[PDFBOX-5629] Custom AcroForm Field renaming function in PDFMergerUtility

• Type: Wish
• Status: Open
• Priority: Minor
• Resolution: Unresolved
• Affects Version/s: 2.0.28
• Component/s: Utilities
• Assignee: Unassigned
• Reporter: Stefan Ziegler
• Created: 28/Jun/23 09:08
• Updated: 14/Nov/23 12:43

Description

In PDFMergerUtility field name collisions are handled. The second field gets a new name like dummyFieldName1. The number is incremented with each new collision.

Could you please add a new protected method where these collisions are handled? Then we can extend PDFMergerUtility and override the method and give it a better name.

dummyFieldName doesn't say much currently and there you can't derive what this was about. We would find better e.g. the following version:

For example, if there is a field named XXX in both documents, then during collision resolution the second field could get one like XXX+1. We would then increment the number at the end if there would be further fields with the name.

The advantage is that the actual name XXX is kept, because this name contains the information about what kind of field it was. With dummyFieldName you don't know anymore what it was about.

In the end, you would simply create the following method, which then contains the current implementation:

 

protected void handleFieldNameCollision(PDAcroForm destAcroForm, PDAcroForm srcAcroForm, PDField srcField, COSDictionary dstField);

 

The default implementation of the method above can simply be the following, which is what you currently do:

protected void handleFieldNameCollision(PDAcroForm destAcroForm, PDAcroForm srcAcroForm, PDField srcField, COSDictionary dstField) {
    final String prefix = "dummyFieldName";
    for(int i=1;; i++) {
        String newName = prefix + i;
        if (destAcroForm.getField(newName) == null) {
            dstField.setString(COSName.T, newName);
            break;
        }
    }
}

Then we can simplify the acroFormLegacyMode function like this:
 

private void acroFormLegacyMode(PDFCloneUtility cloner, PDAcroForm destAcroForm, PDAcroForm srcAcroForm) throws IOException
{
    List<PDField> srcFields = srcAcroForm.getFields();
    COSArray destFields;    if (!srcFields.isEmpty())
    {        
        // get the destinations root fields. Could be that the entry doesn't exist
        // or is of wrong type
        COSBase base = destAcroForm.getCOSObject().getItem(COSName.FIELDS);
        if (base instanceof COSArray)
        {
            destFields = (COSArray) base;
        }
        else
        {
            destFields = new COSArray();
        }
        
        for (PDField srcField : srcAcroForm.getFields())
        {               
            COSDictionary dstField = (COSDictionary) cloner.cloneForNewDocument(srcField.getCOSObject());
            // if the form already has a field with this name then we need to rename this field
            // to prevent merge conflicts.
            if (destAcroForm.getField(srcField.getFullyQualifiedName()) != null)
            {
                handleFieldNameCollision(destAcroForm, srcAcroForm, srcField, dstField);
            }
            destFields.add(dstField);
        }
        destAcroForm.getCOSObject().setItem(COSName.FIELDS,destFields);
    }
}

 

This would be the same as the current implementation but this gives us the ability to override handleFieldNameCollision so that we can handle name collisions in a different way.