[GitHub] [drill] vvysotskyi commented on a change in pull request #2023: DRILL-7640: EVF-based JSON Loader

[GitBox <gi...@apache.org>]

vvysotskyi commented on a change in pull request #2023: DRILL-7640: EVF-based JSON Loader
URL: https://github.com/apache/drill/pull/2023#discussion_r392384985
 
 

 ##########
 File path: exec/java-exec/src/main/java/org/apache/drill/exec/physical/resultSet/impl/TupleState.java
 ##########
 @@ -92,6 +92,153 @@
 public abstract class TupleState extends ContainerState
   implements AbstractTupleWriter.TupleWriterListener {
 
+  /**
+   * The set of columns added via the writers: includes both projected
+   * and unprojected columns. (The writer is free to add columns that the
+   * query does not project; the result set loader creates a dummy column
+   * and dummy writer, then does not project the column to the output.)
+   */
+  protected final List<ColumnState> columns = new ArrayList<>();
+
+  /**
+   * Internal writer schema that matches the column list.
+   */
+  protected final TupleMetadata schema = new TupleSchema();
+
+  /**
+   * Metadata description of the output container (for the row) or map
+   * (for map or repeated map.)
+   * <p>
+   * Rows and maps have an output schema which may differ from the internal schema.
+   * The output schema excludes unprojected columns. It also excludes
+   * columns added in an overflow row.
+   * <p>
+   * The output schema is built slightly differently for maps inside a
+   * union vs. normal top-level (or nested) maps. Maps inside a union do
+   * not defer columns because of the muddy semantics (and infrequent use)
+   * of unions.
+   */
+  protected TupleMetadata outputSchema;
+
+  private int prevHarvestIndex = -1;
+
+  protected TupleState(LoaderInternals events,
+      ResultVectorCache vectorCache,
+      ProjectionFilter projectionSet) {
+    super(events, vectorCache, projectionSet);
+  }
+
+  protected void bindOutputSchema(TupleMetadata outputSchema) {
+    this.outputSchema = outputSchema;
+  }
+
+  /**
+   * Returns an ordered set of the columns which make up the tuple.
+   * Column order is the same as that defined by the map's schema,
+   * to allow indexed access. New columns always appear at the end
+   * of the list to preserve indexes.
+   *
+   * @return ordered list of column states for the columns within
+   * this tuple
+   */
+  public List<ColumnState> columns() { return columns; }
+
+  public TupleMetadata schema() { return writer().tupleSchema(); }
+
+  public abstract AbstractTupleWriter writer();
+
+  @Override
+  public boolean isProjected(String colName) {
+    return projectionSet.isProjected(colName);
+  }
+
+  @Override
+  public ObjectWriter addColumn(TupleWriter tupleWriter, MaterializedField column) {
+    return addColumn(tupleWriter, MetadataUtils.fromField(column));
+  }
+
+  @Override
+  public ObjectWriter addColumn(TupleWriter tupleWriter, ColumnMetadata columnSchema) {
+    return BuildFromSchema.instance().buildColumn(this, columnSchema);
+  }
+
+  @Override
+  protected void addColumn(ColumnState colState) {
+    columns.add(colState);
+  }
+
+  public boolean hasProjections() {
+    for (final ColumnState colState : columns) {
+      if (colState.isProjected()) {
+        return true;
+      }
+    }
+    return false;
+  }
+
+  @Override
+  protected Collection<ColumnState> columnStates() {
+    return columns;
+  }
+
+  protected void updateOutput(int curSchemaVersion) {
+
+    // Scan all columns
+    for (int i = 0; i < columns.size(); i++) {
+      final ColumnState colState = columns.get(i);
+
+      // Ignore unprojected columns
+      if (! colState.writer().isProjected()) {
+        continue;
+      }
+
+      // If this is a new column added since the lastoutput, then we may have
 
 Review comment:
   ```suggestion
         // If this is a new column added since the last output, then we may have
   ```

----------------------------------------------------------------
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:
users@infra.apache.org


With regards,
Apache Git Services