Hi, > I would like to add spreadsheet function to my Pivot app. great, at least could be good for some blog articles ...
> As a beginner, I really want to get your valuable comment on. ok, I'll try (for what I can :-) ) ... > Should I start from Tableview or TablePane? it depends even on how many data you think to handle ... But I'd start with TableView, because otherwise with TablePane you need to put components in any cell (unless a trick to put a TextInput where you are editing, and keep Labels in all other cells ... but I'm not sure this can work here). > Is there an existing for building formula recognition? no, but thare are many open source frameworks for this. Probably from the Client (Pivot app) side you need to implement your own table/row editors and renderers. > What data model can help to reduce the complexity? ... it depends even on how many data you think to handle ... Nota that with enough memory in the client, TableView doesn't have problem with 1 million of records, like in our related demo. It depends even on how many data you send to the server ... I'd try with row-oriented data structures, or cell-oriented to minimize data transfers. Note that maybe you could need transformation functions between Pivot data structures and your data structures. Many of us knows that I like very much Scala, it has fantastic data structures (and many other powerful features like implicits, actors, etc), and it works good with Pivot ... try to take a look if you have some time, but it requires to learn some things. There is some sample of this interaction in Pivot and in our subprojects under Apache-Extras. > etc that you can think of really it seems an interesting application. You need a good balance between functions at client side and at server side, but to me this look at a good approach. Keep us updated. Bye, Sandro
