手搓责任链框架 6：高级应用

杨杨杨大侠2025-09-05 9:17

高级应用

概述

责任链框架的高级应用涵盖了过滤器链、中断处理机制、性能优化等多个方面。这些高级特性使得框架能够应对复杂的业务场景，提供更强大的功能和更好的性能表现。本章将详细介绍这些高级应用的设计与实现。

过滤器链实现

1. Web过滤器链

graph LR A[HTTP请求] --> B[CORS过滤器] B --> C[认证过滤器] C --> D[授权过滤器] D --> E[限流过滤器] E --> F[日志过滤器] F --> G[业务处理器] G --> H[HTTP响应] B -.->|OPTIONS请求| H C -.->|认证失败| H D -.->|权限不足| H E -.->|超出限制| H
java 复制代码 
/**
 * Web过滤器链实现
 */
public class WebFilterChain {
    
    private final ChainExecutor<HttpRequest, HttpResponse> executor;
    
    public WebFilterChain() {
        this.executor = buildFilterChain();
    }
    
    /**
     * 构建Web过滤器链
     */
    private ChainExecutor<HttpRequest, HttpResponse> buildFilterChain() {
        ChainRegistry<HttpRequest, HttpResponse> registry = new ChainRegistry<>();
        
        // 构建过滤器链
        Handler chain = new AdvancedChainBuilder<HttpRequest, HttpResponse>()
            .addHandler(new CorsFilter())
            .addHandler(new AuthenticationFilter())
            .addHandler(new AuthorizationFilter())
            .addHandler(new RateLimitFilter())
            .addHandler(new LoggingFilter())
            .addHandler(new BusinessHandler())
            .build();
        
        registry.registerChain("web-filter", chain);
        return new ChainExecutor<>(registry);
    }
    
    /**
     * 处理HTTP请求
     */
    public HttpResponse doFilter(HttpRequest request) {
        Context<HttpRequest, HttpResponse> context = new DefaultContext<>(request);
        
        try {
            return executor.execute("web-filter", context);
        } catch (ChainExecutionException e) {
            return createErrorResponse(e);
        }
    }
    
    private HttpResponse createErrorResponse(ChainExecutionException e) {
        return HttpResponse.builder()
            .status(500)
            .body("Internal Server Error: " + e.getMessage())
            .build();
    }
}

/**
 * CORS过滤器
 */
public class CorsFilter extends AbstractHandler<HttpRequest, HttpResponse> {
    
    @Override
    protected boolean doHandle(Context<HttpRequest, HttpResponse> context) {
        HttpRequest request = context.getRequest();
        
        // 处理预检请求
        if ("OPTIONS".equals(request.getMethod())) {
            HttpResponse response = HttpResponse.builder()
                .status(200)
                .header("Access-Control-Allow-Origin", "*")
                .header("Access-Control-Allow-Methods", "GET, POST, PUT, DELETE, OPTIONS")
                .header("Access-Control-Allow-Headers", "Content-Type, Authorization")
                .build();
            
            context.setResponse(response);
            return false; // 预检请求直接返回，不继续处理
        }
        
        // 设置CORS头
        context.setAttribute("cors.origin", "*");
        context.setAttribute("cors.methods", "GET, POST, PUT, DELETE");
        
        return true;
    }
}

/**
 * 限流过滤器
 */
public class RateLimitFilter extends AbstractHandler<HttpRequest, HttpResponse> {
    
    private final RateLimiter rateLimiter;
    
    public RateLimitFilter() {
        this.rateLimiter = RateLimiter.create(100.0); // 每秒100个请求
    }
    
    @Override
    protected boolean doHandle(Context<HttpRequest, HttpResponse> context) {
        if (!rateLimiter.tryAcquire()) {
            HttpResponse response = HttpResponse.builder()
                .status(429)
                .body("Too Many Requests")
                .build();
            
            context.setResponse(response);
            return false; // 限流，直接返回
        }
        
        return true;
    }
}

2. 数据处理管道

java 复制代码 
/**
 * 数据处理管道
 */
public class DataProcessingPipeline {
    
    private final ChainExecutor<DataPacket, ProcessingResult> executor;
    
    public DataProcessingPipeline() {
        this.executor = buildPipeline();
    }
    
    /**
     * 构建数据处理管道
     */
    private ChainExecutor<DataPacket, ProcessingResult> buildPipeline() {
        ChainRegistry<DataPacket, ProcessingResult> registry = new ChainRegistry<>();
        
        Handler chain = new AdvancedChainBuilder<DataPacket, ProcessingResult>()
            .addHandler(new DataValidationProcessor())
            .addHandler(new DataTransformationProcessor())
            .addHandler(new DataEnrichmentProcessor())
            .addHandler(new DataPersistenceProcessor())
            .addHandler(new DataNotificationProcessor())
            .build();
        
        registry.registerChain("data-processing", chain);
        return new ChainExecutor<>(registry);
    }
    
    /**
     * 处理数据包
     */
    public ProcessingResult process(DataPacket dataPacket) {
        Context<DataPacket, ProcessingResult> context = new DefaultContext<>(dataPacket);
        
        try {
            return executor.execute("data-processing", context);
        } catch (ChainExecutionException e) {
            return ProcessingResult.failure(e.getMessage());
        }
    }
}

/**
 * 数据验证处理器
 */
public class DataValidationProcessor extends AbstractHandler<DataPacket, ProcessingResult> {
    
    @Override
    protected boolean doHandle(Context<DataPacket, ProcessingResult> context) {
        DataPacket packet = context.getRequest();
        
        // 数据格式验证
        if (!isValidFormat(packet)) {
            context.setResponse(ProcessingResult.failure("Invalid data format"));
            return false;
        }
        
        // 数据完整性验证
        if (!isDataComplete(packet)) {
            context.setResponse(ProcessingResult.failure("Incomplete data"));
            return false;
        }
        
        // 业务规则验证
        if (!isBusinessRuleValid(packet)) {
            context.setResponse(ProcessingResult.failure("Business rule validation failed"));
            return false;
        }
        
        context.setAttribute("validation.passed", true);
        return true;
    }
    
    private boolean isValidFormat(DataPacket packet) {
        // 实现数据格式验证逻辑
        return packet.getData() != null && !packet.getData().isEmpty();
    }
    
    private boolean isDataComplete(DataPacket packet) {
        // 实现数据完整性验证逻辑
        return packet.getHeaders() != null && packet.getTimestamp() != null;
    }
    
    private boolean isBusinessRuleValid(DataPacket packet) {
        // 实现业务规则验证逻辑
        return true;
    }
}

/**
 * 数据转换处理器
 */
public class DataTransformationProcessor extends AbstractHandler<DataPacket, ProcessingResult> {
    
    @Override
    protected boolean doHandle(Context<DataPacket, ProcessingResult> context) {
        DataPacket packet = context.getRequest();
        
        try {
            // 数据格式转换
            String transformedData = transformDataFormat(packet.getData());
            
            // 数据标准化
            String normalizedData = normalizeData(transformedData);
            
            // 数据清洗
            String cleanedData = cleanData(normalizedData);
            
            // 更新数据包
            packet.setData(cleanedData);
            context.setAttribute("transformation.completed", true);
            
            return true;
        } catch (Exception e) {
            context.setResponse(ProcessingResult.failure("Data transformation failed: " + e.getMessage()));
            return false;
        }
    }
    
    private String transformDataFormat(String data) {
        // 实现数据格式转换逻辑
        return data.toUpperCase();
    }
    
    private String normalizeData(String data) {
        // 实现数据标准化逻辑
        return data.trim();
    }
    
    private String cleanData(String data) {
        // 实现数据清洗逻辑
        return data.replaceAll("[^a-zA-Z0-9\\s]", "");
    }
}

中断处理机制

1. 优雅中断

java 复制代码 
/**
 * 可中断的责任链执行器
 */
public class InterruptibleChainExecutor<T, R> extends ChainExecutor<T, R> {
    
    private final AtomicBoolean interrupted = new AtomicBoolean(false);
    private final Map<String, CompletableFuture<R>> runningExecutions = new ConcurrentHashMap<>();
    
    public InterruptibleChainExecutor(ChainRegistry<T, R> chainRegistry) {
        super(chainRegistry);
    }
    
    @Override
    public R execute(String chainId, Context<T, R> context) throws ChainExecutionException {
        String executionId = generateExecutionId();
        
        try {
            // 检查是否已被中断
            checkInterrupted();
            
            // 注册执行
            CompletableFuture<R> future = CompletableFuture.supplyAsync(() -> {
                try {
                    return super.execute(chainId, context);
                } catch (ChainExecutionException e) {
                    throw new CompletionException(e);
                }
            });
            
            runningExecutions.put(executionId, future);
            
            // 等待执行完成或中断
            return future.get();
            
        } catch (InterruptedException e) {
            Thread.currentThread().interrupt();
            throw new ChainExecutionException("Chain execution interrupted", e);
        } catch (ExecutionException e) {
            if (e.getCause() instanceof ChainExecutionException) {
                throw (ChainExecutionException) e.getCause();
            } else {
                throw new ChainExecutionException("Chain execution failed", e.getCause());
            }
        } finally {
            runningExecutions.remove(executionId);
        }
    }
    
    /**
     * 中断所有正在执行的链
     */
    public void interrupt() {
        interrupted.set(true);
        
        // 取消所有正在执行的任务
        runningExecutions.values().forEach(future -> future.cancel(true));
        runningExecutions.clear();
    }
    
    /**
     * 检查是否被中断
     */
    private void checkInterrupted() throws ChainExecutionException {
        if (interrupted.get() || Thread.currentThread().isInterrupted()) {
            throw new ChainExecutionException("Chain execution interrupted");
        }
    }
    
    /**
     * 重置中断状态
     */
    public void resetInterrupt() {
        interrupted.set(false);
    }
}

/**
 * 可中断的处理器基类
 */
public abstract class InterruptibleHandler<T, R> extends AbstractHandler<T, R> {
    
    @Override
    protected final boolean doHandle(Context<T, R> context) {
        try {
            // 检查中断状态
            checkInterrupted();
            
            // 执行具体处理逻辑
            return doInterruptibleHandle(context);
            
        } catch (InterruptedException e) {
            Thread.currentThread().interrupt();
            throw new ChainExecutionException("Handler interrupted", e);
        }
    }
    
    /**
     * 可中断的处理逻辑
     */
    protected abstract boolean doInterruptibleHandle(Context<T, R> context) throws InterruptedException;
    
    /**
     * 检查中断状态
     */
    protected void checkInterrupted() throws InterruptedException {
        if (Thread.currentThread().isInterrupted()) {
            throw new InterruptedException("Handler execution interrupted");
        }
    }
}

2. 超时中断

java 复制代码 
/**
 * 超时处理器装饰器
 */
public class TimeoutHandlerDecorator<T, R> implements Handler {
    
    private final Handler delegate;
    private final long timeoutMillis;
    private final ExecutorService executorService;
    
    public TimeoutHandlerDecorator(Handler delegate, long timeoutMillis) {
        this.delegate = delegate;
        this.timeoutMillis = timeoutMillis;
        this.executorService = Executors.newCachedThreadPool();
    }
    
    @Override
    public boolean handle(Context context) {
        CompletableFuture<Boolean> future = CompletableFuture.supplyAsync(
            () -> delegate.handle(context), executorService);
        
        try {
            return future.get(timeoutMillis, TimeUnit.MILLISECONDS);
        } catch (TimeoutException e) {
            future.cancel(true);
            throw new ChainExecutionException("Handler execution timeout", e);
        } catch (InterruptedException e) {
            Thread.currentThread().interrupt();
            future.cancel(true);
            throw new ChainExecutionException("Handler execution interrupted", e);
        } catch (ExecutionException e) {
            throw new ChainExecutionException("Handler execution failed", e.getCause());
        }
    }
    
    @Override
    public void setNext(Handler handler) {
        delegate.setNext(handler);
    }
    
    @Override
    public Handler getNext() {
        return delegate.getNext();
    }
}

/**
 * 超时感知的链构建器
 */
public class TimeoutAwareChainBuilder<T, R> extends AdvancedChainBuilder<T, R> {
    
    private long defaultTimeoutMillis = 5000; // 默认5秒超时
    
    /**
     * 设置默认超时时间
     */
    public TimeoutAwareChainBuilder<T, R> setDefaultTimeout(long timeoutMillis) {
        this.defaultTimeoutMillis = timeoutMillis;
        return this;
    }
    
    /**
     * 添加带超时的处理器
     */
    public TimeoutAwareChainBuilder<T, R> addHandlerWithTimeout(Handler handler, long timeoutMillis) {
        Handler timeoutHandler = new TimeoutHandlerDecorator(handler, timeoutMillis);
        return (TimeoutAwareChainBuilder<T, R>) addHandler(timeoutHandler);
    }
    
    /**
     * 添加使用默认超时的处理器
     */
    @Override
    public ChainBuilder<T, R> addHandler(Handler handler) {
        return addHandlerWithTimeout(handler, defaultTimeoutMillis);
    }
}

性能优化建议

1. 内存优化

java 复制代码 
/**
 * 内存优化的上下文实现
 */
public class MemoryOptimizedContext<T, R> implements Context<T, R> {
    
    private final T request;
    private R response;
    private Map<String, Object> attributes;
    
    public MemoryOptimizedContext(T request) {
        this.request = request;
        // 延迟初始化attributes，节省内存
    }
    
    @Override
    public void setAttribute(String key, Object value) {
        if (attributes == null) {
            attributes = new HashMap<>();
        }
        
        if (value == null) {
            attributes.remove(key);
            // 如果attributes为空，释放内存
            if (attributes.isEmpty()) {
                attributes = null;
            }
        } else {
            attributes.put(key, value);
        }
    }
    
    @Override
    public Object getAttribute(String key) {
        return attributes != null ? attributes.get(key) : null;
    }
    
    @Override
    public void clearAttributes() {
        if (attributes != null) {
            attributes.clear();
            attributes = null; // 释放内存
        }
    }
    
    // 其他方法实现...
}

/**
 * 轻量级处理器基类
 */
public abstract class LightweightHandler<T, R> implements Handler {
    
    private Handler nextHandler;
    
    @Override
    public final boolean handle(Context context) {
        try {
            boolean result = doHandle(context);
            
            if (result && nextHandler != null) {
                return nextHandler.handle(context);
            }
            
            return result;
        } catch (Exception e) {
            handleError(context, e);
            throw e;
        }
    }
    
    /**
     * 轻量级处理逻辑
     */
    protected abstract boolean doHandle(Context<T, R> context);
    
    /**
     * 简化的错误处理
     */
    protected void handleError(Context<T, R> context, Exception e) {
        // 默认空实现，减少内存占用
    }
    
    @Override
    public void setNext(Handler handler) {
        this.nextHandler = handler;
    }
    
    @Override
    public Handler getNext() {
        return nextHandler;
    }
}

2. 并发优化

java 复制代码 
/**
 * 并发优化的链执行器
 */
public class ConcurrentChainExecutor<T, R> extends ChainExecutor<T, R> {
    
    private final ForkJoinPool forkJoinPool;
    private final int parallelismThreshold;
    
    public ConcurrentChainExecutor(ChainRegistry<T, R> chainRegistry, 
                                  int parallelism, 
                                  int parallelismThreshold) {
        super(chainRegistry);
        this.forkJoinPool = new ForkJoinPool(parallelism);
        this.parallelismThreshold = parallelismThreshold;
    }
    
    /**
     * 并行执行多个请求
     */
    public List<R> executeParallel(String chainId, List<Context<T, R>> contexts) {
        if (contexts.size() < parallelismThreshold) {
            // 请求数量较少，使用串行执行
            return contexts.stream()
                .map(context -> {
                    try {
                        return execute(chainId, context);
                    } catch (ChainExecutionException e) {
                        throw new RuntimeException(e);
                    }
                })
                .collect(Collectors.toList());
        } else {
            // 请求数量较多，使用并行执行
            return contexts.parallelStream()
                .map(context -> {
                    try {
                        return execute(chainId, context);
                    } catch (ChainExecutionException e) {
                        throw new RuntimeException(e);
                    }
                })
                .collect(Collectors.toList());
        }
    }
    
    /**
     * 异步批量执行
     */
    public CompletableFuture<List<R>> executeBatchAsync(String chainId, List<Context<T, R>> contexts) {
        List<CompletableFuture<R>> futures = contexts.stream()
            .map(context -> executeAsync(chainId, context))
            .collect(Collectors.toList());
        
        return CompletableFuture.allOf(futures.toArray(new CompletableFuture[0]))
            .thenApply(v -> futures.stream()
                .map(CompletableFuture::join)
                .collect(Collectors.toList()));
    }
    
    @Override
    public void shutdown() {
        super.shutdown();
        forkJoinPool.shutdown();
    }
}

/**
 * 无锁的链注册器
 */
public class LockFreeChainRegistry<T, R> implements ChainRegistry<T, R> {
    
    private final ConcurrentHashMap<String, Handler> chains = new ConcurrentHashMap<>();
    private final AtomicReference<Map<String, Handler>> chainsRef = new AtomicReference<>(new HashMap<>());
    
    @Override
    public void registerChain(String chainId, Handler chain) {
        chains.put(chainId, chain);
        
        // 更新不可变映射
        Map<String, Handler> currentChains = chainsRef.get();
        Map<String, Handler> newChains = new HashMap<>(currentChains);
        newChains.put(chainId, chain);
        chainsRef.set(Collections.unmodifiableMap(newChains));
    }
    
    @Override
    public Handler getChain(String chainId) {
        return chainsRef.get().get(chainId);
    }
    
    @Override
    public Set<String> getChainIds() {
        return chainsRef.get().keySet();
    }
}

3. 缓存优化

java 复制代码 
/**
 * 缓存感知的链执行器
 */
public class CacheAwareChainExecutor<T, R> extends ChainExecutor<T, R> {
    
    private final Cache<String, R> responseCache;
    private final Function<Context<T, R>, String> cacheKeyGenerator;
    
    public CacheAwareChainExecutor(ChainRegistry<T, R> chainRegistry,
                                  Cache<String, R> responseCache,
                                  Function<Context<T, R>, String> cacheKeyGenerator) {
        super(chainRegistry);
        this.responseCache = responseCache;
        this.cacheKeyGenerator = cacheKeyGenerator;
    }
    
    @Override
    public R execute(String chainId, Context<T, R> context) throws ChainExecutionException {
        // 生成缓存键
        String cacheKey = generateCacheKey(chainId, context);
        
        // 尝试从缓存获取
        R cachedResponse = responseCache.getIfPresent(cacheKey);
        if (cachedResponse != null) {
            logger.debug("Cache hit for key: {}", cacheKey);
            return cachedResponse;
        }
        
        // 缓存未命中，执行链
        R response = super.execute(chainId, context);
        
        // 缓存响应
        if (response != null && isCacheable(context, response)) {
            responseCache.put(cacheKey, response);
            logger.debug("Cached response for key: {}", cacheKey);
        }
        
        return response;
    }
    
    private String generateCacheKey(String chainId, Context<T, R> context) {
        if (cacheKeyGenerator != null) {
            return cacheKeyGenerator.apply(context);
        } else {
            return chainId + ":" + context.getRequest().hashCode();
        }
    }
    
    private boolean isCacheable(Context<T, R> context, R response) {
        // 检查响应是否可缓存
        Boolean cacheable = context.getAttribute("cacheable", Boolean.class);
        return cacheable == null || cacheable;
    }
}

/**
 * 缓存处理器
 */
public class CacheHandler<T, R> extends AbstractHandler<T, R> {
    
    private final Cache<String, Object> cache;
    private final Function<T, String> keyGenerator;
    
    public CacheHandler(Cache<String, Object> cache, Function<T, String> keyGenerator) {
        this.cache = cache;
        this.keyGenerator = keyGenerator;
    }
    
    @Override
    protected boolean doHandle(Context<T, R> context) {
        T request = context.getRequest();
        String cacheKey = keyGenerator.apply(request);
        
        // 尝试从缓存获取
        Object cachedValue = cache.getIfPresent(cacheKey);
        if (cachedValue != null) {
            context.setAttribute("cached.value", cachedValue);
            context.setAttribute("cache.hit", true);
        } else {
            context.setAttribute("cache.hit", false);
        }
        
        return true;
    }
    
    @Override
    protected void onAfter(Context<T, R> context) {
        // 在处理完成后缓存结果
        Boolean cacheHit = context.getAttribute("cache.hit", Boolean.class);
        if (Boolean.FALSE.equals(cacheHit)) {
            Object valueToCache = context.getAttribute("cache.value");
            if (valueToCache != null) {
                String cacheKey = keyGenerator.apply(context.getRequest());
                cache.put(cacheKey, valueToCache);
            }
        }
    }
}

监控和诊断

1. 性能监控

java 复制代码 
/**
 * 性能监控处理器
 */
public class PerformanceMonitoringHandler<T, R> extends AbstractHandler<T, R> {
    
    private final MeterRegistry meterRegistry;
    private final Timer.Sample sample;
    
    public PerformanceMonitoringHandler(MeterRegistry meterRegistry) {
        this.meterRegistry = meterRegistry;
        this.sample = Timer.start(meterRegistry);
    }
    
    @Override
    protected void onBefore(Context<T, R> context) {
        // 记录开始时间
        context.setAttribute("performance.start.time", System.nanoTime());
        
        // 增加请求计数
        Counter.builder("chain.requests")
            .tag("handler", this.getClass().getSimpleName())
            .register(meterRegistry)
            .increment();
    }
    
    @Override
    protected boolean doHandle(Context<T, R> context) {
        // 实际的业务逻辑
        return processRequest(context);
    }
    
    @Override
    protected void onAfter(Context<T, R> context) {
        // 记录执行时间
        Long startTime = context.getAttribute("performance.start.time", Long.class);
        if (startTime != null) {
            long duration = System.nanoTime() - startTime;
            
            Timer.builder("chain.handler.duration")
                .tag("handler", this.getClass().getSimpleName())
                .register(meterRegistry)
                .record(duration, TimeUnit.NANOSECONDS);
        }
        
        // 记录成功计数
        Counter.builder("chain.success")
            .tag("handler", this.getClass().getSimpleName())
            .register(meterRegistry)
            .increment();
    }
    
    @Override
    protected void onError(Context<T, R> context, Exception e) {
        // 记录错误计数
        Counter.builder("chain.errors")
            .tag("handler", this.getClass().getSimpleName())
            .tag("error.type", e.getClass().getSimpleName())
            .register(meterRegistry)
            .increment();
    }
    
    protected boolean processRequest(Context<T, R> context) {
        // 子类实现具体的业务逻辑
        return true;
    }
}

/**
 * 健康检查处理器
 */
public class HealthCheckHandler extends AbstractHandler<HealthCheckRequest, HealthCheckResponse> {
    
    private final List<HealthIndicator> healthIndicators;
    
    public HealthCheckHandler(List<HealthIndicator> healthIndicators) {
        this.healthIndicators = healthIndicators;
    }
    
    @Override
    protected boolean doHandle(Context<HealthCheckRequest, HealthCheckResponse> context) {
        Map<String, Health> healthDetails = new HashMap<>();
        boolean overallHealthy = true;
        
        for (HealthIndicator indicator : healthIndicators) {
            try {
                Health health = indicator.health();
                healthDetails.put(indicator.getName(), health);
                
                if (health.getStatus() != Health.Status.UP) {
                    overallHealthy = false;
                }
            } catch (Exception e) {
                Health errorHealth = Health.down()
                    .withDetail("error", e.getMessage())
                    .build();
                healthDetails.put(indicator.getName(), errorHealth);
                overallHealthy = false;
            }
        }
        
        HealthCheckResponse response = HealthCheckResponse.builder()
            .status(overallHealthy ? "UP" : "DOWN")
            .details(healthDetails)
            .timestamp(Instant.now())
            .build();
        
        context.setResponse(response);
        return true;
    }
}

2. 分布式追踪

java 复制代码 
/**
 * 分布式追踪处理器
 */
public class TracingHandler<T, R> extends AbstractHandler<T, R> {
    
    private final Tracer tracer;
    
    public TracingHandler(Tracer tracer) {
        this.tracer = tracer;
    }
    
    @Override
    protected boolean doHandle(Context<T, R> context) {
        Span span = tracer.nextSpan()
            .name("chain.handler." + this.getClass().getSimpleName())
            .tag("handler.type", this.getClass().getSimpleName())
            .start();
        
        try (Tracer.SpanInScope ws = tracer.withSpanInScope(span)) {
            // 添加追踪信息到上下文
            context.setAttribute("trace.id", span.context().traceId());
            context.setAttribute("span.id", span.context().spanId());
            
            // 执行实际的处理逻辑
            boolean result = processWithTracing(context);
            
            // 添加结果标签
            span.tag("result", String.valueOf(result));
            
            return result;
        } catch (Exception e) {
            span.tag("error", e.getMessage());
            throw e;
        } finally {
            span.end();
        }
    }
    
    protected boolean processWithTracing(Context<T, R> context) {
        // 子类实现具体的业务逻辑
        return true;
    }
}

扩展机制

1. 插件系统

java 复制代码 
/**
 * 插件接口
 */
public interface ChainPlugin {
    
    /**
     * 插件名称
     */
    String getName();
    
    /**
     * 插件版本
     */
    String getVersion();
    
    /**
     * 初始化插件
     */
    void initialize(ChainPluginContext context);
    
    /**
     * 销毁插件
     */
    void destroy();
    
    /**
     * 插件是否启用
     */
    boolean isEnabled();
}

/**
 * 插件管理器
 */
public class ChainPluginManager {
    
    private final Map<String, ChainPlugin> plugins = new ConcurrentHashMap<>();
    private final ChainPluginContext pluginContext;
    
    public ChainPluginManager(ChainPluginContext pluginContext) {
        this.pluginContext = pluginContext;
    }
    
    /**
     * 注册插件
     */
    public void registerPlugin(ChainPlugin plugin) {
        if (plugins.containsKey(plugin.getName())) {
            throw new IllegalArgumentException("Plugin already registered: " + plugin.getName());
        }
        
        plugins.put(plugin.getName(), plugin);
        
        if (plugin.isEnabled()) {
            plugin.initialize(pluginContext);
        }
    }
    
    /**
     * 卸载插件
     */
    public void unregisterPlugin(String pluginName) {
        ChainPlugin plugin = plugins.remove(pluginName);
        if (plugin != null) {
            plugin.destroy();
        }
    }
    
    /**
     * 获取插件
     */
    public ChainPlugin getPlugin(String pluginName) {
        return plugins.get(pluginName);
    }
    
    /**
     * 获取所有插件
     */
    public Collection<ChainPlugin> getAllPlugins() {
        return plugins.values();
    }
    
    /**
     * 启用插件
     */
    public void enablePlugin(String pluginName) {
        ChainPlugin plugin = plugins.get(pluginName);
        if (plugin != null && !plugin.isEnabled()) {
            plugin.initialize(pluginContext);
        }
    }
    
    /**
     * 禁用插件
     */
    public void disablePlugin(String pluginName) {
        ChainPlugin plugin = plugins.get(pluginName);
        if (plugin != null && plugin.isEnabled()) {
            plugin.destroy();
        }
    }
}

2. 自定义扩展点

java 复制代码 
/**
 * 扩展点接口
 */
public interface ExtensionPoint {
    
    /**
     * 扩展点名称
     */
    String getName();
    
    /**
     * 执行扩展逻辑
     */
    void execute(ExtensionContext context);
}

/**
 * 扩展点管理器
 */
public class ExtensionPointManager {
    
    private final Map<String, List<ExtensionPoint>> extensionPoints = new ConcurrentHashMap<>();
    
    /**
     * 注册扩展点
     */
    public void registerExtension(String pointName, ExtensionPoint extension) {
        extensionPoints.computeIfAbsent(pointName, k -> new ArrayList<>()).add(extension);
    }
    
    /**
     * 执行扩展点
     */
    public void executeExtensions(String pointName, ExtensionContext context) {
        List<ExtensionPoint> extensions = extensionPoints.get(pointName);
        if (extensions != null) {
            for (ExtensionPoint extension : extensions) {
                try {
                    extension.execute(context);
                } catch (Exception e) {
                    logger.error("Extension execution failed: {}", extension.getName(), e);
                }
            }
        }
    }
}

总结

责任链框架的高级应用提供了强大的功能扩展：

  1. 过滤器链：支持Web请求过滤、数据处理管道等场景
  2. 中断处理：提供优雅中断、超时控制等机制
  3. 性能优化：内存优化、并发优化、缓存优化等策略
  4. 监控诊断：性能监控、健康检查、分布式追踪等功能
  5. 扩展机制：插件系统、自定义扩展点等扩展能力

通过这些高级特性，责任链框架能够满足复杂企业级应用的需求，提供高性能、高可用、易扩展的解决方案。

最佳实践总结

  1. 合理设计处理器粒度：避免过于细粒度或粗粒度的处理器
  2. 注意性能影响：监控链的执行时间，避免过长的处理链
  3. 完善错误处理：提供多层次的错误处理和恢复机制
  4. 使用缓存优化：对于重复计算，合理使用缓存提升性能
  5. 监控和诊断：建立完善的监控体系，便于问题定位
  6. 文档和测试：提供清晰的文档和完整的测试用例

通过遵循这些最佳实践，可以构建出高质量、高性能的责任链框架应用。

相关推荐
砖厂小工7 小时前
用 GLM + OpenClaw 打造你的 AI PR Review Agent — 让龙虾帮你审代码
android·github
程序员鱼皮7 小时前
又一个新项目完结，我要出海了！
ai·github·开源项目
徐小夕8 小时前
pxcharts-vue：一款专为 Vue3 打造的开源多维表格解决方案
前端·vue.js·github
Moment8 小时前
想要长期陪伴你的助理？先从部署一个 OpenClaw 开始 😍😍😍
前端·后端·github
我叫黑大帅8 小时前
前端如何利用 GitHub Actions 自动构建并发布到 GitHub Pages？
前端·面试·github
SimonKing9 小时前
OpenCode AI辅助编程，不一样的编程思路，不写一行代码
java·后端·程序员
FastBean9 小时前
Jackson View Extension Spring Boot Starter
java·后端
HelloGitHub10 小时前
这个年轻的开源项目，想让每个人都能拥有自己的专业级 AI 智能体
开源·github·agent
Seven9710 小时前
剑指offer-79、最⻓不含重复字符的⼦字符串
java
热门推荐
01GitHub 镜像站点02OpenClaw 使用和管理 MCP 完全指南03OpenClaw + 飞书（Feishu）环境搭建指南04Claude Code + GLM4.7 避坑指南：解决 Unable to connect to Anthropic services05Clawdbot部署教程：解决‘gateway token missing’授权问题的完整步骤06OpenClaw优化飞书API 额度已耗尽问题07【OpenClaw 本地实战 Ep.3】突破瓶颈：强制修改 openclaw.json 解锁 32k 上下文记忆08小黑课堂计算机二级WPSoffice题库软件下载安装教程（2026年3月最新版）09Window 10部署openclaw报错node.exe : npm error code 12810OpenClaw大龙虾机器人完整安装教程