tsreaper commented on code in PR #281: URL: https://github.com/apache/flink-table-store/pull/281#discussion_r962503790
########## flink-table-store-connector/src/main/java/org/apache/flink/table/store/connector/lookup/PrimaryKeyLookupTable.java: ########## @@ -0,0 +1,86 @@ +/* + * 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.connector.lookup; + +import org.apache.flink.table.data.RowData; +import org.apache.flink.table.runtime.typeutils.InternalSerializers; +import org.apache.flink.table.store.utils.KeyProjectedRowData; +import org.apache.flink.table.store.utils.TypeUtils; +import org.apache.flink.table.types.logical.RowType; +import org.apache.flink.types.RowKind; + +import java.io.IOException; +import java.util.Collections; +import java.util.Iterator; +import java.util.List; +import java.util.function.Predicate; + +/** A {@link LookupTable} for primary key table. */ +public class PrimaryKeyLookupTable implements LookupTable { + + protected final RocksDBValueState tableState; + + protected final Predicate<RowData> recordFilter; + + protected int[] primaryKeyMapping; + + protected final KeyProjectedRowData primaryKey; + + public PrimaryKeyLookupTable( + RocksDBStateFactory stateFactory, + RowType rowType, + List<String> primaryKey, + Predicate<RowData> recordFilter, + long lruCacheSize) + throws IOException { + List<String> fieldNames = rowType.getFieldNames(); + this.primaryKeyMapping = primaryKey.stream().mapToInt(fieldNames::indexOf).toArray(); + this.primaryKey = new KeyProjectedRowData(primaryKeyMapping); + this.tableState = + stateFactory.valueState( + "table", + InternalSerializers.create(TypeUtils.project(rowType, primaryKeyMapping)), + InternalSerializers.create(rowType), + lruCacheSize); + this.recordFilter = recordFilter; + } + + @Override + public List<RowData> get(RowData key) throws IOException { + RowData value = tableState.get(key); + return value == null ? Collections.emptyList() : Collections.singletonList(value); + } + + @Override + public void refresh(Iterator<RowData> incremental) throws IOException { + while (incremental.hasNext()) { + RowData row = incremental.next(); + primaryKey.replaceRow(row); + if (row.getRowKind() == RowKind.INSERT || row.getRowKind() == RowKind.UPDATE_AFTER) { + if (recordFilter.test(row)) { + tableState.put(primaryKey, row); Review Comment: Let's say the filter condition is `key < 10 and value < 10`. In `TableStreamingReader`, this filter won't be applied, because it contains conditions both for keys and for values. However in `FileStoreLookupFunction`, this filter can be transformed to `recordFilter` because it does not restrict the filter to contain conditions only for keys. Let's now consider these two change logs: `INSERT (1, 1)`, `UPDATE_AFTER (1, 100)`. These two change logs won't be filtered out by `TableStreamingReader` (because there is actually no filter) so they'll both reach this line. However, only `INESRT (1, 1)` can pass `recordFilter.test`. So the new changes won't be applied to the state and users will always see the old record, which is incorrect. -- 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-unsubscr...@flink.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
