Github user paul-rogers commented on a diff in the pull request:
https://github.com/apache/drill/pull/644#discussion_r86384912
--- Diff:
exec/java-exec/src/main/java/org/apache/drill/exec/store/parquet/Metadata.java
---
@@ -944,14 +940,16 @@ public
ParquetTableMetadata_v2(ParquetTableMetadataBase parquetTable,
this.directories = directories;
this.columnTypeInfo = ((ParquetTableMetadata_v2)
parquetTable).columnTypeInfo;
this.drillVersion = DrillVersionInfo.getVersion();
- this.isDateCorrect = true;
+ this.writerVersion = ParquetWriter.WRITER_VERSION;
}
public ParquetTableMetadata_v2(List<ParquetFileMetadata_v2> files,
List<String> directories,
ConcurrentHashMap<ColumnTypeMetadata_v2.Key,
ColumnTypeMetadata_v2> columnTypeInfo) {
this.files = files;
this.directories = directories;
this.columnTypeInfo = columnTypeInfo;
+ this.drillVersion = DrillVersionInfo.getVersion();
+ this.writerVersion = ParquetWriter.WRITER_VERSION;
--- End diff --
Given a set of files and directories, we're assuming the writer version?
This is supposed to be the version of the Drill Parquet writer that wrote the
actual Parquet files, right? Not the version of the metadata files that hold
information about the Parquet files? Or, are we using the same writer version
for both purposes? If we need two versions (one for Parquet, another for
metadata), then we should introduce a separate Parquet metadata version. (Or,
I'm misunderstanding the code...)
---
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.
---
