Question

我正在尝试使用运动学分析为实时排行榜做一个概念证明，并尝试使用他们的入门示例，演示股票行情自动收录器数据流中的时间段示例。但是，当我尝试模拟按价格排序的页首横幅时，出现错误Leading column in ORDER BY must be monotonic。

SELECT from and INSERT into like a TABLE
    -- PUMP: an entity used to continuously 'SELECT ... FROM' a source STREAM, and INSERT SQL results into an output STREAM
    -- Create output stream, which can be used to send to a destination
    CREATE OR REPLACE STREAM "DESTINATION_SQL_STREAM" (ticker_symbol VARCHAR(4), price REAL);
    -- Create a pump which continuously selects from a source stream (SOURCE_SQL_STREAM_001)
    -- performs an aggregate count that is grouped by columns ticker over a 10-second tumbling window
    -- and inserts into output stream (DESTINATION_SQL_STREAM)
    CREATE OR REPLACE  PUMP "STREAM_PUMP" AS INSERT INTO "DESTINATION_SQL_STREAM"
    -- Aggregate function COUNT|AVG|MAX|MIN|SUM|STDDEV_POP|STDDEV_SAMP|VAR_POP|VAR_SAMP)
    SELECT STREAM  ticker_symbol, price
    FROM "SOURCE_SQL_STREAM_001"
    ORDER BY price desc
    -- Uses a 10-second tumbling time window
    GROUP BY ticker_symbol, price, FLOOR(("SOURCE_SQL_STREAM_001".ROWTIME - TIMESTAMP '1970-01-01 00:00:00') SECOND / 10 TO SECOND);

Answer 1

According to the documentation，除单调外，ORDER BY还必须基于时间：

如果流查询的前导表达式为，则可以使用ORDER BY 基于时间的单调

在您的情况下，您似乎正在计算GROUP BY中的时间戳，但是您必须将相同的计算结果添加到ORDER BY中（或者可能想出一种方法来在SELECT中进行选择并赋予它列别名）。

如何使用Kinesis Analytics创建排行榜查询

1 个答案: