ODPS SQL <for 数据操作语言DML>
基本操作:
查询:
SELECT [ALL | DISTINCT] select_expr, select_expr, ...
FROM table_reference
[WHERE where_condition]
[GROUP BY col_list]
[ORDER BY order_condition]
[DISTRIBUTE BY distribute_condition [SORT BY sort_condition] ]
[LIMIT number]
更新:
INSERT OVERWRITE|INTO TABLE tablename [PARTITION (partcol1=val1, partcol2=val2 ...)] [(col1,col2 ...)]
select_statement
FROM from_statement;
表关联:
join_table:
table_reference join table_factor [join_condition]
| table_reference {left outer|right outer|full outer|inner} join table_reference join_condition
table_reference:
table_factor
| join_table
table_factor:
tbl_name [alias]
| table_subquery alias
| ( table_references )
join_condition:
on equality_expression ( and equality_expression )
演示内容:
使用DML:
*查询已有数据
*使用表连接查询数据
*覆盖更新
*追加更新
//上传演示数据
odps@ sdrtest>tunnel upload /root/.odpscmd/t_people.txt t_people;
//查看演示数据
odps@ sdrtest>select * from t_people;
+------------+------+
| id | name |
+------------+------+
| 1 | Michael Jordan |
| 2 | Angela Dorthea Merkel |
| 3 | Bruce Willis |
| 4 | Kim Kardashian |
| 5 | Jhon Knight |
| 6 | Maria Sharapova |
| 7 | Chiang Kai-shek |
| 8 | Jennifer Aniston |
| 9 | David Beckham |
| 10 | Dragon Lady |
+------------+------+
odps@ sdrtest>select count(*),id from t_people group by id having count(*) > 0; ID = 20190414051444876gfk3c692
Job Queueing.
----------------------------------------------------------------------------------------------
STAGES STATUS TOTAL COMPLETED RUNNING PENDING BACKUP
M1_job_0 ................. TERMINATED 1 1 0 0 0
----------------------------------------------------------------------------------------------
STAGES STATUS TOTAL COMPLETED RUNNING PENDING BACKUP----------
M1_job_0 ................. TERMINATED 1 1 0 0 0----------------------------------------------------------------------------------
R2_1_job_0 ............... TERMINATED 1 1 0 0 0
----------------------------------------------------------------------------------------------
STAGES: 02/02 [==========================>>] 100% ELAPSED TIME: 8.17 s
----------------------------------------------------------------------------------------------
Summary:
resource cost: cpu 0.00 Core * Min, memory 0.00 GB * Min
inputs:
sdrtest.t_people: 10 (776 bytes)
outputs:
Job run time: 5.000
Job run mode: fuxi job
Job run engine: execution engine
M1:
instance count: 1
run time: 3.000
instance time:
min: 0.000, max: 0.000, avg: 0.000
input records:
TableScan1: 10 (min: 10, max: 10, avg: 10)
output records:
StreamLineWrite1: 10 (min: 10, max: 10, avg: 10)
writer dumps:
StreamLineWrite1: (min: 0, max: 0, avg: 0)
R2_1:
instance count: 1
run time: 5.000
instance time:
min: 0.000, max: 0.000, avg: 0.000
input records:
StreamLineRead1: 10 (min: 10, max: 10, avg: 10)
output records:
AdhocSink1: 10 (min: 10, max: 10, avg: 10)
reader dumps:
StreamLineRead1: (min: 0, max: 0, avg: 0) +------------+------------+
| _c0 | id |
+------------+------------+
| 1 | 1 |
| 1 | 2 |
| 1 | 3 |
| 1 | 4 |
| 1 | 5 |
| 1 | 6 |
| 1 | 7 |
| 1 | 8 |
| 1 | 9 |
| 1 | 10 |
+------------+------------+
//使用表连接查询数据
odps@ sdrtest>select t1.* from t_people t1 join t_people t2 on t1.id=t2.id; ID = 20190414052009767gcpxmnim Job Queueing.
----------------------------------------------------------------------------------------------
STAGES STATUS TOTAL COMPLETED RUNNING PENDING BACKUP
M1_job_0 ................. TERMINATED 1 1 0 0 0
----------------------------------------------------------------------------------------------
STAGES STATUS TOTAL COMPLETED RUNNING PENDING BACKUP----------
M1_job_0 ................. TERMINATED 1 1 0 0 0----------------------------------------------------------------------------------
J2_1_job_0 ............... TERMINATED 1 1 0 0 0
----------------------------------------------------------------------------------------------
STAGES: 02/02 [==========================>>] 100% ELAPSED TIME: 8.56 s
----------------------------------------------------------------------------------------------
Summary:
resource cost: cpu 0.00 Core * Min, memory 0.00 GB * Min
inputs:
sdrtest.t_people: 10 (776 bytes)
outputs:
Job run time: 6.000
Job run mode: fuxi job
Job run engine: execution engine
M1:
instance count: 1
run time: 3.000
instance time:
min: 0.000, max: 0.000, avg: 0.000
input records:
TableScan1: 10 (min: 10, max: 10, avg: 10)
output records:
StreamLineWrite1: 10 (min: 10, max: 10, avg: 10)
StreamLineWrite2: 10 (min: 10, max: 10, avg: 10)
writer dumps:
StreamLineWrite1: (min: 0, max: 0, avg: 0)
StreamLineWrite2: (min: 0, max: 0, avg: 0)
J2_1:
instance count: 1
run time: 6.000
instance time:
min: 0.000, max: 0.000, avg: 0.000
input records:
StreamLineRead1: 10 (min: 10, max: 10, avg: 10)
StreamLineRead2: 10 (min: 10, max: 10, avg: 10)
output records:
AdhocSink1: 10 (min: 10, max: 10, avg: 10)
reader dumps:
StreamLineRead1: (min: 0, max: 0, avg: 0)
StreamLineRead2: (min: 0, max: 0, avg: 0) +------------+------+
| id | name |
+------------+------+
| 1 | Michael Jordan |
| 2 | Angela Dorthea Merkel |
| 3 | Bruce Willis |
| 4 | Kim Kardashian |
| 5 | Jhon Knight |
| 6 | Maria Sharapova |
| 7 | Chiang Kai-shek |
| 8 | Jennifer Aniston |
| 9 | David Beckham |
| 10 | Dragon Lady |
+------------+------+
//创建新表用于演示追加更新:
odps@ sdrtest>create table t_people_new like t_people;
odps@ sdrtest>read t_people_new;
+------------+------------+
| id | name |
+------------+------------+
+------------+------------+
odps@ sdrtest>insert into table t_people_new select * from t_people;
//查看新建的表内写入的数据:
odps@ sdrtest>read t_people_new
>;
+------------+------------+
| id | name |
+------------+------------+
| 1 | Michael Jordan |
| 2 | Angela Dorthea Merkel |
| 3 | Bruce Willis |
| 4 | Kim Kardashian |
| 5 | Jhon Knight |
| 6 | Maria Sharapova |
| 7 | Chiang Kai-shek |
| 8 | Jennifer Aniston |
| 9 | David Beckham |
| 10 | Dragon Lady |
+------------+------------+
//演示追加更新 <into>
odps@ sdrtest>insert into table t_people_new select * from t_people;
Job Queueing.
----------------------------------------------------------------------------------------------
STAGES STATUS TOTAL COMPLETED RUNNING PENDING BACKUP
M1_job_0 ................. TERMINATED 1 1 0 0 0
----------------------------------------------------------------------------------------------
STAGES: 01/01 [==========================>>] 100% ELAPSED TIME: 5.26 s
----------------------------------------------------------------------------------------------
Summary:
resource cost: cpu 0.00 Core * Min, memory 0.00 GB * Min
inputs:
sdrtest.t_people: 10 (776 bytes)
outputs:
sdrtest.t_people_new: 10 (776 bytes)
Job run time: 3.000
Job run mode: fuxi job
Job run engine: execution engine
M1:
instance count: 1
run time: 3.000
instance time:
min: 0.000, max: 0.000, avg: 0.000
input records:
TableScan1: 10 (min: 10, max: 10, avg: 10)
output records:
TableSink1: 10 (min: 10, max: 10, avg: 10) OK
odps@ sdrtest>read t_people_new;
+------------+------------+
| id | name |
+------------+------------+
| 1 | Michael Jordan |
| 2 | Angela Dorthea Merkel |
| 3 | Bruce Willis |
| 4 | Kim Kardashian |
| 5 | Jhon Knight |
| 6 | Maria Sharapova |
| 7 | Chiang Kai-shek |
| 8 | Jennifer Aniston |
| 9 | David Beckham |
| 10 | Dragon Lady |
| 1 | Michael Jordan |
| 2 | Angela Dorthea Merkel |
| 3 | Bruce Willis |
| 4 | Kim Kardashian |
| 5 | Jhon Knight |
| 6 | Maria Sharapova |
| 7 | Chiang Kai-shek |
| 8 | Jennifer Aniston |
| 9 | David Beckham |
| 10 | Dragon Lady |
+------------+------------+ //演示覆盖更新 <overwrite>
odps@ sdrtest>insert overwrite table t_people_new select * from t_people; ----------------------------------------------------------------------------------------------
STAGES STATUS TOTAL COMPLETED RUNNING PENDING BACKUP
M1_job_0 ................. TERMINATED 1 1 0 0 0
----------------------------------------------------------------------------------------------
STAGES: 01/01 [==========================>>] 100% ELAPSED TIME: 5.24 s
----------------------------------------------------------------------------------------------
Summary:
resource cost: cpu 0.00 Core * Min, memory 0.00 GB * Min
inputs:
sdrtest.t_people: 10 (776 bytes)
outputs:
sdrtest.t_people_new: 10 (776 bytes)
Job run time: 3.000
Job run mode: fuxi job
Job run engine: execution engine
M1:
instance count: 1
run time: 3.000
instance time:
min: 0.000, max: 0.000, avg: 0.000
input records:
TableScan1: 10 (min: 10, max: 10, avg: 10)
output records:
TableSink1: 10 (min: 10, max: 10, avg: 10) OK
odps@ sdrtest>read t_people_new;
+------------+------------+
| id | name |
+------------+------------+
| 1 | Michael Jordan |
| 2 | Angela Dorthea Merkel |
| 3 | Bruce Willis |
| 4 | Kim Kardashian |
| 5 | Jhon Knight |
| 6 | Maria Sharapova |
| 7 | Chiang Kai-shek |
| 8 | Jennifer Aniston |
| 9 | David Beckham |
| 10 | Dragon Lady |
+------------+------------+
odps@ sdrtest>
others....
ODPS SQL <for 数据操作语言DML>的更多相关文章
- ODPS SQL <for 数据定义语言 DDL>
数据定义语言:(DDL) 建表语句: CREATE TABLE [IF NOT EXISTS] table_name [(col_name data_type [COMMENT col_comment ...
- MySQL之数据定义语言(DDL)
写在前面 本文中 [ 内容 ] 代表啊可选项,即可写可不写. SQL语言的基本功能介绍 SQL是一种结构化查询语言,主要有如下几个功能: 数据定义语言(DDL):全称Data Definition L ...
- <MySQL>入门三 数据定义语言 DDL
-- DDL 数据定义语言 /* 库和表的管理 一.库的管理:创建.修改.删除 二.表的管理:创建.修改.删除 创建:create 修改:alter 删除:drop */ 1.库的管理 -- 库的管理 ...
- 30441数据定义语言DDL
数据定义:指对数据库对象的定义.删除和修改操作. 数据库对象主要包括数据表.视图.索引等. 数据定义功能通过CREATE.ALTER.DROP语句来完成. 按照操作对象分类来介绍数据定义的SQL语法. ...
- SQL语句整理(二) 数据定义语言DDL
前言: 这是我学数据库时整理的学习资料,基本上包括了所以的SQL语句的知识点. 我的教材是人大王珊老师的<数据库系统概论>. 因为是手打的,所以会用一些细节打错了,但都挺明显也不多(考完试 ...
- oracle 数据定义语言(DDL)语法
DDL语言包括数据库对象的创建(create).删除(drop)和修改(alter)的操作 1.创建表语法 create table table_name( column_name datatype ...
- 【MySQL笔记】数据定义语言DDL
1.创建基本表 create table <表名> (<列名><数据类型>[列级完整性约束条件] ...
- SQLite基础-4.数据定义语言(DDL)
目录 一.创建数据库 1. 创建方式 2. 数据库命名规范 二. 创建表 1. 基本用法 2. 数据表命名规范 3. 字段命名规范 三. 删除表 一.创建数据库 1. 创建方式 在第二章中我们讲了如何 ...
- MySQL SQL DLL (数据定义语言)
CREATE CREATE DATABASE CREATE DATABASE 用于创建数据库 CREATE DATABASE new_database_name; CREATE TABLE CREAT ...
随机推荐
- 支持续传功能的ASP.NET WEB API文件下载服务
先把原文地址放上来,随后翻译
- RSA加密解密实现(JAVA)
1.关于RSA算法的原理解析参考:http://www.ruanyifeng.com/blog/2013/06/rsa_algorithm_part_one.html 2.RSA密钥长度.明文长度和密 ...
- 自动化测试-6.selenium的css定位
前言 大部分人在使用selenium定位元素时,用的是xpath定位,因为xpath基本能解决定位的需求.css定位往往被忽略掉了,其实css定位也有它的价值,css定位更快,语法更简洁.这一篇css ...
- python3+requests+unittest:接口自动化测试(一)
转载请表明出处:https://www.cnblogs.com/shapeL/p/9179484.html 简单介绍框架的实现逻辑,参考代码的git地址: https://github.com/zha ...
- nf_conntrack
How to Fix Nf_conntrack Table Full Dropping Packet Linux CentOS PacketLoss 2 minutes to read (234 w ...
- git教程: 查看文件状态与修改内容
转载:时光机穿梭 我们已经成功地添加并提交了一个readme.txt文件,现在,是时候继续工作了,于是,我们继续修改readme.txt文件,改成如下内容: Git is a distributed ...
- pycharm中添加PATH变量
最近在pycharm中run程序,终端terminal没有问题,在pycharm找不到$PATH中的变量值,如下图所示 同样的命令,在终端敲就没毛病,终端echo $PATH的时候,显示的是有cuda ...
- SSH设置秘钥登录
设置 SSH 通过密钥登录 我们一般使用 PuTTY 等 SSH 客户端来远程管理 Linux 服务器.但是,一般的密码方式登录,容易有密码被暴力破解的问题.所以,一般我们会将 SSH 的端口设置为默 ...
- 自定义textview
#import <UIKit/UIKit.h> @class FSTextView; typedef void(^FSTextViewHandler)(FSTextView *textVi ...
- 创建ApplicationContext与BeanFactory时的区别-Spring源码学习之容器的基本实现
传送门 可以加载XML两种方法 使用 BeanFactory 加载 XML BeanFactory bf = new XmlBeanFactory(new ClassPathResource(&quo ...