Question about OpenDocument API development

Hi all, I'm making progress in understanding the OpenDocument package, reorganizing it and fixing bugs in it. While tinkering with new content types (e.g. lists and images), I found that he current API approach does not suffice. Let me explain: - class OpenDocument_Text has two methods regarding adding content: createHeading and createParagraph - OpenDocument_Element_Paragraph has createSpan, createHyperlink, createText - OpenDocument_Element_Heading has createSpan, createHyperlink, createText - all of them should support createBookmark which they don't do yet. Potentially all of them will be able to get images added to. Currently, every single element class (+ the opendocument one) implements the create*() methods itself, redirecting the calls to the element::instance() methods. What I don't like is that, when adding support for a new element - e.g. lists - I have to do this in every single element class that supports it! So there needs to be a better way - for example using __call and throwing exceptions in the cases the element-to-be-created is not supported. I could also try to find class hierarchies with elements that are supported in certain classes only, and base them all off a common element class. The problem with that approach is that I don't know yet if that hierarchy is tree-like or not. I could also do take the quickform approach and use a general create('elementtype', $params) function - as flexible as __call, but not very usable from a user's point of view in regard of capturing the supported parameters. What would you do? -- Regards/Mit freundlichen Grüßen Christian Weiske