feat(system.fileName): file extension count

[xDivisionByZerox]

Implementation is discussed in #395 (direct comment ref).

This is implemented in a non-breaking manner.

[xDivisionByZerox]

Since @pkuczynski somewhat provided the implementation in his PR, I want to make them coauthor in this PR. How would this be possible? Is assigning them to the PR enough?

[codecov]

Codecov Report

Merging #1101 (eb94851) into main (316f61f) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##             main    #1101   +/-   ##
=======================================
  Coverage   99.64%   99.64%           
=======================================
  Files        2146     2146           
  Lines      230438   230462   +24     
  Branches      976      979    +3     
=======================================
+ Hits       229625   229649   +24     
  Misses        792      792           
  Partials       21       21           

Impacted Files	Coverage Δ
src/modules/system/index.ts	100.00% <100.00%> (ø)

[prisis]

Since @pkuczynski somewhat provided the implementation in his PR, I want to make them coauthor in this PR. How would this be possible? Is assigning them to the PR enough?

It is possible if you merge the commit to add a coauthor

[xDivisionByZerox]

So by adding new test cases, existing ones (not the new one) are randomly failing? This doesn't seem right to me. I also can reproduce the values in the test by setting the seed to the one in the test.

[ST-DDT]

Please add a seeded test, that uses the new option.

[ST-DDT]

So by adding new test cases, existing ones (not the new one) are randomly failing? This doesn't seem right to me. I also can reproduce the values in the test by setting the seed to the one in the test.

The random seeded tests are less random than one would like. I will try to have a look at the randomness issue soon.
The test previously passing was probably a false positive.

[xDivisionByZerox]

Please add a seeded test, that uses the new option.

Honestly, no. I don't see the point in the seeded tests. No other function has an extra seeded test case for their argument, why would this have? I already added some test cases that check for the argument options in the non-seeded test runs. They are much more valuable IMO.

[xDivisionByZerox]

The failing test is expecting a common file extension to be included in an array provided in the test. According to our implementation of system.commonFileExt() all of the following would be valid common file types

[
  'gif', 'htm', 'html', 'jpe',
  'jpeg', 'jpg', 'm1v', 'm2a',
  'm2v', 'm3a', 'mp2', 'mp2a',
  'mp3', 'mp4', 'mp4v', 'mpe',
  'mpeg', 'mpg', 'mpg4', 'mpga',
  'pdf', 'png', 'shtml', 'wav',
]

But our expected values are only:

[
  'gif', 'htm', 'html', 'jpeg',
  'm2a', 'm2v', 'mp2', 'mp3',
  'mp4', 'mp4v', 'mpeg', 'mpg',
  'pdf', 'png', 'shtml', 'wav',
]

[ST-DDT]

Then update the expected values!?
I'm not sure whether I would consider all of them "common".

[xDivisionByZerox]

[...] I would consider all of them "common".

That's where I wasn't sure about and wanted to have some "backup" from you.

[ST-DDT]

I have rarily/never seen m2a/v mp2 or shtml.
So I would remove them. Lets discuss this tomorrow with everyone.
I miss json and txt.

[xDivisionByZerox]

For now, I updated the test expectation. In the future we should rework the function to pick extensions from a custom-defined array or something like that.

[xDivisionByZerox]

This is non-blocking. Can we merge?