# Flink窗口机制

## 窗口是什么

窗口是流处理中将无限流分割为有限块进行计算的机制。

## 窗口类型

### 1. 滚动窗口（Tumbling Window）

固定大小、不重叠的窗口。

```java
stream
    .keyBy(0)
    .window(TumblingProcessingTimeWindows.of(Time.seconds(5)))
    .sum(1);
```

### 2. 滑动窗口（Sliding Window）

固定大小、可重叠的窗口。

```java
stream
    .keyBy(0)
    .window(SlidingProcessingTimeWindows.of(Time.seconds(10), Time.seconds(5)))
    .sum(1);
```

### 3. 会话窗口（Session Window）

基于会话间隔的窗口。

```java
stream
    .keyBy(0)
    .window(ProcessingTimeSessionWindows.withGap(Time.seconds(30)))
    .sum(1);
```

### 4. 全局窗口（Global Window）

所有数据在一个窗口，需自定义触发器。

```java
stream
    .keyBy(0)
    .window(GlobalWindows.create())
    .trigger(CountTrigger.of(10))
    .sum(1);
```

## 窗口函数

### 1. ReduceFunction

```java
stream
    .keyBy(0)
    .window(TumblingProcessingTimeWindows.of(Time.seconds(5)))
    .reduce((a, b) -> Tuple2.of(a.f0, a.f1 + b.f1));
```

### 2. AggregateFunction

```java
stream
    .keyBy(0)
    .window(TumblingProcessingTimeWindows.of(Time.seconds(5)))
    .aggregate(new AggregateFunction<Tuple2<String, Integer>, Integer, Integer>() {
        public Integer createAccumulator() { return 0; }
        public Integer add(Tuple2<String, Integer> value, Integer acc) { return acc + value.f1; }
        public Integer getResult(Integer acc) { return acc; }
        public Integer merge(Integer a, Integer b) { return a + b; }
    });
```

### 3. ProcessWindowFunction

```java
stream
    .keyBy(0)
    .window(TumblingProcessingTimeWindows.of(Time.seconds(5)))
    .process(new ProcessWindowFunction<Tuple2<String, Integer>, String, Tuple, TimeWindow>() {
        public void process(Tuple key, Context ctx, Iterable<Tuple2<String, Integer>> elements, Collector<String> out) {
            int count = 0;
            for (Tuple2<String, Integer> e : elements) {
                count += e.f1;
            }
            out.collect("Window: " + ctx.window() + ", Count: " + count);
        }
    });
```

## 触发器与驱逐器

### 触发器（Trigger）

决定何时触发窗口计算。

### 驱逐器（Evictor）

在计算前后移除窗口元素。

## 总结

窗口是流处理的核心机制，选择合适的窗口类型对业务逻辑至关重要。


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://qiangrens-organization.gitbook.io/qkd90/12flink-shi-shi-ji-suan/flink-chuang-kou-ji-zhi.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.