chore: resolve minor feedback from #4685

[divmain]

Details

#4685 got merged automatically with a few comments still pending. However, only very minor changes were remaining. Those comments are addressed here.

Does this pull request introduce a breaking change?

• 😮‍💨 No, it does not introduce a breaking change.

Does this pull request introduce an observable change?

• 🤞 No, it does not introduce an observable change.

GUS work item

W-17029474

[wjhsf]

This feels like it's gonna get outdated real fast. 🤷

[nolanlawson]

Agreed, could we just have placeholder/stubs here instead? Literally just an async function* tmpl() {} should be enough to test the basics of transmogrification. Alternatively, call the compiler itself to generate this string.

[nolanlawson]

Although now that I look at it... this is pretty hello-world-ish. So maybe it's fine.

[divmain]

Yeah, I took the most basic fixture we had, modified the customer code to add things like customer-written async generator functions, compiled, and stripped out the parts of the compiled output that I didn't really care about. What is left is not actually runnable code, but retains the pieces that I wanted to make assertions about.

[divmain]

In the future, it might be worth adding some of these customer-written async generator things (or whatever) into a fixture somewhere and actually consuming that stuff programmatically in a way that would effect rendering. Those might be useful additional guardrails that would be a bit more "evergreen" compared to this snippet of code.

[nolanlawson]

I'd still like to see ssrMode folded into the main set of options, but it's not a big deal since this mode may be temporary anyway.

[divmain]

I'd still like to see ssrMode folded into the main set of options, but it's not a big deal since this mode may be temporary anyway.

I'll take a look at this in another follow-up PR. It'll touch packages other than the SSR-related ones, and will probably be easier to review atomically.