fix: return correct bounds on will-resize

[deermichel]

Description of Change

Fixes #19611.

Use gfx::ScreenRectFromNSRect to refer to the window's top-left corner as origin instead of the bottom-left one.

No tests since will-resize cannot be triggered programmatically - correct me if i'm wrong :)

cc @codebytere @pmkary

Checklist

• PR description included and stakeholders cc'd
• npm test passes
• PR title follows semantic commit guidelines
• PR release notes describe the change in a way relevant to app developers, and are capitalized, punctuated, and past tense.

Release Notes

Notes: Fixed BrowserWindow's will-resize event returning wrong bounds on macOS.

[alitaheri]

Will this also take position change into account?

I'm not a expert at reading the electron code base. What I'm worried about is that when the user grabs one of the top, top-left, top-right, left, bottom-left corners and tries to resize from there. it's not only the size that changes. the position will change too.

For example:

If I have a window at (x=10, y=10, width=20, height=30) and drag the left edge and move my mouse 5 pixels the newBounds passed to the will-resize callback must be (x=5, y=10, width=25, height=30) this is to keep the right edge at it's place. and it's crucial for our application to actually receive the correct newBounds. otherwise it won't be able to figure out which edge is being dragged.

So, please check this and make sure this behavior is provided. it is on windows.

Thank you very much 👍

[deermichel]

I think so:

[alitaheri]

Perfect 😍 Thank you

[release-clerk]

Release Notes Persisted

Fixed BrowserWindow's will-resize event returning wrong bounds on macOS.

[trop]

I was unable to backport this PR to "5-0-x" cleanly;
you will need to perform this backport manually.

[trop]

I was unable to backport this PR to "6-0-x" cleanly;
you will need to perform this backport manually.

[trop]

I have automatically backported this PR to "7-0-x", please check out #19680

[pouyakary]

So nice, thanks a lot @deermichel 😍

[trop]

A maintainer has manually backported this PR to "6-0-x", please check out #19704

[trop]

A maintainer has manually backported this PR to "5-0-x", please check out #19705

[davej]

I believe there is a bug here. The y value changes on newBounds when the window is dragged from the bottom edge. It can be seen in @deermichel's gif above. This behaviour is also inconsistent with Windows where the y value doesn't change (correctly IMO).

[nrajinikanth]

+1 to @davej comment. When resizing from the bottom edge, I see y origin in newBounds differs from y origin in current bounds. Is there any way to workaround this issue and/or a plan to fix? Looking for some way to detect that a user has dragged the browser window from the bottom edge.