Re: [office] table:style-name vs table:default-cell-style-name

[Eike Rathke <erack@sun.com>]

Hi Andreas,

On Tuesday, 2009-07-07 08:47:41 -0600, Andreas J. Guelzow wrote:

> > style given as table:default-cell-style-name attribute of table:table-row
> > style given as table:style-name attribute of table:table-row
> 
> I still don't understand the difference between the two. In the list you
> have quoted above, couldn't everything that is specified in
> table:default-cell-style-name of table:table-row be given in
> table:style-name of table:table-row instead? This still looks like
> unnecessary duplication to me.

It seems the original intention was to have table:style-name name the
style that applies row-wide settings such as row height, that would not
be overridden by cell styles, and no cell styles. Apparently that wasn't
well specified and at least one implementation wrote cell styles to
table:style-name, so indeed we now have some sort of duplication.
A reading implementation will have to inspect both for cell styles,
a writing implementation should write cell styles with
table:default-cell-style-name.


> > > Why are we jumping over the table level, ie. do not have
> > > table:style-name or table:default-cell-style-name attributes for
> > > table:table.
> > 
> > It isn't really needed. In fact, having the default styles at the
> > columns (and/or rows) is an optimization to not have to write and read
> > a style information for each cell in case an entire column is formatted
> > identically. We could define that for the table level as well, but it
> > wouldn't save much, maybe a few columns' attributes.
> 
> I would think it may save a huge number of column attributes. If the
> column switch between two styles we now have to specify a style for
> every column rather than just a default style and a style for every
> second column.

Ok, if a large amount of columns of a sheet were formatted that way it
would save some.

> Moreover, Gnumeric has a default style for each table. Currently this
> cannot be saved in an ODF file.

I see. I guess we'll have to address that in ODF-Next.

  Eike

-- 
 OpenOffice.org / StarOffice Calc core developer and i18n transpositionizer.
 SunSign   0x87F8D412 : 2F58 5236 DB02 F335 8304  7D6C 65C9 F9B5 87F8 D412
 OpenOffice.org Engineering at Sun: http://blogs.sun.com/GullFOSS

PGP signature