change Image example

[JAicewizard]

This changes the image example to showcase FillStrat and Interpolation mode.

[cmyr]

Cool, I think showing more off here makes sense. If we wanted to get fancy we could also do something here like we do in examples/flex.rs, where we actually let the user choose the FillStrat and InterpolationMode from some radio buttons, and then rebuild the widget in response to that?

[JAicewizard]

I reworked it to be more like the flex, that was a great idea. You can now play around with width/height interpolation mode and fillstrat dynamically

[cmyr]

Cool, I think this is a great improvement.

Playing around with it: the default settings might be confusing, since changing many of the options doesn't actually change anything.

I think we can maybe make it a bit clearer? Here's what I would probably do:

• remove the 'debug layout' option, and just always paint the border of the image widget, which should make it easier to see what's going on.
• I would check 'fix width' and 'fix height' by default, and give 'height' and 'width' some non-zero starting values; maybe something like 200 & 320. (making them non-uniform will make it more obvious what's happening with the various fill strategies.

Other than that, the code looks good, just a couple of notes in-line.

Thanks!

[cmyr]

we should be able to derive this, like:

#[derive(Clone, Data)]
 struct InterpolationModeData(
    #[data(same_fn="PartialEq::eq")]
    InterpolationMode
);

I would also be okay with us adding a Data impl to InterpolationMode, I did this for the various types in examples/flex.rs for this same reason.

[JAicewizard]

yes that would be great! I think it makes sense to have InterpolationMode in data for some things, its not something we want to prevent people from putting in there.

[cmyr]

Did we make this change? If we did we can also probably get rid of this InterpolationModeData type, and just use InterpolationMode directly?

[JAicewizard]

InterpolationMode does not have the data trait implemented for it, yet. It would be nice if that would be possible to do.
I can create a PR for it later?

EDIT: I can also put it in this PR, its not a big change and then I don't have to rebase.

[cmyr]

@JAicewizard yes feel free to add it to this PR. Also if merge in the latest master it should fix that problem with CI failing.

[JAicewizard]

had to rebase off master but forgot to pull this first, there was a commit I hadn't pulled apparently

[JAicewizard]

the failing ones are getrandom not being compatible with wasm(?)

[cmyr]

that getrandom error is strange, since we don't see it on master and I don't think anything should have changed??

[cmyr]

Okay one fixup and one question about just using InterpolationMode, but other than that (and assuming we can figure out why CI is failing) this looks good. Thanks!

[cmyr]

Thanks!