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

    quick-king-64105

    04/17/2019, 3:30 AM
    It's a little temperamental in the flixel backend, but as far as I can tell, it's in good shape in most places. And even in the flixel backend, it's one of the cleanest UI libraries I've dealt with in a minute. If you want my opinion, it's worth trying, shows promise, and there are far worse things that I've used in the past.
  • c

    cool-psychiatrist-49311

    04/17/2019, 5:22 AM
    The better question is: when will haxeui get out of "alpha" stage.
  • c

    cool-psychiatrist-49311

    04/17/2019, 5:23 AM
    perhaps pick some "core" backends and make a beta release. it is not possible for a single man to maintain a dozen of backends
  • h

    handsome-television-62908

    04/17/2019, 7:22 AM
    he seems to be doing a pretty decent job besides that though . . . but yeah. . I imagine most people want to use this for building real applications and having so many backends detracts from the primary product of "building an app"
  • q

    quick-king-64105

    04/17/2019, 7:39 AM
    > implying that games aren't real applications > implying that many Haxe developers aren't game developers BRB, dialing the BBS on the party line.
  • q

    quick-king-64105

    04/17/2019, 7:40 AM
    Having so many backends is because Haxe is a fairly open ecosystem - there's no "narrow focus" so to speak.
  • q

    quick-king-64105

    04/17/2019, 7:41 AM
    A UI library that speaks the native color and behavior of whatever backend you're using is extremely useful - it means that you have to learn just the extra bit that is the UI library, not every little thing about it. (If you've never been alienated by a UI library's API, you've never tried to understand gobject.)
  • h

    handsome-television-62908

    04/17/2019, 7:41 AM
    You know what I mean lol
  • h

    handsome-television-62908

    04/17/2019, 7:41 AM
    i also get what you're saying haha
  • q

    quick-king-64105

    04/17/2019, 7:42 AM
    Frankly, I've yet to meet a UI library I like, but at least this one is clearly trying, and trying hard.
  • q

    quick-king-64105

    04/17/2019, 7:43 AM
    As far as maintaining backends, that's a complicated picture I don't understand, but as I do understand the component methods were changed recently enough to have a branch entitled "new component method".
  • q

    quick-king-64105

    04/17/2019, 7:43 AM
    Once the expected underlying implementation is stabilized, it's fairly good odds that implementations will arise.
  • q

    quick-king-64105

    04/17/2019, 7:44 AM
    Nor are implementations that large.
  • q

    quick-king-64105

    04/17/2019, 7:44 AM
    https://github.com/haxeui/haxeui-flixel/tree/master/haxe/ui/backend
  • q

    quick-king-64105

    04/17/2019, 7:44 AM
    Most of those files are fulla one-liners last I checked.
  • q

    quick-king-64105

    04/17/2019, 7:46 AM
    Beyond just this, mixing GUI libraries is bad. If you want a de-facto library for all of haxe, it must mix well with anything else managing graphics heavily in the ui.
  • q

    quick-king-64105

    04/17/2019, 7:46 AM
    I should know - my current project uses two GUI libraries, and it's dreadful.
  • q

    quick-king-64105

    04/17/2019, 7:46 AM
    *eyebrow wiggle
  • q

    quick-king-64105

    04/17/2019, 7:47 AM
    I'mma go back to my coffee and not dumping long pages of data at people now.
  • b

    bright-gpu-74537

    04/17/2019, 11:54 AM
    I agree with pretty much all points above... esp. the "pick a few backends and go with a beta"... that said, having all the different backends is actually pretty useful as it means the core lib (haxeui-core) is doing its job well of abstracting a UI
  • b

    bright-gpu-74537

    04/17/2019, 11:55 AM
    the biggest friction with all the backends is if you change (or want to change) something stucturally... you have then broken ALL backends, sometimes this just cant be avoided, but its a real pain.... ... ... i think its something ive improved upon however, but im sure it will be gripe in the future from time to time also
  • b

    bright-gpu-74537

    04/17/2019, 11:57 AM
    also regarding backends: for the most part i rarely touch them (really)... i mean, from time to time - again, structurally, but for the most part they are pretty static
  • b

    bright-gpu-74537

    04/17/2019, 11:58 AM
    bulk of the time is just adding, enhancing the main core... which itself is time consuming (UI libs being pretty big), and time is a valuable resource that i dont have a huge amount of... that all said and done, i certainly agree with Kevin in "get it into beta" as well as "stop pissing around with new-component-method branch and make it master"... ... then PRs can start "rolling in" (?) 😃
  • b

    bitter-family-72722

    04/17/2019, 12:00 PM
    it definitely feels a bit like most people are waiting on the sidelines for this to get out of alpha 😃
  • b

    bitter-family-72722

    04/17/2019, 12:00 PM
    and if you want to try it out but are not comfortable with installing stuff from GitHub, the haxelib releases aren't very up to date
  • b

    bright-gpu-74537

    04/17/2019, 12:01 PM
    totally, i dont even remember when there was a haxelib release... in fact, i cant even make one (as "haxeui") as the password was generated and was on my old SSD - so poof gone
  • b

    bright-gpu-74537

    04/17/2019, 12:01 PM
    will have to get that reset somehow
  • b

    bitter-family-72722

    04/17/2019, 12:02 PM
    always use LastPass / some password manager
  • b

    bitter-family-72722

    04/17/2019, 12:02 PM
    btw, you might get some competition soon 😉 (I assume you've seen that FeathersUI announcement)
  • b

    bright-gpu-74537

    04/17/2019, 12:02 PM
    yeah, well, if i had a proper backup system (which i have now) it will all be academic
1...567...1687Latest