How do I find which image field in PDF has image inserted and which one has no images attached using PDFbox 1.8.11? - itext

Question

I have a PDF that has image fields inside it. I am not using a PDPushButton with javascript to attach pictures because if I do that the button's top layer gets replaced with the picture that I am attaching which is not what I want. So I am explicitly using a ImageField that is available in Adobe LiveCycle Designer. I am able to extract the files attached on it using PDFBox but I am not able to find any way of seeing which image fields have files attached to them and which ones do not. For example if I have the following code here:
ImageField[1], ImageField[2], ImageField[3]
I want to see something like
ImageField[1]: null,
ImageField[2]: true,
ImageField[3]: trueenter code here
etc assuming ImageField[2] and ImageField[3] has images attached to them.
Below is the code that I was working on:
I have a constant:
Then I am looping through the whole set of image field names and see which field is a instance of PDXObjectImage and then if it is a PDXObjectImage then I check if that object.getRGBImage().getHeight() > 0 assuming that only files uploaded have a height > 1 which means a file has been attached.
private static String[] IMAGE_FIELD_ROW = {"ImageField1[0]","ImageField2[0]",....} => 100 rows of string values such as "ImageField3[0]", "ImageField4[0]", ...etc.
for(int i = 0; i<IMAGE_FIELD_ROW.length; i++)
{
if(field.getPartialName().equals(IMAGE_FIELD_ROW[i]))
{
Map<String, PDAppearanceStream> stateAppearances = field.getWidget().getAppearance().getNormalAppearance();
for (Map.Entry<String, PDAppearanceStream> entry: stateAppearances.entrySet())
{
PDAppearanceStream appearance = entry.getValue();
PDResources resources = appearance.getResources();
if (resources == null)
return;
Map<String, PDXObject> xObjects = resources.getXObjects();
if (xObjects == null)
return;
for (Map.Entry<String, PDXObject> entryNew : xObjects.entrySet())
{
PDXObject xObject = entryNew.getValue();
System.out.println("printing out the xobject name: "+ entryNew.getKey());
if (xObject instanceof PDXObjectForm)
{
PDXObjectForm form = (PDXObjectForm)xObject;
PDResources resources2 = form.getResources();
if (resources2 == null)
return;
Map<String, PDXObject> xObjects2 = resources2.getXObjects();
if (xObjects2 == null)
{
return;
}
for (Map.Entry<String, PDXObject> entry2 : xObjects2.entrySet())
{
PDXObject xObject2 = entry2.getValue();
if (xObject2 instanceof PDXObjectForm)
{
continue;
}
else if (xObject2 instanceof PDXObjectImage)
{
PDXObjectImage ig = (PDXObjectImage)xObject2;
if(ig.getRGBImage().getHeight() > 0)
{
images.put(field.getPartialName(), "true");
}
else
{
images.put(field.getPartialName(), null);
}
//imageIds.add(imageId);
}
else
{
continue;
}
}
}
}
}
}
}
Images is a map variable: Mapimages.
Also my code file is large and so I didn't want to overwhelm anybody by pasting the whole file. Below is the dropbox link for the sample PDF file that I am using:
https://www.dropbox.com/s/g2wqm8ipsp8t8l5/GSA%20500%20PDF_v4.pdf?dl=0

Answer 1

Your PDF is a hybrid AcroForm/XFA document; where the XFA part uses fields with an imageEdit user interface, the AcroForm part uses pushbutton fields.
Thus, it allows you two ways to check whether an image field is set: Either you look at the AcroForm buttons and inspect their appearances for images, or you retrieve the XFA XML and inspect that.
Checking the XFA XML
Initially I did overlook the PDFBox version in the question title and implemented this for PDFBox 2.0.x. As it turns out, though, the identical code can be used for PDFBox 1.8.11, merely some additional exceptions may be thrown and, therefore, must be considered.
The latter option, inspecting the XFA XML, actually is a bit easier for the document at hand. Simply search the XML for an element with the name in question and check its contents. As an additional sanity check one can verify the content type attribute of the element:
boolean isFieldFilledXfa(Document xfaDom, String fieldName) {
NodeList fieldElements = xfaDom.getElementsByTagName(fieldName);
for (int i = 0; i < fieldElements.getLength(); i++) {
Node node = fieldElements.item(i);
if (node instanceof Element) {
Element element = (Element) node;
if (element.getAttribute("xfa:contentType").startsWith("image/")) {
return element.getTextContent().length() > 0;
}
}
}
return false;
}
(CheckImageFieldFilled helper method)
With it you can check your document:
PDDocument document = PDDocument.load(SOURCE);
PDAcroForm acroForm = document.getDocumentCatalog().getAcroForm();
Document xfaDom = acroForm.getXFA().getDocument();
System.out.println("Filled image fields from ImageField1..ImageField105:");
for (int i=1; i < 106; i++) {
if (isFieldFilledXfa(xfaDom, "ImageField" + i)) {
System.out.printf("* ImageField%d\n", i);
}
}
(CheckImageFieldFilled test method testCheckXfaGsa500Pdf_v4)
The output:
Filled image fields from ImageField1..ImageField105:
* ImageField1
* ImageField3
* ImageField6
Checking the AcroForm Appearances
The implementation here only works as is for PDFBox 2.0.x. The structure of the content stream parser classes has been considerably overhauled in 2.0.0, making a back-port of this code to 1.8.x a bit tedious.
To check whether the push button appearance actually shows an image (and not only has an image in its resources), one can use a simple PDFGraphicsStreamEngine subclass like this:
public class WidgetImageChecker extends PDFGraphicsStreamEngine
{
public WidgetImageChecker(PDAnnotationWidget widget) {
super(widget.getPage());
this.widget = widget;
}
public boolean hasImages() throws IOException {
count = 0;
PDAppearanceStream normalAppearance = widget.getNormalAppearanceStream();
processChildStream(normalAppearance, widget.getPage());
return count != 0;
}
#Override
public void drawImage(PDImage pdImage) throws IOException {
count++;
}
#Override
public void appendRectangle(Point2D p0, Point2D p1, Point2D p2, Point2D p3) throws IOException { }
#Override
public void clip(int windingRule) throws IOException { }
#Override
public void moveTo(float x, float y) throws IOException { }
#Override
public void lineTo(float x, float y) throws IOException { }
#Override
public void curveTo(float x1, float y1, float x2, float y2, float x3, float y3) throws IOException { }
#Override
public Point2D getCurrentPoint() throws IOException { return null; }
#Override
public void closePath() throws IOException { }
#Override
public void endPath() throws IOException { }
#Override
public void strokePath() throws IOException { }
#Override
public void fillPath(int windingRule) throws IOException { }
#Override
public void fillAndStrokePath(int windingRule) throws IOException { }
#Override
public void shadingFill(COSName shadingName) throws IOException { }
final PDAnnotationWidget widget;
int count = 0;
}
(CheckImageFieldFilled helper class)
With it you can create a check method like this:
boolean isFieldFilledAcroForm(PDAcroForm acroForm, String fieldName) throws IOException {
for (PDField field : acroForm.getFieldTree()) {
if (field instanceof PDPushButton && fieldName.equals(field.getPartialName())) {
for (final PDAnnotationWidget widget : field.getWidgets()) {
WidgetImageChecker checker = new WidgetImageChecker(widget);
if (checker.hasImages())
return true;
}
}
}
return false;
}
(CheckImageFieldFilled helper method)
and use it like this:
PDDocument document = PDDocument.load(SOURCE);
PDAcroForm acroForm = document.getDocumentCatalog().getAcroForm();
System.out.println("Filled image fields (AcroForm) from ImageField1..ImageField105:");
for (int i=1; i < 106; i++) {
if (isFieldFilledAcroForm(acroForm, "ImageField" + i + "[0]")) {
System.out.printf("* ImageField%d\n", i);
}
}
(CheckImageFieldFilled test testCheckAcroFormGsa500Pdf_v4)
The output, just like above:
Filled image fields (AcroForm) from ImageField1..ImageField105:
* ImageField1
* ImageField3
* ImageField6