[S] Rework

cloudsufi · Dec 11, 2023 · edb65ca · edb65ca
1 parent e6ed78f
commit edb65ca
Show file tree

Hide file tree

Showing 3 changed files with 9 additions and 7 deletions.
diff --git a/format-xls/src/main/java/io/cdap/plugin/format/xls/input/XlsInputFormat.java b/format-xls/src/main/java/io/cdap/plugin/format/xls/input/XlsInputFormat.java
@@ -88,17 +88,17 @@ public static class XlsRecordReader extends RecordReader<LongWritable, Structure
     public void initialize(InputSplit genericSplit, TaskAttemptContext context) throws IOException {
 
       CombineFileSplit split = (CombineFileSplit) genericSplit;
-      Configuration job = context.getConfiguration();
+      Configuration jobConf = context.getConfiguration();
       // Path of input file.
       Path file = split.getPath(0);
       String schema = context.getConfiguration().get("schema");
       outputSchema = schema != null ? Schema.parseJson(schema) : null;
-      FileSystem fs = file.getFileSystem(job);
+      FileSystem fs = file.getFileSystem(jobConf);
       fileIn = fs.open(split.getPath(0));
 
-      String sheet = job.get(SHEET_NO);
-      String sheetValue = job.get(SHEET_VALUE, "0");
-      terminateIfEmptyRow = job.getBoolean(TERMINATE_IF_EMPTY_ROW, false);
+      String sheet = jobConf.get(SHEET_NO);
+      String sheetValue = jobConf.get(SHEET_VALUE, "0");
+      terminateIfEmptyRow = jobConf.getBoolean(TERMINATE_IF_EMPTY_ROW, false);
 
       try (Workbook workbook = WorkbookFactory.create(fileIn)) {
         formulaEvaluator = workbook.getCreationHelper().createFormulaEvaluator();
@@ -118,7 +118,7 @@ public void initialize(InputSplit genericSplit, TaskAttemptContext context) thro
       rowIndex = 0;
       isRowNull = false;
 
-      boolean skipFirstRow = job.getBoolean(NAME_SKIP_HEADER, true);
+      boolean skipFirstRow = jobConf.getBoolean(NAME_SKIP_HEADER, true);
       if (skipFirstRow) {
         Preconditions.checkArgument(lastRowNum != -1, "No rows found on sheet %s", sheetValue);
         rowIndex = 1;

diff --git a/format-xls/src/main/java/io/cdap/plugin/format/xls/input/XlsInputFormatDataFormatter.java b/format-xls/src/main/java/io/cdap/plugin/format/xls/input/XlsInputFormatDataFormatter.java
@@ -78,7 +78,7 @@ public String formatCellValue(Cell cell, Schema.Type type) {
       case ERROR:
         return null;
       default:
-        throw new IllegalStateException("Unexpected celltype (" + cellType + ")");
+        throw new IllegalStateException(String.format("Unexpected celltype (%s)", cellType));
     }
   }
 }
diff --git a/format-xls/src/main/java/io/cdap/plugin/format/xls/input/XlsInputFormatProvider.java b/format-xls/src/main/java/io/cdap/plugin/format/xls/input/XlsInputFormatProvider.java
@@ -28,6 +28,7 @@
 import io.cdap.cdap.etl.api.validation.ValidatingInputFormat;
 import io.cdap.plugin.format.input.PathTrackingConfig;
 import io.cdap.plugin.format.input.PathTrackingInputFormatProvider;
+import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
 import org.apache.poi.ss.usermodel.Cell;
 import org.apache.poi.ss.usermodel.DataFormatter;
 import org.apache.poi.ss.usermodel.FormulaEvaluator;
@@ -97,6 +98,7 @@ protected void addFormatProperties(Map<String, String> properties) {
     properties.put(XlsInputFormat.SHEET_VALUE, conf.getSheetValue());
     properties.put(XlsInputFormat.NAME_SKIP_HEADER, String.valueOf(conf.getSkipHeader()));
     properties.put(XlsInputFormat.TERMINATE_IF_EMPTY_ROW, String.valueOf(conf.getTerminateIfEmptyRow()));
+    properties.put(FileInputFormat.SPLIT_MINSIZE, Long.toString(Long.MAX_VALUE));
   }
 
   @Override