Apache
/
druid
mirror of https://github.com/apache/druid.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #1226 from dkharrat/master 

Honor ignoreInvalidRows in reducer of Hadoop indexer
This commit is contained in:
Fangjin Yang 2015-03-20 08:36:09 -07:00
parent 46eae49bf5 3a6dc99384
commit bb9118319f
2 changed files with 19 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
indexing-hadoop/src/main/java/io/druid/indexer/HadoopDruidIndexerMapper.java
View File

@ -17,6 +17,7 @@
package io.druid.indexer;
import com.metamx.common.logger.Logger;
import io.druid.data.input.InputRow;
import io.druid.data.input.impl.InputRowParser;
import io.druid.data.input.impl.MapInputRowParser;
@ -42,6 +43,8 @@ import com.metamx.common.RE;
public abstract class HadoopDruidIndexerMapper<KEYOUT, VALUEOUT> extends Mapper<Writable, Writable, KEYOUT, VALUEOUT>
{
private static final Logger log = new Logger(HadoopDruidIndexerMapper.class);
private HadoopDruidIndexerConfig config;
private InputRowParser parser;
protected GranularitySpec granularitySpec;
@ -77,6 +80,7 @@ public abstract class HadoopDruidIndexerMapper<KEYOUT, VALUEOUT> extends Mapper<
}
catch (Exception e) {
if (config.isIgnoreInvalidRows()) {
log.debug(e, "Ignoring invalid row [%s] due to parsing error", value.toString());
context.getCounter(HadoopDruidIndexerConfig.IndexJobCounters.INVALID_ROW_COUNTER).increment(1);
return; // we're ignoring this invalid row
} else {

18
indexing-hadoop/src/main/java/io/druid/indexer/IndexGeneratorJob.java
View File

@ -17,6 +17,7 @@
package io.druid.indexer;
import com.metamx.common.parsers.ParseException;
import io.druid.collections.StupidPool;
import io.druid.data.input.InputRow;
import io.druid.data.input.Rows;
@ -378,10 +379,21 @@ public class IndexGeneratorJob implements Jobby
for (final Writable value : values) {
context.progress();
final InputRow inputRow = index.formatRow(HadoopDruidIndexerMapper.parseInputRow(value, parser));
allDimensionNames.addAll(inputRow.getDimensions());
int numRows;
try {
final InputRow inputRow = index.formatRow(HadoopDruidIndexerMapper.parseInputRow(value, parser));
allDimensionNames.addAll(inputRow.getDimensions());
int numRows = index.add(inputRow);
numRows = index.add(inputRow);
} catch (ParseException e) {
if (config.isIgnoreInvalidRows()) {
log.debug(e, "Ignoring invalid row [%s] due to parsing error", value.toString());
context.getCounter(HadoopDruidIndexerConfig.IndexJobCounters.INVALID_ROW_COUNTER).increment(1);
continue;
} else {
throw e;
}
}
++lineCount;
if (!index.canAppendRow()) {