【TDengine 使用环境】
测试 环境
【TDengine 版本】3.4
【操作系统以及版本】 debain
【部署方式】 单机版
【描述业务影响】
使用 流式计算 统计 ，结果无法写入结果表中，或者不正确
【遇到的问题：问题现象及影响】
使用 流式计算 统计 ，有联合主键时，结果无法写入结果表中，流式计算的查询部分是有结果的。
没联合主键时，统计结果没有正确写入子表
【资源配置】
测试表sql及流式计算sql 【测试表以商家店铺积分统计为例】
说明：
1 如果目标表有联合主键[COMPOSITE KEY]，结果无法写入目标表，执行任务后，结果表是空的,
2 去掉 【COMPOSITE KEY】 统计不正确，没有把正确的结果写到目标表，多个子表出现相同数据，没有把正确数据写到相应的子表; 另外，[去掉联合主键，多店铺数据统计写入肯定是有问题的，表数据是按时间覆盖，先不论]
drop TABLE IF EXISTS `statistics`.test_record ;
CREATE TABLE IF NOT EXISTS `statistics`.test_record (
ts TIMESTAMP,
id BIGINT COMPOSITE KEY,
order_num BIGINT,
points SMALLINT,
shop_id bigint
) TAGS (
merchant_id BIGINT,
area_code int
);
drop TABLE IF EXISTS statistics_h.test_income_h;
CREATE TABLE IF NOT EXISTS statistics_h.test_income_h
(
ts TIMESTAMP,
shop_id bigint COMPOSITE KEY,
points INT
) TAGS (
merchant_id BIGINT,
area_code int
);
drop stream IF EXISTS statistics.test_str_h_test_per;
CREATE stream statistics.test_str_h_test_per period ( 1m )
FROM
`statistics`.test_record PARTITION BY merchant_id,area_code
INTO statistics_h.test_income_h
output_subtable ( concat( 'test_income_h_', cast( merchant_id AS BINARY ), '_', cast( area_code AS BINARY ) ) )
( ts, shop_id COMPOSITE KEY, points )
tags ( merchant_id BIGINT AS merchant_id, area_code INT AS area_code )
AS SELECT
timetruncate ( ts, 1h, 1 ) AS ts,shop_id,
cast( sum( points ) AS INT ) AS points
FROM
statistics.test_record PARTITION BY ts,
merchant_id,
area_code, shop_id
测试数据
INSERT INTO
`statistics`.test_record_101_110000 USING `statistics`.test_record TAGS(101, 110000)
VALUES
(now - 1d, 1001, 150, 25, 2001)
(now - 2d, 1002, 200, 30, 2002)
(now - 3d, 1003, 120, 20, 2003)
(now - 25h, 1003, 120, 25, 2003)
`statistics`.test_record_102_120000 USING `statistics`.test_record TAGS(102, 120000)
VALUES
(now - 1d, 1004, 180, 35, 2004)
(now - 2d, 1005, 95, 15, 2005)
(now - 3d, 1006, 220, 40, 2006)
`statistics`.test_record_103_130000 USING `statistics`.test_record TAGS(103, 130000)
VALUES
(now - 1d, 1007, 165, 28, 2007)
(now - 2d, 1008, 140, 22, 2008)
(now - 3d, 1009, 190, 32, 2009)
`statistics`.test_record_104_140000 USING `statistics`.test_record TAGS(104, 140000)
VALUES
(now - 1d, 1010, 175, 27, 2010)
(now - 2d, 1011, 88, 18, 2011)
(now - 3d, 1012, 105, 24, 2012)
`statistics`.test_record_105_150000 USING `statistics`.test_record TAGS(105, 150000)
VALUES
(now - 1d, 1013, 75, 12, 2013)
(now - 2d, 1014, 92, 19, 2014)
(now - 3d, 1015, 65, 10, 2015)
(now - 1m, 1015, 65, 10, 2015)
(now - 1h, 1015, 65, 11, 2015)
(now - 2h, 1015, 65, 140, 2015)
`statistics`.test_record_106_160000 USING `statistics`.test_record TAGS(106, 160000)
VALUES
(now - 1d, 1016, 130, 26, 2016)
(now - 2d, 1017, 155, 31, 2017)
(now - 3d, 1018, 145, 23, 2018);
【报错完整截图】（不要大段的粘贴报错代码，论坛直接看报错代码不直观）
