Channels
#announcements
- k
karelpeeters
08/30/2018, 9:50 AM*implicit - d
dharrigan
08/30/2018, 9:50 AMYes, sorry 🙂 That's the word 🙂 - d
dharrigan
08/30/2018, 9:51 AMIn the end, I did this - d
dharrigan
08/30/2018, 9:51 AMCopy codeprivate const val ZERO: Byte = 0 ... ... if (it.value[0] == ZERO).... - k
karelpeeters
08/30/2018, 9:53 AMOr something like
and then useinline val Int.b get() = this.toByte()
for almost-literals.... == 0.b - d
dharrigan
08/30/2018, 9:54 AMI considered using an inline, but it just felt "heavy" as opposed to having a const literal. - d
dharrigan
08/30/2018, 9:54 AM<shrug> different ways of doing the same thing 🙂 - k
karelpeeters
08/30/2018, 9:55 AMthere's an actual command for it
¯\_(ツ)_/¯/shrug👍 1 - d
dharrigan
08/30/2018, 9:55 AMTIL! - d
dharrigan
08/30/2018, 9:55 AM¯\_(ツ)_/¯ - g
Gta
08/30/2018, 12:42 PMHi guys! - h
Hele
08/30/2018, 4:47 PMhey everybody, has anybody ever tried using kotlin on IOS ? - h
Hullaballoonatic
08/30/2018, 9:01 PMsoCopy code
might be written as:override fun compare(a: ByteArray, b: ByteArray): Int { for (i in a.indices) { if (a[i] < b[i]) return -1 if (a[i] > b[i]) return 1 } return 0 }Copy codeoverride fun compare(a: ByteArray, b: ByteArray): Int = 0 { for (i in a.indices) { if (a[i] < b[i]) return -1 if (a[i] > b[i]) return 1 } } - h
Hullaballoonatic
08/30/2018, 9:03 PMof course this syntax might conflict with single line function declarations likeCopy codefun Int.plusTwo() = this + 2 - h
Hullaballoonatic
08/30/2018, 9:06 PMassuming that that wouldn't simply count as a function with a default return value, and thus not requiring functions declared with default return value to have a body - h
Hullaballoonatic
08/30/2018, 9:15 PMbut of course that my be confusing for readability, since you'd have to look at the declaration to see what the return value is in some cases, and those cases may not be immediately clear from the fn body - s
Seri
08/30/2018, 9:26 PMmaybe a new keyword would help? - s
Seri
08/30/2018, 9:26 PMoverride fun compare(a: ByteArray, b: ByteArray): Int = 0 unless { ... }➕ 1🤔 3 - k
karelpeeters
08/30/2018, 9:45 PMOr we could have a special case of a return statement as the last one in the function body being the default return. Oh wait.K 4➕ 4👍 2💕 6😎 1 - p
pablisco
08/31/2018, 10:17 AMWhat do you peeps think? Is this a mad idea? https://twitter.com/pablisc0/status/1035469304697946112 (Posted on wrong channel, oops) - k
karelpeeters
08/31/2018, 10:23 AMYou don't want your code to break all over the place because you added a class somewhere. - k
karelpeeters
08/31/2018, 10:23 AMOr a dependency for that matter. - m
marstran
08/31/2018, 10:29 AMI don't think it could even work. Let's say you have a dependency with a class named "Util", and it's implicitly imported in the library because it's was the only "Util" class when it was compiled. Then you add another dependency, which also happens to have a Util class. That would just break the first dependency. - k
karelpeeters
08/31/2018, 10:31 AMNo, you want packages to be as independent as possible. - p
pablisco
08/31/2018, 10:33 AMMy understanding is that encapsulation in Kotlin favours to use
instead of package private to encourage modularisation...internal - r
robin
08/31/2018, 10:36 AMApart from the obvious implementation problems, why would you want that? Imports are managed transparently by any IDE worth its salt, and any time I do actually have to manually look at the imports it's a situation that would not be solved by your proposal.👍 1 - p
pablisco
08/31/2018, 10:37 AMReduce noise on PRs I guess. 😅 - s
SerVB
08/31/2018, 3:05 PMUnsigned in
are pretty good but what about methods1.3-M2
and.toShort()
? Now I have to do:.toDouble()
.UInt.toLong().toDouble() - k
karelpeeters
08/31/2018, 4:13 PMI think the JVM doesn't support them. - k
karelpeeters
08/31/2018, 4:14 PMAnd sure you could fake it with arrays or classes but then you're hiding performance things.