Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[connector] flink read support for fluss primary key table changelog auxiliary table/ cdc events #510

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
+156 −11
Open

[connector] flink read support for fluss primary key table changelog auxiliary table/ cdc events #510

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0555e07
[connector] flink read support for fluss primary key table changelog …
MehulBatra Mar 1, 2025
86722e8
added license to newly added file & removed unwanted comments
MehulBatra Mar 1, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
47 changes: 47 additions & 0 deletions ...connector-flink/src/main/java/com/alibaba/fluss/connector/flink/catalog/FlinkCatalog.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,6 +32,8 @@
import com.alibaba.fluss.utils.ExceptionUtils;
import com.alibaba.fluss.utils.IOUtils;

import org.apache.flink.table.api.DataTypes;
import org.apache.flink.table.api.Schema;
import org.apache.flink.table.catalog.Catalog;
import org.apache.flink.table.catalog.CatalogBaseTable;
import org.apache.flink.table.catalog.CatalogDatabase;
Expand Down Expand Up @@ -63,6 +65,7 @@

import javax.annotation.Nullable;

import java.util.ArrayList;
import java.util.Collections;
import java.util.HashMap;
import java.util.List;
Expand All @@ -78,6 +81,8 @@ public class FlinkCatalog implements Catalog {

public static final String LAKE_TABLE_SPLITTER = "$lake";

public static final String CHANGELOG_TABLE_SPLITTER = "$changelog";

protected final ClassLoader classLoader;

protected final String catalogName;
Expand Down Expand Up @@ -263,6 +268,48 @@ public CatalogBaseTable getTable(ObjectPath objectPath)
tableName.split("\\" + LAKE_TABLE_SPLITTER)[0])));
}
return getLakeTable(objectPath.getDatabaseName(), tableName);
} else if ((tableName.contains(CHANGELOG_TABLE_SPLITTER))) {
String baseTableName = tableName.split("\\" + CHANGELOG_TABLE_SPLITTER)[0];
TablePath baseTablePath = TablePath.of(objectPath.getDatabaseName(), baseTableName);
tableInfo = admin.getTableInfo(baseTablePath).get();
if (!tableInfo.hasPrimaryKey()) {
throw new UnsupportedOperationException(
String.format(
"\"Table %s has no primary key, only primary key tables support changelog.",
baseTableName));
}
CatalogTable originalTable = FlinkConversions.toFlinkTable(tableInfo);
Schema originalSchema = originalTable.getUnresolvedSchema();
List<Schema.UnresolvedColumn> newColumns =
new ArrayList<>(originalSchema.getColumns());
newColumns.add(
new Schema.UnresolvedPhysicalColumn("_change_type", DataTypes.STRING()));
newColumns.add(
new Schema.UnresolvedPhysicalColumn("_log_offset", DataTypes.BIGINT()));
newColumns.add(
new Schema.UnresolvedPhysicalColumn(
"_commit_timestamp", DataTypes.TIMESTAMP_WITH_LOCAL_TIME_ZONE()));
Schema changeLogSchema =
Schema.newBuilder()
.fromColumns(newColumns)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

we can call column(..).withComment(..) here to make the code clean and attach addition column comments.

.primaryKey(
originalSchema
.getPrimaryKey()
.orElse(null)
.getColumnNames())
.build();

Map<String, String> options = new HashMap<>(originalTable.getOptions());
options.put(BOOTSTRAP_SERVERS.key(), bootstrapServers);
options.put("changelog", "true");

return CatalogTable.newBuilder()
.schema(changeLogSchema)
.comment(originalTable.getComment())
.options(options)
.partitionKeys(originalTable.getPartitionKeys())
.build();

} else {
tableInfo = admin.getTableInfo(tablePath).get();
}
Expand Down
8 changes: 7 additions & 1 deletion ...ctor-flink/src/main/java/com/alibaba/fluss/connector/flink/catalog/FlinkTableFactory.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,6 +51,7 @@
import java.util.HashSet;
import java.util.Set;

import static com.alibaba.fluss.connector.flink.catalog.FlinkCatalog.CHANGELOG_TABLE_SPLITTER;
import static com.alibaba.fluss.connector.flink.catalog.FlinkCatalog.LAKE_TABLE_SPLITTER;
import static com.alibaba.fluss.connector.flink.utils.FlinkConversions.toFlinkOption;

Expand All @@ -69,6 +70,10 @@ public DynamicTableSource createDynamicTableSource(Context context) {
lakeTableFactory = mayInitLakeTableFactory();
return lakeTableFactory.createDynamicTableSource(context, tableName);
}
boolean isChangelog = false;
if (tableName.contains(CHANGELOG_TABLE_SPLITTER)) {
isChangelog = true;
}

FactoryUtil.TableFactoryHelper helper = FactoryUtil.createTableFactoryHelper(this, context);
helper.validateExcept("table.", "client.");
Expand Down Expand Up @@ -129,7 +134,8 @@ public DynamicTableSource createDynamicTableSource(Context context) {
cache,
partitionDiscoveryIntervalMs,
tableOptions.get(toFlinkOption(ConfigOptions.TABLE_DATALAKE_ENABLED)),
tableOptions.get(toFlinkOption(ConfigOptions.TABLE_MERGE_ENGINE)));
tableOptions.get(toFlinkOption(ConfigOptions.TABLE_MERGE_ENGINE)),
isChangelog);
}

@Override
Expand Down
8 changes: 6 additions & 2 deletions ...s-connector-flink/src/main/java/com/alibaba/fluss/connector/flink/source/FlinkSource.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,6 +55,7 @@ public class FlinkSource implements Source<RowData, SourceSplitBase, SourceEnume
private final OffsetsInitializer offsetsInitializer;
private final long scanPartitionDiscoveryIntervalMs;
private final boolean streaming;
private boolean isChangelog;

public FlinkSource(
Configuration flussConf,
Expand All @@ -65,7 +66,8 @@ public FlinkSource(
@Nullable int[] projectedFields,
OffsetsInitializer offsetsInitializer,
long scanPartitionDiscoveryIntervalMs,
boolean streaming) {
boolean streaming,
boolean isChangelog) {
this.flussConf = flussConf;
this.tablePath = tablePath;
this.hasPrimaryKey = hasPrimaryKey;
Expand All @@ -75,6 +77,7 @@ public FlinkSource(
this.offsetsInitializer = offsetsInitializer;
this.scanPartitionDiscoveryIntervalMs = scanPartitionDiscoveryIntervalMs;
this.streaming = streaming;
this.isChangelog = isChangelog;
}

@Override
Expand Down Expand Up @@ -136,6 +139,7 @@ public SourceReader<RowData, SourceSplitBase> createReader(SourceReaderContext c
sourceOutputType,
context,
projectedFields,
flinkSourceReaderMetrics);
flinkSourceReaderMetrics,
isChangelog);
}
}
11 changes: 8 additions & 3 deletions ...nector-flink/src/main/java/com/alibaba/fluss/connector/flink/source/FlinkTableSource.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -96,6 +96,7 @@ public class FlinkTableSource
// will be empty if no partition key
private final int[] partitionKeyIndexes;
private final boolean streaming;
private final boolean isChangelog;
private final FlinkConnectorOptionsUtils.StartupOptions startupOptions;

// options for lookup source
Expand Down Expand Up @@ -137,7 +138,8 @@ public FlinkTableSource(
@Nullable LookupCache cache,
long scanPartitionDiscoveryIntervalMs,
boolean isDataLakeEnabled,
@Nullable MergeEngineType mergeEngineType) {
@Nullable MergeEngineType mergeEngineType,
boolean isChangelog) {
this.tablePath = tablePath;
this.flussConfig = flussConfig;
this.tableOutputType = tableOutputType;
Expand All @@ -155,6 +157,7 @@ public FlinkTableSource(
this.scanPartitionDiscoveryIntervalMs = scanPartitionDiscoveryIntervalMs;
this.isDataLakeEnabled = isDataLakeEnabled;
this.mergeEngineType = mergeEngineType;
this.isChangelog = isChangelog;
}

@Override
Expand Down Expand Up @@ -261,7 +264,8 @@ public boolean isBounded() {
projectedFields,
offsetsInitializer,
scanPartitionDiscoveryIntervalMs,
streaming);
streaming,
isChangelog);

if (!streaming) {
// return a bounded source provide to make planner happy,
Expand Down Expand Up @@ -350,7 +354,8 @@ public DynamicTableSource copy() {
cache,
scanPartitionDiscoveryIntervalMs,
isDataLakeEnabled,
mergeEngineType);
mergeEngineType,
isChangelog);
source.producedDataType = producedDataType;
source.projectedFields = projectedFields;
source.singleRowFilter = singleRowFilter;
Expand Down
9 changes: 7 additions & 2 deletions ...nk/src/main/java/com/alibaba/fluss/connector/flink/source/emitter/FlinkRecordEmitter.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,7 @@
import com.alibaba.fluss.connector.flink.source.reader.RecordAndPos;
import com.alibaba.fluss.connector.flink.source.split.HybridSnapshotLogSplitState;
import com.alibaba.fluss.connector.flink.source.split.SourceSplitState;
import com.alibaba.fluss.connector.flink.utils.ChangelogRowConverter;
import com.alibaba.fluss.connector.flink.utils.FlussRowToFlinkRowConverter;
import com.alibaba.fluss.types.RowType;

Expand All @@ -48,8 +49,12 @@ public class FlinkRecordEmitter implements RecordEmitter<RecordAndPos, RowData,

private LakeRecordRecordEmitter lakeRecordRecordEmitter;

public FlinkRecordEmitter(RowType rowType) {
this.converter = new FlussRowToFlinkRowConverter(rowType);
public FlinkRecordEmitter(RowType rowType, boolean isChangeLogMode) {
if (!isChangeLogMode) {
this.converter = new FlussRowToFlinkRowConverter(rowType);
} else {
this.converter = new ChangelogRowConverter(rowType);
}
}

@Override
Expand Down
7 changes: 4 additions & 3 deletions ...link/src/main/java/com/alibaba/fluss/connector/flink/source/reader/FlinkSourceReader.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,15 +48,15 @@
public class FlinkSourceReader
extends SingleThreadMultiplexSourceReaderBase<
RecordAndPos, RowData, SourceSplitBase, SourceSplitState> {

public FlinkSourceReader(
FutureCompletingBlockingQueue<RecordsWithSplitIds<RecordAndPos>> elementsQueue,
Configuration flussConfig,
TablePath tablePath,
RowType sourceOutputType,
SourceReaderContext context,
@Nullable int[] projectedFields,
FlinkSourceReaderMetrics flinkSourceReaderMetrics) {
FlinkSourceReaderMetrics flinkSourceReaderMetrics,
boolean isChangelog) {
super(
elementsQueue,
new FlinkSourceFetcherManager(
Expand All @@ -69,7 +69,8 @@ public FlinkSourceReader(
projectedFields,
flinkSourceReaderMetrics),
(ignore) -> {}),
new FlinkRecordEmitter(sourceOutputType),
// InternalRow into Flink RowData with the additional metadata columns
new FlinkRecordEmitter(sourceOutputType, isChangelog),
context.getConfiguration(),
context);
}
Expand Down
64 changes: 64 additions & 0 deletions ...or-flink/src/main/java/com/alibaba/fluss/connector/flink/utils/ChangelogRowConverter.java
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,64 @@
/*
* Copyright (c) 2024 Alibaba Group Holding Ltd.
*
* Licensed 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 com.alibaba.fluss.connector.flink.utils;

import com.alibaba.fluss.client.table.scanner.ScanRecord;
import com.alibaba.fluss.types.RowType;

import org.apache.flink.table.data.GenericRowData;
import org.apache.flink.table.data.RowData;

public class ChangelogRowConverter extends FlussRowToFlinkRowConverter {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It would be better to use the composite design pattern instead of extending in this case, as ChangelogRowConverter does not represent a direct row-to-row conversion.

We can introduce a pure RecordToFlinkRowConverter interface to abstract the conversion of ScanRecord to RowData. This interface would have two implementations:

  1. One implementation (e.g., DirectRowConverter or PlainRowConverter) directly invokes FlussRowToFlinkRowConverter#toFlinkRowData(ScanRecord) for the conversion.
  2. Another implementation (e.g., ChangelogRowConverter) wraps FlussRowToFlinkRowConverter to generate the baseRowData and enrich it with additional metadata columns.

public ChangelogRowConverter(RowType rowType) {
super(rowType);
}

public RowData toFlinkRowData(ScanRecord scanRecord) {
RowData baseRowData = super.toFlinkRowData(scanRecord);
GenericRowData rowWithMetadata = new GenericRowData(baseRowData.getArity() + 3);
rowWithMetadata.setRowKind(baseRowData.getRowKind());

for (int i = 0; i < baseRowData.getArity(); i++) {
rowWithMetadata.setField(i, baseRowData.getRawValue(i));
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

getRawValue will throw ClassCastException here. We can use Flink JoinedRowData to join the baseRowData and metadataRowData.

}

int baseArity = baseRowData.getArity();
String changeType;
switch (scanRecord.getRowKind()) {
case INSERT:
changeType = "+I";
break;
case UPDATE_BEFORE:
changeType = "-U";
break;
case UPDATE_AFTER:
changeType = "+U";
break;
case DELETE:
changeType = "-D";
break;
default:
changeType = "+I";
break;
}
rowWithMetadata.setField(baseArity, changeType);
rowWithMetadata.setField(baseArity + 1, scanRecord.logOffset());
rowWithMetadata.setField(baseArity + 2, scanRecord.timestamp());

return rowWithMetadata;
}
}
13 changes: 13 additions & 0 deletions fluss-server/src/main/java/com/alibaba/fluss/server/coordinator/CoordinatorService.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,6 +65,8 @@
import javax.annotation.Nullable;

import java.io.UncheckedIOException;
import java.util.Arrays;
import java.util.List;
import java.util.Map;
import java.util.concurrent.CompletableFuture;
import java.util.function.Supplier;
Expand Down Expand Up @@ -143,6 +145,17 @@ public CompletableFuture<CreateTableResponse> createTable(CreateTableRequest req
TableDescriptor tableDescriptor;
try {
tableDescriptor = TableDescriptor.fromJsonBytes(request.getTableJson());
// Validate reserved column names
List<String> reservedColumns =
Arrays.asList("_change_type", "_log_offset", "_commit_timestamp");
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Let's do the check in TableDescriptorValidation#validateTableDescriptor, and make the reserved columns as static fields.

for (String columnName : tableDescriptor.getSchema().getColumnNames()) {
if (reservedColumns.contains(columnName)) {
throw new InvalidTableException(
String.format(
"Column name '%s' is reserved for system use.", columnName));
}
}

} catch (Exception e) {
if (e instanceof UncheckedIOException) {
throw new InvalidTableException(
Expand Down
Loading