Migration to PDFBox 2.0.0


PDFBox 2.0.0 requires at least Java 6


There are some significant changes to the package structure of PDFBox:

  • Jempbox is no longer supported and was removed in favour of Xmpbox
  • the package org.apache.pdfbox.pdmodel.edit was removed. The only class contained PDPageContentStream was moved to the parent package.
  • all examples were moved to the new package “pdfbox-examples”
  • all commandline tools were moved to the new package “pdfbox-tools”
  • all debugger related stuff was moved to the new package “pdfbox-debugger”
  • the new package “debugger-app” provides a standalone pre built binary for the debugger

Dependency Updates

All libraries on which PDFBox depends are updated to their latest stable versions:

  • Bouncy Castle 1.53
  • Apache Commons Logging 1.2

For test support the libraries are updated to

  • JUnit 4.12
  • JAI Image Core 1.3.1
  • JAI JPEG2000 1.3.0
  • Levigo JBIG ImageIO Plugin 1.6.3

For PDFBox Preflight

  • Apache Commons IO 2.4

Breaking Changes to the Library

Deprecated API calls

Most deprecated API calls in PDFBox 1.8.x have been removed for PDFBox 2.0.0

API Changes

The API changes are reflected in the Javadoc for PDFBox 2.0.0. The most notable changes are:

  • getCOSDictionary() is no longer used. Instead getCOSObject now returns the matching COSBase subtype.
  • PDXObjectForm was renamed to PDFormXObject to be more in line with the PDF specification.
  • PDXObjectImage was renamed to PDImageXObject to be more in line with the PDF specification.
  • PDPage.getContents().createInputStream()was simplified to PDPage.getContents().
  • PDPageContentStream was moved to org.apache.pdfbox.pdmodel.

General Behaviour

PDFBox 2.0.0 is now parsing PDF files following the Xref information in the PDF. This is similar to the functionality using PDDocument.loadNonSeq with PDFBox 1.8.x. Users still using PDDocument.load with PDFBox 1.8.x might experience different results when switching to PDFBox 2.0.0.

Font Handling

Font handling now has full Unicode support and supports font subsetting.

TrueType fonts shall now be loaded using


to leverage that.

PDAfmPfbFont has been removed. To load such a font pass the pfb file to PDType1Font. Loading the afm file is no longer required.

PDF Resources Handling

The individual calls to add resources such as PDResources.addFont(PDFont font) and PDResources.addXObject(PDXObject xobject, String prefix) have been replaced with PDResources.add(resource type) where resource type represents the different resource classes such as PDFont, PDAbstractPattern and so on. The add method now supports all the different type of resources available.

Instead of returning a Map like with PDResources.getFonts() or PDResources.getXObjects() in 2.0 an Iterable<COSName> of references shall be retrieved with PDResources.getFontNames() or PDResources.getXObjectNames(). The individual item can be retrieved with PDResources.getFont(COSName fontName) or PDResources.getXObject(COSName xObjectName).

Working with Images

The individual classes PDJpeg(), PDPixelMap() and PDCCitt() to import images have been replaced with PDImageXObject.createFromFile which works for JPG, TIFF (only G4 compression), PNG, BMP and GIF.

In addition there are some specialized classes:

  • JPEGFactory.createFromStream which preserve the JPEG data and embed it in the PDF file without modification. (This is best if you have a JPEG file).
  • CCITTFactory.createFromFile (for bitonal TIFF images with G4 compression).
  • LosslessFactory.createFromImage (this is best if you start with a BufferedImage).

Parsing the Page Content

Getting the content for a page has been simplified.

Prior to PDFBox 2.0 parsing the page content was done using

PDStream contents = page.getContents();
PDFStreamParser parser = new PDFStreamParser(contents.getStream());
List<Object> tokens = parser.getTokens();

With PDFBox 2.0 the code is reduced to

PDFStreamParser parser = new PDFStreamParser(page);
List<Object> tokens = parser.getTokens();

In addition this also works if the page content is defined as an array of content streams.

Iterating Pages

With PDFBox 2.0.0 the prefered way to iterate through the pages of a document is

for(PDPage page : document.getPages())
    ... (do something)

PDF Rendering

With PDFBox 2.0.0 PDPage.convertToImage and PDFImageWriter have been removed. Instead the new PDFRenderer class shall be used.

PDDocument document = PDDocument.load(new File(pdfFilename));
PDFRenderer pdfRenderer = new PDFRenderer(document);
int pageCounter = 0;
for (PDPage page : document.getPages())
    // note that the page number parameter is zero based
    BufferedImage bim = pdfRenderer.renderImageWithDPI(pageCounter, 300, ImageType.RGB);

    // suffix in filename will be used as the file format
    ImageIOUtil.writeImage(bim, pdfFilename + "-" + (pageCounter++) + ".png", 300);

ImageIOUtil has been moved into the org.apache.pdfbox.tools.imageio package. This is in the pdfbox-tools download. If you are using maven, the artifactId has the same name.

Important notice when using PDFBox with Java 8

Due to the change of the java color management module towards “LittleCMS”, users can experience slow performance in color operations. Solution: disable LittleCMS in favour of the old KCMS (Kodak Color Management System):

  • start with -Dsun.java2d.cmm=sun.java2d.cmm.kcms.KcmsServiceProvideror call
  • System.setProperty("sun.java2d.cmm", "sun.java2d.cmm.kcms.KcmsServiceProvider");


Since PDFBox 2.0.4

PDFBox 2.0.4 introduced a new command line setting


which may improve the performance of rendering PDFs on some systems especially if there are a lot of images on a page.

PDF Printing

With PDFBox 2.0.0 PDFPrinter has been removed.

Users of PDFPrinter.silentPrint() should now use this code:

PrinterJob job = PrinterJob.getPrinterJob();
job.setPageable(new PDFPageable(document));

While users of PDFPrinter.print() should now use this code:

PrinterJob job = PrinterJob.getPrinterJob();
job.setPageable(new PDFPageable(document));
if (job.printDialog()) {

Advanced use case examples can be found in th examples package under org/apache/pdfbox/examples/printing/Printing.java

Text Extraction

In 1.8, to get the text colors, one method was to pass an expanded .properties file to the PDFStripper constructor. To achieve the same in PDFBox 2.0 you can extend PDFTextStripperand add the following Operators to the constructor:

addOperator(new SetStrokingColorSpace());
addOperator(new SetNonStrokingColorSpace());
addOperator(new SetStrokingDeviceCMYKColor());
addOperator(new SetNonStrokingDeviceCMYKColor());
addOperator(new SetNonStrokingDeviceRGBColor());
addOperator(new SetStrokingDeviceRGBColor());
addOperator(new SetNonStrokingDeviceGrayColor());
addOperator(new SetStrokingDeviceGrayColor());
addOperator(new SetStrokingColor());
addOperator(new SetStrokingColorN());
addOperator(new SetNonStrokingColor());
addOperator(new SetNonStrokingColorN());

Interactive Forms

Large parts of the support for interactive forms (AcroForms) have been rewritten. The most notable change from 1.8.x is that there is a clear distinction between fields and the annotations representing them visually. Intermediate nodes in a field tree are now represented by the PDNonTerminalField class.

With PDFBox 2.0.0 the prefered way to iterate through the fields is now

PDAcroForm form;
for (PDField field : form.getFieldTree())
    ... (do something)

Most PDField subclasses now accept Java generic types such as String as parameters instead of the former COSBase subclasses.

PDField.getWidget() removed

As form fields do support multiple annotations PDField.getWidget() has been removed in favour of PDField.getWidgets()which returns all annotations associated with a field.

PDUnknownField removed

The PDUnknownField class has been removed, such fields are treated as null see PDFBOX-2885.

Document Outline

The method PDOutlineNode.appendChild() has been renamed to PDOutlineNode.addLast(). There is now also a complementary method PDOutlineNode.addFirst().

Why was the ReplaceText example removed?

The ReplaceText example has been removed as it gave the incorrect illusion that text can be replaced easily. Words are often split, as seen by this excerpt of a content stream:

[ (Do) -29 (c) -1 (umen) 30 (tation) ] TJ

Other problems will appear with font subsets: for example, if only the glyphs for a, b and c are used, these would be encoded as hex 0, 1 and 2, so you won’t find “abc”. Additionally, you can’t replace “c” with “d” because it isn’t part of the subset.

You could also have problems with ligatures, e.g. “ff”, “fl”, “fi”, “ffi”, “ffl”, which can be represented by a single code in many fonts. To understand this yourself, view any file with PDFDebugger and have a look at the “Contents” entry of a page.

See also https://stackoverflow.com/questions/35420609/pdfbox-2-0-rc3-find-and-replace-text