added an experiment of compositional map/filter style functions on nullables to mimick Scala's Option behaviour without any wrapper classes

This commit is contained in:
James Strachan
2012-03-17 09:35:35 +00:00
parent d66077ea06
commit ed6ffe34fa
2 changed files with 202 additions and 18 deletions
+184
View File
@@ -0,0 +1,184 @@
package kotlin.nullable
import java.util.*
import kotlin.util.getOrPut
/** Returns true if the element is not null and matches the given predicate */
inline fun <T> T?.any(predicate: (T)-> Boolean): Boolean {
return this != null && predicate(this)
}
/** Returns true if the element is not null and matches the given predicate */
inline fun <T> T?.all(predicate: (T)-> Boolean): Boolean {
return this != null && predicate(this)
}
/** Returns the 1 if the element is not null else 0 */
inline fun <T> T?.count(predicate: (T)-> Boolean): Int {
return if (this != null) 1 else 0
}
/** Returns the first item which matches the predicate if this element is not null else null */
inline fun <T> T?.find(predicate: (T)-> Boolean): T? {
return if (this != null && predicate(this)) this else null
}
/** Returns a new List containing all elements in this collection which match the given predicate */
inline fun <T> T?.filter(predicate: (T)-> Boolean): T? = find(predicate)
/** Filters all elements in this collection which match the given predicate into the given result collection */
inline fun <T, C: Collection<in T>> T?.filterTo(result: C, predicate: (T)-> Boolean): C {
if (this != null && predicate(this))
result.add(this)
return result
}
/** Returns a List containing all the non null elements in this collection */
inline fun <T> T?.filterNotNull(): Collection<T> = filterNotNullTo(java.util.ArrayList<T>())
/** Filters all the null elements in this collection winto the given result collection */
inline fun <T, C: Collection<in T>> T?.filterNotNullTo(result: C): C {
if (this != null) {
result.add(this)
}
return result
}
/** Returns a new collection containing all elements in this collection which do not match the given predicate */
inline fun <T> T?.filterNot(predicate: (T)-> Boolean): Collection<T> = filterNotTo(ArrayList<T>(), predicate)
/** Returns a new collection containing all elements in this collection which do not match the given predicate */
inline fun <T, C: Collection<in T>> T?.filterNotTo(result: C, predicate: (T)-> Boolean): C {
if (this != null && !predicate(this)) {
result.add(this)
}
return result
}
/**
* Returns the result of transforming each item in the collection to a one or more values which
* are concatenated together into a single collection
*/
inline fun <T, R> T?.flatMap(transform: (T)-> Collection<R>): Collection<R> {
return flatMapTo(ArrayList<R>(), transform)
}
/**
* Returns the result of transforming each item in the collection to a one or more values which
* are concatenated together into a single collection
*/
inline fun <T, R> T?.flatMapTo(result: Collection<R>, transform: (T)-> Collection<R>): Collection<R> {
if (this != null) {
val coll = transform(this)
if (coll != null) {
for (r in coll) {
result.add(r)
}
}
}
return result
}
/** Performs the given operation on each element inside the collection */
inline fun <T> T?.forEach(operation: (element: T) -> Unit) {
if (this != null) {
operation(this)
}
}
/**
* Folds all the values from from left to right with the initial value to perform the operation on sequential pairs of values
*
* For example to sum together all numeric values in a collection of numbers it would be
* {code}val total = numbers.fold(0){(a, b) -> a + b}{code}
*/
inline fun <T> T?.fold(initial: T, operation: (it: T, it2: T) -> T): T {
return if (this != null) {
operation(initial, this)
} else {
initial
}
}
/**
* Folds all the values from right to left with the initial value to perform the operation on sequential pairs of values
*/
inline fun <T> T?.foldRight(initial: T, operation: (it: T, it2: T) -> T): T {
// maximum size is 1 so it makes no difference :)
return fold(initial, operation)
}
/**
* Iterates through the collection performing the transformation on each element and using the result
* as the key in a map to group elements by the result
*/
inline fun <T, K> T?.groupBy(result: Map<K, List<T>> = HashMap<K, List<T>>(), toKey: (T)-> K): Map<K, List<T>> {
if (this != null) {
val key = toKey(this)
val list = result.getOrPut(key){ ArrayList<T>() }
list.add(this)
}
return result
}
/** Creates a String from the nullable or item with the given prefix and postfix if supplied */
inline fun <T> T?.join(separator: String, prefix: String = "", postfix: String = ""): String {
val buffer = StringBuilder(prefix)
var first = true
if (this != null) {
buffer.append(this)
}
buffer.append(postfix)
return buffer.toString().sure()
}
/** Returns the nullable result of transforming this with the given transformation function */
inline fun <T, R> T?.map(transform : (T) -> R) : R? {
return if (this != null) {
transform(this)
} else {
null
}
}
/** Transforms each element of this collection with the given function then adds the results to the given collection */
inline fun <T, R, C: Collection<in R>> T?.mapTo(result: C, transform : (T) -> R) : C {
if (this != null) {
result.add(transform(this))
}
return result
}
/** Returns itself since it can't be reversed as it can contain at most one item */
inline fun <T> T?.reverse(): T? {
return this
}
/** Copies the collection into the given collection */
inline fun <T, C: Collection<T>> T?.to(result: C): C {
if (this != null)
result.add(this)
return result
}
/** Converts the collection into a LinkedList */
inline fun <T> T?.toLinkedList(): LinkedList<T> = this.to(LinkedList<T>())
/** Converts the collection into a List */
inline fun <T> T?.toList(): List<T> = this.to(ArrayList<T>())
/** Converts the collection into a Set */
inline fun <T> T?.toSet(): Set<T> = this.to(HashSet<T>())
/** Converts the collection into a SortedSet */
inline fun <T> T?.toSortedSet(): SortedSet<T> = this.to(TreeSet<T>())
/**
TODO figure out necessary variance/generics ninja stuff... :)
inline fun <in T> T?.toSortedList(transform: fun(T) : java.lang.Comparable<*>) : List<T> {
val answer = this.toList()
answer.sort(transform)
return answer
}
*/
@@ -1,5 +1,7 @@
package language.scala
import kotlin.nullable.*
import junit.framework.TestCase
import kotlin.test.assertEquals
@@ -94,32 +96,30 @@ class OptionTest: TestCase() {
fun testFunctionComposition() {
assertEquals("", foo3(Request(null)))
assertEquals("", foo3(Request(" ")))
assertEquals("BAR", foo3(Request(" bar ")))
}
fun foo3(request: Request): String {
/* Scala:
val name:Option[String] = request.getParameter("name")
val upper = name map { _.trim } filter { _.length != 0 } map { _.toUpperCase }
println(upper.getOrElse(""))
*/
/** TODO
The following would work if we implemented the filter/map methods on T?
val name = request.getParameter("name")
val upper = name.map<String, String>{ it.trim() }.filter{ it.length != 0 }.map<String, String>{ it.toUpperCase() }
return upper ?: ""
fun foo(name: String?): String {
val upper = name.map<String,String>{ it.trim() }.filter{ it.length != 0 }.map { it.toUpperCase() }
return upper ?: ""
}
assertEquals("", foo(null))
assertEquals("", foo(" "))
assertEquals("BAR", foo(" bar "))
*/
// TODO...
// TODO when http://youtrack.jetbrains.com/issue/KT-1145 is fixed
// we can get rid of the unnecessary <String, String> on map
}
fun testCompositionWithFor() {
fun foo3(request: Request): String {
fun foo4(request: Request): String {
/* Scala:
val upper = for {
@@ -143,8 +143,8 @@ class OptionTest: TestCase() {
return ""
}
assertEquals("", foo3(Request(null)))
assertEquals("", foo3(Request("")))
assertEquals("BAR", foo3(Request(" bar ")))
assertEquals("", foo4(Request(null)))
assertEquals("", foo4(Request("")))
assertEquals("BAR", foo4(Request(" bar ")))
}
}