kotlinlang #arrow

A JNI error has occurred, please check your installation and try again Exception in thread "main" java.lang.VerifyError: Bad type on operand stack Exception Details: Location: personal/MainKt.main()V @267: invokevirtual

pakoito

03/06/2019, 2:02 PM

lol

pakoito

03/06/2019, 2:02 PM

this is beyond Arrow's scope

pakoito

03/06/2019, 2:02 PM

try cleaning up your build, check what JDK version you're using etc

Bob Glamm

03/06/2019, 2:07 PM

that almost seems like a bug in the Kotlin or Java compilers emitting incorrect bytecode

raulraja

03/06/2019, 2:27 PM

@Bob Glamm This works on https://try.arrow-kt.io:80

Copy code

import arrow.core.Some
import arrow.core.None

val remoteInput = Some("blue")
        .flatMap { if(it.isNullOrBlank()) None else Some(it.trim()) }
        .map { it.toLowerCase() }
    
val vi = remoteInput.toEither { Exception("No valid input") }

fun main(args: Array<String>) {
    println(remoteInput)
    println(vi)
}

This is most likely a JDK bug or Kotlin compiler bug in codegen. Do you have a full example including imports?

sitepodmatt

03/06/2019, 2:36 PM

```

sitepodmatt

03/06/2019, 2:36 PM

Copy code

package personal

import arrow.core.None
import arrow.core.Some

fun main() {

    val remoteInput = Some("blue")
        .flatMap { if(it.isNullOrBlank()) None else Some(it.trim()) }
        .map { it.toLowerCase() }
    println(remoteInput)
    val vn = remoteInput.toEither { Exception("No valid input") }
    println(vn)

}

sitepodmatt

03/06/2019, 2:37 PM

kotlin 1.3.21, arrow 0.9.0-snapshot, openjdk 11.0.1 2018-10-16 OpenJDK Runtime Environment (build 11.0.1+13-Ubuntu-3ubuntu3.18.10.1) OpenJDK 64-Bit Server VM (build 11.0.1+13-Ubuntu-3ubuntu3.18.10.1, mixed mode, sharing)

sitepodmatt

03/06/2019, 2:38 PM

gradle_build_kts.scala

Bob Glamm

03/06/2019, 2:42 PM

Hmm, I can run that exact code just fine on Kotlin 1.3.21, arrow-0.8.2, JDK 10

sitepodmatt

03/06/2019, 2:53 PM

okay maybe something with my setup, ill try again in the morning

sitepodmatt

03/06/2019, 3:03 PM

in Arrow's either case when mapping/handling the left/sad side, and you're able to recover and rejoin the the right side (for futher mapping) how would you do this with arrow

sitepodmatt

03/06/2019, 3:11 PM

.handleError { "NewRightSideValue" } seems to do it

sitepodmatt

03/06/2019, 3:52 PM

Copy code

val remoteInput = Option.fromNullable(getRemoteInput())
        val output  = remoteInput
            .toEither { Exception("No input") }
            .flatMap { input ->
                if(input.isNullOrBlank()) Exception("Blank input").left() else input.trim().right()
            }
            .handleErrorWith {
                // do some logging
              Option.fromNullable(getAlternativeRemoteInput()).toEither { Exception("No input")}... and so on
            }

sitepodmatt

03/06/2019, 3:52 PM

what about something like this where we are about repeat ourselves for the retry attempt

pakoito

03/06/2019, 4:01 PM

i'd use tailrecM for that

pakoito

03/06/2019, 4:01 PM

you call yourself until you get a value

pakoito

03/06/2019, 4:01 PM

or your retries count down

sitepodmatt

03/06/2019, 4:08 PM

thanks

pakoito

03/06/2019, 4:11 PM

no, I'd always use

Functor.map

Applicative.map

sitepodmatt

03/06/2019, 4:12 PM

okay, np, I'm confused by ap and was trying to find an example in flatMap/map adventures

pakoito

03/06/2019, 4:12 PM

myEither.map { bla }

map(oneEither, otherEither) { one, two -> bla(one + two) }

pakoito

03/06/2019, 4:13 PM

if you're using functions that return functions maybe it'd work, but it's uncommon in kotlin

Hexa

03/07/2019, 7:13 PM

is there a more efficient way of doing this using Kotlin arrows?

Copy code

// event: List<KinesisEvent.KinesisEventRecord>

 event.records.forEach {
            try {
                val putRecordsRequest = PutRecordsRequest()
                putRecordsRequest.streamName = "test-stream" 
                val putRecordsRequestEntryList = ArrayList<PutRecordsRequestEntry>()
                for (i in 0..499) {
                    val putRecordsRequestEntry = PutRecordsRequestEntry()
                    putRecordsRequestEntry.data = it.kinesis.data
                    putRecordsRequestEntry.partitionKey = it.kinesis.partitionKey
                    putRecordsRequestEntryList.add(putRecordsRequestEntry)
                }
                putRecordsRequest.setRecords(putRecordsRequestEntryList)
                val putRecordsResult = kinesisClient.putRecords(putRecordsRequest)

            } catch (e: InterruptedException) {
                //log error here
            }

Hexa

03/07/2019, 7:14 PM

is it worth calling

kinesisEvent.records.asSequence()

before calling foreach?

pakoito

03/07/2019, 7:41 PM

there’s a lot to unpack there

pakoito

03/07/2019, 7:41 PM

so you have a lists of records

pakoito

03/07/2019, 7:42 PM

and you want to send 500 of them into a request