Windows

原文链接

Windows是处理无限流的核心。Windows将流分割为有限大小的“桶”，我们可以在其中应用计算。本文档关注的是如何在Flink中执行窗口操作，以及程序员如何从其提供的功能中获益最多。

下面给出了一个窗口Flink程序的一般结构。第一个片段指的是keyed流，而第一个代码片段是指non-keyed流。正如你所看到的，唯一的区别是keyed流调用keyBy(…)，并且对于non-keyed流window(…)变成了windowAll(…)。这也将成为本页其余部分的路线图。

Keyed Windows

stream
       .keyBy(...)               <-  keyed versus non-keyed windows
       .window(...)              <-  required: "assigner"
      [.trigger(...)]            <-  optional: "trigger" (else default trigger)
      [.evictor(...)]            <-  optional: "evictor" (else no evictor)
      [.allowedLateness(...)]    <-  optional: "lateness" (else zero)
      [.sideOutputLateData(...)] <-  optional: "output tag" (else no side output for late data)
       .reduce/aggregate/fold/apply()      <-  required: "function"
      [.getSideOutput(...)]      <-  optional: "output tag"

Non-Keyed Windows

stream
       .windowAll(...)           <-  required: "assigner"
      [.trigger(...)]            <-  optional: "trigger" (else default trigger)
      [.evictor(...)]            <-  optional: "evictor" (else no evictor)
      [.allowedLateness(...)]    <-  optional: "lateness" (else zero)
      [.sideOutputLateData(...)] <-  optional: "output tag" (else no side output for late data)
       .reduce/aggregate/fold/apply()      <-  required: "function"
      [.getSideOutput(...)]      <-  optional: "output tag"