After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 556470 - Orca should not land on empty (cell-less) tables when using structural navigation by table
Orca should not land on empty (cell-less) tables when using structural naviga...
Status: RESOLVED FIXED
Product: orca
Classification: Applications
Component: general
2.24.x
Other All
: Normal normal
: 2.24.3
Assigned To: Joanmarie Diggs (IRC: joanie)
Orca Maintainers
Depends on:
Blocks: 404403
 
 
Reported: 2008-10-15 21:07 UTC by Mesar Hameed
Modified: 2008-11-04 14:52 UTC
See Also:
GNOME target: ---
GNOME version: 2.23/2.24


Attachments
promised sample page (438 bytes, text/html)
2008-10-15 21:32 UTC, Mesar Hameed
  Details
proposed patch (890 bytes, patch)
2008-11-02 20:35 UTC, Joanmarie Diggs (IRC: joanie)
none Details | Review
revision 2: also check for bogus row count (1.00 KB, patch)
2008-11-04 00:37 UTC, Joanmarie Diggs (IRC: joanie)
committed Details | Review

Description Mesar Hameed 2008-10-15 21:07:01 UTC
Please describe the problem:
First experienced this when i was looking at my bugzilla page, and i was jumping between the tables. Some of the headings and tables, do not have any data in them (such as need info), the html code is simply <table> </table>
Orca accurately reports this as a table with 0 rows and 0 columns, but on pressing t to get pass this table we seem to get stuck.

please See to be attached mini page that illustrates this.

1. load attached page.
2. press t to navigate between tables
on reaching 0 rows 0 columns, pressing further t does not move us on to the next table.
This is true for both t and shift+t.
Visually, if a table is empty, is there anything to show this at all?
depending on previous answer, if a table is 0 rows 0 columns, should we still be reporting it?

(running firefox that comes with ibex, and trunk orca)

Thank you

Steps to reproduce:


Actual results:


Expected results:


Does this happen every time?
yes

Other information:
Comment 1 Mesar Hameed 2008-10-15 21:32:43 UTC
Created attachment 120682 [details]
promised sample page
Comment 2 Joanmarie Diggs (IRC: joanie) 2008-11-02 20:25:32 UTC
Bug 558516 seems to boil down to our poking our way through and/or around Gignormous tables (like with 15,000 children) grinds us to a halt and sometimes causes Firefox to segfault. Figure I might as well deal with the opposite end of the spectrum while I'm at it.

Assigning this one to myself.
Comment 3 Joanmarie Diggs (IRC: joanie) 2008-11-02 20:35:29 UTC
Created attachment 121828 [details] [review]
proposed patch

If the table has no children and hence no cells, it's arguably not a table.
Comment 4 Mesar Hameed 2008-11-03 11:44:20 UTC
Hi Joanie,

Thank you, that gets rid of the 0,0 tables.
For some strange reason we seem to get a table with -1 rows, -1 columns though, which we also get stuck on.
on my bugs page, under the "Patches: reviewed" heading, i hear that we have a table of 1 rows, 4 columns.
Pressing t once more, we arrive on the -1 rows -1 columns table, which we get stuck on. Can you reproduce this?
I tried to extract just that bit of html, but then the error does not appear.


Comment 5 Joanmarie Diggs (IRC: joanie) 2008-11-04 00:37:17 UTC
Created attachment 121926 [details] [review]
revision 2: also check for bogus row count

If we don't have at least one row, it ain't a table. :-)
Comment 6 Mesar Hameed 2008-11-04 11:23:24 UTC
Excellent, thank you, onwards and forwards :)
Comment 7 Joanmarie Diggs (IRC: joanie) 2008-11-04 14:52:10 UTC
Thanks Jon.

Patch committed to both the gnome-2-24 branch and trunk. Given the safe/simple nature of this change, closing as FIXED.