deprecated DataColumn.createFrameColumn with startIndices, moved it to chunked

Author: Jolanrensen

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/DataColumn.kt

@@ -103,7 +103,11 @@ public interface DataColumn<out T> : BaseColumn<T> {
          */
         public fun <T> createColumnGroup(name: String, df: DataFrame<T>): ColumnGroup<T> = ColumnGroupImpl(name, df)
 
-        // TODO this shouldn't be here
+        @Deprecated(
+            message = CREATE_FRAME_COLUMN,
+            replaceWith = ReplaceWith(CREATE_FRAME_COLUMN_REPLACE, CREATE_FRAME_COLUMN_IMPORT),
+            level = DeprecationLevel.WARNING,
+        )
         public fun <T> createFrameColumn(name: String, df: DataFrame<T>, startIndices: Iterable<Int>): FrameColumn<T> =
             FrameColumnImpl(name, df.splitByIndices(startIndices.asSequence()).toList(), lazy { df.schema() })
 

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/api/chunked.kt

@@ -7,12 +7,28 @@ import org.jetbrains.kotlinx.dataframe.columns.ColumnGroup
 import org.jetbrains.kotlinx.dataframe.columns.FrameColumn
 import org.jetbrains.kotlinx.dataframe.columns.ValueColumn
 import org.jetbrains.kotlinx.dataframe.impl.getListType
+import org.jetbrains.kotlinx.dataframe.impl.splitByIndices
 import org.jetbrains.kotlinx.dataframe.nrow
 import org.jetbrains.kotlinx.dataframe.type
 
+/**
+ * Creates a [FrameColumn] from [this] by splitting the dataframe into
+ * smaller ones, based on the given [startIndices].
+ */
+public fun <T> DataFrame<T>.chunked(startIndices: Iterable<Int>, name: String = "groups"): FrameColumn<T> =
+    DataColumn.createFrameColumn(
+        name = name,
+        groups = this.splitByIndices(startIndices.asSequence()).toList(),
+        schema = lazy { this.schema() },
+    )
+
+/**
+ * Creates a [FrameColumn] from [this] by splitting the dataframe into
+ * smaller ones, with their number of rows at most [size].
+ */
 public fun <T> DataFrame<T>.chunked(size: Int, name: String = "groups"): FrameColumn<T> {
     val startIndices = (0 until nrow step size)
-    return DataColumn.createFrameColumn(name, this, startIndices)
+    return this.chunked(startIndices, name)
 }
 
 public fun <T> DataColumn<T>.chunked(size: Int): ValueColumn<List<T>> {

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/impl/api/groupBy.kt

@@ -1,12 +1,12 @@
 package org.jetbrains.kotlinx.dataframe.impl.api
 
 import org.jetbrains.kotlinx.dataframe.ColumnsSelector
-import org.jetbrains.kotlinx.dataframe.DataColumn
 import org.jetbrains.kotlinx.dataframe.DataFrame
 import org.jetbrains.kotlinx.dataframe.DataRow
 import org.jetbrains.kotlinx.dataframe.api.GroupBy
 import org.jetbrains.kotlinx.dataframe.api.GroupedDataRow
 import org.jetbrains.kotlinx.dataframe.api.cast
+import org.jetbrains.kotlinx.dataframe.api.chunked
 import org.jetbrains.kotlinx.dataframe.api.getColumnsWithPaths
 import org.jetbrains.kotlinx.dataframe.api.getRows
 import org.jetbrains.kotlinx.dataframe.api.indices
@@ -62,7 +62,7 @@ internal fun <T> DataFrame<T>.groupByImpl(moveToTop: Boolean, columns: ColumnsSe
     }
 
     val groupedColumnName = keyColumnsDf.nameGenerator().addUnique(GroupBy.groupedColumnAccessor.name())
-    val groupedColumn = DataColumn.createFrameColumn(groupedColumnName, sorted, startIndices.asIterable())
+    val groupedColumn = sorted.chunked(startIndices.asIterable(), groupedColumnName)
 
     val df = keyColumnsDf + groupedColumn
     return GroupByImpl(df, groupedColumn, columns)

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/impl/io/readJson.kt

@@ -23,6 +23,7 @@ import org.jetbrains.kotlinx.dataframe.DataRow
 import org.jetbrains.kotlinx.dataframe.api.JsonPath
 import org.jetbrains.kotlinx.dataframe.api.KeyValueProperty
 import org.jetbrains.kotlinx.dataframe.api.cast
+import org.jetbrains.kotlinx.dataframe.api.chunked
 import org.jetbrains.kotlinx.dataframe.api.columnOf
 import org.jetbrains.kotlinx.dataframe.api.concat
 import org.jetbrains.kotlinx.dataframe.api.dataFrameOf
@@ -298,11 +299,12 @@ internal fun fromJsonListAnyColumns(
                     )
                 }
 
-                else -> DataColumn.createFrameColumn(
-                    name = ARRAY_COLUMN_NAME, // will be erased
-                    df = parsed.unwrapUnnamedColumns(),
-                    startIndices = startIndices,
-                )
+                else ->
+                    parsed.unwrapUnnamedColumns()
+                        .chunked(
+                            startIndices = startIndices,
+                            name = ARRAY_COLUMN_NAME, // will be erased
+                        )
             }
             listOf(UnnamedColumn(res))
         }
@@ -640,7 +642,7 @@ internal fun fromJsonListArrayAndValueColumns(
                                 )
                             }
 
-                            else -> DataColumn.createFrameColumn(colName, parsed.unwrapUnnamedColumns(), startIndices)
+                            else -> parsed.unwrapUnnamedColumns().chunked(startIndices, colName)
                         }
                         UnnamedColumn(res)
                     }

core/src/main/kotlin/org/jetbrains/kotlinx/dataframe/util/deprecationMessages.kt

@@ -15,6 +15,10 @@ internal const val DF_READ_NO_CSV = "This function is deprecated and should be r
 internal const val DF_READ_NO_CSV_REPLACE =
     "this.readCSV(fileOrUrl, delimiter, header, colTypes, skipLines, readLines, duplicate, charset)"
 
+internal const val CREATE_FRAME_COLUMN = "Replaced by df.chunked(). $MESSAGE_0_16"
+internal const val CREATE_FRAME_COLUMN_REPLACE = "df.chunked(startIndices, name)"
+internal const val CREATE_FRAME_COLUMN_IMPORT = "org.jetbrains.kotlinx.dataframe.api.chunked"
+
 // endregion
 
 // region WARNING in 0.16, ERROR in 0.17