Integration Test Failures in xArchive

[PlagueHO]

Details of the scenario you tried and the problem that is occurring

Integration tests are failing in xArchive.

Verbose logs showing the problem

https://ci.appveyor.com/project/PowerShell/xpsdesiredstateconfiguration/builds/19771449?fullLog=true#L2834

Suggested solution to the issue

I'll review and correct the unit tests.

The DSC configuration that is used to reproduce the issue (as detailed as possible)

N/A

The operating system the target node is running

AppVeyor.

Version and build of PowerShell the target node is running

AppVeyor

Version of the DSC module that was used ('dev' if using current dev branch)

dev

[PlagueHO]

I'm finding a lot of problems with the archive tests in general. Things like the tests being dependent on the machine date format and failing when run on machines not set to US date format. I'll need to continue to refactor these to try and make them more robust and reliable.

[mhendric]

@PlagueHO , I already opened a similar issue (#457). I also opened up #458 for the other tests that are failing. I spent a decent amount of time trying to troubleshoot these, but was unable to find a root cause. I still suspect something changed on the AppVeyor side, as these work fine locally. I almost think we should just disable these tests until we have a fix, then we can re-enable them. Else the CI is going to continually throw errors for everyone's PR's.

But, until we can get a new maintainer (#466), we can't really move forward :(

[PlagueHO]

Hi @mhendric - I have managed to track this issue down I think. It relates to different behaviors coming out of the System.IO.Compression libraries. Some versions return archive folder paths ending with a forward slash rather than a backslash. This causes the detection of whether or not to create a folder or not.

So this actually relates to (and is solved by) issue #469.

I did see your issue, but only after I'd logged another one (my bad). I do think we should avoid disabling integration tests though.

There is a bunch of others I needed to fix as well that relate to date localization problems. So I wanted to refactor the code to make it easier to maintain and test.

@mgreenegit has offered to help out with some of these critical reviews till we get some maintainers lined up.

[PlagueHO]

Closing - duplicate of #457