🐛 Modules with Trailing Zeroes do not install correctly

[raandree]

Fixes #98

[JustinGrote]

Thanks for your submission!

This is a good start, but I really didn't want to touch disk to read the manifests after each install/update, for the times where someone is doing something like Az or whatnot and installing hundreds of modules. We probably want to at the very least add a -NoTrailingZeroModules switch or something to bypass this check if you know you're not using modules with trailing quad zero modules.

I also don't know if I currently handle the case for if a trailing zero module is already installed and a new trailing zero is existing, both the same version and an older version, so we'll need Pester tests to make sure those all work correctly.

[JustinGrote]

@raandree oops I thought this merged into main, I'll get that done and get you credit.

[raandree]

I thought I should add something like a NoTrailingZeroModules switch first so it doesn't negatively affect performance. Should I still do that?

[JustinGrote]

I thought I should add something like a NoTrailingZeroModules switch first so it doesn't negatively affect performance. Should I still do that?

If you check, I did some updates to switch to just doing a raw parse of the .psd1 file that has way less overhead. Since it occurs at the end of installation and it's part of the parallel process, it's a very minimal overhead compared to doing a full parse of the psd, so I decided it wasn't necessary. My testing was like 5ms in most cases.