Re: [office] Proposal update: data cache, field name, source field id fordata pilot tables

[Ming Fei Jia <jiamingf@cn.ibm.com>]

Hi Kohei,

Kohei Yoshida <kyoshida@novell.com> wrote on 12/13/2008 01:42:11 AM:
>
> Re: [office] Proposal update: data cache, field name, source field
> id for data pilot tables

> > Of course, in case of that, the attribute <table:field-name> in the
> > element <table:data-pilot-field-reference> should also be renamed to
> > <table:field-display-name> since they have the same meaning.
>
> I would be extremely careful with renaming an existing attribute name,
> for none other than backward compatibility reasons.  The original
> attribute name may use some improvement, but not at the expense of
> breaking backward compatibility IMO, especially if this proposal is to
> make it into 1.2 as opposed to 2.0.
>
> Given this, I have to disagree with this renaming.
Make sense, renaming an existing attribute surely has backward compatibility risk at this phase. I just concern different places use different attribute names for the same meaning. But now I find it does not matter since different attributes names are used in different contexts although often they are referring to the same thing.

I just saw your extended proposal for the display names in data pilot. I think if from consideration of the interoperability with MS Excel, for the <table:data-field> in <table:data-pilot-display-info> and <table:data-pilot-sort-info> maybe also need a unique name. I suggest if you think it necessary, you can add them to your proposal. But interoperability with MS Excel is not all, maybe we have other concerns or comments.

>
> > Till now, I know Kohei propose the new name, so I'm waiting for
> > others' comments.
>
> Yes.  Using a new name is safer in terms of backward compatibility.
>
> >
> > (2) Add a new user scenario for the data pilot table enhancement. That
> > is to add an attribute <table:cell-auto-resize> to the
> > <table:data-pilot-table> so that user can control whether or not the
> > cells of data pilot table is resized automatically according to the
> > height and width of cell contents when the data pilot table is
> > refreshed.
>
> The functionality of auto-resizing will be undoubtedly useful, but I'm
> not sure if there is a merit in storing this flag in the file itself...
> Is there an interoperability reason for this proposal?
Yes, MS Excel has such feature and store this flag to the data pilot table internal. The interoperability is surely a reason for the proposal. Of course, I mainly think this feature is useful.
>
> I know at least Excel file format stores all sorts of flags.  I'm just
> trying to see if this is one of them.
>
> Thanks.
>
> Kohei
>