Github user jackylk commented on a diff in the pull request:
https://github.com/apache/incubator-carbondata/pull/263#discussion_r87122653
--- Diff:
processing/src/main/java/org/apache/carbondata/processing/newflow/parser/impl/RowParserImpl.java
---
@@ -18,22 +18,80 @@
*/
package org.apache.carbondata.processing.newflow.parser.impl;
+import
org.apache.carbondata.processing.newflow.CarbonDataLoadConfiguration;
+import org.apache.carbondata.processing.newflow.DataField;
+import
org.apache.carbondata.processing.newflow.constants.DataLoadProcessorConstants;
+import org.apache.carbondata.processing.newflow.parser.CarbonParserFactory;
import org.apache.carbondata.processing.newflow.parser.GenericParser;
import org.apache.carbondata.processing.newflow.parser.RowParser;
public class RowParserImpl implements RowParser {
private GenericParser[] genericParsers;
- public RowParserImpl(GenericParser[] genericParsers) {
- this.genericParsers = genericParsers;
+ private int[] outputMapping;
+
+ private int[] inputMapping;
+
+ private int numberOfColumns;
+
+ public RowParserImpl(DataField[] output, CarbonDataLoadConfiguration
configuration) {
+ String[] complexDelimiters =
+ (String[])
configuration.getDataLoadProperty(DataLoadProcessorConstants.COMPLEX_DELIMITERS);
+ String nullFormat =
+
configuration.getDataLoadProperty(DataLoadProcessorConstants.SERIALIZATION_NULL_FORMAT)
+ .toString();
+ DataField[] input = getInput(configuration);
+ genericParsers = new GenericParser[input.length];
+ for (int i = 0; i < genericParsers.length; i++) {
+ genericParsers[i] =
+ CarbonParserFactory.createParser(input[i].getColumn(),
complexDelimiters, nullFormat);
+ }
+ outputMapping = new int[output.length];
+ for (int i = 0; i < input.length; i++) {
+ for (int j = 0; j < output.length; j++) {
+ if (input[i].getColumn().equals(output[j].getColumn())) {
+ outputMapping[i] = j;
+ break;
+ }
+ }
+ }
+ }
+
+ public DataField[] getInput(CarbonDataLoadConfiguration configuration) {
+ DataField[] fields = configuration.getDataFields();
+ String[] header = configuration.getHeader();
+ numberOfColumns = header.length;
+ DataField[] input = new DataField[fields.length];
+ inputMapping = new int[input.length];
+ int k = 0;
+ for (int i = 0; i < numberOfColumns; i++) {
+ for (int j = 0; j < fields.length; j++) {
+ if
(header[i].equalsIgnoreCase(fields[j].getColumn().getColName())) {
+ input[k] = fields[j];
+ inputMapping[k] = i;
+ k++;
+ break;
+ }
+ }
+ }
+ return input;
}
@Override
public Object[] parseRow(Object[] row) {
- for (int i = 0; i < row.length; i++) {
- row[i] = genericParsers[i].parse(row[i].toString());
+ // If number of columns are less in a row then create new array with
same size of he
--- End diff --
last word of this sentence is not correct
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
