Merge pull request #606 from datazip-inc/staging

chore: release version v0.2.9

Author: hash-data

.github/workflows/security-ci.yaml

@@ -29,7 +29,7 @@ jobs:
       - uses: actions/setup-go@v3
         with:
           check-latest: "true"
-          go-version: "1.23.x"
+          go-version: "1.24.x"
       - name: Install govulncheck
         run: go install golang.org/x/vuln/cmd/govulncheck@latest
       - name: Run vulnerability checks

.github/workflows/unit-tests.yml

@@ -0,0 +1,23 @@
+name: Unit Tests
+
+on:
+  push:
+  pull_request:
+
+jobs:
+  unit-tests:
+    name: Run Unit Tests
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+
+      - uses: actions/setup-go@v3
+        with:
+          check-latest: true
+          go-version: 1.24.x
+
+      - name: Install Dependencies
+        run: go mod download
+
+      - name: Run All Unit Tests
+        run: go test -v ./...

constants/constants.go

@@ -25,6 +25,7 @@ const (
 	ConfigFolder           = "CONFIG_FOLDER"
 	StatePath              = "STATE_PATH"
 	StreamsPath            = "STREAMS_PATH"
+	DifferencePath         = "DIFFERENCE_STREAMS_PATH"
 	// DestinationDatabasePrefix is used as prefix for destination database name
 	DestinationDatabasePrefix = "DESTINATION_DATABASE_PREFIX"
 	// EffectiveParquetSize is the effective size in bytes considering 256mb targeted parquet size, compression ratio as 8

destination/iceberg/iceberg.go

@@ -613,17 +613,52 @@ func (i *Iceberg) parsePartitionRegex(pattern string) error {
 	return nil
 }
 
-func (i *Iceberg) DropStreams(_ context.Context, _ []string) error {
-	logger.Info("iceberg destination not support clear destination, skipping clear operation")
+// drop streams required for clear destination
+func (i *Iceberg) DropStreams(ctx context.Context, dropStreams []types.StreamInterface) error {
+	i.options = &destination.Options{
+		ThreadID: "iceberg_destination_drop",
+	}
+	if len(dropStreams) == 0 {
+		logger.Info("No streams selected for clearing Iceberg destination, skipping operation")
+		return nil
+	}
 
-	// logger.Infof("Clearing Iceberg destination for %d selected streams: %v", len(selectedStreams), selectedStreams)
+	// server setup for dropping tables
+	server, err := newIcebergClient(i.config, []PartitionInfo{}, i.options.ThreadID, false, false, "")
+	if err != nil {
+		return fmt.Errorf("failed to setup iceberg server for dropping streams: %s", err)
+	}
 
-	// TODO: Implement Iceberg table clearing logic
-	// 1. Connect to the Iceberg catalog
-	// 2. Use Iceberg's delete API or drop/recreate the table
-	// 3. Handle any Iceberg-specific cleanup
+	// to close client properly
+	i.server = server
+	defer func() {
+		i.Close(ctx)
+	}()
+
+	logger.Infof("Starting Clear Iceberg destination for %d selected streams", len(dropStreams))
+
+	// process each stream
+	for _, stream := range dropStreams {
+		destDB := stream.GetDestinationDatabase(&i.config.IcebergDatabase)
+		destTable := stream.GetDestinationTable()
+		dropTable := fmt.Sprintf("%s.%s", destDB, destTable)
+
+		logger.Infof("Dropping Iceberg table: %s", dropTable)
+
+		request := proto.IcebergPayload{
+			Type: proto.IcebergPayload_DROP_TABLE,
+			Metadata: &proto.IcebergPayload_Metadata{
+				DestTableName: dropTable,
+				ThreadId:      i.server.serverID,
+			},
+		}
+		_, err := i.server.sendClientRequest(ctx, &request)
+		if err != nil {
+			return fmt.Errorf("failed to drop table %s: %s", dropTable, err)
+		}
+	}
 
-	// logger.Info("Successfully cleared Iceberg destination for selected streams")
+	logger.Info("Successfully cleared Iceberg destination for selected streams")
 	return nil
 }
 

destination/iceberg/java_client.go

@@ -114,16 +114,11 @@ func getServerConfigJSON(config *Config, partitionInfo []PartitionInfo, port int
 		logger.Warnf("No region explicitly provided for Glue catalog, the Java process will attempt to use region from AWS environment")
 	}
 
-	// Configure custom endpoint for S3-compatible services (like MinIO)
 	if config.S3Endpoint != "" {
 		serverConfig["s3.endpoint"] = config.S3Endpoint
-		serverConfig["io-impl"] = "org.apache.iceberg.io.ResolvingFileIO"
-		// Set SSL/TLS configuration
-		serverConfig["s3.ssl-enabled"] = utils.Ternary(config.S3UseSSL, "true", "false").(string)
 	}
-
-	// Configure S3 or GCP file IO
-	serverConfig["io-impl"] = utils.Ternary(strings.HasPrefix(config.IcebergS3Path, "gs://"), "org.apache.iceberg.gcp.gcs.GCSFileIO", "org.apache.iceberg.aws.s3.S3FileIO")
+	serverConfig["io-impl"] = "org.apache.iceberg.io.ResolvingFileIO"
+	serverConfig["s3.ssl-enabled"] = utils.Ternary(config.S3UseSSL, "true", "false").(string)
 
 	// Marshal the config to JSON
 	return json.Marshal(serverConfig)

destination/iceberg/olake-iceberg-java-writer/src/main/java/io/debezium/server/iceberg/IcebergUtil.java

@@ -276,4 +276,19 @@ public static int partitionId() {
     return Integer.parseInt(dtFormater.format(Instant.now()));
   }
 
+  public static boolean dropIcebergTable(String namespace, String tableName, Catalog icebergCatalog) {
+    try{
+      TableIdentifier tableID = TableIdentifier.of(namespace, tableName);
+      // Check if table exists
+      if (!icebergCatalog.tableExists(tableID)) {
+        LOGGER.warn("Table not found: {}", tableID.toString());
+        return false;
+      }
+      return icebergCatalog.dropTable(tableID, false);
+    } catch(Exception e){
+      LOGGER.error("Failed to drop table {}.{}: {}", namespace, tableName, e.getMessage());
+      throw new RuntimeException("Failed to drop table: " + namespace + "." + tableName, e);
+    }
+  }
+
 }

destination/iceberg/olake-iceberg-java-writer/src/main/java/io/debezium/server/iceberg/rpc/OlakeRowsIngester.java

@@ -62,7 +62,7 @@ public void sendRecords(IcebergPayload request, StreamObserver<RecordIngest.Reco
                 throw new Exception("Destination table name not present in metadata");
             }
 
-            if (this.icebergTable == null) {
+            if (this.icebergTable == null && request.getType() != IcebergPayload.PayloadType.DROP_TABLE) {
                 SchemaConvertor schemaConvertor = new SchemaConvertor(identifierField, schemaMetadata);
                 this.icebergTable = loadIcebergTable(TableIdentifier.of(icebergNamespace, destTableName), 
                                         schemaConvertor.convertToIcebergSchema());
@@ -111,10 +111,24 @@ public void sendRecords(IcebergPayload request, StreamObserver<RecordIngest.Reco
                     break;
 
                 case DROP_TABLE:
-                    LOGGER.warn("{} Table {} not dropped, drop table not implemented", requestId, destTableName);
-                    sendResponse(responseObserver, "Drop table not implemented");
+                    String dropTable = metadata.getDestTableName();
+                    String[] parts = dropTable.split("\\.", 2);
+                    if (parts.length != 2) {
+                        throw new IllegalArgumentException("Invalid destination table name: " + dropTable);
+                    }
+                    String namespace = parts[0], tableName = parts[1];
+                    
+                    LOGGER.warn("{} Dropping table {}.{}", requestId, namespace, tableName);
+
+                    boolean dropped = IcebergUtil.dropIcebergTable(namespace, tableName, icebergCatalog);
+                    if (dropped) {
+                        sendResponse(responseObserver, "Successfully dropped table " + tableName);
+                        LOGGER.info("{} Table {} dropped", requestId, tableName);
+                    } else {
+                        sendResponse(responseObserver, "Table " + tableName + " does not exist");
+                        LOGGER.warn("{} Table {} not dropped, table does not exist", requestId, tableName);
+                    }
                     break;
-                
                 default:
                     throw new IllegalArgumentException("Unknown payload type: " + request.getType());
             }

destination/interface.go

@@ -32,6 +32,6 @@ type Writer interface {
 	// Need to pass olakeTimestamp as end argument to get the correct partition path based on record ingestion time.
 	EvolveSchema(ctx context.Context, globalSchema, recordsSchema any) (any, error)
 	// DropStreams is used to clear the destination before re-writing the stream
-	DropStreams(ctx context.Context, selectedStream []string) error
+	DropStreams(ctx context.Context, dropStreams []types.StreamInterface) error
 	Close(ctx context.Context) error
 }

destination/parquet/parquet.go

@@ -456,57 +456,61 @@ func (p *Parquet) getPartitionedFilePath(values map[string]any, olakeTimestamp t
 	return filepath.Join(p.basePath, strings.TrimSuffix(result, "/"))
 }
 
-func (p *Parquet) DropStreams(ctx context.Context, selectedStreams []string) error {
+func (p *Parquet) DropStreams(ctx context.Context, selectedStreams []types.StreamInterface) error {
+	// check for s3 writer configuration
+	err := p.initS3Writer()
+	if err != nil {
+		return err
+	}
+
 	if len(selectedStreams) == 0 {
-		logger.Infof("Thread[%s]: no streams selected for clearing, skipping clear operation", p.options.ThreadID)
+		logger.Infof("no streams selected for clearing, skipping clear operation")
 		return nil
 	}
 
-	logger.Infof("Thread[%s]: clearing destination for %d selected streams: %v", p.options.ThreadID, len(selectedStreams), selectedStreams)
+	paths := make([]string, 0, len(selectedStreams))
+	for _, stream := range selectedStreams {
+		paths = append(paths, stream.GetDestinationDatabase(nil)+"."+stream.GetDestinationTable())
+	}
 
 	if p.s3Client == nil {
-		if err := p.clearLocalFiles(selectedStreams); err != nil {
+		if err := p.clearLocalFiles(paths); err != nil {
 			return fmt.Errorf("failed to clear local files: %s", err)
 		}
 	} else {
-		if err := p.clearS3Files(ctx, selectedStreams); err != nil {
+		if err := p.clearS3Files(ctx, paths); err != nil {
 			return fmt.Errorf("failed to clear S3 files: %s", err)
 		}
 	}
-
-	logger.Infof("Thread[%s]: successfully cleared destination for selected streams", p.options.ThreadID)
 	return nil
 }
 
-func (p *Parquet) clearLocalFiles(selectedStreams []string) error {
-	for _, streamID := range selectedStreams {
+func (p *Parquet) clearLocalFiles(paths []string) error {
+	for _, streamID := range paths {
 		parts := strings.SplitN(streamID, ".", 2)
 		if len(parts) != 2 {
-			logger.Warnf("Thread[%s]: invalid stream ID format: %s, skipping", p.options.ThreadID, streamID)
+			logger.Warnf("invalid stream ID format: %s, skipping", streamID)
 			continue
 		}
+		namespace, tableName := parts[0], parts[1]
+		streamPath := filepath.Join(p.config.Path, namespace, tableName)
 
-		namespace, streamName := parts[0], parts[1]
-		streamPath := filepath.Join(p.config.Path, namespace, streamName)
-
-		logger.Infof("Thread[%s]: clearing local path: %s", p.options.ThreadID, streamPath)
+		logger.Infof("clearing local path: %s", streamPath)
 
 		if _, err := os.Stat(streamPath); os.IsNotExist(err) {
-			logger.Debugf("Thread[%s]: local path does not exist, skipping: %s", p.options.ThreadID, streamPath)
+			logger.Debugf("local path does not exist, skipping: %s", streamPath)
 			continue
 		}
 
 		if err := os.RemoveAll(streamPath); err != nil {
 			return fmt.Errorf("failed to remove local path %s: %s", streamPath, err)
 		}
-
-		logger.Debugf("Thread[%s]: successfully cleared local path: %s", p.options.ThreadID, streamPath)
 	}
 
 	return nil
 }
 
-func (p *Parquet) clearS3Files(ctx context.Context, selectedStreams []string) error {
+func (p *Parquet) clearS3Files(ctx context.Context, paths []string) error {
 	deleteS3PrefixStandard := func(filtPath string) error {
 		iter := s3manager.NewDeleteListIterator(p.s3Client, &s3.ListObjectsInput{
 			Bucket: aws.String(p.config.Bucket),
@@ -519,21 +523,21 @@ func (p *Parquet) clearS3Files(ctx context.Context, selectedStreams []string) er
 		return nil
 	}
 
-	for _, streamID := range selectedStreams {
+	for _, streamID := range paths {
 		parts := strings.SplitN(streamID, ".", 2)
 		if len(parts) != 2 {
-			logger.Warnf("Thread[%s]: invalid stream ID format: %s, skipping", p.options.ThreadID, streamID)
+			logger.Warnf("invalid stream ID format: %s, skipping", streamID)
 			continue
 		}
+		namespace, tableName := parts[0], parts[1]
+		s3TablePath := filepath.Join(p.config.Prefix, namespace, tableName, "/")
 
-		namespace, streamName := parts[0], parts[1]
-		s3TablePath := filepath.Join(p.config.Prefix, namespace, streamName, "/")
-		logger.Debugf("Thread[%s]: clearing S3 prefix: s3://%s/%s", p.options.ThreadID, p.config.Bucket, s3TablePath)
+		logger.Debugf("clearing S3 prefix: s3://%s/%s", p.config.Bucket, s3TablePath)
 		if err := deleteS3PrefixStandard(s3TablePath); err != nil {
 			return fmt.Errorf("failed to clear S3 prefix %s: %s", s3TablePath, err)
 		}
 
-		logger.Debugf("Thread[%s]: successfully cleared S3 prefix: s3://%s/%s", p.options.ThreadID, p.config.Bucket, s3TablePath)
+		logger.Debugf("successfully cleared S3 prefix: s3://%s/%s", p.config.Bucket, s3TablePath)
 	}
 	return nil
 }

destination/writers.go

@@ -83,7 +83,7 @@ func WithThreadID(threadID string) ThreadOptions {
 	}
 }
 
-func NewWriterPool(ctx context.Context, config *types.WriterConfig, syncStreams, dropStreams []string, batchSize int64) (*WriterPool, error) {
+func NewWriterPool(ctx context.Context, config *types.WriterConfig, syncStreams []string, batchSize int64) (*WriterPool, error) {
 	newfunc, found := RegisteredWriters[config.Type]
 	if !found {
 		return nil, fmt.Errorf("invalid destination type has been passed [%s]", config.Type)
@@ -99,12 +99,6 @@ func NewWriterPool(ctx context.Context, config *types.WriterConfig, syncStreams,
 		return nil, fmt.Errorf("failed to test destination: %s", err)
 	}
 
-	if dropStreams != nil {
-		if err := adapter.DropStreams(ctx, dropStreams); err != nil {
-			return nil, fmt.Errorf("failed to clear destination: %s", err)
-		}
-	}
-
 	pool := &WriterPool{
 		stats: &Stats{
 			TotalRecordsToSync: atomic.Int64{},
@@ -279,3 +273,22 @@ func (wt *WriterThread) Close(ctx context.Context) error {
 		return wt.writer.Close(ctx)
 	}
 }
+
+func ClearDestination(ctx context.Context, config *types.WriterConfig, dropStreams []types.StreamInterface) error {
+	newfunc, found := RegisteredWriters[config.Type]
+	if !found {
+		return fmt.Errorf("invalid destination type has been passed [%s]", config.Type)
+	}
+
+	adapter := newfunc()
+	if err := utils.Unmarshal(config.WriterConfig, adapter.GetConfigRef()); err != nil {
+		return err
+	}
+
+	if dropStreams != nil {
+		if err := adapter.DropStreams(ctx, dropStreams); err != nil {
+			return fmt.Errorf("failed to drop the streams: %s", err)
+		}
+	}
+	return nil
+}