So I guess what I’m saying is that for me what would help the most the current workflow used by the xwiki core dev team is: * the introduction of the xar:fetch and xar:deploy mojos * the validation (to help prevent mistakes) * the extraction of the attachments as standard files on the file system so that they can be replaced easily Plus at least extract the JS-extensions and CSS-extensions as JS and CSS files at least or? Yes those are content for me (what I meant by content is doc content + all xproperties of type textarea). ok, so that's deterministic at least (but sometimes too much?) Also, how do you determine the extension ? (seems like the object nature would dictate that...). And the same holds for velocity and groovy code or? But here, I cannot find a way that would make it clear that a given object property or page content is in velocity, groovy, or xxxx. Is there a way? They’re not in groovy or velocity, they’re in some markup syntax (e.g. XWiki Syntax 2.1). Mmh, here I am really not convinced. I've seen many pages be only designed to be used using parseGroovyFromPage. Is this something that is deprecated now? I've also seen velocity-based content to be the core of the UI of most applications and be contained in the content of pages.
From the description you make, the files would not be .vm or .groovy (actually, there would be none, right?) but using some wiki-syntax-extension. Right?
To me, it seems like having .vm and .grv or .groovy files is essential. I would be alone? Paul