Skip to content

Resolve "Set Karma coverage threshold"

Noel Simmel requested to merge 1335-set-karma-coverage-threshold into main

Closes #1335 (closed)

The following numbers are based on the most recent commit fc532ff5 (October 29, 2024 at 1:11pm)

GitLab pipeline output:

=============================== Coverage summary ===============================
Statements   : 35.66% ( 3021/8470 )
Branches     : 22.78% ( 541/2374 )
Functions    : 27.32% ( 723/2646 )
Lines        : 36.01% ( 2689/7466 )
================================================================================

But running ng test locally sometimes gives different results. Possible reasons:

  1. Test order of execution.
  2. Re-runs of the tests may use cached results instead of calculating everything from scratch. ng cache clear and/or deleting the .angular folder didn't produce the same result as the GL pipeline, though.

I haven't been able to pinpoint what the problem is exactly.

After running ng test locally about 10 times, with and without clearing the cache in between, this is the lowest result I have found:

=============================== Coverage summary ===============================
Statements   : 35.58% ( 3014/8470 )
Branches     : 22.78% ( 541/2374 )
Functions    : 27.17% ( 719/2646 )
Lines        : 35.93% ( 2683/7466 )
================================================================================

This is without clearing the cache first, which I think is a more realistic scenario for developing, since clearing the cache makes the tests take longer. So I propose this as a baseline for future code coverage.

Merge request reports