RFR: 8292353: TableRow vs. TreeTableRow: inconsistent visuals in cell selection mode [v10]

[Andy Goryachev]

On Wed, 14 Sep 2022 10:32:19 GMT, Jeanette Winzenburg <fastegal at openjdk.org> wrote:

>> Is it possible to have two or more rows with the same row index?  i would imagine that will break a lot of things.
>> 
>> I am not sure why you think the lookup is brittle - after all, VirtualFlowTestUtils uses lookup to get a pointer to the VirtualFlow, is it not (line 333)?  Is it because the lookup may not return a cell if it is outside of the current view port area and therefore has not been created (while the .getCell() guarantees to create one)? 
>> 
>> Is this the only objection to the changes in this PR?  I suppose i could change the tests to use VirtualFlowTestUtils, though my preference would be still to test close to the real world scenario, using public APIs (as opposed to getting into internals using some internal test utils), given the fact that I *do* expect these cells to be visible given the dimensions of the table and columns.
>
>> Is it possible to have two or more rows with the same row index? i would imagine that will break a lot of things.
> 
> don't imagine, verify ;) What happens when you run the test? What happens when you look at the scenegraph (f.i. with ScenicView)? Don't know why they are there but they are - well outside the viewport.
> 
>> 
>> Is this the only objection to the changes in this PR?
> 
> I honestly cannot understand how you possibly could have reached the conclusion that this is the _only_ objection given my comments .. 
> 
> Anyway, I'm off here - don't know how to abort a review, but not going to waste more time and nerves on who doesn't seem to listen (that's my personal feeling, of course ;)

sorry, what I really meant is - are the code changes ok and we are only discussing updating tests?

my goal here is to fix as many bugs as I possibly can, while listening to the feedback of more experienced team members and addressing their concerns.  it is not always clear when an exchange of ideas ends and marching orders begin, especially in situations where several possibilities exist.  perhaps a clearly annunciated request for specific changes would work?

I feel really uncomfortable with thousands of defects logged against openjfx, and all I want is to help.

-------------

PR: https://git.openjdk.org/jfx/pull/875