ensure 5.x and 4.4 can use the same logic

[mollux]

No description provided.

[escopecz]

👍

[codecov-commenter]

Codecov Report

Merging #311 (39e53c6) into main (e90d511) will not change coverage.
The diff coverage is n/a.

❗ Current head 39e53c6 differs from pull request most recent head 9a45170. Consider uploading reports for the commit 9a45170 to get more accurate results

@@            Coverage Diff            @@
##               main     #311   +/-   ##
=========================================
  Coverage     52.49%   52.49%           
  Complexity      390      390           
=========================================
  Files            30       30           
  Lines          1042     1042           
=========================================
  Hits            547      547           
  Misses          495      495           

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

[escopecz]

I've run the M4.4.10 release on this branch and it ended up with several failures:

https://github.com/mautic/api-library/actions/runs/5736933636/job/15547552524

[RCheesley]

@mollux are you able to take a look at this so we can make the 4.4.10 release?

[mollux]

@escopecz

I found a solution to run the tests both on M4 and M5:

• M5: https://github.com/mautic/api-library/actions/runs/5923005394
• M4: https://github.com/mautic/api-library/actions/runs/5923007453 (using 4.4.10 bump mautic#12617 as reference PR)

changes since last time are:

• awareness of the mautic version and some small test data changes
• removal of paratest to prevent a race condition in M4.
  This results in 30 - 50 seconds more test time, but that's still a huge difference compared to before this PR

The last change should also solve the random test failure on the daily tests, as it's the same race condition

[fedys]

Thanks @mollux! 👍

[escopecz]

Amazing work. Thanks Mattias!