Can I merge the required changes into Trunk's SUnit package? :-) It's basically a few additions to TestResult and RestTunner. Best, Marcel |
> On 03.09.2019, at 10:35, Marcel Taeumel <[hidden email]> wrote: > > Can I merge the required changes into Trunk's SUnit package? :-) It's basically a few additions to TestResult and RestTunner. > :D Yes please, at least something in that direction. -t |
In reply to this post by marcel.taeumel
On Tue, Sep 03, 2019 at 10:35:01AM +0200, Marcel Taeumel wrote:
> Can I merge the required changes into Trunk's SUnit package? :-) It's basically a few additions to TestResult and RestTunner. > > Yes please, this is a useful enhancement. Dave |
In reply to this post by Tobias Pape
> On 2019-09-03, at 1:38 AM, Tobias Pape <[hidden email]> wrote: > > >> On 03.09.2019, at 10:35, Marcel Taeumel <[hidden email]> wrote: >> >> Can I merge the required changes into Trunk's SUnit package? :-) It's basically a few additions to TestResult and RestTunner. >> > > :D > Yes please, at least something in that direction. Likewise tim -- tim Rowledge; [hidden email]; http://www.rowledge.org/tim "Bother," said Pooh. "Eeyore - ready two photon torpedoes and lock phasers on the Heffalump. Piglet, meet me in Transporter Room Three." |
In reply to this post by marcel.taeumel
A great start to detecting "performance failures". +1 Now imagine if we added a variable to TestCase, say, 'timings', a Dictionary of #testSelector -> lastRunDuration 's. Then, any test more than ~10% off its baseline would raise an Error (or.. PerformanceFailure?, whatever). There'd be a method, #resetBaselines to reset this Dictionary under the same condition that we re-ask for developer initials -- when image detects it has moved to a new location. Since this this could mean possibly different hardware, new baselines would need established to avoid spurious failures. From a UI perspective, access to #resetBaselines should be sufficient -- once all are addressed or "okay'ed", the #resetBaselines would allow new baselines to be captured on the next run... - Chris On Tue, Sep 3, 2019 at 3:35 AM Marcel Taeumel <[hidden email]> wrote:
|
Hmm... at the moment, it would be a new entry in the TestCase's history dict besides #passes, #failures, etc.: How to compute a reliable baseline? You know about SMark and BechmarkRunner? A TestCase's history dict is the mechanism to store such kind of data, I suppose. Even though I am not sure why there are no TestResults kept in such a history record. Objects! :-D Best, Marcel
|
Free forum by Nabble | Edit this page |