Alex, I would not use the information from tfoot, just thead.
Thanks, RG -----Original Message----- From: Alexander Surkov [mailto:[email protected]] Sent: Thursday, May 28, 2009 10:00 AM To: Rob Gallo Cc: [email protected]; Willie Walker; IA2 List Subject: Re: [Accessibility-ia2] rowHeader/columnHeader implementationproposal Right because of thead and tfoot usage the same time. Could you give me idea how column header table might look and what is startingRowIndex? Does this table contain rows from thead plus tfoot? Alex. On Thu, May 28, 2009 at 9:53 PM, Rob Gallo <[email protected]> wrote: > Are you asking because there is a tfoot? > > This seems pretty straight forward to me; so I'd like to know why the > confusion. > > > > Thanks, > RG > > > -----Original Message----- > From: Alexander Surkov [mailto:[email protected]] > Sent: Thursday, May 28, 2009 9:44 AM > To: Rob Gallo > Cc: [email protected]; Willie Walker; IA2 List > Subject: Re: [Accessibility-ia2] rowHeader/columnHeader > implementationproposal > > By the way, what kind of table should columnHeader return for the case > > <table> > <thead> > <tr> > <th>col</th><th>col</th> > </tr> > </thead> > <tbody> > <tr> > <td>cell</td><td>cell</td> > </tr> > </thead> > <tfoot> > <tr> > <th>col</th><th>col</th> > </tr> > </tfoot> > </table> > > Alex. > > > On Thu, May 28, 2009 at 9:41 PM, Alexander Surkov > <[email protected]> wrote: >> Hi, Rob. >> >> This index is where header table is started relative main table. I >> think it might make sense for example like >> >> <table> >> <tr> >> <td>cell</td><td>cell</td> >> </tr> >> <tfoot> >> <tr> >> <th>col</th><th>col</th> >> </tr> >> </tfoot> >> </table> >> >> If we would consider dual method to rowHeader, then out argument >> startingRowIndex will be 1. >> >> Alex. >> >> >> On Thu, May 28, 2009 at 9:22 PM, Rob Gallo >> <[email protected]> > wrote: >>> Alex, >>> >>> I think this is good. >>> >>> But what is the purpose of the second parameter? I'm not sure this >>> is supposed to be an [out] parameter. >>> >>> HRESULT rowHeader( >>> [out] IAccessibleTable **accessibleTable, >>> [out, retval] long * startingColumnIndex) [get] >>> >>> >>> >>> >>> Thanks, >>> RG >>> >>> >>> -----Original Message----- >>> From: [email protected] >>> [mailto:[email protected]] On >>> Behalf Of Alexander Surkov >>> Sent: Wednesday, May 27, 2009 7:59 PM >>> To: [email protected] >>> Cc: Willie Walker; IA2 List >>> Subject: Re: [Accessibility-ia2] rowHeader/columnHeader >>> implementationproposal >>> >>> Hi, Pete. >>> >>> It's really good question. I can see two cases. >>> >>> 1. If we operate with HTML table where rowspans and colspans are >>> used then I believe you're right. In this case row and column >>> indexes are in sync with main table. >>> 2. If we deal with pure ARIA grid where there is no way to specify >>> row/col spans then we could return header table of the same >>> width/height like for primary table. For example, >>> >>> <div role="grid"> >>> <div role="row"> >>> <span role="gridcell">cell</span><span >>> role="columnheader">cell</span> >>> </div> >>> <div role="row"> >>> <span role="gridcell">cell</span><span >>> role="gridcell">cell</span> >>> </div> >>> </div> >>> >>> I think we should return table column header containing one row with >>> two cells (gridcell and columnheader). >>> >>> Does it makes sense? >>> >>> Alex. >>> >>> >>> On Fri, May 22, 2009 at 10:53 PM, Pete Brunet <[email protected]> wrote: >>>> Alex, A question that came up offline: When the column headers >>>> are not the full width of the table what is the relation between >>>> the row offset in the primary table and the header table? >>>> >>>> What I am remembering now is that you will get back a table, and >>>> most of the time the table will have one cell. A table is provided >>>> because there might be n rows in the header portion in which case >>>> the table would be one cell wide and with n rows. Or there might >>>> be a cell in the header which spans n columns. In that case I >>>> think you should get back a table with a single cell with >>>> columnExtentAt indicating it spans n columns. You could also have >>>> the case of a header table with two rows, with the first row >>>> containing a spanning cell and the second row containing n normal >>>> cells under the spanning > cell. >>>> >>>> Also, one feedback I got was that an AT would not expect to >>>> manipulate the cell in the table header expecting it to reflect >>>> that change in the primary table. >>>> >>>> I am ccing Will Walker in case he has comments regarding AT-SPI as >>>> we will want to remain in harmony with it and also Xing Li since >>>> his team implemented table headers in Symphony. >>>> >>>> Pete >>>> -- >>>> Pete Brunet >>>> >>>> a11ysoft - Accessibility Architecture and Development >>>> (512) 238-6967 >>>> pete @ a11ysoft.com >>>> http://www.a11ysoft.com/about/ >>>> http://www.linkedin.com/in/petebrunet >>>> Ionosphere: WS4G >>>> >>>> Alexander Surkov wrote: >>>> >>>> Hi. >>>> >>>> I summarized all our discussions with Pete about IAccessibleTalbe >>>> rowHeader/columnHeader implementation - >>>> https://wiki.mozilla.org/Accessibility/TableHeaders. Feedback on >>>> this proposal is needed. >>>> >>>> Thank you. >>>> Alex. >>>> >>>> >>>> >>>> >>>> _______________________________________________ >>>> Accessibility-ia2 mailing list >>>> [email protected] >>>> https://lists.linux-foundation.org/mailman/listinfo/accessibility-i >>>> a >>>> 2 >>>> >>>> >>> _______________________________________________ >>> Accessibility-ia2 mailing list >>> [email protected] >>> https://lists.linux-foundation.org/mailman/listinfo/accessibility-ia >>> 2 >>> >>> >> > > _______________________________________________ Accessibility-ia2 mailing list [email protected] https://lists.linux-foundation.org/mailman/listinfo/accessibility-ia2
