On Wed, 26 Sep 2012 17:35:21 +0200 Frederic Parrenin <[email protected]> wrote: > > Thanks for the information, I did not know that. > However, it seems there is a bug. > If I enter your formula, place 1 in Sheet2!A1, the result is 1. > Then if I move Sheet3 before Sheet1, the result is still 1. > Forcing the recalc fixes it.
Yes, this is a bug. I suspect that reordering worksheets doesn't lead to any checks for changes to any formulas' dependency chains. > > I've got no idea what a 4D formula would look like, or how you'd > > arrange the data in a spreadsheet. What were you thinking about for > > this one? > > Cells reference could be of the form X1Y3Z2T5. > Of course you can view in a screen only 2D slices of this hyper-cube. > You could select which pair of axes you want to view, e.g. (X,Y). > Then you could move the other 2 variables (Z and T in this case) using > 2 cursors. This doesn't sound much like a spreadsheet any more. What sort of data are you modelling? Using a database for this might be far simpler, they can easily store multi-dimensional data. - olly _______________________________________________ gnumeric-list mailing list [email protected] https://mail.gnome.org/mailman/listinfo/gnumeric-list
