sharding-jdbc中SQLExecutionHook的用法

本篇內容主要講解“sharding-jdbc中SQLExecutionHook的用法”，感興趣的朋友不妨來看看。本文介紹的方法操作簡單快捷，實用性強。下面就讓小編來帶大家學習“sharding-jdbc中SQLExecutionHook的用法”吧!

序

本文主要研究一下sharding-jdbc的SQLExecutionHook

SQLExecutionHook

incubator-shardingsphere-4.0.0-RC1/sharding-core/sharding-core-execute/src/main/java/org/apache/shardingsphere/core/execute/hook/SQLExecutionHook.java

public interface SQLExecutionHook {
    
    /**
     * Handle when SQL execution started.
     * 
     * @param routeUnit route unit to be executed
     * @param dataSourceMetaData data source meta data
     * @param isTrunkThread is execution in trunk thread
     * @param shardingExecuteDataMap sharding execute data map
     */
    void start(RouteUnit routeUnit, DataSourceMetaData dataSourceMetaData, boolean isTrunkThread, Map<String, Object> shardingExecuteDataMap);
    
    /**
     * Handle when SQL execution finished success.
     */
    void finishSuccess();
    
    /**
     * Handle when SQL execution finished failure.
     *
     * @param cause failure cause
     */
    void finishFailure(Exception cause);
}

• SQLExecutionHook接口定義了start、finishSuccess、finishFailure方法

SPISQLExecutionHook

incubator-shardingsphere-4.0.0-RC1/sharding-core/sharding-core-execute/src/main/java/org/apache/shardingsphere/core/execute/hook/SPISQLExecutionHook.java

public final class SPISQLExecutionHook implements SQLExecutionHook {
    
    private final Collection<SQLExecutionHook> sqlExecutionHooks = NewInstanceServiceLoader.newServiceInstances(SQLExecutionHook.class);
    
    static {
        NewInstanceServiceLoader.register(SQLExecutionHook.class);
    }
    
    @Override
    public void start(final RouteUnit routeUnit, final DataSourceMetaData dataSourceMetaData, final boolean isTrunkThread, final Map<String, Object> shardingExecuteDataMap) {
        for (SQLExecutionHook each : sqlExecutionHooks) {
            each.start(routeUnit, dataSourceMetaData, isTrunkThread, shardingExecuteDataMap);
        }
    }
    
    @Override
    public void finishSuccess() {
        for (SQLExecutionHook each : sqlExecutionHooks) {
            each.finishSuccess();
        }
    }
    
    @Override
    public void finishFailure(final Exception cause) {
        for (SQLExecutionHook each : sqlExecutionHooks) {
            each.finishFailure(cause);
        }
    }
}

• SPISQLExecutionHook實現了SQLExecutionHook接口；它使用NewInstanceServiceLoader注冊了SQLExecutionHook；sqlExecutionHooks集合由NewInstanceServiceLoader.newServiceInstances創建；start方法遍歷sqlExecutionHooks，執行其start方法；finishSuccess方法則遍歷sqlExecutionHooks，執行其finishSuccess方法；finishFailure方法則遍歷sqlExecutionHooks，執行其finishFailure方法

OpenTracingSQLExecutionHook

incubator-shardingsphere-4.0.0-RC1/sharding-opentracing/src/main/java/org/apache/shardingsphere/opentracing/hook/OpenTracingSQLExecutionHook.java

public final class OpenTracingSQLExecutionHook implements SQLExecutionHook {
    
    private static final String OPERATION_NAME = "/" + ShardingTags.COMPONENT_NAME + "/executeSQL/";
    
    private ActiveSpan activeSpan;
    
    private Span span;
    
    @Override
    public void start(final RouteUnit routeUnit, final DataSourceMetaData dataSourceMetaData, final boolean isTrunkThread, final Map<String, Object> shardingExecuteDataMap) {
        if (!isTrunkThread) {
            activeSpan = ((ActiveSpan.Continuation) shardingExecuteDataMap.get(OpenTracingRootInvokeHook.ACTIVE_SPAN_CONTINUATION)).activate();
        }
        span = ShardingTracer.get().buildSpan(OPERATION_NAME)
                .withTag(Tags.COMPONENT.getKey(), ShardingTags.COMPONENT_NAME)
                .withTag(Tags.SPAN_KIND.getKey(), Tags.SPAN_KIND_CLIENT)
                .withTag(Tags.PEER_HOSTNAME.getKey(), dataSourceMetaData.getHostName())
                .withTag(Tags.PEER_PORT.getKey(), dataSourceMetaData.getPort())
                .withTag(Tags.DB_TYPE.getKey(), "sql")
                .withTag(Tags.DB_INSTANCE.getKey(), routeUnit.getDataSourceName())
                .withTag(Tags.DB_STATEMENT.getKey(), routeUnit.getSqlUnit().getSql())
                .withTag(ShardingTags.DB_BIND_VARIABLES.getKey(), toString(routeUnit.getSqlUnit().getParameters())).startManual();
        
    }
    
    private String toString(final List<Object> parameterSets) {
        return parameterSets.isEmpty() ? "" : String.format("[%s]", Joiner.on(", ").join(parameterSets));
    }
    
    @Override
    public void finishSuccess() {
        span.finish();
        if (null != activeSpan) {
            activeSpan.deactivate();
        }
    }
    
    @Override
    public void finishFailure(final Exception cause) {
        ShardingErrorSpan.setError(span, cause);
        span.finish();
        if (null != activeSpan) {
            activeSpan.deactivate();
        }
    }
}

• OpenTracingSQLExecutionHook實現了SQLExecutionHook接口，其start方法創建并啟動span、activeSpan；finishSuccess及finishFailure方法都會執行span.finish()及activeSpan.deactivate()，只是finishFailure則會標記span的exception信息

SQLExecuteCallback

incubator-shardingsphere-4.0.0-RC1/sharding-core/sharding-core-execute/src/main/java/org/apache/shardingsphere/core/execute/sql/execute/SQLExecuteCallback.java

@RequiredArgsConstructor
public abstract class SQLExecuteCallback<T> implements ShardingGroupExecuteCallback<StatementExecuteUnit, T> {
    
    private final DatabaseType databaseType;
    
    private final boolean isExceptionThrown;
    
    @Override
    public final Collection<T> execute(final Collection<StatementExecuteUnit> statementExecuteUnits, final boolean isTrunkThread,
                                       final Map<String, Object> shardingExecuteDataMap) throws SQLException {
        Collection<T> result = new LinkedList<>();
        for (StatementExecuteUnit each : statementExecuteUnits) {
            result.add(execute0(each, isTrunkThread, shardingExecuteDataMap));
        }
        return result;
    }
    
    private T execute0(final StatementExecuteUnit statementExecuteUnit, final boolean isTrunkThread, final Map<String, Object> shardingExecuteDataMap) throws SQLException {
        ExecutorExceptionHandler.setExceptionThrown(isExceptionThrown);
        DataSourceMetaData dataSourceMetaData = DataSourceMetaDataFactory.newInstance(databaseType, statementExecuteUnit.getStatement().getConnection().getMetaData().getURL());
        SQLExecutionHook sqlExecutionHook = new SPISQLExecutionHook();
        try {
            sqlExecutionHook.start(statementExecuteUnit.getRouteUnit(), dataSourceMetaData, isTrunkThread, shardingExecuteDataMap);
            T result = executeSQL(statementExecuteUnit.getRouteUnit(), statementExecuteUnit.getStatement(), statementExecuteUnit.getConnectionMode());
            sqlExecutionHook.finishSuccess();
            return result;
        } catch (final SQLException ex) {
            sqlExecutionHook.finishFailure(ex);
            ExecutorExceptionHandler.handleException(ex);
            return null;
        }
    }
    
    protected abstract T executeSQL(RouteUnit routeUnit, Statement statement, ConnectionMode connectionMode) throws SQLException;
}

• SQLExecuteCallback的execute0方法在執行前創建SPISQLExecutionHook，然后調用sqlExecutionHook.start方法，執行成功之后執行sqlExecutionHook.finishSuccess方法，捕獲到SQLException則執行sqlExecutionHook.finishFailure方法

小結

SQLExecutionHook接口定義了start、finishSuccess、finishFailure方法；SPISQLExecutionHook實現了SQLExecutionHook接口；它使用NewInstanceServiceLoader注冊了SQLExecutionHook；sqlExecutionHooks集合由NewInstanceServiceLoader.newServiceInstances創建；start方法遍歷sqlExecutionHooks，執行其start方法；finishSuccess方法則遍歷sqlExecutionHooks，執行其finishSuccess方法；finishFailure方法則遍歷sqlExecutionHooks，執行其finishFailure方法

到此，相信大家對“sharding-jdbc中SQLExecutionHook的用法”有了更深的了解，不妨來實際操作一番吧！這里是億速云網站，更多相關內容可以進入相關頻道進行查詢，關注我們，繼續學習！