python3-pip-skeleton with no arguments now returns help

[tizayi]

python3-pip-skeleton with no arguments now prints help. resolves #79

[codecov]

Codecov Report

Merging #80 (53b09c4) into main (c784df0) will decrease coverage by 1.41%.
The diff coverage is 50.00%.

❗ Current head 53b09c4 differs from pull request most recent head f547bca. Consider uploading reports for the commit f547bca to get more accurate results

@@             Coverage Diff             @@
##              main      #80      +/-   ##
===========================================
- Coverage   100.00%   98.59%   -1.41%     
===========================================
  Files            2        2              
  Lines          138      142       +4     
===========================================
+ Hits           138      140       +2     
- Misses           0        2       +2     

Impacted Files	Coverage Δ
src/python3_pip_skeleton/__main__.py	98.55% <50.00%> (-1.45%)	⬇️

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[coretl]

Looks good, please can we have a test for it too?

[tizayi]

Writing one now what should i compare the output to.

[AlexanderWells-diamond]

I'd argue that comparing the text output isn't great, as it'll be a bit of a maintenance burden to constantly have to update the test every time you change a command line argument.

Instead, I'd prefer to see a test that simply tries to execute the command line with zero parameters, and then confirm that the parser.print_help() function is called. There's a variety of ways to patch sys.argv, e.g. mock.patch, monkeypatch or unittest.mock, and MagicMocks are good for checking a function has been called. However I do note that the current code structure will make this difficult - as the parser object is created inside the function you can't directly mock it, you'd have to mock the ArgumentParser() constructor to return a mock. Which is doable, but a bit less elegant.