Re: [office-formula] Semantics

[Eike Rathke <erack@sun.com>]

Hi Tomas,

On Wed, Mar 01, 2006 at 09:37:32 +0100, Tomas Mecir wrote:

> > I think all spreadsheet applications handle this the same: SUM, COUNT,
> > AVERAGE, and all the like, get passed what OpenFormula called
> > a NumberSequence. This by definition ignores string cell content,
> 
> They currently do, yeah. So the question is - do we want it to be like
> that ?

Yes.

> And do we even have a better solution ?

No.

> I realise that I may be playing a devil's advocate here a bit :)

What for? Those functions work on numbers, text isn't a number. I wish
all function parameters / operands to numeric operators would behave
like that..


> > We'd then end up with a definition of "how does it the one application
> > of the big player". Which is not our goal, though will be congruent in
> > many but not all cases.
> 
> Isn't this what OpenFormula has been doing so far ? With tests being
> adjusted for OOo to pass them and all ...

Well, thanks for the flowers, but with the big player I wasn't referring
to OOo ;-)


> > > I think this introduces the problem of our goal: shall we design a
> > > spec based on real spreadsheets, or also put in new/changed things, if
> > > we feel that it makes sense ?
> >
> > Based on real applications, with changes only when and where necessary
> > for clear benefits.
> 
> Well, those are the points that I'm trying to raise - where the
> current implementations could be considered ... a bit dirty.

I don't consider ignoring text in a range to be SUM'ed dirty. Not at
all.

> Everyone agrees with the complex number problem. Then we have datatype
> conversion, and I think that's about all there is to difficult
> problems.

You're oversimplifying things. We have many areas where different
applications currently do not agree.

  Eike