Polyfill SubmitEvent for PaleMoon

[wxiaoguang]

Fix #28319

It only polyfills if there is no "SubmitEvent" class, so it has no side effect for most users.

[silverwind]

Maybe outsource the polyfill to https://github.com/idea2app/event-submitter-polyfill?

[silverwind]

Also, I think we should create a web_src/js/polyfills.js and move existing polyfills to it, like this one for Intl.NumberFormat:

https://github.com/go-gitea/gitea/blob/main/web_src/js/webcomponents/polyfill.js

[wxiaoguang]

Maybe outsource the polyfill to https://github.com/idea2app/event-submitter-polyfill?

No, event-submitter-polyfill has many problems. The biggest problem is that IIRC PaleMoon doesn't seem having addEventListener(capture) support.

Also, I think we should create a web_src/js/polyfills.js and move existing polyfills to it, like this one for Intl.NumberFormat:

No, it's not feasible at the moment. Our polyfill patch needs some dependencies between functions. For example, the _submitter, and the index.js code needs to call a helper function. Putting the code together is more maintainable. If you put the code into the webcomponents/polyfill.js, then some polyfill code appears in the webcomponents.js, some polyfill code appears in index.js, it's difficult to read or maintain.

---

[silverwind]

If you put the code into the webcomponents/polyfill.js, then some polyfill code appears in the webcomponents.js, some polyfill code appears in index.js, it's difficult to read or maintain.

Not into webcomponents, but a separate polyfill file in the root. If polyfills are spread out in the code base, it will be harder to remove them later. Prefer to have them all in one file. Ideally only export functions from that file. Can be either ponyfills or polyfills.

[wxiaoguang]

If polyfills are spread out in the code base, it will be harder to remove them later.

This polyfill is special, it already "spreads" because there is no transparent approach, the callers have directly dependency with this change. So I do not see it's worth to move the code again.

If you have better ideas, feel free to edit this PR directly, I do not see necessary changes at the moment.