As you ask this question in general (not PDFBox specific) terms, I'll start by answering similarly. PDFBox is versatile enough to implement the concepts in question.
First, since multiple signatures from different people can be added to the document, the position of the signatures is important due to aesthetics and document printing if needed. Having said this, I would like to know an approach to handle this. What I was thinking was adding/append an additional page at the end of the documents and assign to it some kind of identifier like "doc_signatures", so when the second person opens the document for signature, it detects it already has a "doc_signatures" page created, and just add the signature and save the document using the increment option in PDFBox. Is this a good approach?
Whether this is a good approach or not depends on the nature of the documents to be signed and your influence on the pre-signing workflow of the document.
Paper documents often have dedicated positions for signatures of persons in a specific role. If you are buying something and as part of the sales contract acknowledge receipt, your signature has to clearly also sign the receipt part while the signature of the vendor needs not.
In digital PDF signatures you can alternatively make this clear by means of the Reason entry of the signature field value, but as you also want to print the documents, that might not suffice: In print there is no signature field value, only its appearance.
In such a situation the document to sign should already be prepared with empty signature fields positioned appropriately in the document and named or otherwise flagged to signal the role of the person to sign it. This, by the way, would also be the interoperable way, empty signature fields can easily be signed in e.g. Adobe Reader.
If this is not possible, though, and if the software for signing the document has a GUI, this GUI might provide the capabilities for each signer to position his signature appropriately for his signing reason and role.
Otherwise your extra signature page approach would be the approach of choice.
If all signers have the same role, though, or if there at least is no special appropriate position for any of the signing roles, your extra page approach might not merely be a last resort. It even kind of looks like a document resulting from a notarial act.
If it is, is there a way to identify the "doc_signatures" page so I don't append it again.
For PDFs according to the current ISO 32000-1 norm, you could do this using a page-piece dictionary:
A page-piece dictionary may be used to hold private conforming product data. The data may be
associated with a page or form XObject by means of the optional PieceInfo entry in the page object or form dictionary.
(section 14.5 of ISO 32000-1)
It looks like these piece dictionaries will be deprecated in the upcoming ISO 32000-2, though. Thus, a more future-proof approach would be for you to register a developer prefix and use your own key for that endeavor:
Developer
prefixes shall be used to identify extensions to PDF that use First Class names (see below) and that are
intended for public use.
(annex E of ISO 32000-1)
These custom keys don't seem to become deprecated in ISO 32000-2.
Also, can I add like signature fields to that "doc_signatures" page, with a position each one, so when I open the PDF, I detect it has "doc_signatures" already created and that it already has a signature on that page on "Field 1"(with its own X,Y coordinates) so place the second signature on "Field 2" on "doc_signatures" page and "Field 3" for the third signature, and also some type of limmit of the amount of signatures on the document?
You can easily inspect the annotations on your extra page and especially determine their location and extent. Consequently you can arrange additional signatures to your liking on an individual basis. Alternatively you can prepare a fixed number of empty signature fields on that extra page when you create it, arranging the signatures to your liking in one go.
All the above is only possible if the source document has not been signed before! If it already has been signed, adding a new page usually is considered a disallowed change of the document, effectively invalidating that first signature. For allowed and disallowed changes of signed document, see this answer.
Lets say each document type has a specific amount of signatures, for example, A sales document, with seller and buyer signatures, so the approach would be adding too signing fields to the documents and then place the signatures on those fields.. am I correct?
Exactly that is what I would propose: If you know the number and roles of the signers beforehand, prepare empty signature fields for them. In that case you do not even have to mark a signature page or something.
Now, sorry to bother you, with PDFBox will I be able to create signature fields and add signatures to those fields? Is there any example code for that?
Both is possible with PDFBox, but in particular adding a signature to an existing empty signature field may require some own coding.
