Fix String.offsetByCodePoints for unpaired surrogates

[tanishiking]

fix #3470

Previously, String.offsetByCodePoints had an error when the offsetByCodePoints ends with the unpaired higher surrogate.

❯ cat test.scala
@main def main = {
    println("\uD800".offsetByCodePoints(0, 1))
}

❯ scala-cli test.scala
Compiling project (Scala 3.3.0, JVM)
Compiled project (Scala 3.3.0, JVM)
1

❯ scala-cli test.scala --native
Compiling project (Scala 3.3.0, Scala Native)
Compiled project (Scala 3.3.0, Scala Native)
[info] Linking (756 ms)
[info] Discovered 680 classes and 3771 methods
[info] Optimizing (debug mode) (889 ms)
[info] Generating intermediate code (856 ms)
[info] Produced 8 files
[info] Compiling to native code (1649 ms)
[info] Total (4257 ms)
java.lang.ArrayIndexOutOfBoundsException: 1
        at java.lang.StackTrace$.currentStackTrace$$anonfun$1(Unknown Source)
        at java.lang.StackTrace$$$Lambda$2.applyVoid(Unknown Source)
        at scala.runtime.function.JProcedure1.apply(Unknown Source)
        at scala.scalanative.unsafe.Zone$.apply(Unknown Source)
        at java.lang.StackTrace$.currentStackTrace(Unknown Source)
        at java.lang.Throwable.fillInStackTrace(Unknown Source)
        at scala.scalanative.runtime.package$.throwOutOfBounds(Unknown Source)
        at java.lang.Character$.offsetByCodePoints(Unknown Source)
        at java.lang.String.offsetByCodePoints(Unknown Source)
        at test$package$.main(Unknown Source)
        at main.main(Unknown Source)
        at <none>.main(Unknown Source)

This commit fix the bound check in the offsetByCodePoints not to check the next index is the lower surrogate after the higher surrogate.

[tanishiking]

Though there shouldn't be a difference, it might be easier to understand by

if (next <= end && next < seq.length && isLowerSurrogate(seq(next))) {

?

---

There shouldn't be a difference because we already have a check prior to this line that end <= seq.length which entails next < end <= seq.length

scala-native/javalib/src/main/scala/java/lang/Character.scala

Line 364 in 64b07d6

if (start < 0 || count < 0 || end > seq.length || index < start

[ekrich]

It might be a good idea to check Scala.js code too as most of this was ported from Scala.js. Edit: I checked and it works as expected. It is a good idea to keep that code in sync with Scala.js if able.

[tanishiking]

That's a good idea. I also checked the Scala.js's implementation and it seems they don't have an implementation Character#offsetByCodePoints and there's only a String.offsetByCodePoints.

It might be good to sync after we implement Character.offsetByCodePoints on scala.js side 👍