郭元方 created FLINK-26307: --------------------------- Summary: 当事件时间在(Long.MIN_VALUE ,-windowsize)时,不能正确的求到开启窗口的起始时间 Key: FLINK-26307 URL: https://issues.apache.org/jira/browse/FLINK-26307 Project: Flink Issue Type: Improvement Components: API / DataStream Reporter: 郭元方
对于底层实现窗口时,求其起始时间的方法: public static long getWindowStartWithOffset(long timestamp, long offset, long windowSize) { return timestamp - (timestamp - offset + windowSize) % windowSize; } 例如:窗口大小为10,事件时间分别为1s,7s,12s得到的起始时间,正确。 => timestamp - (timestamp - offset + windowSize) % windowSize => 1 - (1 -0 + 10) % 10 => start = 0s => 7 - (7 -0 + 10) % 10 => start = 0s => 12 - (12 - 0 + 10) % 10 => start = 10s 但是当事件时间为 -1s,-7s,-12s,-15s,-17s,-19s时,窗口的起始时间却也都为-10s,是否由于在java中的取余规则,正数对正数取余的商值向0逼近,而负数对正数取余的商值也向0逼近 导致该算法对事件时间在(Long.MIN_VALUE,-windowsize)的这个范围内的数据不能正确的求到开启窗口的起始时间呢? => timestamp - (timestamp - offset + windowSize) % windowSize => -1 - (-1 -0 + 10) % 10 => start = -10s => -7 - (-7 -0 + 10) % 10 => start = -10s => -12 - (-12 - 0 + 10) % 10 => start = -10s => -15 - (-15 - 0 + 10) % 10 => start = -10s => -17 - (-17 - 0 + 10) % 10 => start = -10s => -19 - (-19 - 0 + 10) % 10 => start = -10s 和同学邓子琦的讨论,并没有经历过业务,所以此处事件时间求到了负值,现实世界来说不一定,但逻辑上来说应该可以求到负值吧。 -- This message was sent by Atlassian Jira (v8.20.1#820001)