[html4all] Headers= up

[Leif Halvard Silli]

On 2007-09-08 07:30:36 +0200 Robert Burns <rob at robburns.com> wrote:
> On Sep 8, 2007, at 12:10 AM, Leif Halvard Silli wrote:
>> On 2007-09-08 06:36:23 +0200 Robert Burns <rob at robburns.com> wrote:

>>> The HTML 4.01 algorithm says:
>>>
>>> "The search in a given direction stops when the edge of the table is
>>> reached or when a data cell is found after a header cell."

[...]
>> I am not certain you are correct here. What you quoted above is merly
>> step 1 of 5 possible steps in that algorithm.
>
> This is the only relevant step in the algorithm for tables without
> headers= or axis=. The entire association happens in that step.

Well ... see below.

>> What is a direction stop?
>
> It is not a 'direction stop' 

Aha... ;-) «the search [in a direction] stops/ends». 

> The sentence is saying that when
> searching a column for example search upwards in the column for header
> cell until: either 1) the edge of the table is reached or 2) a data
> cell is reached after one or more header cells. In other words in
> Anne's table the data cells only get associated with the headers on
> the same tbody. The algorithm does not address the issue of a THEAD,
> but I think it only makes sense to include the THEAD cells as header
> cells too. In other words when searching for column header cells, the
> algorithm should also add all of the THEAD cells that span that cells
> column (in Anne's table the "remarks" heading).

Agree about THEAD. But the 3rd step is «Column headers are inserted after row headers, in the order they appear in the table, from top to bottom.» This 3rd step is a needed step - also with a «ungenerous» reading of the basic algorithm of HTML4 (and e.g. in Ferg's tables). Agree?

With regard to THEAD, here is another thing which supports a «generous» readin of the basic algorithm: The HTML4 speaks  about THEAD being independent from the tbody/-ies. [Quote from HTML4]:

        Table rows may be grouped into a table head, table foot, and one
        or more table body sections, using the THEAD, TFOOT and TBODY
        elements, respectively. This division enables user agents to
        support scrolling of table bodies independently of the table
        head and foot.

The purpose of the scrolling of bodies independently of table head and foot, is of course to make the association of TH-cells in THEAD/TFOOT with cells in the TBODY/-ies mor simple for sighted users.

>> Could it mean that it has reached a kind of «axis»?
>
> No, I don't think axis has anything to do with that step of the
> algorithm.

Agree. Forget my «direction stop» thing. 

>> (You could very well be correct, though, because Ferg is not focused
>> on such tables as Anne's in his table explanations.)
>>
>> How about having a look at that table with the table inspector of of
>> juicy studio?`
>
> I took a look and it look to me that the this juicy studio inspector
> shows the association I would expect with the HTMl 4 algorithm. It
> doesn't show the global heading association (associating the THEAD
> headers with the data cells)), but as I said HTML 4.01 doesn't
> explicitly call for this (though I think its an oversight more than
> intentional).


[Quote from HTML4] <http://www.w3.org/TR/html401/struct/tables.html#edef-THEAD>
-- 
leif halvard silli