I have a List<object> and this list contains thousands of record. I want to generated pdf using itextsharp. and Pdfptable to generated pdf it is working fine, but but I want only 10 records per page in pdf.
How can I do it?
Another way to set the number of rows per page:
using System.Diagnostics;
using System.IO;
using iTextSharp.text;
using iTextSharp.text.pdf;
namespace RowsCountSample
{
class Program
{
static void Main(string[] args)
{
using (var pdfDoc = new Document(PageSize.A4))
{
var pdfWriter = PdfWriter.GetInstance(pdfDoc, new FileStream("Test.pdf", FileMode.Create));
pdfDoc.Open();
var table1 = new PdfPTable(3);
table1.HeaderRows = 2;
table1.FooterRows = 1;
//header row
var headerCell = new PdfPCell(new Phrase("header"));
headerCell.Colspan = 3;
headerCell.HorizontalAlignment = Element.ALIGN_CENTER;
table1.AddCell(headerCell);
//footer row
var footerCell = new PdfPCell(new Phrase("footer"));
footerCell.Colspan = 3;
footerCell.HorizontalAlignment = Element.ALIGN_CENTER;
table1.AddCell(footerCell);
//adding some rows
for (int i = 0; i < 70; i++)
{
//adds a new row
table1.AddCell(new Phrase("Cell[0], Row[" + i + "]"));
table1.AddCell(new Phrase("Cell[1], Row[" + i + "]"));
table1.AddCell(new Phrase("Cell[2], Row[" + i + "]"));
//sets the number of rows per page
if (i > 0 && table1.Rows.Count % 7 == 0)
{
pdfDoc.Add(table1);
table1.DeleteBodyRows();
pdfDoc.NewPage();
}
}
pdfDoc.Add(table1);
}
//open the final file with adobe reader for instance.
Process.Start("Test.pdf");
}
}
}
In the most recent version of iTextSharp (5.3.3), new functionality was added allowing you to define breakpoints: SetBreakPoints(int[] breakPoints)
If you define an array of multiples of 10, you can use this to get the desired effect.
If you have an older version, you should loop over the list and create a new PdfPTable for every 10 objects. Note that this is the better solution if you want to keep the memory use of your application low.
Related
I added a sample project to reproduce the problem to the GitHub repo, I hope this helps:
Click to go to the Github repo
I need to add text and a number of page to a PDF, this PDF is from an invoice.
I already tried using events (but the examples are for java, and are incomplete, or I think they are), and it's not working, I'm getting the same error, and in that way text cannot be aligned.
As I said, I've trying to achieve this using a table and a canvas, the code works fine if the PDF has only one page.
But with more than one page I get this error:
This exception was originally thrown at this call stack: KernelExtensions.Get<TKey,
TValue>(System.Collections.Generic.IDictionary, TKey)
iText.Kernel.Pdf.PdfDictionary.Get(iText.Kernel.Pdf.PdfName, bool)
iText.Kernel.Pdf.PdfDictionary.GetAsNumber(iText.Kernel.Pdf.PdfName)
iText.Kernel.Pdf.PdfPage.GetRotation()
iText.Kernel.Pdf.Canvas.PdfCanvas.PdfCanvas(iText.Kernel.Pdf.PdfPage)
BoarGiveMeMoar.Invoices.InvoiceToPdf.AddFooter(iText.Layout.Document)
in InvoiceToPdf.cs
BoarGiveMeMoar.Invoices.InvoiceToPdf.FillPdf.AnonymousMethod__4() in InvoiceToPdf.cs
System.Threading.Tasks.Task.InnerInvoke() in Task.cs
System.Threading.Tasks.Task..cctor.AnonymousMethod__274_0(object) in Task.cs
System.Threading.ExecutionContext.RunFromThreadPoolDispatchLoop(System.Threading.Thread,
System.Threading.ExecutionContext, System.Threading.ContextCallback,
object) in ExecutionContext.cs
...
[Call Stack Truncated]
This is my code, but I get the error from above:
private readonly int smallFontSize = 6;
private readonly int stdFontSize = 7;
public void FillPdf(string filename)
{
using var pdfWriter = new PdfWriter(filename);
using var pdfDoc = new PdfDocument(pdfWriter);
using var doc = new Document(pdfDoc, PageSize.LETTER);
doc.SetMargins(12, 12, 36, 12);
AddData(doc);
AddFooter(doc);
doc.Close();
}
private void AddData(Document doc)
{
Table table = new Table(UnitValue.CreatePercentArray(5)).UseAllAvailableWidth();
PdfFont bold = PdfFontFactory.CreateFont(StandardFonts.HELVETICA_BOLD);
for (int i = 0; i < 250; i++)
{
var cell = new Cell(1, 5)
.Add(new Paragraph($"My Favorite animals are boars and hippos")
.SetFontSize(stdFontSize).SetFont(bold));
cell.SetBorder(Border.NO_BORDER);
cell.SetPadding(0);
table.AddCell(cell);
}
doc.Add(table);
}
private void AddFooter(Document doc)
{
if (doc is null)
return;
Table table = new Table(UnitValue.CreatePercentArray(60)).UseAllAvailableWidth();
int numberOfPages = doc.GetPdfDocument().GetNumberOfPages();
for (int i = 1; i <= numberOfPages; i++)
{
PdfPage page = doc.GetPdfDocument().GetPage(i);
PdfCanvas pdfCanvas = new PdfCanvas(page);
Rectangle rectangle = new Rectangle(
0,
0,
page.GetPageSize().GetWidth(),
15);
Canvas canvas = new Canvas(pdfCanvas, doc.GetPdfDocument(), rectangle);
var cell = new Cell(1, 20).SetFontSize(smallFontSize);
cell.SetBorder(Border.NO_BORDER);
cell.SetPadding(0);
table.AddCell(cell);
cell = new Cell(1, 20).Add(new Paragraph("This document is an invoice")
.SetTextAlignment(TextAlignment.CENTER)).SetFontSize(smallFontSize);
cell.SetBorder(Border.NO_BORDER);
cell.SetPadding(0);
table.AddCell(cell);
cell = new Cell(1, 10).SetFontSize(smallFontSize);
cell.SetBorder(Border.NO_BORDER);
cell.SetPadding(0);
table.AddCell(cell);
cell = new Cell(1, 7)
.Add(new Paragraph($"Page {string.Format(CultureInfo.InvariantCulture, "{0:#,0}", i)} of {string.Format(CultureInfo.InvariantCulture, "{0:#,0}", numberOfPages)} ")
.SetTextAlignment(TextAlignment.RIGHT)).SetFontSize(smallFontSize);
cell.SetBorder(Border.NO_BORDER);
cell.SetPadding(0);
table.AddCell(cell);
cell = new Cell(1, 3).SetFontSize(smallFontSize);
cell.SetBorder(Border.NO_BORDER);
cell.SetPadding(0);
table.AddCell(cell);
canvas.Add(table).SetFontSize(smallFontSize);
canvas.Close();
}
}
Example way to call the code:
new InvoiceToPdf()
.FillPdf(#$"{Environment.GetFolderPath(Environment.SpecialFolder.Desktop)}\StackOverflow Invoice Test.pdf");
I get an error on this line that creates a new PdfCanvas
PdfCanvas pdfCanvas = new PdfCanvas(page);
So, anyone of you has a solution?
By default to decrease memory consumption a bit the pages of the PDF document you are creating are flushed as they fill up.
It means that when the content is written e.g. on page 3 of the document, the content of the page 1 is already written to the disk and it's too late to add more content there.
One option is to first create a document and close it, then open it with reader and writer, i.e.: new PdfDocument(PdfReader, PdfWriter), create Document around PdfDocument again and append content to the document as you do now:
public void FillPdf(string filename)
{
{
using var pdfWriter = new PdfWriter(tempFilename);
using var pdfDoc = new PdfDocument(pdfWriter);
using var doc = new Document(pdfDoc, PageSize.LETTER);
doc.SetMargins(12, 12, 36, 12);
AddData(doc);
doc.Close();
}
{
using var pdfWriter = new PdfWriter(filename);
using var pdfReader = new PdfReader(tempFilename);
using var pdfDoc = new PdfDocument(pdfReader, pdfWriter);
using var doc = new Document(pdfDoc, PageSize.LETTER);
doc.SetMargins(12, 12, 36, 12);
AddFooter(doc);
doc.Close();
}
}
Second option is not to flush the content as soon as possible and keep it in memory instead. You can do that by only making a slight modification in your code: pass third parameter to Document constructor
using var doc = new Document(pdfDoc, PageSize.LETTER, false);
Please keep in mind that second option might result in additional empty page being created at the end of your document in some corner cases (e.g. when you add such a large image to your document that it doesn't even fit into a full page), so use it carefully. It should not create any problems with you are dealing with regular simple content though
I want to insert a paragraph (which will word-wrap over several lines) into a PDF document using iTextSharp, but I want to restrict the width of the paragraph to the left half of the page. I see no "width" property for the Paragraph class, but surely there is a way to do this, stimmt?
UPDATE
The supposed answer doesn't work for me because it uses iText (Java) stuff apparently unavailable in iTextSharp (C#). Specifically (to begin with, there might be more):
ct.setSimpleColumn(myText, 60, 750, document.getPageSize().getWidth()
Although there is a "SetSimpleColumn" for *Sharp (uppercased initial 's'), there is no "GetPageSize".
UPDATE 2
I'm beginning to think that what I really need to do may be to create a 'borderless table' as suggested, and as articulated in "BestiTextQuestionsOnStackOverflowFull.pdf"
This is one way to do it - a borderless, single-row table with WidthPercentage set to 50 and Horizontal Alignment sent to Left:
using (var ms = new MemoryStream())
{
using (var doc = new Document(PageSize.A4, 50, 50, 25, 25)) {
//Create a writer that's bound to our PDF abstraction and our stream
using (var writer = PdfWriter.GetInstance(doc, ms))
{
//Open the document for writing
doc.Open();
var courier9RedFont = FontFactory.GetFont("Courier", 9, BaseColor.RED);
var importantNotice = new Paragraph("Sit on a potato pan Otis - if you don't agree that that's the best palindrome ever, I will sic Paladin on you, or at least say, 'All down but nine - set 'em up on the other alley, pard'", courier9RedFont);
importantNotice.Leading = 0;
importantNotice.MultipliedLeading = 0.9F; // reduce the width between lines in the paragraph with these two settings
// Add a single-cell, borderless, left-aligned, half-page, table
PdfPTable table = new PdfPTable(1);
PdfPCell cellImportantNote = new PdfPCell(importantNotice);
cellImportantNote.BorderWidth = PdfPCell.NO_BORDER;
table.WidthPercentage = 50;
table.HorizontalAlignment = Element.ALIGN_LEFT;
table.AddCell(cellImportantNote);
doc.Add(table);
doc.Close();
}
var bytes = ms.ToArray();
String PDFTestOutputFileName = String.Format("iTextSharp_{0}.pdf", DateTime.Now.ToShortTimeString());
PDFTestOutputFileName = PDFTestOutputFileName.Replace(":", "_");
var testFile = Path.Combine(Environment.GetFolderPath(Environment.SpecialFolder.Desktop), PDFTestOutputFileName);
File.WriteAllBytes(testFile, bytes);
MessageBox.Show(String.Format("{0} written", PDFTestOutputFileName));
}
}
I'm trying to add a table to the PDF file I'm generating. I can add stuff "directly" but want to put the various paragraphs or phrases in table cells to get everything to align nicely.
The following code should add three paragraphs "free-form" first, then the same three in a table. However, only the first three display - the table is nowhere to be seen. Here's the code:
try
{
using (var ms = new MemoryStream())
{
using (var doc = new Document(PageSize.A4, 50, 50, 25, 25))
{
using (var writer = PdfWriter.GetInstance(doc, ms))
{
doc.Open();
var titleFont = FontFactory.GetFont(FontFactory.COURIER_BOLD, 11, BaseColor.BLACK);
var docTitle = new Paragraph("UCSC Direct - Direct Payment Form", titleFont);
doc.Add(docTitle);
var subtitleFont = FontFactory.GetFont("Times Roman", 9, BaseColor.BLACK);
var subTitle = new Paragraph("(not to be used for reimbursement of services)", subtitleFont);
doc.Add(subTitle);
var importantNoticeFont = FontFactory.GetFont("Courier", 9, BaseColor.RED);
var importantNotice = new Paragraph("Important: Form must be filled out in Adobe Reader or Acrobat Professional 8.1 or above. To save completed forms, Acrobat Professional is required. For technical and accessibility assistance, contact the Campus Controller's Office.", importantNoticeFont);
importantNotice.Leading = 0;
importantNotice.MultipliedLeading = 0.9F; // reduce the width between lines in the paragraph with these two settings
importantNotice.ExtraParagraphSpace = 4; // ? test this....
doc.Add(importantNotice);
// Add a table
PdfPTable table = new PdfPTable(10); // the arg is the number of columns
// Row 1
PdfPCell cell = new PdfPCell(docTitle);
cell.Colspan = 3;
cell.BorderWidth = 0;
//cell.BorderColor = BaseColor.WHITE; <= this works for me, but if background is something other than white, it wouldn't
cell.HorizontalAlignment = 0; //0=Left, 1=Centre, 2=Right
table.AddCell(cell);
// Row 2
PdfPCell cellCaveat = new PdfPCell(subTitle);
cellCaveat.Colspan = 2;
cellCaveat.BorderWidth = 0;
table.AddCell(cellCaveat);
// Row 3
PdfPCell cellImportantNote = new PdfPCell(importantNotice);
cellImportantNote.Colspan = 5;
cellImportantNote.BorderWidth = 0;
table.AddCell(importantNotice);
doc.Add(table);
doc.Close();
}
var bytes = ms.ToArray();
String PDFTestOutputFileName = String.Format("iTextSharp_{0}.pdf", DateTime.Now.ToShortTimeString());
PDFTestOutputFileName = PDFTestOutputFileName.Replace(":", "_");
var testFile = Path.Combine(Environment.GetFolderPath(Environment.SpecialFolder.Desktop), PDFTestOutputFileName);
File.WriteAllBytes(testFile, bytes);
MessageBox.Show(String.Format("{0} written", PDFTestOutputFileName));
}
}
}
catch (DocumentException dex)
{
throw (dex);
}
catch (IOException ioex)
{
throw (ioex);
}
catch (Exception ex)
{
String exMsg = ex.Message;
MessageBox.Show(String.Format("Boo-boo!: {0}", ex.Message));
}
Why is my borderless table invisible or nonexistant?
UPDATE
Based on Bruno's answer, and applying it to the output that I really need, this works:
using (var ms = new MemoryStream())
{
using (var doc = new Document(PageSize.A4, 50, 50, 25, 25)) {
//Create a writer that's bound to our PDF abstraction and our stream
using (var writer = PdfWriter.GetInstance(doc, ms))
{
//Open the document for writing
doc.Open();
// Mimic the appearance of Direct_Payment.pdf
var courierBold11Font = FontFactory.GetFont(FontFactory.COURIER_BOLD, 11, BaseColor.BLACK);
var docTitle = new Paragraph("UCSC - Direct Payment Form", courierBold11Font);
doc.Add(docTitle);
var timesRoman9Font = FontFactory.GetFont("Times Roman", 9, BaseColor.BLACK);
var subTitle = new Paragraph("(not to be used for reimbursement of services)", timesRoman9Font);
doc.Add(subTitle);
var courier9RedFont = FontFactory.GetFont("Courier", 9, BaseColor.RED);
var importantNotice = new Paragraph("Important: Form must be filled out in Adobe Reader or Acrobat Professional 8.1 or above. To save completed forms, Acrobat Professional is required. For technical and accessibility assistance, contact the Campus Controller's Office.", courier9RedFont);
importantNotice.Leading = 0;
importantNotice.MultipliedLeading = 0.9F; // reduce the width between lines in the paragraph with these two settings
// Add a table
PdfPTable table = new PdfPTable(1);
PdfPCell cellImportantNote = new PdfPCell(importantNotice);
cellImportantNote.BorderWidth = PdfPCell.NO_BORDER;
table.WidthPercentage = 50;
table.HorizontalAlignment = Element.ALIGN_LEFT;
table.AddCell(cellImportantNote);
doc.Add(table);
doc.Close();
}
var bytes = ms.ToArray();
String PDFTestOutputFileName = String.Format("iTextSharp_{0}.pdf", DateTime.Now.ToShortTimeString());
PDFTestOutputFileName = PDFTestOutputFileName.Replace(":", "_");
var testFile = Path.Combine(Environment.GetFolderPath(Environment.SpecialFolder.Desktop), PDFTestOutputFileName);
File.WriteAllBytes(testFile, bytes);
MessageBox.Show(String.Format("{0} written", PDFTestOutputFileName));
}
}
Please take a look at the SimpleTable7 example and compare the Java code with your [?] code:
This part is relevant:
PdfPCell cellImportantNote = new PdfPCell(importantNotice);
cellImportantNote.setColspan(5);
cellImportantNote.setBorder(PdfPCell.NO_BORDER);
table.addCell(cellImportantNote);
If I would convert your [?] code to Java, you'd have:
PdfPCell cellImportantNote = new PdfPCell(importantNotice);
cellImportantNote.setColspan(5);
cellImportantNote.setBorder(PdfPCell.NO_BORDER);
table.addCell(importantNotice);
Do you see the difference? You create a cellImportantNote instance, but you aren't using it anywhere. Instead of adding cellImportantNote to the table, you add the importantNotice paragraph.
This means that you are creating a table with 10 columns that has a single row of which only 6 cells are taken (because you have 1 cell with colspan 3, 1 cell with colspan 2 and 1 cell with colspan 1).
By default, iText doesn't render any rows that aren't complete, and since your table doesn't have any complete row, the table isn't being rendered.
If you look at the resulting PDF, simple_table7.pdf, you'll notice that I also added a second table. This table has only three columns, but it looks identical to the table you constructed. Instead of improper use of the colspan functionality, I defined relative widths for the three columns:
table = new PdfPTable(3);
table.setWidths(new int[]{3, 2, 5});
cell.setColspan(1);
table.addCell(cell);
cellCaveat.setColspan(1);
table.addCell(cellCaveat);
cellImportantNote.setColspan(1);
table.addCell(cellImportantNote);
document.add(table);
Note that I also avoid to use meaningless numbers. For instance, instead of:
cell.setHorizontalAlignment(0); // 0 means left
I use:
cell.setHorizontalAlignment(Element.ALIGN_LEFT);
This way, people can read what this line means without having to depend on comments.
Furtermore, I replaced:
cell.setBorderWidth(0);
with:
cell.setBorder(PdfPCell.NO_BORDER);
That too, is only a matter of taste, but I think it's important if you want to keep your code maintainable.
I have a class that build the content for my table of contents and that works, fine and dandy.
Here's the code:
public void Build(string center,IDictionary<string,iTextSharp.text.Image> images)
{
iTextSharp.text.Image image = null;
XPathDocument rapportTekst = new XPathDocument(Application.StartupPath + #"..\..\..\RapportTemplates\RapportTekst.xml");
XPathNavigator nav = rapportTekst.CreateNavigator();
XPathNodeIterator iter;
iter = nav.Select("//dokument/brevhoved");
iter.MoveNext();
var logo = iTextSharp.text.Image.GetInstance(iter.Current.GetAttribute("url", ""));
iter = nav.Select("//dokument/gem_som");
iter.MoveNext();
string outputPath = iter.Current.GetAttribute("url", "")+center+".pdf";
iter = nav.Select("//dokument/titel");
iter.MoveNext();
this.titel = center;
Document document = new Document(PageSize.A4, 30, 30, 100, 30);
var outputStream = new FileStream(outputPath, FileMode.Create);
var pdfWriter = PdfWriter.GetInstance(document, outputStream);
pdfWriter.SetLinearPageMode();
var pageEventHandler = new PageEventHandler();
pageEventHandler.ImageHeader = logo;
pdfWriter.PageEvent = pageEventHandler;
DateTime timeOfReport = DateTime.Now.AddMonths(-1);
pageWidth = document.PageSize.Width - (document.LeftMargin + document.RightMargin);
pageHight = document.PageSize.Height - (document.TopMargin + document.BottomMargin);
document.Open();
var title = new Paragraph(titel, titleFont);
title.Alignment = Element.ALIGN_CENTER;
document.Add(title);
List<TableOfContentsEntry> _contentsTable = new List<TableOfContentsEntry>();
nav.MoveToRoot();
iter = nav.Select("//dokument/indhold/*");
Chapter chapter = null;
int chapterCount = 1;
while (iter.MoveNext())
{
_contentsTable.Add(new TableOfContentsEntry("Test", pdfWriter.CurrentPageNumber.ToString()));
XPathNodeIterator innerIter = iter.Current.SelectChildren(XPathNodeType.All);
chapter = new Chapter("test", chapterCount);
while(innerIter.MoveNext())
{
if (innerIter.Current.Name.ToString().ToLower().Equals("billede"))
{
image = images[innerIter.Current.GetAttribute("navn", "")];
image.Alignment = Image.ALIGN_CENTER;
image.ScaleToFit(pageWidth, pageHight);
chapter.Add(image);
}
if (innerIter.Current.Name.ToString().ToLower().Equals("sektion"))
{
string line = "";
var afsnit = new Paragraph();
line += (innerIter.Current.GetAttribute("id", "") + " ");
innerIter.Current.MoveToFirstChild();
line += innerIter.Current.Value;
afsnit.Add(line);
innerIter.Current.MoveToNext();
afsnit.Add(innerIter.Current.Value);
chapter.Add(afsnit);
}
}
chapterCount++;
document.Add(chapter);
}
document = CreateTableOfContents(document, pdfWriter, _contentsTable);
document.Close();
}
I'm then calling the method CreateTableOfContents(), and as such it is doing what it is supposed to do. Here's the code for the method:
public Document CreateTableOfContents(Document _doc, PdfWriter _pdfWriter, List<TableOfContentsEntry> _contentsTable)
{
_doc.NewPage();
_doc.Add(new Paragraph("Table of Contents", FontFactory.GetFont("Arial", 18, Font.BOLD)));
_doc.Add(new Chunk(Environment.NewLine));
PdfPTable _pdfContentsTable = new PdfPTable(2);
foreach (TableOfContentsEntry content in _contentsTable)
{
PdfPCell nameCell = new PdfPCell(_pdfContentsTable);
nameCell.Border = Rectangle.NO_BORDER;
nameCell.Padding = 6f;
nameCell.Phrase = new Phrase(content.Title);
_pdfContentsTable.AddCell(nameCell);
PdfPCell pageCell = new PdfPCell(_pdfContentsTable);
pageCell.Border = Rectangle.NO_BORDER;
pageCell.Padding = 6f;
pageCell.Phrase = new Phrase(content.Page);
_pdfContentsTable.AddCell(pageCell);
}
_doc.Add(_pdfContentsTable);
_doc.Add(new Chunk(Environment.NewLine));
/** Reorder pages so that TOC will will be the second page in the doc
* right after the title page**/
int toc = _pdfWriter.PageNumber - 1;
int total = _pdfWriter.ReorderPages(null);
int[] order = new int[total];
for (int i = 0; i < total; i++)
{
if (i == 0)
{
order[i] = 1;
}
else if (i == 1)
{
order[i] = toc;
}
else
{
order[i] = i;
}
}
_pdfWriter.ReorderPages(order);
return _doc;
}
The problem is however. I want to insert a page break before the table of contents, for the sake of reordering the pages, so that the table of contents is the first page, naturally. But the output of the pdf-file is not right.
Here's a picture of what it looks like:
It seems like the _doc.NewPage() in the CreateTableOfContents() method does not execute correctly. Meaning that the image and the table of contents is still on the same page when the method starts the reordering of pages.
EDIT: To clarify the above, the _doc.NewPage() gets executed, but the blank page is added after the picture and the table of contents.
I've read a couple of places that this could be because one is trying to insert a new page after an already blank page. But this is not the case.
I'll just link to the pdf files aswell, to better illustrate the problem.
The pdf with table of contents: with table of contents
The pdf without table of contents: without table of contents
Thank you in advance for your help :)
I have a pdf document that has form fields that I'm filling out programatically with c#. Depending on three conditions, I need to trim (delete) some of the pages from that document.
Is that possible to do?
for condition 1: I need to keep pages 1-4 but delete pages 5 and 6
for condition 2: I need to keep pages 1-4 but delete 5 and keep 6
for condition 3: I need to keep pages 1-5 but delete 6
Use PdfReader.SelectPages() combined with PdfStamper. The code below uses iTextSharp 5.5.1.
public void SelectPages(string inputPdf, string pageSelection, string outputPdf)
{
using (PdfReader reader = new PdfReader(inputPdf))
{
reader.SelectPages(pageSelection);
using (PdfStamper stamper = new PdfStamper(reader, File.Create(outputPdf)))
{
stamper.Close();
}
}
}
Then you call this method with the correct page selection for each condition.
Condition 1:
SelectPages(inputPdf, "1-4", outputPdf);
Condition 2:
SelectPages(inputPdf, "1-4,6", outputPdf);
or
SelectPages(inputPdf, "1-6,!5", outputPdf);
Condition 3:
SelectPages(inputPdf, "1-5", outputPdf);
Here's the comment from the iTextSharp source code on what makes up a page selection. This is in the SequenceList class which is used to process a page selection:
/**
* This class expands a string into a list of numbers. The main use is to select a
* range of pages.
* <p>
* The general systax is:<br>
* [!][o][odd][e][even]start-end
* <p>
* You can have multiple ranges separated by commas ','. The '!' modifier removes the
* range from what is already selected. The range changes are incremental, that is,
* numbers are added or deleted as the range appears. The start or the end, but not both, can be ommited.
*/
Instead of deleting pages in a document what you actually do is create a new document and only import the pages that you want to keep. Below is a full working WinForms app that does that (targetting iTextSharp 5.1.1.0). The last parameter to the function removePagesFromPdf is an array of pages to keep.
The code below works off of physical files but would be very easy to convert to something based on streams so that you don't have to write to disk if you don't want to.
using System;
using System.ComponentModel;
using System.IO;
using System.Linq;
using System.Windows.Forms;
using iTextSharp.text.pdf;
using iTextSharp.text;
namespace Full_Profile1
{
public partial class Form1 : Form
{
public Form1()
{
InitializeComponent();
}
private void Form1_Load(object sender, EventArgs e)
{
//The files that we are working with
string sourceFolder = Environment.GetFolderPath(Environment.SpecialFolder.Desktop);
string sourceFile = Path.Combine(sourceFolder, "Test.pdf");
string destFile = Path.Combine(sourceFolder, "TestOutput.pdf");
//Remove all pages except 1,2,3,4 and 6
removePagesFromPdf(sourceFile, destFile, 1, 2, 3, 4, 6);
this.Close();
}
public void removePagesFromPdf(String sourceFile, String destinationFile, params int[] pagesToKeep)
{
//Used to pull individual pages from our source
PdfReader r = new PdfReader(sourceFile);
//Create our destination file
using (FileStream fs = new FileStream(destinationFile, FileMode.Create, FileAccess.Write, FileShare.None))
{
using (Document doc = new Document())
{
using (PdfWriter w = PdfWriter.GetInstance(doc, fs))
{
//Open the desitination for writing
doc.Open();
//Loop through each page that we want to keep
foreach (int page in pagesToKeep)
{
//Add a new blank page to destination document
doc.NewPage();
//Extract the given page from our reader and add it directly to the destination PDF
w.DirectContent.AddTemplate(w.GetImportedPage(r, page), 0, 0);
}
//Close our document
doc.Close();
}
}
}
}
}
}
Here is the code I use to copy all but the last page of an existing PDF. Everything is in memory streams. The variable pdfByteArray is a byte[] of the original pdf obtained using ms.ToArray(). pdfByteArray is overwritten with the new PDF.
PdfReader originalPDFReader = new PdfReader(pdfByteArray);
using (MemoryStream msCopy = new MemoryStream())
{
using (Document docCopy = new Document())
{
using (PdfCopy copy = new PdfCopy(docCopy, msCopy))
{
docCopy.Open();
for (int pageNum = 1; pageNum <= originalPDFReader.NumberOfPages - 1; pageNum ++)
{
copy.AddPage(copy.GetImportedPage(originalPDFReader, pageNum ));
}
docCopy.Close();
}
}
pdfByteArray = msCopy.ToArray();
I know it's an old post, Simply I extend the #chris-haas solution to the next level.
Delete the selected pages after that save them into the separate pdf file.
//ms is MemoryStream and fs is FileStream
ms.CopyTo(fs);
Save the Stream to a separate pdf file. 100% working without any error.
pageRange="5"
pageRange="2,15-20"
pageRange="1-5,15-20"
You can pass the pageRange vales like the above-given samples.
private void DeletePagesNew(string pageRange, string SourcePdfPath, string OutputPdfPath, string Password = "")
{
try
{
var pagesToDelete = new List<int>();
if (pageRange.IndexOf(",") != -1)
{
var tmpHold = pageRange.Split(',');
foreach (string nonconseq in tmpHold)
{
if (nonconseq.IndexOf("-") != -1)
{
var rangeHold = nonconseq.Split('-');
for (int i = Convert.ToInt32(rangeHold[0]), loopTo = Convert.ToInt32(rangeHold[1]); i <= loopTo; i++)
pagesToDelete.Add(i);
}
else
{
pagesToDelete.Add(Convert.ToInt32(nonconseq));
}
}
}
else if (pageRange.IndexOf("-") != -1)
{
var rangeHold = pageRange.Split('-');
for (int i = Convert.ToInt32(rangeHold[0]), loopTo1 = Convert.ToInt32(rangeHold[1]); i <= loopTo1; i++)
pagesToDelete.Add(i);
}
else
{
pagesToDelete.Add(Convert.ToInt32(pageRange));
}
var Reader = new PdfReader(SourcePdfPath);
int[] pagesToKeep;
pagesToKeep = Enumerable.Range(1, Reader.NumberOfPages).ToArray();
using (var ms = new MemoryStream())
{
using (var fs = new FileStream(OutputPdfPath, FileMode.Create, FileAccess.Write, FileShare.None))
{
using (var doc = new Document())
{
using (PdfWriter w = PdfWriter.GetInstance(doc, fs))
{
doc.Open();
foreach (int p in pagesToKeep)
{
if (pagesToDelete.FindIndex(s => s == p) != -1)
{
continue;
}
// doc.NewPage()
// w.DirectContent.AddTemplate(w.GetImportedPage(Reader, p), 0, 0)
//
doc.SetPageSize(Reader.GetPageSize(p));
doc.NewPage();
PdfContentByte cb = w.DirectContent;
PdfImportedPage pageImport = w.GetImportedPage(Reader, p);
int rot = Reader.GetPageRotation(p);
if (rot == 90 || rot == 270)
{
cb.AddTemplate(pageImport, 0, -1.0f, 1.0f, 0, 0, Reader.GetPageSizeWithRotation(p).Height);
}
else
{
cb.AddTemplate(pageImport, 1.0f, 0, 0, 1.0f, 0, 0);
}
cb = default;
pageImport = default;
rot = default;
}
ms.CopyTo(fs);
fs.Flush();
doc.Close();
}
}
}
}
pagesToDelete = null;
Reader.Close();
Reader = default;
}
catch (Exception ex)
{
MessageBox.Show(ex.Message);
}
}