文章目录

• 1. 介绍
• 2. 策略模式结构
• 3. 策略模式使用
• • 3.1. 场景一: 表迁移
  • 3.2. flink connector类型转换

1. 介绍

当if else过多时，可以通过策略模式来进行重构。先定义一个接口，各else处理分支实现这个接口，并定义 < 条件 , 处理类 > 的映射关系，然后根据条件找到响应的处理类执行即可。

当有新的分支时，只需要新增一个实现类，增加一个映射关系即可。

2. 策略模式结构

• 策略接口：提供了一个函数式接口，提供了分支逻辑实现的多种方法：匿名类、普通类等
• context类：通过组合策略接口，并执行统一方法，实现面向接口的动态方法。

先来一个官方的伪代码：

策略接口和实现类

// 一个函数式的策略接口
interface Strategy ismethod execute(a, b)// 实现类：实现了+、-、* 算法。
class ConcreteStrategyAdd implements Strategy ismethod execute(a, b) isreturn a + bclass ConcreteStrategySubtract implements Strategy ismethod execute(a, b) isreturn a - bclass ConcreteStrategyMultiply implements Strategy ismethod execute(a, b) isreturn a * b

上下文类的逻辑

class Context is// 引用策略接口：面向策略private strategy: Strategy//传入策略实现类，或者维护一个mapmethod setStrategy(Strategy strategy) isthis.strategy = strategy//根据传入的实现类，执行不同的逻辑method executeStrategy(int a, int b) isreturn strategy.execute(a, b)

client

class ExampleApplication ismethod main() is。。。创建上下文对象if (action == addition) thencontext.setStrategy(new ConcreteStrategyAdd())if (action == subtraction) thencontext.setStrategy(new ConcreteStrategySubtract())if (action == multiplication) thencontext.setStrategy(new ConcreteStrategyMultiply())result = context.executeStrategy(First number, Second number)

3. 策略模式使用

3.1. 场景一: 表迁移

有一个需求：项目版本升级（不同版本之间数据表结构不同），需要迁移数据库到新的版本，要求出一个框架，适配项目中所有的数据库的迁移。

1. 首先可以使用策略模式，一个表对应一个实现类，通过**类名子串（“t1,t2,t3”）**标识要迁移的表有哪些，然后遍历实例化这些实现类，并放到一个list中，最后遍历这个list，完成某个库的迁移；
2. 其次数据迁移涉及到库的连接和库的关闭，所以在创建迁移逻辑之前和迁移之后要维护好这些逻辑，可以统一放到一个context类中。

大致逻辑：

1.策略接口和实现

 迁移时的sql逻辑接口@FunctionalInterfacepublic interface JdbcDataMigrateFunction {/*** 所有迁移和被迁移的表都在一个mysql节点下*/void dataMigrate(Connection dbConn, JdbcPros pros);}@Log4j2
public class FlinkJobAndInstanceFunction implements JdbcDataMigrateFunction {
...
}

2.context的逻辑

//构造分支的运行逻辑：1. 根据字符串去创建分支逻辑类；2. 分支运行逻辑构建：连接数据库，遍历执行分支逻辑，关闭连接。
public class JdbcTableBuilder {private JdbcPros jdbcPros;private JdbcDataMigrateFunction[] jdbcDataMigrateFunctions;private Connection datalakeDbConn;public JdbcTableBuilder(String propertiesPath) {jdbcPros = getJdbcPros(propertiesPath);String datalakeMigrationTables = jdbcPros.getDatalakeMigrationTables();String[] datalakeTableNames = datalakeMigrationTables.split(",");jdbcDataMigrateFunctions = new JdbcDataMigrateFunction[datalakeTableNames.length];for (int i = 0; i < datalakeTableNames.length; i++) {jdbcDataMigrateFunctions[i] = createJdbcDataMigrateFunction(datalakeTableNames[i]);}}JdbcDataMigrateFunction createJdbcDataMigrateFunction(String dataflowTableName) {switch (dataflowTableName) {case "table1":return new FlinkJobFunction();// todo:添加其他表的迁移逻辑
//            case ""
//                ;default:throw new RuntimeException("No logical instance of the table migration could be found");}}//CONTEXT的逻辑public void runDataMigrate() {open(jdbcPros);//runfor (JdbcDataMigrateFunction function : jdbcDataMigrateFunctions) {function.dataMigrate(datalakeDbConn, dodpDbConn);}//closecloseConn();}/*** 1、jdbc链接*/public void open(JdbcPros pros) {try {DriverManager.getDrivers();Class.forName(pros.getDrivername());dodpDbConn = DriverManager.getConnection(pros.getDodpDbURL(),pros.getDodpDbUsername(),pros.getDodpDbPassword());} catch (SQLException | ClassNotFoundException e) {e.printStackTrace();}}/*** 关闭Connection*/public void closeConn() {if (dodpDbConn != null) {try {dodpDbConn.close();} catch (SQLException se) {} finally {dodpDbConn = null;}}}

3.2. flink connector类型转换

/** Base class for all converters that convert between JDBC object and Flink internal object. */public class JdbcRowConverterextends AbstractRowConverter {private static final long serialVersionUID = 1L;//context逻辑：根据flink sql中字段类型来构建类型转换规则public JdbcRowConverter(RowType rowType) {super(rowType);for (int i = 0; i < rowType.getFieldCount(); i++) {toInternalConverters.add(wrapIntoNullableInternalConverter(createInternalConverter(rowType.getTypeAt(i))));toExternalConverters.add(wrapIntoNullableExternalConverter(createExternalConverter(fieldTypes[i]), fieldTypes[i]));}}@Overrideprotected ISerializationConverterwrapIntoNullableExternalConverter(ISerializationConverter serializationConverter,LogicalType type) {final int sqlType =JdbcTypeUtil.typeInformationToSqlType(TypeConversions.fromDataTypeToLegacyInfo(TypeConversions.fromLogicalToDataType(type)));return (val, index, statement) -> {if (val == null|| val.isNullAt(index)|| LogicalTypeRoot.NULL.equals(type.getTypeRoot())) {statement.setNull(index, sqlType);} else {serializationConverter.serialize(val, index, statement);}};}//内部转换：将数据转换为flink能进行处理的数据@Overridepublic RowData toInternal(ResultSet resultSet) throws Exception {GenericRowData genericRowData = new GenericRowData(rowType.getFieldCount());for (int pos = 0; pos < rowType.getFieldCount(); pos++) {Object field = resultSet.getObject(pos + 1);genericRowData.setField(pos, toInternalConverters.get(pos).deserialize(field));}return genericRowData;}@Overridepublic RowData toInternalLookup(JsonArray jsonArray) throws Exception {GenericRowData genericRowData = new GenericRowData(rowType.getFieldCount());for (int pos = 0; pos < rowType.getFieldCount(); pos++) {Object field = jsonArray.getValue(pos);genericRowData.setField(pos, toInternalConverters.get(pos).deserialize(field));}return genericRowData;}//外部数据转换@Overridepublic FieldNamedPreparedStatement toExternal(RowData rowData, FieldNamedPreparedStatement statement) throws Exception {for (int index = 0; index < rowData.getArity(); index++) {toExternalConverters.get(index).serialize(rowData, index, statement);}return statement;}//具体策略的实现类：由flink sql字段名匹配类型转换规则实例@Overrideprotected IDeserializationConverter createInternalConverter(LogicalType type) {switch (type.getTypeRoot()) {case NULL:return val -> null;case BOOLEAN:case FLOAT:case DOUBLE:case INTERVAL_YEAR_MONTH:case INTERVAL_DAY_TIME:case INTEGER:case BIGINT:return val -> val;case TINYINT:return val -> ((Integer) val).byteValue();case SMALLINT:// Converter for small type that casts value to int and then return short value,// since// JDBC 1.0 use int type for small values.return val -> val instanceof Integer ? ((Integer) val).shortValue() : val;case DECIMAL:final int precision = ((DecimalType) type).getPrecision();final int scale = ((DecimalType) type).getScale();// using decimal(20, 0) to support db type bigint unsigned, user should define// decimal(20, 0) in SQL,// but other precision like decimal(30, 0) can work too from lenient consideration.return val ->val instanceof BigInteger? DecimalData.fromBigDecimal(new BigDecimal((BigInteger) val, 0), precision, scale): DecimalData.fromBigDecimal((BigDecimal) val, precision, scale);case DATE:return val ->(int) ((Date.valueOf(String.valueOf(val))).toLocalDate().toEpochDay());case TIME_WITHOUT_TIME_ZONE:return val ->(int)((Time.valueOf(String.valueOf(val))).toLocalTime().toNanoOfDay()/ 1_000_000L);case TIMESTAMP_WITH_TIME_ZONE:case TIMESTAMP_WITHOUT_TIME_ZONE:return val -> TimestampData.fromTimestamp((Timestamp) val);case CHAR:case VARCHAR:return val -> StringData.fromString(val.toString());case BINARY:case VARBINARY:return val -> (byte[]) val;case ARRAY:case ROW:case MAP:case MULTISET:case RAW:default:throw new UnsupportedOperationException("Unsupported type:" + type);}}@Overrideprotected ISerializationConverter createExternalConverter(LogicalType type) {switch (type.getTypeRoot()) {case BOOLEAN:return (val, index, statement) ->statement.setBoolean(index, val.getBoolean(index));case TINYINT:return (val, index, statement) -> statement.setByte(index, val.getByte(index));case SMALLINT:return (val, index, statement) -> statement.setShort(index, val.getShort(index));case INTEGER:case INTERVAL_YEAR_MONTH:return (val, index, statement) -> statement.setInt(index, val.getInt(index));case BIGINT:case INTERVAL_DAY_TIME:return (val, index, statement) -> statement.setLong(index, val.getLong(index));case FLOAT:return (val, index, statement) -> statement.setFloat(index, val.getFloat(index));case DOUBLE:return (val, index, statement) -> statement.setDouble(index, val.getDouble(index));case CHAR:case VARCHAR:// value is BinaryStringreturn (val, index, statement) ->statement.setString(index, val.getString(index).toString());case BINARY:case VARBINARY:return (val, index, statement) -> statement.setBytes(index, val.getBinary(index));case DATE:return (val, index, statement) ->statement.setDate(index, Date.valueOf(LocalDate.ofEpochDay(val.getInt(index))));case TIME_WITHOUT_TIME_ZONE:return (val, index, statement) ->statement.setTime(index,Time.valueOf(LocalTime.ofNanoOfDay(val.getInt(index) * 1_000_000L)));case TIMESTAMP_WITH_TIME_ZONE:case TIMESTAMP_WITHOUT_TIME_ZONE:final int timestampPrecision = ((TimestampType) type).getPrecision();return (val, index, statement) ->statement.setTimestamp(index, val.getTimestamp(index, timestampPrecision).toTimestamp());case DECIMAL:final int decimalPrecision = ((DecimalType) type).getPrecision();final int decimalScale = ((DecimalType) type).getScale();return (val, index, statement) ->statement.setBigDecimal(index,val.getDecimal(index, decimalPrecision, decimalScale).toBigDecimal());case ARRAY:case MAP:case MULTISET:case ROW:case RAW:default:throw new UnsupportedOperationException("Unsupported type:" + type);}}
}

参考：
策略模式