The methods that change those properties are just shorthands, like

pos(x, y);

to set

x

and

y

etc...

sure, but i guess you would need to refer to the docs to understand "why my visual is still there after i

remove

it".. but i get it though, i mean, you might also what to remove a visual and not hide it... (though i guess that would be more rare? 🤔 )

But yeah I have been hesitating about disabling visuals with no parents

I need to see what would be the implications to that, and would still need to keep a define around to use previous behaviour, if I do this change.

Put it another way add()/remove()/children is a hierarchy thing, not a visibility thing. They are unrelated

A visual is visible by default, regardless of where it is in hierarchy, that's the idea

But I'm not 100% sûre it's the best behaviour, idk. Maybe it's ok. At least it's been working for me so far

yeah, i suppose it make sense when you put it like that, and its fine if you "know", just not sure its what 99% of people would expect (without looking at docs)

not that it matters, now we "know"... so the solution is simple enough

Anyway that's something I have been wondering about, regarding haxe API, without being able to take a definitive decision 😅

i think the problem with automatically deactivating orphaned visuals is that 1% of the time when you didnt want it to happen.. that could be a pain then to work around

but, im now wonder if it is EVER desired behaviour?

like for instance, in your own code, have you ever .remove'd something and NOT set .active=false directly after?

Well a hierarchy has to start somewhere

so a visual without parent is just the root of a visual hierarchy

Scenes don't have parent, so technically they are visual without parent that are hierarchy roots

Displaying a visual without parent is also the most efficient option performance wise

@full-journalist-82607 :

🤣

lol I felt inspired !

they are funny, didnt notice them the first time around 😄

It makes me think of this blog post I read https://ironicsans.substack.com/p/the-strangest-computer-manual-ever

ha, thats funny also

@full-journalist-82607 - https://github.com/haxeui/haxeui-guides/blob/master/001-Guides/004-Styling.md made a few tiny edits... lemme know if its not clear, or not "right" 🙂

No, they are great. I 'll try to use the same formatting next time 🙂

ah okay, that's cool. I switched things to this

I do get why you're doing it this way now tho so thanks for explaining!

HI guys, I'm brand new to haxeui, i started last week and I already love it ❤️ Well i got a question about my actual app (cf image.png) The

TableView

table (id: table) you see in the screenshot is bound directly to a datasource:

table.dataSource = ArrayDataSource.fromArray(mySource);

I am using an Item Renderer to display a property of my source data (

has_stream

) as a checkbox:

<item-renderer verticalAlign="center">
  <checkbox id="has_stream" />
</item-renderer>

That's cool but actually I'd like to display an icon instead, let's say icons/green.png if has_stream == true else

red.png

. Is it possible to do it directly in the .xml view without changing my source structure? By doing some sort of binding expression? Otherwise the only solution I see is to make a wrapper structure (

IDataItem

..) And wrap my mySource data to the wrapper and then bind the data directly but I find it heavy... Thanks you

the syntax from a haxeui css pov is

hx
.checkbox .checkbox-value:selected {
    icon: "image.png";
}

here's a custom theme you can look at as a reference