0
点赞
收藏
分享

微信扫一扫

定义HarmonyOS IDL接口

自由情感小屋 2024-02-03 阅读 29
数据分析clickhousesql

文章标题

一、WITH语法-定义变量

SELECT pow(2, 2);

┌─pow(2, 2)─┐
│         4 │
└───────────┘

SELECT pow(pow(2, 2), 2);

┌─pow(pow(2, 2), 2)─┐
│                16 │
└───────────────────┘

WITH
   pow(2, 2) AS a
SELECT pow(a, 2);

┌─pow(a, 2)─┐
│        16 │
└───────────┘

1.1 定义变量

# tb_mysql- 创建数据
DROP TABLE IF EXISTS tb_mysql;
CREATE TABLE tb_mysql (
        id UInt8,
        name String,
        age UInt8
)ENGINE = MergeTree()
ORDER BY id;
INSERT INTO tb_mysql VALUES ('1', 'tom', 23);
INSERT INTO tb_mysql VALUES ('2', 'lisa', 33);
INSERT INTO tb_mysql VALUES ('3', 'henry', 44);
INSERT INTO tb_mysql VALUES ('1', 'linda', 23);
INSERT INTO tb_mysql VALUES ('2', 'ross', 33);
INSERT INTO tb_mysql VALUES ('1', 'julie', 23);
INSERT INTO tb_mysql VALUES ('2', 'niki', 33);

# 数据分析
WITH 
    1  AS constant
SELECT
    id + constant,
    name
FROM
    tb_mysql;
    
┌─plus(id, constant)─┬─name─┐
│                  3 │ niki │
└────────────────────┴──────┘
┌─plus(id, constant)─┬─name──┐
│                  2 │ tom   │
│                  2 │ linda │
│                  2 │ julie │
│                  3 │ lisa  │
│                  3 │ ross  │
│                  4 │ henry │
└────────────────────┴───────┘

1.2 调用函数

# tb_partition-创造数据
DROP TABLE IF EXISTS tb_partition;
CREATE TABLE tb_partition (
      id UInt8,
      name String,
      birthday String
)ENGINE = MergeTree()
ORDER BY id;
INSERT INTO tb_partition VALUES ('1', 'x1', '2024-05-20 10:50:46');
INSERT INTO tb_partition VALUES ('2', 'xy', '2024-05-20 11:17:47');
INSERT INTO tb_partition VALUES ('3', 'xf', '2024-05-19 11:11:12');


# 数据分析
WITH 
    toDate(birthday) AS bday
SELECT
    id,
    name,
    bday
FROM
    tb_partition;

┌─id─┬─name─┬───────bday─┐
│  1 │ x1   │ 2024-05-20 │
└────┴──────┴────────────┘
┌─id─┬─name─┬───────bday─┐
│  2 │ xy   │ 2024-05-20 │
└────┴──────┴────────────┘
┌─id─┬─name─┬───────bday─┐
│  3 │ xf   │ 2024-05-19 │
└────┴──────┴────────────┘

1.3 子查询

WITH 
    (SELECT 
         *
     FROM 
         tb_partition
     WHERE tb_partition.id = '1') AS sub
SELECT
    * ,
    sub
FROM 
    tb_partition;

┌─id─┬─name─┬─birthday────────────┬─sub────────────────────────────┐
│  3 │ xf   │ 2024-05-19 11:11:12 │ (1,'x1','2024-05-20 10:50:46') │
└────┴──────┴─────────────────────┴────────────────────────────────┘
┌─id─┬─name─┬─birthday────────────┬─sub────────────────────────────┐
│  2 │ xy   │ 2024-05-20 11:17:47 │ (1,'x1','2024-05-20 10:50:46') │
└────┴──────┴─────────────────────┴────────────────────────────────┘
┌─id─┬─name─┬─birthday────────────┬─sub────────────────────────────┐
│  1 │ x1   │ 2024-05-20 10:50:46 │ (1,'x1','2024-05-20 10:50:46') │
└────┴──────┴─────────────────────┴────────────────────────────────┘

二、GROUP BY子句（结合WITH ROLLUP、CUBE、TOTALS）

# tb_with-创建数据
DROP TABLE IF EXISTS tb_with;
CREATE TABLE tb_with (
    id UInt8,
    vist UInt8,
    province String,
    city String,
    area String
)ENGINE = MergeTree()
ORDER BY id;
INSERT INTO tb_with VALUES (1, 12 ,'湖北', '黄冈', '武穴');
INSERT INTO tb_with VALUES (2, 12 ,'湖北', '黄冈', '黄州');
INSERT INTO tb_with VALUES (3, 12 ,'湖北', '黄冈', '麻城');
INSERT INTO tb_with VALUES (4, 32 ,'湖北', '黄冈', '黄梅');
INSERT INTO tb_with VALUES (5, 12 ,'湖北', '黄石', '下陆');
INSERT INTO tb_with VALUES (6, 54 ,'湖北', '黄石', '铁山');
INSERT INTO tb_with VALUES (7, 12 ,'湖北', '黄石', '石灰窑');
INSERT INTO tb_with VALUES (8, 89 ,'湖北', '荆州', '荆门');
INSERT INTO tb_with VALUES (9, 99 ,'湖北', '荆州', '钟祥');


# 数据分析
SELECT 
    province,
    city,
    area,
    sum(vist) AS total_visit
FROM tb_with
GROUP BY province, city, area WITH ROLLUP ;

┌─province─┬─city─┬─area───┬─total_visit─┐
│ 湖北     │ 黄冈 │ 黄梅   │          32 │
│ 湖北     │ 荆州 │ 钟祥   │          99 │
│ 湖北     │ 黄冈 │ 麻城   │          12 │
│ 湖北     │ 荆州 │ 荆门   │          89 │
│ 湖北     │ 黄冈 │ 黄州   │          12 │
│ 湖北     │ 黄石 │ 下陆   │          12 │
│ 湖北     │ 黄石 │ 石灰窑 │          12 │
│ 湖北     │ 黄石 │ 铁山   │          54 │
│ 湖北     │ 黄冈 │ 武穴   │          12 │
└──────────┴──────┴────────┴─────────────┘
┌─province─┬─city─┬─area─┬─total_visit─┐
│ 湖北     │ 黄石 │      │          78 │
│ 湖北     │ 荆州 │      │         188 │
│ 湖北     │ 黄冈 │      │          68 │
└──────────┴──────┴──────┴─────────────┘
┌─province─┬─city─┬─area─┬─total_visit─┐
│ 湖北     │      │      │         334 │
└──────────┴──────┴──────┴─────────────┘
┌─province─┬─city─┬─area─┬─total_visit─┐
│          │      │      │         334 │
└──────────┴──────┴──────┴─────────────┘

三、FORM语法

3.1表函数

请添加图片描述

3.1.1 file

请添加图片描述

-- 数据文件必须在指定的目录下 /var/lib/clickhouse/user_files

SELECT 
	*
FROM file('demo.csv', 'CSV', 'id Int8,name String, age UInt8')
-- 文件夹下任意的文件
SELECT
	*
FROM file('*', 'CSV', 'id Int8, name String, age UInt8')

3.1.2 numbers

-- numbers(N) – 返回一个包含单个 ‘number’ 列(UInt64)的表，其中包含从0到N-1的整数。
-- numbers(N, M) - 返回一个包含单个 ‘number’ 列(UInt64)的表，其中包含从N到(N+M-1)的整数。
SELECT 
	*
FROM 
	numbers(10);

SELECT 
	*
FROM 
	numbers(2, 10);

SELECT 
	*
FROM 
	numbers(2, 10) 
LIMIT 3;

SELECT 
	toDate('2020-01-01') + number AS d
FROM 
numbers(365);

3.1.3 mysql

-- CH可以直接从mysql服务中查询数据
mysql('host:port', 'database', 'table', 'user', 'password'[, replace_query, 'on_duplicate_clause']);

SELECT
	*
FROM 
	mysql('linux01:3306', 'db_doit_ch', 'emp', 'root', 'root');

请添加图片描述

3.1.4 hdfs

SELECT 
	*
FROM
	hdfs('hdfs://hdfs1:9000/test', 'TSV', 'column1 UInt32, column2 UInt32') 
LIMIT 2;


SELECT
	*
FROM
	hdfs('hdfs://linux01:8020/demo.csv', 'CSV', 'id Int8, name String, age Int8')

请添加图片描述

四、ARRAY JOIN语法（区别于arrayJoin(arr)函数）

# ARRAY JOIN-创建数据
DROP TABLE IF EXISTS tb_array_join;
CREATE TABLE tb_array_join (
    id Int8,
    hobby Array(String)
)ENGINE = Log;

INSERT INTO tb_array_join VALUES (1, ['eat', 'drink', 'sleep']), (2, ['study', 'sport', 'read']), (2, ['study', 'sport']);

# 查询数据
SELECT * FROM tb_array_join;
┌─id─┬─hobby────────────────────┐
│  1 │ ['eat','drink','sleep']  │
│  2 │ ['study','sport','read'] │
│  2 │ ['study','sport']        │
└────┴──────────────────────────┘

# 分析1
SELECT
	id,
	hobby,
	hobby_expand
FROM
	tb_array_join
ARRAY JOIN
hobby AS hobby_expand;
┌─id─┬─hobby────────────────────┬─hobby_expand─┐
│  1 │ ['eat','drink','sleep']  │ eat          │
│  1 │ ['eat','drink','sleep']  │ drink        │
│  1 │ ['eat','drink','sleep']  │ sleep        │
│  2 │ ['study','sport','read'] │ study        │
│  2 │ ['study','sport','read'] │ sport        │
│  2 │ ['study','sport','read'] │ read         │
│  2 │ ['study','sport']        │ study        │
│  2 │ ['study','sport']        │ sport        │
└────┴──────────────────────────┴──────────────┘

# 分析2
SELECT
	id,
	hobby,
	arrayEnumerate(hobby) AS indexs
FROM
	tb_array_join;
┌─id─┬─hobby────────────────────┬─indexs──┐
│  1 │ ['eat','drink','sleep']  │ [1,2,3] │
│  2 │ ['study','sport','read'] │ [1,2,3] │
│  2 │ ['study','sport']        │ [1,2]   │
└────┴──────────────────────────┴─────────┘

# 分析3（将hobby展开，并与索引组成一列）
SELECT
	id,
	hobby_expand,
	index
FROM
	tb_array_join
ARRAY JOIN
hobby AS hobby_expand,
arrayEnumerate(hobby) AS index;

┌─id─┬─hobby_expand─┬─index─┐
│  1 │ eat          │     1 │
│  1 │ drink        │     2 │
│  1 │ sleep        │     3 │
│  2 │ study        │     1 │
│  2 │ sport        │     2 │
│  2 │ read         │     3 │
│  2 │ study        │     1 │
│  2 │ sport        │     2 │
└────┴──────────────┴───────┘

五、连续销售案例

# tb_shop-创建数据
# 思路：
# 1、按照name和cdate排序；
# 2、对name和cdate相同的数据只保留一个，按此业务所以保留每天最大的营销额
# 基于上述需求，所以用ReplacingMergeTree引擎实现
# 3、然后基于下方的分析步骤进行逐步分析
DROP TABLE IF EXISTS tb_shop;
CREATE TABLE tb_shop (
        name String,
        cdate Date,
        cost Float64
)ENGINE = ReplacingMergeTree()
ORDER BY (name, cdate);
INSERT INTO tb_shop VALUES ('a', '2024-02-05', 200),
('a', '2024-02-04', 320),
('a', '2024-02-03', 260),
('a', '2024-01-29', 300),
('a', '2024-01-29', 230),
('a', '2024-01-28', 880),
('a', '2024-01-27', 900),
('a', '2024-01-26', 350),
('a', '2024-01-26', 500),
('a', '2024-01-26', 900),

('b', '2024-02-05', 200),
('b', '2024-02-04', 320),
('b', '2024-02-03', 260),
('b', '2024-02-02', 670),

('c', '2024-02-05', 200),
('c', '2024-02-05', 900),
('c', '2024-02-05', 800),
('c', '2024-02-05', 200);

# 按照以下sql一步步排查
# 分析1
SELECT 
	name,
	groupArray(cdate) AS arr
FROM
	tb_shop
GROUP BY name;
┌─name─┬─arr──────────────────────────────────────────────────────────────────────────────────────────┐
│ b    │ ['2024-02-02','2024-02-03','2024-02-04','2024-02-05']                                        │
│ c    │ ['2024-02-05']                                                                               │
│ a    │ ['2024-01-26','2024-01-27','2024-01-28','2024-01-29','2024-02-03','2024-02-04','2024-02-05'] │
└──────┴──────────────────────────────────────────────────────────────────────────────────────────────┘

# 分析2
SELECT
    name,
    ct,
    arr_index
FROM
    (
        SELECT
            name,
            groupArray(cdate) AS arr,
            arrayEnumerate(arr) AS idx
        FROM
            tb_shop
        GROUP BY name
    )
ARRAY JOIN
arr AS ct,
idx AS arr_index;
┌─name─┬─────────ct─┬─arr_index─┐
│ b    │ 2024-02-02 │         1 │
│ b    │ 2024-02-03 │         2 │
│ b    │ 2024-02-04 │         3 │
│ b    │ 2024-02-05 │         4 │
│ c    │ 2024-02-05 │         1 │
│ a    │ 2024-01-26 │         1 │
│ a    │ 2024-01-27 │         2 │
│ a    │ 2024-01-28 │         3 │
│ a    │ 2024-01-29 │         4 │
│ a    │ 2024-02-03 │         5 │
│ a    │ 2024-02-04 │         6 │
│ a    │ 2024-02-05 │         7 │
└──────┴────────────┴───────────┘

# 分析3
SELECT
    name,
    (ct - arr_index) AS diff
FROM
    (
        SELECT
            name,
            groupArray(cdate) AS arr,
            arrayEnumerate(arr) AS idx
        FROM
            tb_shop
        GROUP BY name
    )
ARRAY JOIN
arr AS ct,
idx AS arr_index;
┌─name─┬───────diff─┐
│ b    │ 2024-02-01 │
│ b    │ 2024-02-01 │
│ b    │ 2024-02-01 │
│ b    │ 2024-02-01 │
│ c    │ 2024-02-04 │
│ a    │ 2024-01-25 │
│ a    │ 2024-01-25 │
│ a    │ 2024-01-25 │
│ a    │ 2024-01-25 │
│ a    │ 2024-01-29 │
│ a    │ 2024-01-29 │
│ a    │ 2024-01-29 │
└──────┴────────────┘

# 分析4
SELECT
    name,
    (ct - arr_index) AS diff,
    count(1)
FROM
    (
        SELECT
            name,
            groupArray(cdate) AS arr,
            arrayEnumerate(arr) AS idx
        FROM
            tb_shop
        GROUP BY name
    )
ARRAY JOIN
arr AS ct,
idx AS arr_index
GROUP BY name, diff;
┌─name─┬───────diff─┬─count()─┐
│ c    │ 2024-02-04 │       1 │
│ a    │ 2024-01-25 │       4 │
│ b    │ 2024-02-01 │       4 │
│ a    │ 2024-01-29 │       3 │
└──────┴────────────┴─────────┘
# 分析5
SELECT
    name,
    (ct - arr_index) AS diff,
    count(1) AS cc
FROM
    (
        SELECT
            name,
            groupArray(cdate) AS arr,
            arrayEnumerate(arr) AS idx
        FROM
            tb_shop
        GROUP BY name
    )
ARRAY JOIN
arr AS ct,
idx AS arr_index
GROUP BY name, diff
ORDER BY name, cc DESC;
┌─name─┬───────diff─┬─cc─┐
│ a    │ 2024-01-25 │  4 │
│ a    │ 2024-01-29 │  3 │
│ b    │ 2024-02-01 │  4 │
│ c    │ 2024-02-04 │  1 │
└──────┴────────────┴────┘

# 分析6
SELECT
    name,
    (ct - arr_index) AS diff,
    count(1) AS cc
FROM
    (
        SELECT
            name,
            groupArray(cdate) AS arr,
            arrayEnumerate(arr) AS idx
        FROM
            tb_shop
        GROUP BY name
    )
ARRAY JOIN
arr AS ct,
idx AS arr_index
GROUP BY name, diff
ORDER BY name, cc DESC
LIMIT 1 BY name;
┌─name─┬───────diff─┬─cc─┐
│ a    │ 2024-01-25 │  4 │
│ b    │ 2024-02-01 │  4 │
│ c    │ 2024-02-04 │  1 │
└──────┴────────────┴────┘

六、连接函数

请添加图片描述

6.1 连接精度

# 准备数据
DROP TABLE IF EXISTS yg;
CREATE TABLE yg(
 id Int8,
 name String,
 age UInt8,
 bid Int8
)ENGINE = Log;
INSERT INTO yg VALUES(1, 'AA', 23, 1),
(2, 'BB', 24, 3),
(3, 'VV', 27, 1),
(4, 'CC', 13, 3),
(5, 'KK', 53, 3),
(6, 'MM', 33, 3);

DROP TABLE IF EXISTS bm;
CREATE TABLE bm(
  bid Int8,
  name String
)ENGINE = Log;
INSERT INTO bm VALUES(1, 'x'),(2, 'Y'),(3, 'z');

DROP TABLE IF EXISTS gz;
CREATE TABLE gz(
  id Int8,
  jb Int64,
  jj Int64
)ENGINE = Log;
INSERT INTO gz VALUES(1, 1000, 2000),
(1, 1000, 2000),(2, 2000, 1233),(3, 2000, 3000),(4, 4000, 1000);

6.1.1 ALL

# 分析1
SELECT 
	*
FROM
	yg
ALL INNER JOIN gz ON yg.id = gz.id;
┌─id─┬─name─┬─age─┬─bid─┬─gz.id─┬───jb─┬───jj─┐
│  1 │ AA   │  23 │   1 │     1 │ 1000 │ 2000 │
│  1 │ AA   │  23 │   1 │     1 │ 1000 │ 2000 │
│  2 │ BB   │  24 │   3 │     2 │ 2000 │ 1233 │
│  3 │ VV   │  27 │   1 │     3 │ 2000 │ 3000 │
│  4 │ CC   │  13 │   3 │     4 │ 4000 │ 1000 │
└────┴──────┴─────┴─────┴───────┴──────┴──────┘

6.1.2 ANY

# 分析2
SELECT 
	*
FROM
	yg
ANY INNER JOIN gz ON yg.id = gz.id;
┌─id─┬─name─┬─age─┬─bid─┬─gz.id─┬───jb─┬───jj─┐
│  1 │ AA   │  23 │   1 │     1 │ 1000 │ 2000 │
│  2 │ BB   │  24 │   3 │     2 │ 2000 │ 1233 │
│  3 │ VV   │  27 │   1 │     3 │ 2000 │ 3000 │
│  4 │ CC   │  13 │   3 │     4 │ 4000 │ 1000 │
└────┴──────┴─────┴─────┴───────┴──────┴──────┘

6.1.3 ASOF

DROP TABLE IF EXISTS emp1;
CREATE TABLE emp1(
   id Int8,
   name String,
   ctime DateTime
)ENGINE = Log;
INSERT INTO emp1 VALUES(1, 'AA', '2021-01-03 00:00:00'),(1, 'AA', '2021-01-02 00:00:00'),(2, 'CC', '2021-01-01 00:00:00'),(3, 'DD', '2021-01-01 00:00:00'),(4, 'EE', '2021-01-01 00:00:00');

DROP TABLE IF EXISTS emp2;
CREATE TABLE emp2(
  id Int8,
  name String,
  ctime DateTime
)ENGINE = Log;
INSERT INTO emp2 VALUES (1, 'aa', '2021-01-02 00:00:00'),(1, 'aa', '2021-01-02 00:00:00'),(2, 'cc', '2021-01-02 00:00:00'),(3, 'dd', '2021-01-02 00:00:00');

# 分析
SELECT 
*
FROM emp1
ASOF INNER JOIN emp2
ON (emp1.id = emp2.id) AND (emp1.ctime > emp2.ctime);
┌─id─┬─name─┬───────────────ctime─┬─emp2.id─┬─emp2.name─┬──────────emp2.ctime─┐
│  1 │ AA   │ 2021-01-03 00:00:00 │       1 │ aa        │ 2021-01-02 00:00:00 │
└────┴──────┴─────────────────────┴─────────┴───────────┴─────────────────────┘

七、系统函数介绍

八、JSON解析案例

-- 建表
DROP TABLE IF EXISTS tb_ods_log;
CREATE TABLE tb_ods_log (
    line String
) ENGINE = Log;

INSERT INTO tb_ods_log VALUES('{"account":"14d9TM","deviceId":"Kcjksekjg","ip":"180.12.12.3","sessionId":"sfjkeIGj","eventId":"","properties":{"adId":"6","adCampain":"7"},"timeStamp":18992891918}'),
                           ('{"account":"14d9TM","deviceId":"Kcfafafkjg","ip":"180.12.12.3","sessionId":"sfjkeIGj","eventId":"","properties":{"adId":"6","adCampain":"7"},"timeStamp":189923891918}'),
                           ('{"account":"14faTM","deviceId":"Kcfaekjg","ip":"180.12.12.3","sessionId":"sfjkeIGj","eventId":"","properties":{"adId":"6","adCampain":"5"},"timeStamp":189924891918}');

# json解析
WITH 
  visitParamExtractString(line, 'account') as account,
  visitParamExtractString(line, 'deviceId') as deviceId,
  visitParamExtractString(line, 'sessionId') as sessionId,
  visitParamExtractRaw(line, 'properties') as properties,
  visitParamExtractInt(line, 'timeStamp') as timeStamp
SELECT
	account,
	deviceId,
	sessionId,
	properties,
	timeStamp
FROM
    tb_ods_log 
LIMIT 10;

┌─account─┬─deviceId───┬─sessionId─┬─properties───────────────────┬────timeStamp─┐
│ 14d9TM  │ Kcjksekjg  │ sfjkeIGj  │ {"adId":"6","adCampain":"7"} │  18992891918 │
│ 14d9TM  │ Kcfafafkjg │ sfjkeIGj  │ {"adId":"6","adCampain":"7"} │ 189923891918 │
│ 14faTM  │ Kcfaekjg   │ sfjkeIGj  │ {"adId":"6","adCampain":"5"} │ 189924891918 │
└─────────┴────────────┴───────────┴──────────────────────────────┴──────────────┘

九、实用函数

SELECT
    1 AS b,
    toTypeName(b)
    
┌─b─┬─toTypeName(1)─┐
│ 1 │ UInt8         │
└───┴───────────────┘

十、语法注意事项

  • cklickhouse大小写敏感
  • 实现需求的时候可以先查找ck是否有函数可以支持，如果不支持再去想其他方式

课件学习地址

举报

相关推荐

thrift idl 定义语法小认识

字段数据类型ThriftPython后端开发

web idl 接口定义语言数据类型与 C++绑定关系

c++web语言interfaceconstructorLinux系统/运维

webkit idl 资料

webkitjs引擎perl扩展语言虚拟化云计算

IDL是什么？

数据类型浮点数c++Linux系统/运维

微服务间请求响应定义方式对比 （Apache Dubbo, IDL 定义，RESTful API）

webpackrust前端

使用 CORBA 和 Java IDL

java服务器语言编译器objectJavaScript前端开发

console接口(console接口定义)

网络网线串口ip地址编程语言

接口定义相关

System请求参数表单Java后端开发

typescript接口定义

typescript赋值只读属性TypeScript前端开发

webkit idl parser 模块分析

webkitpreprocessorgenerator正则表达式makefile编程语言
0 条评论