adds kotlin.dataframe.debug property which runs the #713 checks when …

…enabled. Updated contribution guide too. TC will have to be updated manually

CONTRIBUTING.md

@@ -52,8 +52,8 @@ so do familiarize yourself with the following guidelines.
 
 ## PR workflow
 
-0. The contributor builds the library locally and runs all unit tests via the Gradle task `dataframe:test`
-   (see the ["Building"](#building) chapter).
+0. The contributor builds the library locally and runs all unit tests via the Gradle task 
+   `dataframe:test -Pkotlin.dataframe.debug=true` (see the ["Building"](#building) chapter).
 1. The contributor submits the PR if the local build is successful and the tests are green.
 2. The reviewer puts their name in the "Reviewers" section of the proposed PR at the start of the review process.
 3. The reviewer leaves comments or marks the PR with the abbreviation "LGTM" (Looks good to me).
@@ -103,6 +103,8 @@ This library is built with Gradle.
 * Run `./gradlew build` to build. It also runs all the tests and checks the linter.
 * Run `./gradlew <module>:test` to test the module you are looking at to speed
   things up during development.
+* Make sure to pass the extra parameter `-Pkotlin.dataframe.debug=true` to enable debug mode. This flag will
+  make sure some extra checks are run, which are important but too heavy for production.
 
 You can import this project into IDEA, but you have to delegate the build actions
 to Gradle (in Preferences -> Build, Execution, Deployment -> Build Tools -> Gradle -> Runner)

build.gradle.kts

@@ -163,6 +163,7 @@ allprojects {
                 packageName = "org.jetbrains.kotlinx.dataframe"
                 className = "BuildConfig"
                 buildConfigField("VERSION", "${project.version}")
+                buildConfigField("DEBUG", findProperty("kotlin.dataframe.debug")?.toString()?.toBoolean() ?: false)
             }
         } catch (_: UnknownDomainObjectException) {
             logger.warn("Could not set buildConfig on :${this.name}")

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/impl/columns/DataColumnImpl.kt

@@ -1,8 +1,13 @@
 package org.jetbrains.kotlinx.dataframe.impl.columns
 
+import org.jetbrains.kotlinx.dataframe.BuildConfig
 import org.jetbrains.kotlinx.dataframe.DataColumn
 import org.jetbrains.kotlinx.dataframe.api.dataFrameOf
+import org.jetbrains.kotlinx.dataframe.impl.isArray
+import org.jetbrains.kotlinx.dataframe.impl.isPrimitiveArray
+import kotlin.reflect.KClass
 import kotlin.reflect.KType
+import kotlin.reflect.full.isSubclassOf
 
 internal abstract class DataColumnImpl<T>(
     protected val values: List<T>,
@@ -12,6 +17,32 @@ internal abstract class DataColumnImpl<T>(
 ) : DataColumn<T>,
     DataColumnInternal<T> {
 
+    private infix fun <T> T?.matches(type: KType) =
+        when {
+            this == null -> type.isMarkedNullable
+
+            this.isPrimitiveArray ->
+                type.isPrimitiveArray &&
+                    this!!::class.qualifiedName == type.classifier?.let { (it as KClass<*>).qualifiedName }
+
+            this.isArray -> type.isArray
+
+            // cannot check the precise type of array
+            else -> this!!::class.isSubclassOf(type.classifier as KClass<*>)
+        }
+
+    init {
+        /* Check for [Issue #713](https://github.com/Kotlin/dataframe/issues/713).
+         * This only runs with `kotlin.dataframe.debug=true` in gradle.properties
+         */
+        if (BuildConfig.DEBUG) {
+            require(values.all { it matches type }) {
+                val types = values.map { if (it == null) "Nothing?" else it!!::class.simpleName }.distinct()
+                "Values of column '$name' have types '$types' which are not compatible given with column type '$type'"
+            }
+        }
+    }
+
     protected val distinct = distinct ?: lazy { values.toSet() }
 
     override fun name() = name

gradle.properties

@@ -11,3 +11,8 @@ org.gradle.jvmargs=-Xmx4G
 # This makes it mandatory to explicitly apply your own version of the
 # KSP plugin in the modules that use it.
 kotlin.dataframe.add.ksp=false
+
+# Enables debug mode for dataframe.
+# This can make certain tests run that should not be run in production.
+# It can also be turned on from the command line with `-Pkotlin.dataframe.debug=true`
+kotlin.dataframe.debug=false