> here's an example where an empty cell comes in between full cells

[SNIP]

What I could see from that example is, that the empty cell number 7 is
"missing"
I know it *is* there because of the Index, like described below

> Ss:Index is used to keep track of table structure. Try 
> ss:Cell[@ss:Index = '6']/Data

But this only works if the cell actually have the Index set, that is not
always.
One way or another I could end up, with an Excel spreadsheet, where I
neither can get a match for ss:Cell[6]/ss:Data or ss:Cell[@ss:Index =
'6']/Data, even though the data is there.

> >From the documentation: "Specifies the column index of this 
> cell within
> the containing row. If this tag is not specified, the first 
> instance of
> a Cell element within a row has an assumed Index="1". Each additional
> Cell element has an assumed Index that is one higher. "

This I already have figured out without reading it from the documentation ;)

My originally request was to do what you describe below
 
> This use of ss:Index, while irritating, also means that one 
> could make a
> generic table handler which either placed cells in, or left them out,
> dependant on parameters, etc. passed to the application.

Actually, if anyone has an example of such a generic table handler I would
very much like to see it.

<Thomas/>

 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list