https://kotlinlang.org logo
Channels
Powered by
#eap
  • s

    sloth

    01/28/2016, 9:20 PM
    for example
  • v

    vmironov

    01/28/2016, 9:29 PM
  • a

    apatrida

    01/28/2016, 9:36 PM
    @sloth: the repo setup is:
    Copy code
    repositories {
    maven {
        url  "<http://dl.bintray.com/kotlin/kotlin-eap>" 
    }
}
  • s

    sloth

    01/28/2016, 9:37 PM
    SMH wonderful
  • s

    sloth

    01/28/2016, 9:37 PM
    thank you!
  • a

    apatrida

    01/28/2016, 9:39 PM
    @yole, this new warning is supposed to mean something changed for reified type parameters, or just that everyone should now be nervous about them for some reason?
    >> Warning:(28, 13) Kotlin: It may be not safe to use 'uy.kohesive.injekt.tests.TestGithub13.Parser<kotlin.String>' as an argument for a reified type parameter. Use a non-generic type or * if possible
  • a

    apatrida

    01/28/2016, 9:40 PM
    its making all of my libraries that use reified type parameters look dangerous
  • k

    kirillrakhman

    01/28/2016, 9:40 PM
    it means the generic type parameter of your generic type parameter is not actually reified
  • a

    apatrida

    01/28/2016, 9:41 PM
    it is
  • a

    apatrida

    01/28/2016, 9:42 PM
    it was
  • a

    apatrida

    01/28/2016, 9:42 PM
    unless reified type parameters for inline functions has been removed
  • k

    kirillrakhman

    01/28/2016, 9:42 PM
    you can't make an 
    is
    check against it
  • a

    apatrida

    01/28/2016, 9:42 PM
    I’m not 
    is
    checking it
  • a

    apatrida

    01/28/2016, 9:42 PM
    addSingleton(Parser<String>("one-string”))
  • k

    kirillrakhman

    01/28/2016, 9:42 PM
    the compiler doesn't know that
  • a

    apatrida

    01/28/2016, 9:42 PM
    Copy code
    public inline fun <reified T : Any> InjektRegistry.addSingleton(singleInstance: T) {
    addSingleton(fullType<T>(), singleInstance)
}
  • a

    apatrida

    01/28/2016, 9:42 PM
    it knows it
  • a

    apatrida

    01/28/2016, 9:43 PM
    and it is now reporting a useless warning that scares people
  • k

    kirillrakhman

    01/28/2016, 9:43 PM
  • a

    apatrida

    01/28/2016, 9:44 PM
    let me be clear: for all releases of Kotlin since reified type parameters were introduced, these methods existed without warning and worked fine. now they generate a warning that makes no sense. Kotlin guys, why the warning?
  • a

    apatrida

    01/28/2016, 9:45 PM
    is it the 
    String
    or primatives that is the issue?
  • k

    kirillrakhman

    01/28/2016, 9:46 PM
    it's because you instantiate 
    T
    with a generic type
  • a

    apatrida

    01/28/2016, 9:46 PM
    Why the change to add the warning, what does the warning actually mean?
  • k

    kirillrakhman

    01/28/2016, 9:47 PM
  • a

    apatrida

    01/28/2016, 9:47 PM
    6484 talks about things that may or may not be true now from 2014, whether reified are deep by default or not, they have been deeper for a long time now. so this should not be related.
  • a

    apatrida

    01/28/2016, 9:47 PM
    I’m receiving full reified type information fine
  • a

    apatrida

    01/28/2016, 9:47 PM
    but I’m getting a new warning
  • k

    kirillrakhman

    01/28/2016, 9:48 PM
    it was marked as fixed 6 days ago
  • a

    apatrida

    01/28/2016, 9:49 PM
    @udalov: can you answer please, I don’t need guesses, I need to know what is going on with this for sure. It is a problem.
  • a

    apatrida

    01/28/2016, 9:51 PM
    The warning is there for cases that it should not be.
1...848586...106Latest