Fix generic props structs

[WorldSEnder]

Description

Fixup of #2729 for generic property structs. I mistakenly converted the checks guarded by if false to a const _: () check, which makes generic arguments from the context invisible, leading to the problem

This also for some reason changes the order in which errors are reported with trybuild.

Checklist

• I have reviewed my own code
• I have added tests

[github-actions]

Visit the preview URL for this PR (updated for commit b83695f):

https://yew-rs-api--pr2746-prop-inheritance-zq5xlke7.web.app

_{(expires Wed, 29 Jun 2022 20:30:53 GMT)}

_{🔥 via Firebase Hosting GitHub Action 🌎}

[WorldSEnder]

@maurerdietmar if you have time to check, I think this fixes your issue.

@hamza1311 aha! the issue with time updating its msrv to 1.57 actually also means wasm-bindgen-cli doesn't build on 1.56. I suppose bumping is a work-around, otherwise we could also install that always built with the stable version.

[github-actions]

Size Comparison

examples	master (KB)	pull request (KB)	diff (KB)	diff (%)
boids	172.581	172.581	0	0.000%
contexts	109.618	109.618	0	0.000%
counter	86.551	86.551	0	0.000%
counter_functional	87.200	87.200	0	0.000%
dyn_create_destroy_apps	89.692	89.692	0	0.000%
file_upload	102.623	102.623	0	0.000%
function_memory_game	166.885	166.885	0	0.000%
function_router	351.672	351.672	0	0.000%
function_todomvc	161.559	161.559	0	0.000%
futures	226.233	226.233	0	0.000%
game_of_life	107.214	107.214	0	0.000%
inner_html	83.617	83.617	0	0.000%
js_callback	112.844	112.844	0	0.000%
keyed_list	195.104	195.104	0	0.000%
mount_point	86.181	86.181	0	0.000%
nested_list	115.691	115.691	0	0.000%
node_refs	93.596	93.596	0	0.000%
password_strength	1538.154	1538.154	0	0.000%
portals	97.225	97.225	0	0.000%
router	320.455	320.455	0	0.000%
simple_ssr	154.493	154.493	0	0.000%
ssr_router	398.015	398.015	0	0.000%
suspense	110.486	110.486	0	0.000%
timer	89.258	89.258	0	0.000%
todomvc	142.615	142.615	0	0.000%
two_apps	87.157	87.157	0	0.000%
web_worker_fib	153.404	153.404	0	0.000%
webgl	87.439	87.439	0	0.000%

✅ None of the examples has changed their size significantly.

[hamza1311]

Looks good

The comments are unrelated to this PR but should be resolved (in a future PR). I must've missed the output spans while reviewing the previous one

[hamza1311]

This is a bad span. Can we do better?

[WorldSEnder]

I agree in principle, but don't want to block this PR with this. The errors were like this before, even before #2729 . This PR only affects the order in which errors are reported, since they emit from a non-const context again. Don't ask me why that changes the order of error reporting though, idk :)

Note that this bad span is only for field children. Not sure what's a better span here.

[hamza1311]

Span should reference MissingTypeBounds here

[maurerdietmar]

@maurerdietmar if you have time to check, I think this fixes your issue.

Yes, this fixes my issue.