xuchuanyin commented on a change in pull request #3188: [CARBONDATA-3336]
Support configurable decode for loading binary data, support base64 and Hex
decode.
URL: https://github.com/apache/carbondata/pull/3188#discussion_r278954034
##########
File path:
processing/src/main/java/org/apache/carbondata/processing/loading/converter/impl/BinaryFieldConverterImpl.java
##########
@@ -43,33 +47,55 @@
private String nullformat;
private boolean isEmptyBadRecord;
private DataField dataField;
+ private String binaryEncoder;
public BinaryFieldConverterImpl(DataField dataField, String nullformat, int
index,
- boolean isEmptyBadRecord) {
+ boolean isEmptyBadRecord,String binaryEncoder) {
this.dataType = dataField.getColumn().getDataType();
this.dimension = (CarbonDimension) dataField.getColumn();
this.nullformat = nullformat;
this.index = index;
this.isEmptyBadRecord = isEmptyBadRecord;
this.dataField = dataField;
+ this.binaryEncoder = binaryEncoder;
}
@Override
public void convert(CarbonRow row, BadRecordLogHolder logHolder)
throws CarbonDataLoadingException {
- if (row.getObject(index) instanceof String) {
- row.update((((String) row.getObject(index)))
- .getBytes(Charset.forName(CarbonCommonConstants.DEFAULT_CHARSET)),
index);
- } else if (row.getObject(index) instanceof byte[]) {
- row.update(row.getObject(index), index);
- } else {
- throw new CarbonDataLoadingException("Binary only support String and
byte[] data type");
- }
+ row.update(convert(row.getObject(index), logHolder), index);
}
@Override
public Object convert(Object value, BadRecordLogHolder logHolder)
throws RuntimeException {
- return null;
+ if (value instanceof String) {
+ if (binaryEncoder.equalsIgnoreCase(
+ CarbonLoadOptionConstants.CARBON_OPTIONS_BINARY_DECODER_BASE64)) {
+ byte[] parsedValue = (String.valueOf(value))
+ .getBytes(Charset.forName(CarbonCommonConstants.DEFAULT_CHARSET));
+ if (Base64.isArrayByteBase64(parsedValue)) {
+ parsedValue = Base64.decodeBase64(parsedValue);
+ } else {
+ //TODO: it should throw exception
Review comment:
???
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
