implicit conversion from Char to String in DataFrame.convertTo<X> { p…

…arser { ... } }
Kotlin · Dec 11, 2024 · 5c54f58 · 5c54f58
1 parent f86a6cc
commit 5c54f58
Show file tree

Hide file tree

Showing 3 changed files with 55 additions and 2 deletions.
diff --git a/core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/api/convertTo.kt b/core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/api/convertTo.kt
@@ -52,7 +52,7 @@ public class ConverterScope(public val fromType: KType, public val toSchema: Col
  * df.convertTo<SomeSchema> {
  *     // defines how to convert Int? -> String
  *     convert<Int?>().with { it?.toString() ?: "No input given" }
- *     // defines how to convert String -> SomeType
+ *     // defines how to convert String -> SomeType (and Char.toString() -> SomeType)
  *     parser { SomeType(it) }
  *     // fill missing column `sum` with expression `a+b`
  *     fill { sum }.with { a + b }
@@ -102,6 +102,10 @@ public fun <T, C> ConvertToFill<T, C>.with(expr: RowExpression<T, C>) {
 
 /**
  * Defines how to convert `String` values into given type [C].
+ *
+ * This method is a shortcut for `convert<String>().with { }`.
+ *
+ * If no converter is defined for `Char` values, this converter will be used for them as well.
  */
 public inline fun <reified C> ConvertSchemaDsl<*>.parser(noinline parser: (String) -> C): Unit =
     convert<String>().with(parser)

diff --git a/core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/impl/api/convertTo.kt b/core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/impl/api/convertTo.kt
@@ -45,8 +45,10 @@ import org.jetbrains.kotlinx.dataframe.schema.ColumnSchema
 import org.jetbrains.kotlinx.dataframe.schema.DataFrameSchema
 import org.jetbrains.kotlinx.dataframe.size
 import kotlin.reflect.KType
+import kotlin.reflect.full.isSubtypeOf
 import kotlin.reflect.full.withNullability
 import kotlin.reflect.jvm.jvmErasure
+import kotlin.reflect.typeOf
 
 private val logger = KotlinLogging.logger {}
 
@@ -143,7 +145,22 @@ internal fun AnyFrame.convertToImpl(
                         // try to perform any user-specified conversions first
                         val from = originalColumn.type()
                         val to = targetSchema.type
-                        val converter = dsl.getConverter(from, targetSchema)
+                        var converter = dsl.getConverter(from, targetSchema)
+
+                        // special case for Char columns; check if we have any converters for String -> target
+                        // if so, we can convert Char -> String -> target
+                        if (converter == null && from.isSubtypeOf(typeOf<Char?>())) {
+                            val stringConverter = dsl.getConverter(
+                                fromType = typeOf<String>().withNullability(from.isMarkedNullable),
+                                toSchema = targetSchema,
+                            )
+                            if (stringConverter != null) {
+                                converter = Converter(
+                                    transform = { stringConverter.transform(this, (it as Char?)?.toString()) },
+                                    skipNulls = stringConverter.skipNulls,
+                                )
+                            }
+                        }
 
                         val convertedColumn = if (converter != null) {
                             val nullsAllowed = to.isMarkedNullable

diff --git a/core/src/test/kotlin/org/jetbrains/kotlinx/dataframe/api/convertTo.kt b/core/src/test/kotlin/org/jetbrains/kotlinx/dataframe/api/convertTo.kt
@@ -51,6 +51,38 @@ class ConvertToTests {
         df.convertTo<Schema> { parser { A(it.toInt()) } }
             .single()
             .a.value shouldBe 1
+
+        // shortcut for:
+        df.convertTo<Schema> { convert<String>().with { A(it.toInt()) } }
+            .single()
+            .a.value shouldBe 1
+    }
+
+    @Test
+    fun `convert from char with parser`() {
+        val df = dataFrameOf("a")('1')
+
+        shouldThrow<TypeConverterNotFoundException> {
+            df.convertTo<Schema>()
+        }
+
+        // Char -> String -> Target
+        df.convertTo<Schema> { parser { A(it.toInt()) } }
+            .single()
+            .a.value shouldBe 1
+
+        // shortcut for:
+        df.convertTo<Schema> { convert<String>().with { A(it.toInt()) } }
+            .single()
+            .a.value shouldBe 1
+
+        // Char -> Target
+        df.convertTo<Schema> {
+            parser<A> { error("should not be triggered if convert<Char>() is present") }
+            convert<String>().with<_, A> { error("should not be triggered if convert<Char>() is present") }
+
+            convert<Char>().with { A(it.digitToInt()) }
+        }.single().a.value shouldBe 1
     }
 
     @Test