- From: Tim Bagot <tsb-w3-style-0003@earth.li>
- Date: Fri, 29 Jun 2001 22:52:36 +0000 (UTC)
- To: <www-style@w3.org>
At 2001-06-29T17:33-0400, fantasai wrote:- > Bernd Mielke wrote: > > > > Ian, > > > > while you are on it, could you also look how row- and colspans > > should be treated. Its hard for me to admit it, but I like the > > way IE6 renders the following snippet. > > http://bugzilla.mozilla.org/show_bug.cgi?id=88260 > (the snippet & 2 screenshots) Aha. I think that IE's rendering is definitely wrong (w.r.t. CSS2, at any rate), since the cells should allow the columns' backgrounds to show through, and not simply inherit them. (Incidentally, do people see there being a use for more sophisticated selectors for table cells, so that it would be possible to apply styles to the cells of/in a given row or column rather than the row/column itself?) I'm also not quite sure about Mozilla's interpretation: with a little playing, I've convinced myself that it does not let row and column backgrounds "leak" into the inter-row and inter-column cell spacing (respectively). CSS2 seems to be fairly unequivocal that the row boxes cover the *whole* table; it's less strong about columns, but it would seem peculiarly inconsistent to treat the two differently in this respect. OTOH, it also says (17.6.1) that the cell spacing is filled with the table background. The only way I can see to interpret this consistently is to think of deciding backgrounds under the layered model of 17.5.1 with collapsed borders, and then "exploding" the table, inserting gaps where the background of the bottom layer can show through; this, however, would mean that row and column backgrounds should not appear at all in the cell spacing - i.e. each cell should be surrounded on all four sides by the table background. Can any of the CSS authors comment on the original intention of the text? Tim Bagot
Received on Friday, 29 June 2001 18:52:40 UTC