Append dataframe rows based on column names (#6808)

* Append dataframe rows based on column names

* Update DataFrame.cs

---------

Co-authored-by: Michael Sharp <[email protected]>

Author: asmirnov82

src/Microsoft.Data.Analysis/DataFrame.cs

@@ -1,4 +1,4 @@
-// Licensed to the .NET Foundation under one or more agreements.
+// Licensed to the .NET Foundation under one or more agreements.
 // The .NET Foundation licenses this file to you under the MIT license.
 // See the LICENSE file in the project root for more information.
 
@@ -483,6 +483,7 @@ private void ResizeByOneAndAppend(DataFrameColumn column, object value)
         /// </summary> 
         /// <remarks>If an input column's value doesn't match a DataFrameColumn's data type, a conversion will be attempted</remarks> 
         /// <remarks>If a <seealso cref="DataFrameRow"/> in <paramref name="rows"/> is null, a null value is appended to each column</remarks>
+        /// <remarks> Values are appended based on the column names</remarks>
         /// <param name="rows">The rows to be appended to this DataFrame </param> 
         /// <param name="inPlace">If set, appends <paramref name="rows"/> in place. Otherwise, a new DataFrame is returned with the <paramref name="rows"/> appended</param>
         /// <param name="cultureInfo">culture info for formatting values</param>
@@ -491,7 +492,7 @@ public DataFrame Append(IEnumerable<DataFrameRow> rows, bool inPlace = false, Cu
             DataFrame ret = inPlace ? this : Clone();
             foreach (DataFrameRow row in rows)
             {
-                ret.Append(row, inPlace: true, cultureInfo: cultureInfo);
+                ret.Append(row.GetValues(), inPlace: true, cultureInfo: cultureInfo);
             }
             return ret;
         }
@@ -503,7 +504,7 @@ public DataFrame Append(IEnumerable<DataFrameRow> rows, bool inPlace = false, Cu
         /// <remarks>If <paramref name="row"/> is null, a null value is appended to each column</remarks>
         /// <param name="row"></param> 
         /// <param name="inPlace">If set, appends a <paramref name="row"/> in place. Otherwise, a new DataFrame is returned with an appended <paramref name="row"/> </param>
-        /// <param name="cultureInfo">culture info for formatting values</param>
+        /// <param name="cultureInfo">Culture info for formatting values</param>
         public DataFrame Append(IEnumerable<object> row = null, bool inPlace = false, CultureInfo cultureInfo = null)
         {
             if (cultureInfo == null)
@@ -586,8 +587,14 @@ public DataFrame Append(IEnumerable<object> row = null, bool inPlace = false, Cu
         /// <remarks>If a column's value doesn't match its column's data type, a conversion will be attempted</remarks> 
         /// <param name="row">An enumeration of column name and value to be appended</param> 
         /// <param name="inPlace">If set, appends <paramref name="row"/> in place. Otherwise, a new DataFrame is returned with an appended <paramref name="row"/> </param>
-        public DataFrame Append(IEnumerable<KeyValuePair<string, object>> row, bool inPlace = false)
+        /// <param name="cultureInfo">Culture info for formatting values</param>
+        public DataFrame Append(IEnumerable<KeyValuePair<string, object>> row, bool inPlace = false, CultureInfo cultureInfo = null)
         {
+            if (cultureInfo == null)
+            {
+                cultureInfo = CultureInfo.CurrentCulture;
+            }
+
             DataFrame ret = inPlace ? this : Clone();
             if (row == null)
             {
@@ -608,7 +615,7 @@ public DataFrame Append(IEnumerable<KeyValuePair<string, object>> row, bool inPl
                 object value = columnAndValue.Value;
                 if (value != null)
                 {
-                    value = Convert.ChangeType(value, column.DataType);
+                    value = Convert.ChangeType(value, column.DataType, cultureInfo);
                     if (value is null)
                     {
                         throw new ArgumentException(string.Format(Strings.MismatchedValueType, column.DataType), column.Name);

src/Microsoft.Data.Analysis/DataFrameRow.cs

@@ -6,6 +6,7 @@
 using System.Collections;
 using System.Collections.Generic;
 using System.Diagnostics;
+using System.Linq;
 using System.Text;
 
 namespace Microsoft.Data.Analysis
@@ -17,6 +18,7 @@ public class DataFrameRow : IEnumerable<object>
     {
         private readonly DataFrame _dataFrame;
         private readonly long _rowIndex;
+
         internal DataFrameRow(DataFrame df, long rowIndex)
         {
             Debug.Assert(rowIndex < df.Columns.RowCount);
@@ -35,6 +37,11 @@ public IEnumerator<object> GetEnumerator()
             }
         }
 
+        public IEnumerable<KeyValuePair<string, object>> GetValues()
+        {
+            return _dataFrame.Columns.Select(col => new KeyValuePair<string, object>(col.Name, col[_rowIndex]));
+        }
+
         /// <summary>
         /// An indexer to return the value at <paramref name="index"/>.
         /// </summary>

test/Microsoft.Data.Analysis.Tests/DataFrameTests.cs

@@ -3206,6 +3206,35 @@ void Verify(DataFrame ret, DataFrame check1, DataFrame check2)
             Verify(df, dfClone, df2);
         }
 
+        [Fact]
+        public void TestAppendRowsIfColumnAreOutOfOrder()
+        {
+            var dataFrame = new DataFrame(
+                new StringDataFrameColumn("ColumnA", new string[] { "a", "b", "c" }),
+                new Int32DataFrameColumn("ColumnB", new int[] { 1, 2, 3 }),
+                new Int32DataFrameColumn("ColumnC", new int[] { 10, 20, 30 }));
+
+            //ColumnC and ColumnB are swaped
+            var dataFrame2 = new DataFrame(
+                new StringDataFrameColumn("ColumnA", new string[] { "d", "e", "f" }),
+                new Int32DataFrameColumn("ColumnC", new int[] { 40, 50, 60 }),
+                new Int32DataFrameColumn("ColumnB", new int[] { 4, 5, 6 }));
+
+            var resultDataFrame = dataFrame.Append(dataFrame2.Rows);
+
+            Assert.Equal(3, resultDataFrame.Columns.Count);
+            Assert.Equal(6, resultDataFrame.Rows.Count);
+
+            Assert.Equal("c", resultDataFrame["ColumnA"][2]);
+            Assert.Equal("d", resultDataFrame["ColumnA"][3]);
+
+            Assert.Equal(3, resultDataFrame["ColumnB"][2]);
+            Assert.Equal(4, resultDataFrame["ColumnB"][3]);
+
+            Assert.Equal(30, resultDataFrame["ColumnC"][2]);
+            Assert.Equal(40, resultDataFrame["ColumnC"][3]);
+        }
+
         [Fact]
         public void TestAppendRow()
         {