#1606 context loader listener not initialized during controller tests

[nya-elimu]

closes #1606

[codecov]

Codecov Report

Merging #1608 (eaf83f0) into main (b00de2d) will increase coverage by 0.46%.
The diff coverage is 0.00%.

@@             Coverage Diff              @@
##               main    #1608      +/-   ##
============================================
+ Coverage     13.45%   13.91%   +0.46%     
- Complexity      445      453       +8     
============================================
  Files           256      256              
  Lines          8378     8409      +31     
  Branches        845      876      +31     
============================================
+ Hits           1127     1170      +43     
+ Misses         7207     7187      -20     
- Partials         44       52       +8     

Impacted Files	Coverage Δ
.../content/contributor/EditMotivationController.java	9.09% <0.00%> (-0.44%)	⬇️
...imu/web/content/letter/LetterCreateController.java	6.25% <0.00%> (-0.21%)	⬇️
...elimu/web/content/letter/LetterEditController.java	5.71% <0.00%> (-0.17%)	⬇️
...nce/LetterSoundCorrespondenceCreateController.java	5.00% <0.00%> (-0.13%)	⬇️
...dence/LetterSoundCorrespondenceEditController.java	4.25% <0.00%> (-0.10%)	⬇️
...ontent/multimedia/audio/AudioCreateController.java	2.10% <0.00%> (-0.03%)	⬇️
.../content/multimedia/audio/AudioEditController.java	1.12% <0.00%> (-0.01%)	⬇️
...ontent/multimedia/image/ImageCreateController.java	2.24% <0.00%> (-0.03%)	⬇️
.../content/multimedia/image/ImageEditController.java	1.07% <0.00%> (-0.01%)	⬇️
...imu/web/content/number/NumberCreateController.java	4.16% <0.00%> (-0.09%)	⬇️
... and 28 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[nya-elimu]

@tomaszsmy In db8dd40, you'll see that I did some refactoring of tests that stopped passing when I added WordCreateControllerTest::testHandleSubmit_success. The WordCreateControllerTest affected the WordContributionEventDaoTest because they are both looking up words stored in the database.

So to make tests less fragile, I suggest we write them in such a way that they will still work even though other tests are storing similar content in the database. One way to do that, is to not expect a specific number of entities stored in the database, but instead look at the total number stored before and after individual test. This is because it should be okay if the database already contains content before running through individual tests, so no need to delete content after tests either. I hope all this makes sense 🙂