Add unit test to investigate and hopefully verify #2454

[senritsu]

Description

Added a unit test for investigating branch versioning issues, as prompted by @asbjornu in the discussion of #2454.

Related Issue

#2454

Motivation and Context

The test was initially created to reproduce #2454. The test failed when it was initially created, but passes on current main (with some modifications to the version in the issue discussion). There are a few details that I am still not completely clear on:

• Should hotfix branches get a beta prerelease tag? In the new test they don't, but according to this line in another test maybe they should?
• Build metadata does not appear, but that is probably because of the mode CD in the Config used?
• Some of the versioning in the different tests in the file seemed inconsistent when first writing the test, I did not have the time to fully check that this time around, so it could use some more scrutiny
• Some comments need to still be removed before the PR should be merged

Disregarding those details, it seems like #2454 is not an issue anymore (?).

How Has This Been Tested?

The change is a unit test, it was executed

Screenshots (if appropriate):

N/A

Checklist:

• [x] My code follows the code style of this project.
• [ ] ~~My change requires a change to the documentation.~~
• [ ] ~~I have updated the documentation accordingly.~~
• [ ] ~~I have added tests to cover my changes.~~
• [x] All new and existing tests passed.

[asbjornu]

The tests are successful?

[senritsu]

Seems like it, yes. I made a few adjustments because apparently some things changed since the original discussion.

Disregarding the few comments, the test may actually validate that #2454 was unintentionally fixed in the meantime, but I would not hold my breath on it just yet.

[asbjornu]

The successful test may be a false positive as discovered in https://github.com/GitTools/GitVersion/pull/2830#issuecomment-918809996.

[asbjornu]

Can you please rebase against main to see whether the test is still successful?

[senritsu]

I can confirm that the test still passes locally, even with the current changes from main.

I am still unsure if a hotfix branch should have a beta prerelease tag, similar to a release branch. Currently it only has (with versioning mode CD) the same tag like main/master, -ci.n.

When that is clarified I would remove the corresponding comments from the unit test, and the test would seem to confirm that at least one scenario of #2454 appears to work now. Of course that only goes for mode: ContinuousDeployment, and there may be edge cases that are missing.

[senritsu]

Curiously, when trying out 5.7.0 with one of our internal library repositories, the hotfix branch does in fact get a beta tag, albeit without mode: ContinuousDeployment. Maybe this warrants another unit test, for the different modes.

This adds another datapoint and confirms that in at least one of our productive scenarios, #2454 now seems to be fixed.

[senritsu]

Probably an unrelated issue, but the same package that locally works perfectly fine now still fails to properly account for the hotfix on our buildserver (TeamCity). Locally it versions to 0.5.1, on TC it ignores the hotfix and stays on 0.5.0.

[senritsu]

Additional scenarios have been added and both new tests now fail. In both modes the second hotfix is ignored for versioning purposes.

[asbjornu]

Sorry for not having time to investigate or fix this. If you figure out a solution, feel free to update the PR.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. After 30 days from now, it will be closed if no further activity occurs. Thank you for your contributions.

[asbjornu]

As the accompanying issue #2454 is closed with a workaround, is this PR relevant anymore?

[arturcic]

Closes in favor of #3445