rdblue commented on a change in pull request #206: Add option to load column
stats with data files.
URL: https://github.com/apache/incubator-iceberg/pull/206#discussion_r291349662
##########
File path: core/src/main/java/org/apache/iceberg/GenericDataFile.java
##########
@@ -174,12 +174,20 @@ private GenericDataFile(GenericDataFile toCopy) {
this.fileSizeInBytes = toCopy.fileSizeInBytes;
this.fileOrdinal = toCopy.fileOrdinal;
this.sortColumns = copy(toCopy.sortColumns);
- // TODO: support lazy conversion to/from map
- this.columnSizes = copy(toCopy.columnSizes);
- this.valueCounts = copy(toCopy.valueCounts);
- this.nullValueCounts = copy(toCopy.nullValueCounts);
- this.lowerBounds =
SerializableByteBufferMap.wrap(copy(toCopy.lowerBounds));
- this.upperBounds =
SerializableByteBufferMap.wrap(copy(toCopy.upperBounds));
+ if (fullCopy) {
+ // TODO: support lazy conversion to/from map
+ this.columnSizes = copy(toCopy.columnSizes);
Review comment:
That's because it isn't required to run the filters in `FilteredManifest`.
Those columns are the ones that need to be projected for the stats filter to
work, but may not have been requested by the caller.
----------------------------------------------------------------
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
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
