Regarding Proposal

[Jaroslaw Staniek]

On 16 March 2012 19:04, Marijn Kruisselbrink <mkruisselbrink at kde.org> wrote:
> On Friday, March 16, 2012 09:56:27 AM Jaroslaw Staniek wrote:
>
>> Jigar,
>> Cool, that's normal - many ideas, food for thought :_)
>>
>> To add even more spice here - pivot tables for database apps (Kexi)
>> are really appealing (but I think reuse of Sheets pivot tables would
>> be rather limited because of different data models and requirements -
>> spreadsheets usually have all the data in memory already). Another
>> challenge :)
> Actually, that is not really true. To support all of pivot tables that are
> defined in odf, having an actual database as source of the data is also
> something that would be possible. So hopefully any pivot table implementation
> would be generic enough to also allow for those usecases.

I know that's possible but it works best if the data is fetched to the
client first, pushed into Sheets' structures and we let Sheets do the
magic. But please consider cross-tab queries/sparse matrices
processing that can be executed on the server side. Depends on the
power of the db engine - IIRC PostgreSQL or Oracle are capable of
that.

I am not saying there's not much to reuse in Kexi from Sheets. There's
whole set of great reusable components, starting with ODF formulas.
Also round-tripping regarding result of pivot operations between Kexi
and Sheets would be a task on its own. I see Sheets being close to
Kexi because of the integration needs - Kexi by design complements the
use cases that spreadsheets offer.

-- 
regards / pozdrawiam, Jaroslaw Staniek
 http://www.linkedin.com/in/jstaniek
 Kexi & Calligra (kexi-project.org, identi.ca/kexi, calligra-suite.org)
 KDE Software Development Platform on MS Windows (windows.kde.org)