[Toaster] Review request for "Bug 5917 - Task Detail page needs support for "Empty" outcome"

[Reyna, David]

Hi Belen,

> When you say sort, do you mean that the 'Empty task' option in the Outcome
> filter works? Which it does, by the way :)

Yes. Ravi's first pass had "Task.OUTCOME_NA" for the filter for "Empty", so that was one of the patch's updates.

- ('Empty Tasks', 'outcome:%d'%Task.OUTCOME_NA),
+ ('Empty Tasks', 'outcome:%d'%Task.OUTCOME_EMPTY),

- David

> -----Original Message-----
> From: Barros Pena, Belen [mailto:belen.barros.pena at intel.com]
> Sent: Friday, March 07, 2014 2:14 AM
> To: Reyna, David
> Cc: toaster at yoctoproject.org
> Subject: Re: Review request for "Bug 5917 - Task Detail page needs support
> for "Empty" outcome"
> 
> Thanks for fixing this, David!
> 
> On 07/03/2014 06:26, "Reyna, David" <david.reyna at windriver.com> wrote:
> 
> >Hi Belen,
> >
> >I have fixed Bug 5917 by adding support for the "Empty" outcome value to
> >the task pages.
> >
> >  dreyna/tasks_empty_outcome_5917
> >
> >This patch in on top of a cherry-pick of your related fix:
> >
> >    bitbake: toaster: Fix help text for task details outcome
> >(d20034c9be9094af31cf63fc9ce56e19cfa87c82)
> >
> >This patch adds these fixes:
> >
> >  1) In the tasks page, you can now sort by "Empty"
> 
> When you say sort, do you mean that the 'Empty task' option in the Outcome
> filter works? Which it does, by the way :)
> 
> >
> >  2) In the task detail page, the help icon now has the text "This task
> >has no executable content"
> 
> This works, thanks!
> 
> >
> >  3) In the task detail page, you now have the <div> for "empty" outcomes
> >of:
> >
> >     "This task is empty because it has the <code>noexec</code> flag set
> >to <code>1</code>"
> 
> I have been told there are at least 2 reasons why a task might no have
> executable content:
> 
> 1. Because it has the noexec flag set
> 2. Because the task function is empty
> 
> Problem is: we don't know which one applies to a certain empty task. So
> although I added that notification to the design, we should probably not
> use it, since right now such notification would be really just a guess
> (not something we are actually checking for).
> 
> >
> >Thanks,
> >David
> >
>