[s] fixed typo and Copyright date

format-xls/pom.xml

@@ -1,7 +1,7 @@
 <project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
          xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">
   <!--
-   ~ Copyright © 2023 Cask Data, Inc.
+   ~ Copyright © 2024 Cask Data, Inc.
    ~
    ~ Licensed under the Apache License, Version 2.0 (the "License"); you may not
    ~ use this file except in compliance with the License. You may obtain a copy of

format-xls/src/main/java/io/cdap/plugin/format/xls/input/XlsInputFormatDataFormatter.java → format-xls/src/main/java/io/cdap/plugin/format/xls/input/XlsDataFormatter.java

@@ -1,5 +1,5 @@
 /*
- * Copyright © 2023 Cask Data, Inc.
+ * Copyright © 2024 Cask Data, Inc.
  *
  * Licensed under the Apache License, Version 2.0 (the "License"); you may not
  * use this file except in compliance with the License. You may obtain a copy of
@@ -26,7 +26,7 @@
 /**
  * Formats the cell value of an Excel file.
  */
-public class XlsInputFormatDataFormatter {
+public class XlsDataFormatter {
   private static final DataFormatter dataFormatter = new DataFormatter();
   private final FormulaEvaluator evaluator;
 
@@ -35,7 +35,7 @@ public class XlsInputFormatDataFormatter {
    *
    * @param evaluator the formula evaluator
    */
-  public XlsInputFormatDataFormatter(FormulaEvaluator evaluator) {
+  public XlsDataFormatter(FormulaEvaluator evaluator) {
     this.evaluator = evaluator;
   }
 

format-xls/src/main/java/io/cdap/plugin/format/xls/input/XlsInputFormat.java

@@ -1,5 +1,5 @@
 /*
- * Copyright © 2023 Cask Data, Inc.
+ * Copyright © 2024 Cask Data, Inc.
  *
  * Licensed under the Apache License, Version 2.0 (the "License"); you may not
  * use this file except in compliance with the License. You may obtain a copy of
@@ -17,7 +17,6 @@
 package io.cdap.plugin.format.xls.input;
 
 import com.google.common.base.Preconditions;
-import com.google.common.base.Strings;
 import io.cdap.cdap.api.data.format.StructuredRecord;
 import io.cdap.cdap.api.data.schema.Schema;
 import org.apache.hadoop.conf.Configuration;
@@ -32,8 +31,6 @@
 import org.apache.hadoop.mapreduce.lib.input.CombineFileSplit;
 import org.apache.hadoop.mapreduce.lib.input.TextInputFormat;
 import org.apache.poi.ss.usermodel.Cell;
-import org.apache.poi.ss.usermodel.CellType;
-import org.apache.poi.ss.usermodel.DataFormatter;
 import org.apache.poi.ss.usermodel.FormulaEvaluator;
 import org.apache.poi.ss.usermodel.Row;
 import org.apache.poi.ss.usermodel.Sheet;
@@ -67,7 +64,7 @@ public RecordReader<LongWritable, StructuredRecord> createRecordReader(InputSpli
    */
   public static class XlsRecordReader extends RecordReader<LongWritable, StructuredRecord> {
     // DataFormatter to format and get each cell's value as String
-    XlsInputFormatDataFormatter formatter;
+    XlsDataFormatter formatter;
     FormulaEvaluator formulaEvaluator;
     // Map key that represents the row index.
     private LongWritable key;
@@ -103,15 +100,15 @@ public void initialize(InputSplit genericSplit, TaskAttemptContext context) thro
       try (Workbook workbook = WorkbookFactory.create(fileIn)) {
         formulaEvaluator = workbook.getCreationHelper().createFormulaEvaluator();
         formulaEvaluator.setIgnoreMissingWorkbooks(true);
-        formatter = new XlsInputFormatDataFormatter(formulaEvaluator);
+        formatter = new XlsDataFormatter(formulaEvaluator);
         // Check if user wants to access with name or number
         if (sheet.equals(XlsInputFormatConfig.SHEET_NUMBER)) {
           workSheet = workbook.getSheetAt(Integer.parseInt(sheetValue));
         } else {
           workSheet = workbook.getSheet(sheetValue);
         }
       } catch (Exception e) {
-        throw new IllegalArgumentException("Exception while reading excel sheet. " + e.getMessage(), e);
+        throw new IOException("Exception while reading excel sheet. " + e.getMessage(), e);
       }
 
       lastRowNum = workSheet.getLastRowNum();
@@ -153,7 +150,7 @@ public boolean nextKeyValue() {
       for (int cellIndex = 0; cellIndex < row.getLastCellNum(); cellIndex++) {
         if (cellIndex >= fields.size()) {
          throw new IllegalArgumentException(
-           String.format("Schema contains less fields than the number of columns in the excel file. " +
+           String.format("Schema contains fewer fields than the number of columns in the excel file. " +
                            "Schema fields: %s, Excel columns: %s", fields.size(), row.getLastCellNum()));
         }
         Cell cell = row.getCell(cellIndex, Row.MissingCellPolicy.RETURN_BLANK_AS_NULL);

format-xls/src/main/java/io/cdap/plugin/format/xls/input/XlsInputFormatConfig.java

@@ -1,5 +1,5 @@
 /*
- * Copyright © 2023 Cask Data, Inc.
+ * Copyright © 2024 Cask Data, Inc.
  *
  * Licensed under the Apache License, Version 2.0 (the "License"); you may not
  * use this file except in compliance with the License. You may obtain a copy of
@@ -47,7 +47,7 @@ public class XlsInputFormatConfig extends PathTrackingConfig {
   public static final String NAME_SAMPLE_SIZE = "sampleSize";
 
   public static final String DESC_SKIP_HEADER =
-    "Whether to skip the first line of each file. The default value is false.";
+    "Whether to skip the first line of each sheet. The default value is false.";
   public static final String DESC_SHEET = "Select the sheet by name or number. Default is 'Sheet Number'.";
   public static final String DESC_SHEET_VALUE = "Specifies the value corresponding to 'sheet' input. " +
     "Can be either sheet name or sheet no; for example: 'Sheet1' or '0' in case user selects 'Sheet Name' or " +
@@ -80,7 +80,6 @@ public class XlsInputFormatConfig extends PathTrackingConfig {
   @Description(DESC_SHEET_VALUE)
   private String sheetValue;
 
-
   @Macro
   @Nullable
   @Name(NAME_SKIP_HEADER)
@@ -170,6 +169,7 @@ public Map<String, Schema> getOverride() throws IllegalArgumentException {
   public static Builder builder() {
     return new Builder();
   }
+
   /**
    * Builder for building a {@link XlsInputFormatConfig}.
    */
@@ -194,6 +194,7 @@ public Builder setSheetValue(String sheetValue) {
       this.sheetValue = sheetValue;
       return this;
     }
+
     public Builder setSkipHeader(Boolean skipHeader) {
       this.skipHeader = skipHeader;
       return this;
@@ -203,6 +204,7 @@ public Builder setTerminateIfEmptyRow(Boolean terminateIfEmptyRow) {
       this.terminateIfEmptyRow = terminateIfEmptyRow;
       return this;
     }
+
     public XlsInputFormatConfig build() {
      return new XlsInputFormatConfig(schema, sheet, sheetValue, skipHeader, terminateIfEmptyRow);
     }

format-xls/src/main/java/io/cdap/plugin/format/xls/input/XlsInputFormatProvider.java

@@ -1,5 +1,5 @@
 /*
- * Copyright © 2023 Cask Data, Inc.
+ * Copyright © 2024 Cask Data, Inc.
  *
  * Licensed under the Apache License, Version 2.0 (the "License"); you may not
  * use this file except in compliance with the License. You may obtain a copy of

format-xls/src/main/java/io/cdap/plugin/format/xls/input/XlsInputFormatSchemaDetector.java

@@ -1,5 +1,5 @@
 /*
- * Copyright © 2023 Cask Data, Inc.
+ * Copyright © 2024 Cask Data, Inc.
  *
  * Licensed under the Apache License, Version 2.0 (the "License"); you may not
  * use this file except in compliance with the License. You may obtain a copy of

format-xls/src/main/java/io/cdap/plugin/format/xls/input/XlsInputFormatUtils.java

@@ -1,5 +1,5 @@
 /*
- * Copyright © 2023 Cask Data, Inc.
+ * Copyright © 2024 Cask Data, Inc.
  *
  * Licensed under the Apache License, Version 2.0 (the "License"); you may not
  * use this file except in compliance with the License. You may obtain a copy of
@@ -16,10 +16,6 @@
 
 package io.cdap.plugin.format.xls.input;
 
-import io.cdap.cdap.api.data.schema.Schema;
-import org.apache.poi.ss.usermodel.Cell;
-import org.apache.poi.ss.usermodel.CellType;
-
 import java.util.ArrayList;
 import java.util.HashMap;
 import java.util.List;

format-xls/src/test/java/io/cdap/plugin/format/xls/input/XlsInputFormatDataFormatterTest.java → format-xls/src/test/java/io/cdap/plugin/format/xls/input/XlsDataFormatterTest.java

@@ -1,5 +1,5 @@
 /*
- * Copyright © 2023 Cask Data, Inc.
+ * Copyright © 2024 Cask Data, Inc.
  *
  * Licensed under the Apache License, Version 2.0 (the "License"); you may not
  * use this file except in compliance with the License. You may obtain a copy of
@@ -31,9 +31,9 @@
 import java.io.IOException;
 
 /**
- * Unit tests for {@link XlsInputFormatDataFormatter}
+ * Unit tests for {@link XlsDataFormatter}
  */
-public class XlsInputFormatDataFormatterTest {
+public class XlsDataFormatterTest {
   Workbook workbook;
   Sheet sheet;
 
@@ -66,15 +66,15 @@ public void testFormatCellValue() {
     Cell errorCell = row.createCell(++testColumn);
     errorCell.setCellErrorValue(FormulaError.DIV0.getCode());
 
-    XlsInputFormatDataFormatter xlsInputFormatDataFormatter = new XlsInputFormatDataFormatter(
+    XlsDataFormatter xlsDataFormatter = new XlsDataFormatter(
             workbook.getCreationHelper().createFormulaEvaluator());
 
-    Assert.assertNull(xlsInputFormatDataFormatter.formatCellValue(blankCell, Schema.Type.STRING));
-    Assert.assertEquals("TRUE", xlsInputFormatDataFormatter.formatCellValue(booleanCell, Schema.Type.BOOLEAN));
-    Assert.assertEquals("1.0", xlsInputFormatDataFormatter.formatCellValue(numericCell, Schema.Type.DOUBLE));
-    Assert.assertEquals("test", xlsInputFormatDataFormatter.formatCellValue(stringCell, Schema.Type.STRING));
-    Assert.assertNull(xlsInputFormatDataFormatter.formatCellValue(errorCell, Schema.Type.STRING));
-    Assert.assertNull(xlsInputFormatDataFormatter.formatCellValue(null, Schema.Type.STRING));
+    Assert.assertNull(xlsDataFormatter.formatCellValue(blankCell, Schema.Type.STRING));
+    Assert.assertEquals("TRUE", xlsDataFormatter.formatCellValue(booleanCell, Schema.Type.BOOLEAN));
+    Assert.assertEquals("1.0", xlsDataFormatter.formatCellValue(numericCell, Schema.Type.DOUBLE));
+    Assert.assertEquals("test", xlsDataFormatter.formatCellValue(stringCell, Schema.Type.STRING));
+    Assert.assertNull(xlsDataFormatter.formatCellValue(errorCell, Schema.Type.STRING));
+    Assert.assertNull(xlsDataFormatter.formatCellValue(null, Schema.Type.STRING));
   }
 
   @Test
@@ -105,10 +105,10 @@ public void testFormatCellValueWithCachedFormulaResult() {
     // Cache the formula results
     evaluator.evaluateAll();
 
-    XlsInputFormatDataFormatter xlsInputFormatDataFormatter = new XlsInputFormatDataFormatter(evaluator);
+    XlsDataFormatter xlsDataFormatter = new XlsDataFormatter(evaluator);
     Assert.assertEquals("3.0",
-            xlsInputFormatDataFormatter.formatCellValue(formulaCell, Schema.Type.DOUBLE));
+            xlsDataFormatter.formatCellValue(formulaCell, Schema.Type.DOUBLE));
     Assert.assertEquals("helloworld",
-            xlsInputFormatDataFormatter.formatCellValue(formulaCell2, Schema.Type.STRING));
+            xlsDataFormatter.formatCellValue(formulaCell2, Schema.Type.STRING));
   }
 }

format-xls/src/test/java/io/cdap/plugin/format/xls/input/XlsInputFormatProviderTest.java

@@ -1,5 +1,5 @@
 /*
- * Copyright © 2023 Cask Data, Inc.
+ * Copyright © 2024 Cask Data, Inc.
  *
  * Licensed under the Apache License, Version 2.0 (the "License"); you may not
  * use this file except in compliance with the License. You may obtain a copy of

format-xls/src/test/java/io/cdap/plugin/format/xls/input/XlsInputFormatSchemaDetectorTest.java

@@ -1,5 +1,5 @@
 /*
- * Copyright © 2023 Cask Data, Inc.
+ * Copyright © 2024 Cask Data, Inc.
  *
  * Licensed under the Apache License, Version 2.0 (the "License"); you may not
  * use this file except in compliance with the License. You may obtain a copy of

format-xls/src/test/java/io/cdap/plugin/format/xls/input/XlsInputFormatUtilsTest.java

@@ -1,5 +1,5 @@
 /*
- * Copyright © 2023 Cask Data, Inc.
+ * Copyright © 2024 Cask Data, Inc.
  *
  * Licensed under the Apache License, Version 2.0 (the "License"); you may not
  * use this file except in compliance with the License. You may obtain a copy of

pom.xml

@@ -39,13 +39,13 @@
     <module>format-orc</module>
     <module>format-parquet</module>
     <module>format-text</module>
+    <module>format-xls</module>
     <module>hbase-plugins</module>
     <module>http-plugins</module>
     <module>mongodb-plugins</module>
     <module>solrsearch-plugins</module>
     <module>spark-plugins</module>
     <module>transform-plugins</module>
-    <module>format-xls</module>
   </modules>
 
   <licenses>