refactor(websockets): improve type safety for WebSocketGateway options

[sapenlei]

Closes #14303

PR Checklist

Please check if your PR fulfills the following requirements:

• The commit message follows our guidelines: https://github.com/nestjs/nest/blob/master/CONTRIBUTING.md
• Tests for the changes have been added (for bug fixes / features)
• Docs have been added / updated (for bug fixes / features)

PR Type

What kind of change does this PR introduce?

• Bugfix
• Feature
• Code style update (formatting, local variables)
• Refactoring (no functional changes, no api changes)
• Build related changes
• CI related changes
• Other... Please describe:

What is the current behavior?

Issue Number: #14303

What is the new behavior?

improve type safety for WebSocketGateway options

Does this PR introduce a breaking change?

• Yes
• No

Other information

[coveralls]

Pull Request Test Coverage Report for Build c9edbf1d-dcc3-4eba-8f47-e0946aebec59

Details

• 1 of 1 (100.0%) changed or added relevant line in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 91.912%

---

Totals
Change from base Build bcb74b8d-8301-431a-a72f-c672ad1a017b:	0.0%
Covered Lines:	6818
Relevant Lines:	7418

---

💛 - Coveralls

[sapenlei]

@kamilmysliwiec

[micalevisk]

FYI this does introduces a breaking change as we'll narrow the type in a public API

[sapenlei]

FYI this does introduces a breaking change as we'll narrow the type in a public API

I don't think there's a breaking change. Here's where options are really used.

[sapenlei]

The type of WebSocketGateway options is T extends Record<string, any> = GatewayMetadata , There's no difference between this and T extends Record<string, any>

[kamilmysliwiec]

@sapenlei this introduces a breaking change as you added T extends GatewayMetadata which is a socket.io options object, not Record<string, any>

[kamilmysliwiec]

we can't do extends GatewayMetadata here as GatewayMetadata interface is specific to socket.io. What if someone uses ws instead? (a different adapter)

[sapenlei]

Even for ws, it still calls this method

[kamilmysliwiec]

We should update the generic argument of this method then (as T is not always a subset of GatewayMetadata)