best SUnit Runner, logging
Locked
 
|
What is the best way to go as far a some sort of SUnit Runner and logging mechanism goes?
So far using just the Parcel Manager/testing code but open to suggestions. vw7.9.1 Thx for any tips, Ken G. brown _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
|
|
Ken, You‘ll probably get different answers from others, but here are my current preferences. I’m not sure what you mean by “logging mechanism”, but I'll take a guess below. I use SUnitToo for all of my tests, along with the SUnitToo(ls) browser integration. I also use my own SUnitTools-AutoTest package for even faster feedback on my tests. I also use a suite of tools I wrote for running my tests from an automated build server (Jenkins in my case). Part of this suite is a package called StoreCI-Testing which will run a set of tests and save the results in a JUnit-compatible XML format. I'm not sure if this is what you meant by “logging mechanism” above. Hope this helps, On Wed, Sep 18, 2013 at 8:29 AM, Ken G. Brown <[hidden email]> wrote: What is the best way to go as far a some sort of SUnit Runner and logging mechanism goes? Randy Coulman Email: [hidden email]
_______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
Re: best SUnit Runner, logging
|
|
In reply to this post by Ken G. Brown
I use Assessments for that purpose. See Assessments bundle in the
public store repository. On 9/18/13 8:29 AM, Ken G. Brown wrote: > What is the best way to go as far a some sort of SUnit Runner and logging mechanism goes? > > So far using just the Parcel Manager/testing code but open to suggestions. > > vw7.9.1 > > Thx for any tips, > Ken G. brown > _______________________________________________ > vwnc mailing list > [hidden email] > http://lists.cs.uiuc.edu/mailman/listinfo/vwnc > vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
|
|
In reply to this post by Ken G. Brown
Dear Ken,
>What is the best way to go as far a some sort of SUnit Runner and logging mechanism goes? > >So far using just the Parcel Manager/testing code but open to suggestions. ... vw7.9.1 > If (and only if) you load the latest OR trunk versions (currently, that means the latest versions) of packages SUnitPreload, SUnit, SUnitResourcePatterns and RBSUnitExtensions in that order, i.e. avoiding loading your image's 7.9.1 parcels, and then load the latest RBSUnitShowResult and remaining prereq, it will keep the latest outcome of any test you then run in the singleton ResultsData current A test's data in that structure is reposted if the test is rerun, dropped if the test method is recompiled, and all results are flushed if RBSUnitExtensions is unloaded (or you send #flush). The structure holds test times as well as the outcome (and the result properties dictionary could be made to store more if wanted). I use this to show test results in browsers. I have not yet integrated with Jenkins or similar (see Randy's post). However it would be straightforward to write out the results from time to time, or write out those whose most recent run was after a certain time, or whatever. I will post an announcement about RBSUnitShowResult as soon as 7.10.1 is released. (BTW, if you're in the dev programme, so have a 7.10 gold image with the 7.10 parcels, nevertheless use the latest RBSUnitShowResult from the OR - see blessing comments for why.) FYI, the very old logging mechanism in core SUnit is limiting; it only logs failing tests and presupposes you've used #assert:description:, not just #assert:, and written a meaningful string parameter. It's on my to do list to look at. HTH Niall Ross _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
|
|
Niall Ross wrote:
> I will post an announcement about RBSUnitShowResult as soon as 7.10.1 > is released. (correct typo) as soon as 7.10 is released _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
|
|
In reply to this post by Niall Ross
Thx, will be considering.
Ken On 2013-09-18, at 1:31 PM, Niall Ross <[hidden email]> wrote: > Dear Ken, > >> What is the best way to go as far a some sort of SUnit Runner and logging mechanism goes? >> >> So far using just the Parcel Manager/testing code but open to suggestions. ... vw7.9.1 >> > If (and only if) you load the latest OR trunk versions (currently, that means the latest versions) of packages SUnitPreload, SUnit, SUnitResourcePatterns and RBSUnitExtensions in that order, i.e. avoiding loading your image's 7.9.1 parcels, and then load the latest RBSUnitShowResult and remaining prereq, it will keep the latest outcome of any test you then run in the singleton > ResultsData current > A test's data in that structure is reposted if the test is rerun, dropped if the test method is recompiled, and all results are flushed if RBSUnitExtensions is unloaded (or you send #flush). The structure holds test times as well as the outcome (and the result properties dictionary could be made to store more if wanted). > > I use this to show test results in browsers. I have not yet integrated with Jenkins or similar (see Randy's post). However it would be straightforward to write out the results from time to time, or write out those whose most recent run was after a certain time, or whatever. > > I will post an announcement about RBSUnitShowResult as soon as 7.10.1 is released. (BTW, if you're in the dev programme, so have a 7.10 gold image with the 7.10 parcels, nevertheless use the latest RBSUnitShowResult from the OR - see blessing comments for why.) > > FYI, the very old logging mechanism in core SUnit is limiting; it only logs failing tests and presupposes you've used #assert:description:, not just #assert:, and written a meaningful string parameter. It's on my to do list to look at. > > HTH > Niall Ross > > > _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
|
|
In reply to this post by Randy Coulman
By logging mechanism, I meant whatever it takes to get a log of the results of running a bunch of tests including time to run. each test.
Thx, Ken On 2013-09-18, at 12:29 PM, Randy Coulman <[hidden email]> wrote:
_______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
«
Return to VisualWorks
|
1 view|%1 views
| Free forum by Nabble | Edit this page |