Better error message when reference range yields no values.

Better error message when reference range yields no values.

The issue is that since catapult code is now parsing values rather than
recipe code, failing to find any values in the reference range seems
like low confidence. This implements a check to distinguish between
those 2 cases.

R=sullivan@chromium.org,dtu@chromium.org
BUG=664510
Committed: https://chromium.googlesource.com/chromium/tools/build/+/e042d4dbd0ee62abaa3ad5d74e7dd897519a80bc

[sullivan, 2016-11-11 20:11:06]

lgtm

https://codereview.chromium.org/2498693002/diff/1/scripts/slave/recipe_module...
File scripts/slave/recipe_modules/auto_bisect_staging/bisector.py (right):

https://codereview.chromium.org/2498693002/diff/1/scripts/slave/recipe_module...
scripts/slave/recipe_modules/auto_bisect_staging/bisector.py:513: msg = 'No
values were found while testing the reference range.'
Shouldn't you surface a result here too?

[RobertoCN, 2016-11-11 20:13:05]

https://codereview.chromium.org/2498693002/diff/1/scripts/slave/recipe_module...
File scripts/slave/recipe_modules/auto_bisect_staging/bisector.py (right):

https://codereview.chromium.org/2498693002/diff/1/scripts/slave/recipe_module...
scripts/slave/recipe_modules/auto_bisect_staging/bisector.py:513: msg = 'No
values were found while testing the reference range.'
On 2016/11/11 20:11:06, sullivan wrote:
> Shouldn't you surface a result here too?

Well, LO_INIT_CONF is confusing in that case. I can add a new one:
'NO_VALUES_FOUND'  Does that sound good?

[sullivan, 2016-11-11 20:13:57]

On 2016/11/11 20:13:05, RobertoCN wrote:
>
https://codereview.chromium.org/2498693002/diff/1/scripts/slave/recipe_module...
> File scripts/slave/recipe_modules/auto_bisect_staging/bisector.py (right):
> 
>
https://codereview.chromium.org/2498693002/diff/1/scripts/slave/recipe_module...
> scripts/slave/recipe_modules/auto_bisect_staging/bisector.py:513: msg = 'No
> values were found while testing the reference range.'
> On 2016/11/11 20:11:06, sullivan wrote:
> > Shouldn't you surface a result here too?
> 
> Well, LO_INIT_CONF is confusing in that case. I can add a new one:
> 'NO_VALUES_FOUND'  Does that sound good?

sgtm, I think LO_INIT_CONF is confusing because it sounds like we have values
but there is no regression.

[RobertoCN, 2016-11-11 20:15:37]

On 2016/11/11 20:13:05, RobertoCN wrote:
>
https://codereview.chromium.org/2498693002/diff/1/scripts/slave/recipe_module...
> File scripts/slave/recipe_modules/auto_bisect_staging/bisector.py (right):
> 
>
https://codereview.chromium.org/2498693002/diff/1/scripts/slave/recipe_module...
> scripts/slave/recipe_modules/auto_bisect_staging/bisector.py:513: msg = 'No
> values were found while testing the reference range.'
> On 2016/11/11 20:11:06, sullivan wrote:
> > Shouldn't you surface a result here too?
> 
> Well, LO_INIT_CONF is confusing in that case. I can add a new one:
> 'NO_VALUES_FOUND'  Does that sound good?

Scratch that, I think that 'MISSING_METRIC' should be the result code for this
case.

[RobertoCN, 2016-11-11 20:27:46]

The CQ bit was checked by robertocn@chromium.org

[RobertoCN, 2016-11-11 20:27:46]

The patchset sent to the CQ was uploaded after l-g-t-m from
sullivan@chromium.org
Link to the patchset: https://codereview.chromium.org/2498693002/#ps20001
(title: "Surfacing result")

[commit-bot: I haz the power, 2016-11-11 20:27:56]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-11 20:47:46]

Description was changed from

==========
Better error message when reference range yields no values.

The issue is that since catapult code is now parsing values rather than
recipe code, failing to find any values in the reference range seems
like low confidence. This implements a check to distinguish between
those 2 cases.

R=sullivan@chromium.org,dtu@chromium.org
BUG=664510
==========

to

==========
Better error message when reference range yields no values.

The issue is that since catapult code is now parsing values rather than
recipe code, failing to find any values in the reference range seems
like low confidence. This implements a check to distinguish between
those 2 cases.

R=sullivan@chromium.org,dtu@chromium.org
BUG=664510

Committed:
https://chromium.googlesource.com/chromium/tools/build/+/e042d4dbd0ee62abaa3a...
==========

[commit-bot: I haz the power, 2016-11-11 20:47:47]

Committed patchset #2 (id:20001) as
https://chromium.googlesource.com/chromium/tools/build/+/e042d4dbd0ee62abaa3a...