Hello Antonio,

> I did not take into account the fact that longer contents are splitted
> into two rows at spaces. That changes the formula: I would 
> have to find
> "the length of the longest word contained in an element". 
> That would not
> change a lot Jeni's recursive approach, except that it would 
> have to be
> douoble-recursive to find the length of the words. But never 
> mind. This
> is certainly not the best solution, as I have a very irregular
> distribution of maximum "word" lengths. So the final result 
> would not be
> appealing.

Does your table have lots of rows, and are you going to change the column
widths on a row-by-row basis?

I was thinking maybe you could work out the word-lengthiness of content by
using the ratio of its total string-length and the string-length of all its
non-whitespace characters. That would give you a (rough) idea of which
content needed the most elbow-room.

it sounds like you'll have a bit of a fight on your hands regardless...
Tom Weissmann


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