You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by GitBox <gi...@apache.org> on 2022/05/13 03:24:48 UTC

[GitHub] [flink-table-store] tsreaper commented on a diff in pull request #121: Introduce AggregatuibMergeFunction

tsreaper commented on code in PR #121:
URL: https://github.com/apache/flink-table-store/pull/121#discussion_r871965386


##########
flink-table-store-core/src/main/java/org/apache/flink/table/store/file/mergetree/compact/AggregationMergeFunction.java:
##########
@@ -0,0 +1,104 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.flink.table.store.file.mergetree.compact;
+
+import org.apache.flink.table.data.GenericRowData;
+import org.apache.flink.table.data.RowData;
+
+import javax.annotation.Nullable;
+
+/**
+ * A {@link MergeFunction} where key is primary key (unique) and value is the partial record, update
+ * non-null fields on merge.
+ */
+public class AggregationMergeFunction implements MergeFunction {
+
+    private static final long serialVersionUID = 1L;
+
+    private final RowData.FieldGetter[] getters;
+
+    private transient GenericRowData row;
+
+    public AggregationMergeFunction(RowData.FieldGetter[] getters) {
+        this.getters = getters;
+    }
+
+    @Override
+    public void reset() {
+        this.row = new GenericRowData(getters.length);
+    }
+
+    @Override
+    public void add(RowData value) {
+        for (int i = 0; i < getters.length; i++)
+        {
+            Object currentField = getters[i].getFieldOrNull(value);
+            Object oldValue = row.getField(i);
+            Object result = sum(oldValue, currentField);
+            if (result != null)
+            {
+                row.setField(i, result);
+            }
+        }
+    }
+
+    private Object sum(Object oldValue, Object currentField) {
+        if (currentField == null)
+        {
+            return null;
+        }
+        if (oldValue == null)
+        {
+            return currentField;
+        }
+        if (oldValue instanceof Integer && currentField instanceof Integer)
+        {
+            return Integer.sum((Integer) oldValue, (Integer) currentField);
+        }
+        else if (oldValue instanceof Long && currentField instanceof Long)
+        {
+            return Long.sum((Long) oldValue, (Long) currentField);
+        }

Review Comment:
   These `instanceof` checks will be performed on a per-record bases and have a big impact on performance. That's why we use `RowData.FieldGetter` instead of `instanceof` to fetch columns out of a row.
   
   A better approach is to create an `Aggregator` class just like `RowData.FieldGetter` and in the constructor of `AggregationMergeFunction` we choose `Aggregator` for each column according to their types. So on a per-record path we can use the `Aggregator` directly without these checks.
   
   An even better approach is to use Flink's existing code generation systems but that might be too complex for your first contribution, considering its usage and the fact that internal system of Flink table planner is now awkward to call from the outside due to the scala-free change in Flink 1.15. If you're interested see `AggWithoutKeysCodeGenerator` in Flink.



##########
flink-table-store-core/src/main/java/org/apache/flink/table/store/file/FileStoreImpl.java:
##########
@@ -196,18 +197,24 @@ public static FileStoreImpl createWithPrimaryKey(
                                 .collect(Collectors.toList()));
 
         MergeFunction mergeFunction;
+        List<LogicalType> fieldTypes = rowType.getChildren();
+        RowData.FieldGetter[] fieldGetters = new RowData.FieldGetter[fieldTypes.size()];
         switch (mergeEngine) {
             case DEDUPLICATE:
                 mergeFunction = new DeduplicateMergeFunction();
                 break;
             case PARTIAL_UPDATE:
-                List<LogicalType> fieldTypes = rowType.getChildren();
-                RowData.FieldGetter[] fieldGetters = new RowData.FieldGetter[fieldTypes.size()];
                 for (int i = 0; i < fieldTypes.size(); i++) {
                     fieldGetters[i] = RowData.createFieldGetter(fieldTypes.get(i), i);
                 }
                 mergeFunction = new PartialUpdateMergeFunction(fieldGetters);
                 break;
+            case AGGREGATION:
+                for (int i = 0; i < fieldTypes.size(); i++) {
+                    fieldGetters[i] = RowData.createFieldGetter(fieldTypes.get(i), i);
+                }

Review Comment:
   Move this out of the `switch` statement as other branches also need this.



##########
flink-table-store-core/src/main/java/org/apache/flink/table/store/file/mergetree/compact/AggregationMergeFunction.java:
##########
@@ -0,0 +1,104 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.flink.table.store.file.mergetree.compact;
+
+import org.apache.flink.table.data.GenericRowData;
+import org.apache.flink.table.data.RowData;
+
+import javax.annotation.Nullable;
+
+/**
+ * A {@link MergeFunction} where key is primary key (unique) and value is the partial record, update
+ * non-null fields on merge.
+ */
+public class AggregationMergeFunction implements MergeFunction {
+
+    private static final long serialVersionUID = 1L;
+
+    private final RowData.FieldGetter[] getters;
+
+    private transient GenericRowData row;
+
+    public AggregationMergeFunction(RowData.FieldGetter[] getters) {
+        this.getters = getters;
+    }
+
+    @Override
+    public void reset() {
+        this.row = new GenericRowData(getters.length);
+    }
+
+    @Override
+    public void add(RowData value) {
+        for (int i = 0; i < getters.length; i++)
+        {
+            Object currentField = getters[i].getFieldOrNull(value);
+            Object oldValue = row.getField(i);
+            Object result = sum(oldValue, currentField);
+            if (result != null)
+            {
+                row.setField(i, result);
+            }
+        }
+    }
+
+    private Object sum(Object oldValue, Object currentField) {
+        if (currentField == null)
+        {
+            return null;
+        }
+        if (oldValue == null)
+        {
+            return currentField;
+        }
+        if (oldValue instanceof Integer && currentField instanceof Integer)
+        {
+            return Integer.sum((Integer) oldValue, (Integer) currentField);
+        }
+        else if (oldValue instanceof Long && currentField instanceof Long)
+        {
+            return Long.sum((Long) oldValue, (Long) currentField);
+        }
+        else if (oldValue instanceof Double && currentField instanceof Double)
+        {
+            return Double.sum((Double) oldValue, (Double) currentField);
+        }
+        else if (oldValue instanceof Float && currentField instanceof Float)
+        {
+            return Float.sum((Float) oldValue, (Float) currentField);
+        }
+        else if (oldValue instanceof String && currentField instanceof String)

Review Comment:
   Flink has its own internal representation for some of the data types (mainly for performance). For example the internal representation of `String` in Flink is `StringData`. Not sure if there is a document for this. Maybe @JingsongLi will know.



-- 
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.

To unsubscribe, e-mail: issues-unsubscribe@flink.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org