
using System.Threading;
using System.Web;

namespace Common
{
    public class PDFHelper
    {
        public void GeneratePDF(string html, string fileName = "", bool isNormalPDF = false)
        {
            try
            {
                string resourceFolderPath = HttpContext.Current.Server.MapPath("~/Resources");

                int imageIndex = html.IndexOf("data:image/jpeg;base64");
                string imageURL = "";
                //Check if there is an image in the HTML to re-create it and use it in the PDF and will remove it later
                if (imageIndex > -1)
                {
                    try
                    {
                        int imageSourceEndIndex = html.IndexOf("\"", imageIndex);
                        string imageBase64 = html.Substring(imageIndex, imageSourceEndIndex - imageIndex);
                        imageBase64 = imageBase64.Substring(imageBase64.IndexOf(",") + 1);
                        byte[] imageBytes = Convert.FromBase64String(imageBase64);
                        imageURL = HttpContext.Current.Server.MapPath(fileName + ".jpg");
                        File.WriteAllBytes(imageURL, imageBytes);

                        html = html.Remove(imageIndex, imageSourceEndIndex - imageIndex);
                        html = html.Insert(imageIndex, imageURL);
                    }
                    catch (Exception ex)
                    {

                    }
                }

                byte[] bytes;
                iText.Layout.Font.FontProvider fontProvider = new iText.Html2pdf.Resolver.Font.DefaultFontProvider(false, false, false);
                string[] fonts = Directory.GetFiles(resourceFolderPath + @"\Font\");
                foreach (string font in fonts)
                {
                    iText.IO.Font.FontProgram fontProgram = iText.IO.Font.FontProgramFactory.CreateFont(font);
                    fontProvider.AddFont(fontProgram);
                }

                ConverterProperties properties = new ConverterProperties();
                properties.SetCreateAcroForm(true);
                properties.SetFontProvider(fontProvider);
                if (isNormalPDF)
                {
                    using (MemoryStream outputStream = new MemoryStream())
                    {
                        HtmlConverter.ConvertToPdf(html, outputStream, properties);
                        bytes = outputStream.ToArray();
                        outputStream.Close();
                    }
                }
                else
                {
                    using (MemoryStream outputStream = new MemoryStream())
                    {
                        FileStream iccProfile = new FileStream(resourceFolderPath + @"\Color\sRGB_CS_profile.icm", FileMode.Open, FileAccess.Read);

                        PdfWriter writer = new PdfWriter(outputStream);
                        iText.Pdfa.PdfADocument pdf = new iText.Pdfa.PdfADocument(writer, PdfAConformanceLevel.PDF_A_3A, new PdfOutputIntent("Custom", "", "https://www.color.org",
                            "sRGB IEC61966-2.1", iccProfile));
                        pdf.SetTagged();

                        // Embed the XML file
                        string xmlString = "";
                        //Fill xml embedded file content, just the XML

                        if (!string.IsNullOrWhiteSpace(xmlString))
                        {
                            //Create XML file in memory and add it to the PDF
                            using (var ms = new MemoryStream())
                            {
                                using (TextWriter tw = new StreamWriter(ms))
                                {
                                    tw.Write(xmlString);
                                    tw.Flush();
                                    ms.Position = 0;

                                    PdfDictionary parameters = new PdfDictionary();
                                    parameters.Put(PdfName.ModDate, new PdfDate().GetPdfObject());

                                    PdfFileSpec fileSpec = PdfFileSpec.CreateEmbeddedFileSpec(pdf, ms.ToArray(), "invoice.xml", "invoice.xml", PdfName.ApplicationXml, parameters, PdfName.Data);
                                    fileSpec.Put(new PdfName("AFRelationship"), new PdfName("Data"));
                                    pdf.AddFileAttachment("Invoice Xml", fileSpec);
                                }
                            }
                        }

                        HtmlConverter.ConvertToPdf(html, pdf, properties);
                        bytes = outputStream.ToArray();
                    }
                }

                // Clears all content output from the buffer stream
                HttpContext.Current.Response.Clear();
                HttpContext.Current.Response.ContentType = "application/pdf";
                HttpContext.Current.Response.AddHeader("content-disposition", "attachment;filename=" + fileName + ".pdf");
                HttpContext.Current.Response.Cache.SetCacheability(HttpCacheability.NoCache);
                HttpContext.Current.Response.BinaryWrite(bytes);
                // Delete the temp image.
                if (!string.IsNullOrWhiteSpace(imageURL))
                    File.Delete(imageURL);

                HttpContext.Current.Response.End();
            }
            catch (ThreadAbortException)
            {
            }
            catch (Exception ex)
            {

            }
        }
    }
}