paul-rogers commented on a change in pull request #2026: DRILL-7330: Implement
metadata usage for all format plugins
URL: https://github.com/apache/drill/pull/2026#discussion_r394696805
##########
File path:
exec/java-exec/src/main/java/org/apache/drill/exec/store/dfs/easy/EasyGroupScan.java
##########
@@ -333,21 +338,73 @@ public boolean canPushdownProjects(List<SchemaPath>
columns) {
}
@JsonProperty
+ @JsonIgnore(value = false)
public TupleMetadata getSchema() {
return getTableMetadata().getSchema();
}
@Override
- @JsonIgnore
- public TableMetadata getTableMetadata() {
- if (tableMetadata == null) {
- tableMetadata = metadataProvider.getTableMetadata();
+ public AnalyzeInfoProvider getAnalyzeInfoProvider() {
+ return new AnalyzeFileInfoProviderImpl(formatPlugin.getName());
+ }
+
+ @Override
+ protected GroupScanWithMetadataFilterer<?> getFilterer() {
+ return new EasyGroupScanFilterer(this);
+ }
+
+ @Override
+ protected FileTableMetadataProviderBuilder<?>
tableMetadataProviderBuilder(MetadataProviderManager source) {
+ if (source.usesMetastore()) {
+ return new
MetastoreFileTableMetadataProvider.Builder<>((MetastoreMetadataProviderManager)
source);
+ } else {
+ return defaultTableMetadataProviderBuilder(source);
}
- return tableMetadata;
}
@Override
- public TableMetadataProvider getMetadataProvider() {
- return metadataProvider;
+ protected FileTableMetadataProviderBuilder<?>
defaultTableMetadataProviderBuilder(MetadataProviderManager source) {
+ return new SimpleFileTableMetadataProvider.Builder(source);
+ }
+
+ /**
+ * Implementation of GroupScanWithMetadataFilterer which uses {@link
EasyGroupScan} as source and
+ * builds {@link EasyGroupScan} instance with filtered metadata.
+ */
+ private static class EasyGroupScanFilterer extends
GroupScanWithMetadataFilterer<EasyGroupScanFilterer> {
+
+ EasyGroupScanFilterer(EasyGroupScan source) {
+ super(source);
+ }
+
+ @Override
+ public AbstractGroupScanWithMetadata<?> build() {
+ EasyGroupScan newScan = new EasyGroupScan((EasyGroupScan) source);
+ newScan.tableMetadata = tableMetadata;
+ // updates common row count and nulls counts for every column
+ if (newScan.getTableMetadata() != null && files != null &&
newScan.getFilesMetadata().size() != files.size()) {
+ newScan.tableMetadata =
TableMetadataUtils.updateRowCount(newScan.getTableMetadata(), files.values());
+ }
+ newScan.partitions = partitions;
+ newScan.segments = segments;
+ newScan.files = files;
+ newScan.matchAllMetadata = matchAllMetadata;
+ newScan.nonInterestingColumnsMetadata = nonInterestingColumnsMetadata;
+
+ newScan.fileSet = newScan.getFilesMetadata().keySet();
+ newScan.selection = FileSelection.create(null, new
ArrayList<>(newScan.fileSet), newScan.selectionRoot);
+ try {
+ newScan.initFromSelection(newScan.selection, newScan.formatPlugin);
+ } catch (IOException e) {
+ throw new RuntimeException("Failed to initialize scan from the
selection.", e);
+ }
+
+ return newScan;
+ }
+
+ @Override
+ protected EasyGroupScanFilterer self() {
+ return this;
+ }
Review comment:
Cool. I could really use this feature. Can you point me to an example of
where you've used it?
----------------------------------------------------------------
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