pdfitextxfa

Detect dynamic XFA with iText


I need to make pre-populated PDF/XFA forms read-only (as in no inputs, such as text, checkboxes, radio buttons etc. can have their values changed).

For regular AcroForms PDFs and static XFA forms, I can accomplish this by calling setFormFlattening(true) on the PdfStamper instance. For dynamic XFA forms, I have to set an access attribute of the XDP's field node to be readOnly.

The problem is, how do I detect if a form is dynamic XFA? isXfaPresent doesn't differentiate between static or dynamic XFA forms, so isn't useful.


Solution

  • iText is free as in free speech, not free as in free beer. Read http://lowagie.com/bumodels for more info about the business models we've tried in order to avoid switching from the MPL/LGPL to the AGPL.

    Q1: making dynamic XFA forms read-only

    That's a no-brainer with the most recent version of iText, but not supported in the obsolete version you're using. Please read http://lowagie.com/itext2 to find out why you shouldn't use iText 4.2.0 (which as far as I know isn't even an official release; and I know, because I'm responsible for every single iText release). Maybe you're talking about iTextSharp.

    Q2: flattening dynamic XFA forms

    If you want to flatten a dynamic XFA form, you have two options: either use Adobe LiveCycle ES (which will cost you an arm and a leg), or use iText's XFA Worker (which is a much less expensive closed source product built on top of the F/OSS iText). Given the cost and the number of man hours that went into this product, I don't think you'll find a solution that is free as in free beer. I for one don't know of any such product.

    Q3: how to find out if an XFA form is dynamic

    This is explained in my book, "iText in Action - Second Edition." You already have half of the solution. Condition 1: isXfaPresent() needs to return true. Condition 2: getFields() needs to have an empty key set. See the method readFieldnames() in this example.