This may affect tools not using an actual PS parser, but in general I think the impact will be minor overall, likely requiring some updates to any regex recognition patterns for some syntax highlighting rules.

Yeah this will possibly need a change to EditorSyntax.

@vexx32 There are many formatting and style issue. Please fix issues for changed code and move unrelated ones to another PR (I mean specially build.psm1 and some in other code). It will very help reviewers!

@iSazonov ah, you're right. I'll have to do a bit of a rebase to clear that up, give me a few moments. I think a lot of those came from Dave's original code here, so I'll pare out what's needed and drop the unrelated style changes. Thanks! 💖

@iSazonov that should do it. Time for bed. 😁

@rjmholt Can you review?

This pull request has been automatically marked as Review Needed because it has been there has not been any activity for 7 days.
Mainainer, Please provide feedback and/or mark it as Waiting on Author

@rjmholt Could you review this please?

I realize this question is probably beyond the scope of this PR, but I feel it's still relevant:

Does it make sense to attempt coercion of enumerables passed as arguments to method parameters constrained by generic type parameters when overload resolution is not otherwise ambiguous?

Take the following example:

$linq = [System.Linq.IEnumerable]

# this fails because [object[]] doesn't implement [IEnumerable[int]] (not because of ambiguous overload resolution)
$linq::Distinct[int](@(1,2,3,1,2,3))

# this succeeds
$linq::Distinct[int]([int[]]@(1,2,3,1,2,3))

Edit: as soon as I posted this I realized what I'm really asking for is "can we have better coercion semantics for downcasting to interfaces with covariant generic type parameters?" which is sorta orthogonal to this change and should be chased separately, please ignore this :)

@IISResetMe I agree we should have that, and agree it's out of this scope... but I would really love to have that. Because yeah, in some method calls it does feel like casts should not be necessary. Maybe would be good to file a follow up issue with that? I think it'll be easier to manage with the changes from this PR actually defining some of the types needed for the method call, so PS has less to infer from just the arguments.

We need @andschwa review too (for possible PSES breaking).

What are the concerns around breaking PSES?

We need @andschwa review too (for possible PSES breaking).

What are the concerns around breaking PSES?

Changes in Parser.

Thanks @vexx32 for your contribution!

can we have better coercion semantics for downcasting to interfaces with covariant generic type parameters?

@IISResetMe Feel free to open an issue for discussion about this 😄

@andschwa I guess EditorSyntax would need updates? If so, can you please open an issue tracking that?
@JamesWTruher Not sure how much this would affect PSSA. It would be great if you can help evaluate and open tracking issue as needed.

Just noticed no documentation issue was filed. @vexx32 can you please open a documentation issue for this feature?

Aye, will do!

EDIT: Doc issue filed: MicrosoftDocs/PowerShell-Docs#8533

w00000000.

Congrats @vexx32

@vexx32 Finally! This is awesome!