https://linen.dev logo
Join Discord
Powered by
# haxe-ui
  • b

    bitter-family-72722

    09/17/2019, 5:33 PM
    with the shallow clones that lix does you can also revert to older states
  • b

    bitter-family-72722

    09/17/2019, 5:33 PM
    as long as nobody takes down the project from GitHub / whereever it's hosted
  • u

    user

    09/17/2019, 5:38 PM
    oh ok; good to know
  • r

    rich-plastic-63351

    09/17/2019, 7:42 PM
    not sure it is the case with per-project installation, but when installed globally, the lix clones/version are shared between projects so it saves space
  • b

    bitter-family-72722

    09/17/2019, 7:43 PM
    it is the case for both local and global installations
  • r

    rich-plastic-63351

    09/17/2019, 10:33 PM
    oh nice ! will try the local install workflow then..
  • b

    brave-kangaroo-30399

    09/20/2019, 12:52 AM
    FYI I’m working on updating the flixel backend, but it turns out writing a new base class is not easy
  • b

    brave-kangaroo-30399

    09/20/2019, 1:02 AM
    I’m a bit stuck between FlxSprite and Kha implementations of applying styles...
  • o

    orange-van-60470

    09/20/2019, 1:03 AM
    Lix should not depend on Node, it's really feasible to change this.
  • b

    brave-kangaroo-30399

    09/20/2019, 1:03 AM
    FlxSprite has a bitmap that effects would go onto, Kha uses renderTo to redraw/reapply styles each time
  • o

    orange-van-60470

    09/20/2019, 1:04 AM
    What do you mean by styles I know Kha and flash but not fixel
  • b

    brave-kangaroo-30399

    09/20/2019, 1:05 AM
    HaxeUI styles
  • o

    orange-van-60470

    09/20/2019, 1:06 AM
    If your looking at a fixel target suspect you have to look deeper into lime/openfl
  • o

    orange-van-60470

    09/20/2019, 1:07 AM
    I browsed HaxeUI to see how I could implement it with trilateral
  • o

    orange-van-60470

    09/20/2019, 1:08 AM
    I think maybe the styles abstraction needs some thought it did not at the time seem suitable to abstract over.
  • o

    orange-van-60470

    09/20/2019, 1:10 AM
    Have you got something working in Both yet?
  • b

    brave-kangaroo-30399

    09/20/2019, 1:11 AM
    I have a working flixel backend, I just need to update it to be cleaner and integrate better
  • o

    orange-van-60470

    09/20/2019, 1:12 AM
    Does it need Fixel could it be implemented for any OpenFL or NME project?
  • o

    orange-van-60470

    09/20/2019, 1:13 AM
    I mean add Fixel abilities over but does it actually need fixel just to work?
  • b

    brave-kangaroo-30399

    09/20/2019, 1:23 AM
    There’s a backend for every target
  • b

    brave-kangaroo-30399

    09/20/2019, 1:23 AM
    The one I’m talking about is specifically for flixel
  • b

    brave-kangaroo-30399

    09/20/2019, 1:26 AM
    @bright-gpu-74537 does the kha backend’s
    renderTo
    get called every frame?
  • b

    brave-kangaroo-30399

    09/20/2019, 1:27 AM
    As in, it’s redrawing the entire UI every frame?
  • b

    brave-kangaroo-30399

    09/20/2019, 1:30 AM
    Ultimately I’m wondering if my flixel class should have a bitmapdata backing it (like FlxSprite) or if it should draw to the flixel camera directly
  • w

    wooden-horse-84574

    09/28/2019, 7:09 AM
    Woo event bubbling problems: I have an event on a "root" component for mousedown. Some of it's children also have mousedown events; and these all work as I would expect. HOWEVER, I have input fields, which seem to pass the mouse events down to the root. If I catch them in a parent (ie, a component that contains the input field, but is a child of the root) and cancel the event, the input field stops working as expected.
  • w

    wooden-horse-84574

    09/28/2019, 7:09 AM
    I tried setting bubble=false but this doesn't appear to do anything.
  • w

    wooden-horse-84574

    09/28/2019, 7:10 AM
    What's the "correct" way to fix this, such that any unhandled mousedown events in the container don't get passed down to the root?
  • w

    wooden-horse-84574

    09/28/2019, 7:13 AM
    I also tried adding a check to the "root" to check that target == this, but that always passes
  • w

    wooden-horse-84574

    09/28/2019, 7:23 AM
    Fixed it. Sorta. I'm using the html5 target, which exposes
    _originalEvent
    , which means I can just do
    event._originalEvent.stopPropagation();
    which does the correct thing.
    event.cancel()
    does this, but also calls
    preventDefault()
    which causes the aforementioned problem with input fields.
  • b

    bright-gpu-74537

    09/29/2019, 6:41 AM
    @wooden-horse-84574 - thats interesting, care to open a PR or an issue? 🙂
1...119120121...1687Latest