下面代码报空指针

with `__all_dim__` as (
select
*
from (
select
from_unixtime(unix_timestamp(`__bts__`) -1,'yyyy-MM-dd HH:mm:ss') as `__bts__`
from (
select
concat_ws(' ', `d`.`date`, `t`.`time_of_day`) as `__bts__`
from `ecmp`.`dim_date` as `d`
left join `ecmp`.`dim_time_of_day` as `t` on 1 = 1
where
`d`.`date` >= '2020-01-12'
and `d`.`date` <= '2020-01-13'
) as `__bts___tp1`
where
`__bts__` > '2020-01-12 00:00:00'
and `__bts__` <= '2020-01-13 00:00:00'
and second(`__bts__`) = 0
and minute(`__bts__`) = 0
and hour(`__bts__`) = 0
and pmod(day(`__bts__`), 1) = 0
) as `__time_model__`
cross join (
select
`dd_59282`.`tenant_pk` as `tenant_pk`,
`dd_59282`.`tenant_id` as `tenant_id`,
`dd_59282`.`tenant_name` as `tenant_name`
from `ecmp`.`dim_tenant` as `dd_59282`
) as `tenant_pk`
cross join (
select
'Fatal' as incident_level from system.dual
union all
select
'Error' as incident_level from system.dual
union all
select
'Warning' as incident_level from system.dual
union all
select
'Info' as incident_level from system.dual
) as `incident_level`
)
,`t` as (
select
`tenant_pk`,
`incident_accept_violation_count`,
`incident_level`,
rank() over( partition by `incident_level` order by `incident_accept_violation_count` DESC) as `incident_accept_violation_count_rank`,
rank() over( partition by `incident_level` order by `incident_accept_violation_count` ASC) as `__inverse_rank__`
from (
select
`__all_dim__`.*,-- 经排查发现表名加单引号再.*就是会报空指针,去掉表名的单引号后该问题解决
CAST(round(nvl(`incident_accept_violation_count`, 0), 0) as INT) as `incident_accept_violation_count`
from `__all_dim__`
left join (
select
`incident_level`,
`tenant_pk`,
count(*) as `incident_accept_violation_count`
from `ecmp`.dwd_incident_accept
where
incident_accept_violation_flag = '违规'
and `incident_accept_time` >= '2020-01-12 00:00:00'
AND `incident_accept_time` <= '2020-01-12 23:59:59'
group by
`incident_level`,
`tenant_pk`
) as `t1` on 1 = 1
and `__all_dim__`.`tenant_pk` = `t1`.`tenant_pk`
and `__all_dim__`.`incident_level` = `t1`.`incident_level`
) as `t0`
) select
`__all_dim__`.`__bts__` as `__bts__`,
CAST(SYSDATE as STRING) as `__cts__`,
CAST(dround(nvl(`incident_accept_violation_count`, 0), 0) as INT) as `incident_accept_violation_count`,
CAST(dround(`incident_accept_violation_count_rank`, 0) as INT) as `incident_accept_violation_count_rank`,
CAST(dround(`incident_accept_violation_count_win_rate`, 1) as DOUBLE) as `incident_accept_violation_count_win_rate`,
`__all_dim__`.`incident_level` as `incident_level`,
`__all_dim__`.`tenant_id` as `tenant_id`,
`__all_dim__`.`tenant_name` as `tenant_name`,
`__all_dim__`.`tenant_pk` as `tenant_pk`
from `__all_dim__`
left join (
select
'2020-01-12 23:59:59' as `__bts__`,
`incident_accept_violation_count`,
`incident_accept_violation_count_rank`,
`incident_accept_violation_count_win_rate`,
CAST(coalesce(`tp1`.`incident_level`) as STRING) as `incident_level`,
CAST(coalesce(`tp1`.`tenant_pk`) as STRING) as `tenant_pk`
from (
select
`t`.`tenant_pk`,
`t`.`incident_level`,
`t`.`incident_accept_violation_count`,
`t`.`incident_accept_violation_count_rank`,
if(`c`.ct = 1,null,(`__inverse_rank__` -1) /(`c`.ct -1) * 100) as `incident_accept_violation_count_win_rate`
from `t`
left join (
select
`incident_level`,
count(*) as `ct`
from `t`
group by
`incident_level`
) as `c` on 1 = 1
and `t`.`incident_level` = `c`.`incident_level`
) as `tp1`
) as `__dws__` on `__all_dim__`.`__bts__` = `__dws__`.`__bts__`
and `__all_dim__`.`tenant_pk` = `__dws__`.`tenant_pk`
and `__all_dim__`.`incident_level` = `__dws__`.`incident_level`;

然后修改后直接SELECT * FROM t;是没有问题的,但是用先有的select报错

[Code: 10009, SQL State: 42000] COMPILE FAILED: Semantic error: [Error 10009] Line 54:8 Invalid table alias. Error encountered near token 'all_dim'

with alldim as (
select
*
from (
select
from_unixtime(unix_timestamp(`__bts__`) -1,'yyyy-MM-dd HH:mm:ss') as `__bts__`
from (
select
concat_ws(' ', `d`.`date`, t.`time_of_day`) as `__bts__`
from `ecmp`.`dim_date` as `d`
left join `ecmp`.`dim_time_of_day` as t on 1 = 1
where
`d`.`date` >= '2020-01-12'
and `d`.`date` <= '2020-01-13'
) as `__bts___tp1`
where
`__bts__` > '2020-01-12 00:00:00'
and `__bts__` <= '2020-01-13 00:00:00'
and second(`__bts__`) = 0
and minute(`__bts__`) = 0
and hour(`__bts__`) = 0
and pmod(day(`__bts__`), 1) = 0
) as `__time_model__`
cross join (
select
`dd_59282`.`tenant_pk` as `tenant_pk`,
`dd_59282`.`tenant_id` as `tenant_id`,
`dd_59282`.`tenant_name` as `tenant_name`
from `ecmp`.`dim_tenant` as `dd_59282`
) as `tenant_pk`
cross join (
select
'Fatal' as incident_level from system.dual
union all
select
'Error' as incident_level from system.dual
union all
select
'Warning' as incident_level from system.dual
union all
select
'Info' as incident_level from system.dual
) as `incident_level`
)
, t as (
select
`tenant_pk`,
`incident_accept_violation_count`,
`incident_level`,
rank() over( partition by `incident_level` order by `incident_accept_violation_count` DESC) as `incident_accept_violation_count_rank`,
rank() over( partition by `incident_level` order by `incident_accept_violation_count` ASC) as `__inverse_rank__`
from (
select
--alldim.*,
alldim.`tenant_pk`,--去掉.*,改用需要几列查询几列的方式
alldim.`incident_level`,
CAST(round(nvl(`incident_accept_violation_count`, 0), 0) as INT) as `incident_accept_violation_count`
from alldim
left join (
select
`incident_level`,
`tenant_pk`,
count(*) as `incident_accept_violation_count`
from `ecmp`.dwd_incident_accept
where
incident_accept_violation_flag = '违规'
and `incident_accept_time` >= '2020-01-12 00:00:00'
AND `incident_accept_time` <= '2020-01-12 23:59:59'
group by
`incident_level`,
`tenant_pk`
) as `t1` on 1 = 1
and alldim.`tenant_pk` = `t1`.`tenant_pk`
and alldim.`incident_level` = `t1`.`incident_level`
) as `t0`) select
alldim.`__bts__` as `__bts__`,
CAST(SYSDATE as STRING) as `__cts__`,
CAST(round(nvl(`incident_accept_violation_count`, 0), 0) as INT) as `incident_accept_violation_count`,--dround改成round,手写错误
CAST(round(`incident_accept_violation_count_rank`, 0) as INT) as `incident_accept_violation_count_rank`,
CAST(round(`incident_accept_violation_count_win_rate`, 1) as DOUBLE) as `incident_accept_violation_count_win_rate`,
alldim.`incident_level` as `incident_level`,
alldim.`tenant_id` as `tenant_id`,
alldim.`tenant_name` as `tenant_name`,
alldim.`tenant_pk` as `tenant_pk`
from alldim
left join (
select
'2020-01-12 23:59:59' as `__bts__`,
`incident_accept_violation_count`,
`incident_accept_violation_count_rank`,
`incident_accept_violation_count_win_rate`,
CAST(coalesce(`tp1`.`incident_level`) as STRING) as `incident_level`,
CAST(coalesce(`tp1`.`tenant_pk`) as STRING) as `tenant_pk`
from (
select
t.`tenant_pk`,
t.`incident_level`,
t.`incident_accept_violation_count`,
t.`incident_accept_violation_count_rank`,
if(`c`.ct = 1,null,(`__inverse_rank__` -1) /(`c`.ct -1) * 100) as `incident_accept_violation_count_win_rate`
from t
left join (
select
`incident_level`,
count(*) as `ct`
from t
group by
`incident_level`
) as `c` on 1 = 1
and t.`incident_level` = `c`.`incident_level`
) as `tp1`
)
as `__dws__` on alldim.`__bts__` = `__dws__`.`__bts__`
and alldim.`tenant_pk` = `__dws__`.`tenant_pk`
and alldim.`incident_level` = `__dws__`.`incident_level`;

Inceptor [Code: 40000, SQL State: 42000] COMPILE FAILED: Internal error NullPointerException: [Error 40000] java.lang.NullPointerException的更多相关文章

  1. Mybatis批量删除之Error code 1064, SQL state 42000;

    (一)小小的一次记载. (二):最近的项目都是使用MyBatis,批量新增自己都会写了,但是一次批量删除可把我给折腾了下,写法网上都有,但是照着做就是不行,最后问公司的人,问网友才得到答案,那就是jd ...

  2. MySQL 报错:Translating SQLException with SQL state '42000', error code '1064', message

    MySQL报错详细日志 2019-09-12 16:42:29 [http-nio-80-exec-25] DEBUG [org.springframework.jdbc.support.SQLErr ...

  3. sql与SQL CODE和SQL State相关报错

    操作数据库过程中,遇到许多问题,很多都与SQL CODE和SQL State相关,现在把一个完整的SQLCODE和SQLState错误信息和相关解释作以下说明,一来可以自己参考,对DB2错误自行找出原 ...

  4. hive分区导致FAILED: Hive Internal Error: java.lang.NullPointerException(null)

    写了一条hive sql ,其中条件中存在 dt>=20150101 and dt<=20150228 这样的条件,原来执行没问题,今天就抛出 FAILED: Hive Internal ...

  5. failed with: java.lang.NullPointerException

    failed with: java.lang.NullPointerException 需要在nutch的配置文件 'conf/nutch-site.xml'. 里设置如下,不然就报上面的错误了. 当 ...

  6. 解决spring mvc 上传报错,Field [] isn't an enum value,Failed to convert value of type 'java.lang.String[]' to required type '

    没有选择附件,但是点击上传按钮的时候会报错. 之前不选择文件,直接上传空文件是可以的,后来不知道改了什么就不行了. 错误信息: -- :: [http--] TRACE org.springframe ...

  7. 异常:failed for object com.sdu.crm.pojo.Customer@136a986 [java.lang.NullPointerException]

    异常: failed for object com.sdu.crm.pojo.Customer@136a986 [java.lang.NullPointerException] 大家好,如果大家看到了 ...

  8. Servlet.service() for servlet [jsp] in context with path [/Healthy_manager] threw exception [Unable to compile class for JSP] with root cause java.lang.IllegalArgumentException: Page directive: inval

    严重: Servlet.service() for servlet [jsp] in context with path [/Healthy_manager] threw exception [Una ...

  9. Failed to convert value of type 'java.lang.String' to required type 'java.time.LocalDate';

    springboot jdbc查询使用LocalDate报:Failed to convert value of type 'java.lang.String' to required type 'j ...

随机推荐

  1. C语言的类型大小

    C语言的类型大小 设计程序的时候我们一般会考虑的尽量的周全,尤其是像C这样的静态类型语言. 有一些溢出的问题就源于没有搞清楚变量的大小范围,所以我们编写的时候需要特别注意 C的整形(整数类型)大小 C ...

  2. Linux下的screen和作业任务管理

    一.screen 首先介绍下screen,screen是Linux下的一个任务容器,开启了之后就可以让任务在后台执行而不会被网络中断或者是终端退出而影响到. 在Linux中有一些耗时比较久的操作(例如 ...

  3. VsCode通过SSH连接远程服务器开发

    前言 nil 正文 安装插件 安装VsCode官方插件 Remote - SSH Remote - SSH: Editing Configuration Files WSL(远程桌面连接需要Remot ...

  4. Java基础进阶类名整理

    类名综合 数学类: Math:数学运算 BigDecimal:用于精确计算数据 数组工具类: Arrays:数组工具类,用于对数组的操作 时间操作: JDK8以前: Date:表示一个时间,并面向对象 ...

  5. Centos 6.5 Rabbitmq 安装和集群,镜像部署

    centos 6.5 rabbitmq 安装和集群,镜像部署 安装erlang: yum install gcc glibc-devel make ncurses-devel openssl-deve ...

  6. Desired_Capabilities配置

    appium服务器初始化参数 最全: https://github.com/appium/appium/blob/master/docs/cn/writing-running-appium/caps. ...

  7. 如何构建一个多人(.io) Web 游戏,第 1 部分

    原文:How to Build a Multiplayer (.io) Web Game, Part 1 GitHub: https://github.com/vzhou842/example-.io ...

  8. SpringBoot Logback无法获取配置中心属性

    SpringBoot Logback无法获取配置中心属性 前言 最近在做项目中,需要把项目中的日志信息通过RabbitMQ将规定格式的消息发送到消息队列中,然后ELK系统通过消息队列拿日志并且保存起来 ...

  9. 多视图子空间聚类/表示学习(Multi-view Subspace Clustering/Representation Learning)

    多视图子空间聚类/表示学习(Multi-view Subspace Clustering/Representation Learning) 作者:凯鲁嘎吉 - 博客园 http://www.cnblo ...

  10. 2V转3V的电源芯片电路图,2.4V转3V电路

    两节镍氢电池1.2V+1.2V是2.4V的标称电压,2.4V可以转3V输出电路应用. 在2.4V转3V和2V转3V的应用中,输出电流可最大600MA. 2V的低压输入,可以采用PW5100低压输入专用 ...