Patch for when rowspan or colspan equals 0

[Carlos Licea]

On Lunes 04 Mayo 2009 18:39:39 Germain Garand escribió:
> Le dimanche 26 avril 2009, Carlos Licea a écrit :
> > Hi list,
> > 	I had forgotten to submit this patch.
> > 	Now i believe it should be safe enough to commit.
> > 	Also I'm tired of this code, so I'll stop my work for now and start
> > implementing the DOM Level 3 XPath specification.
>
> Hi Carlos,
> this looks excellent to me.
>
Excellent!

> It passes all testcases I can throw at it, and keeps the regression test
> suite happy.
So happy to hear, I had to really scratch my head a while.

> If you recall, I had concerns about whether this could be applied in all
> rendering modes, seeing how Opera does make a distinction - but I now think
> we may try this in all modes - seeing the old HTML2 meaning of
> colspan/rowspan zero seems hardly ever used anywhere.
So I just commit as is, without checking the mode, right?

> All I'm left to comment about is typos, otherwise I think its ripe for
> commit.. ;)
Oops, even one is an Spanish word, hehe.

> +  //How: we take care of special case when colspan or rowspan equals 0
> +  //after that position the cell, that is just to tell where is it and
> tell +  //other cells where they can't be located (marking the cells as -1)
>
> Up to here, all good. But the three lines below don't bring much clarity I
> think... maybe replace them with a link to the relevant section of the HTML
> 4.01 specification?
What I meant is that the span is really used and that the cellAt() logic is 
only for positioning and not taken into account when painting, I'll try to 
clarify it and also link to the HTML spec.

> +  //later taking the span into account (and in other function) the cell is
> +  //then painted (that's why we need to set the colspan and rowspan
> properly +  //when any of them is cero
>
> cero -> zero
>
> Thank you for implementing a tricky bit of HTML 4, that no other engine
> ever managed to implement correctly yet.
No problem!

> Greetings,
> Germain
>
> P.S: also please remove/comment the debug statements before commiting, as
> they are a bit noisy.
Done, commented them. Don't worry I usually remove them before committing.
Also, I'll, finally, close the bug 41063.
Carlos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.kde.org/mailman/private/kfm-devel/attachments/20090504/9d043160/attachment.htm>