Github user davies commented on a diff in the pull request:
https://github.com/apache/spark/pull/12017#discussion_r58721198
--- Diff:
sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/VectorizedColumnReader.java
---
@@ -200,9 +150,28 @@ void readBatch(int total, ColumnVector column) throws
IOException {
ColumnVector dictionaryIds = column.reserveDictionaryIds(total);
defColumn.readIntegers(
num, dictionaryIds, column, rowId, maxDefLevel,
(VectorizedValuesReader) dataColumn);
- decodeDictionaryIds(rowId, num, column, dictionaryIds);
+
+ if ((column.hasDictionary() || rowId == 0) && (
+ descriptor.getType() == PrimitiveType.PrimitiveTypeName.INT32
||
+ descriptor.getType() == PrimitiveType.PrimitiveTypeName.INT64
||
+ descriptor.getType() == PrimitiveType.PrimitiveTypeName.FLOAT
||
+ descriptor.getType() == PrimitiveType.PrimitiveTypeName.DOUBLE
||
+ descriptor.getType() ==
PrimitiveType.PrimitiveTypeName.BINARY)) {
+ // Column vector supports lazy decoding of dictionary values so
just set the dictionary.
+ // We can't do this if rowId != 0 AND the column doesn't have a
dictionary (i.e. some
+ // non-dictionary encoded values have already been added).
+ column.setDictionary(dictionary);
+ } else {
+ decodeDictionaryIds(rowId, num, column, dictionaryIds);
+ }
} else {
- column.setDictionary(null);
--- End diff --
I think we should always have this, if column.hasDictionary(), or getLong()
will use the dictionary inside the column.
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
