W3C home > Mailing lists > Public > public-html@w3.org > August 2007

table, column group, column, row groups, TABLE, CAPTION, COLGROUP, COL, THEAD, TFOOT, TR, part of my review of 3.15 Tabular data

From: Robert Burns <rob@robburns.com>
Date: Thu, 9 Aug 2007 07:59:30 -0500
Message-Id: <C31663C3-D7DF-45B7-BBA0-6901F2685B1C@robburns.com>
To: HTML WG <public-html@w3.org>

table, column group, column, row groups, TABLE, CAPTION, COLGROUP,  
COL, THEAD, TFOOT, TR, part of my review of 3.15 Tabular data
Re:<http://dev.w3.org/cvsweb/~checkout~/html5/spec/Overview.html? 
rev=1.78#tabular>

SUMMARY
--------------------------------
   editorial suggestions
   consider allowing COL to be child of TABLE
   consider keeping TFOOT before TBODY or TR for document conformance
   consider adding DOM methods for insertBody and deleteBody


TABLE
--------------------------------

Consider allowing COL elements as a child of TABLE instead of only as  
a grandchild (child of COLGROUP). It's not clear what disallowing COL  
as a child of TABLE solves. If it is just to pave a cowpath, where  
authors mistakenly place the TFOOT last, then why not just let that  
mistake go through UA conformance criteria. That way we pave the  
cowpath, but we don't make it a truck route.

Consider reverting the TABLE content model to have TFOOT before TBODY  
as a part of of document conformance. UA conformance can continue to  
allow TFOOT, but I think we should begin to push for better table  
rendering (including better incremental rendering). Certainly other  
things can prevent incremental rending of tables, but why add  
another.  (see, RFC 1942[1])


Right now the ordered list of table contents has:

"5. Either:
Zero or more tbody elements, or
One or more tr elements."

Should we make that one or more TBODY elements, or do we really mean  
to allow a table with only a TFOOT, a THEAD, COLGROUP, and CAPTION  
(or even an empty TABLE since the others are all optional too).

Currently there's no DOM method insertBody or deleteBody. It would be  
useful to have an insertBody that worked similarly to insertRow.  
Especially when working with more complex tables this would give  
easier access to inserting and deleting table bodies.


COLGROUP
--------------------------------

I can't tell what how the error should be handled when a colgroup  
contains col elements and the @span attribute has a positive integer  
value. The algorithm looks like it favors COL child elements over the  
@span attribute. We should make that clear in the COLGROUP subsection  
to inform both authors and UA developers of which takes precedence.

For the paragraph:

"The span DOM attribute must reflect the content attribute of the  
same name, with the exception that on setting, if the new value is 0,  
then an INDEX_SIZE_ERR exception must be raised."

does this really involve a discrepancy between the two attributes:  
i.e., won't 'get' on the DOM attribute still return the same value as  
the content attribute, after the error (since the error will fail to  
set the DOM attribute or the content attribute)? This exception is  
not listed in other paragraphs related to other DOM attributes (with  
similar error handling situations).

COL
--------------------------------

Consider allowing COL as a child of TABLE instead of only as a child  
of COLGROUP

TBODY
--------------------------------

Nothing to report

TR
--------------------------------

Currently, there are two paragraphs that are nearly identical: one  
for the rowIndex element and one for the  rowIndex DOM attribute (is  
this an Easter egg?). Right now there's no definition of a rowIndex  
element.
"The rowIndex element must, if the element has a parent table  
element, or a parent tbody, thead, or tfoot element and a grandparent  
table element, return the index of the tr element in that table  
element's rows collection. If there is no such table element, then  
the attribute must return 0.

"The rowIndex DOM attribute must, if the element has a parent table,  
tbody, thead, or tfoot element, return the index of the tr element in  
the parent element's rows collection (for tables, that's the rows  
collection; for table sections, that's the rows collection). If there  
is no such parent element, then the attribute must return 0."

The sectionRowIndex is not defined. Perhaps that's the difference the  
two paragraphs are supposed to differentiate.


[1]:: Though it has been superseded by RFC 2854, RFC 1942 has a great  
approach to interactive UA table rendering that got lost in the  
updated RFC. See:
  <http://tools.ietf.org/html/rfc1942>
Received on Thursday, 9 August 2007 12:59:48 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:38:48 UTC