We want a simple tool to parse PDF forms, retrieving just the associated field names and user values. PDFSharp would be a good option for us, as it is under the MIT license, but it is a few versions of Acrobat behind - so not going to work. The most popular library (from what my searches have revealed) seems to be iTextSharp. Thus I am introduced to the intricacies of the AGPL.
Now, for the structure of our app. We want something simple and re-usable among any apps we may want to use later that require the same functionality. My plan was to design a simple web service, that takes in the PDF file, and simply returns the Fields and Values in a list of key-value pairs.
It looks like, as long as this is an intranet service, there is no concern from the AGPL, as described in this question: Can I safely use an open source library in an internal closed-source project?
That is our most likely scenario, but I wanted to be aware ahead of time in case we need to expose our service externally (such as for some of our Silverlight clients, for example). If the service were exposed externally, then, under AGPL, would we simply need to provide the source for the service itself? Or, would the source of any consumer of the service also need to be provided?
So, I am looking for 1) confirmation that in the intranet scenario there are no further considerations, and 2) what is necessary if the web service is publicly visible?