postgres Date/Time 学习笔记

一、Date/Time Types

---

参考文档：https://www.postgresql.org/docs/9.2/static/datatype-datetime.html

Types	别名	备注
timestamp [ (p) ] [ without time zone ]		日期+时间
timestamp [ (p) ] with time zone	timestamptz	日期+时间
date		日期
time [ (p) ] [ without time zone ]		时间
time [ (p) ] with time zone	timetz	时间
interval [ fields ] [ (p) ]		时间段

注：

1、sequelize 里常用的 DATE 类型指的是 postgres 的 timestamptz 类型

2、仅 Date 没有时区概念。

3、interval 这里不展开了，待写。

二、Date/Time Input/Output

---

前提：postgres 的时区设置成了 PRC（中国）。

1、普通值

例：2018-08-16 20:12:16+08

//存
INSERT INTO "public"."MemberOrderLasts"("id","mobile","last_time")
VALUES
(1,'13600000000','2018-08-16 20:12:16+08');

//取
select  "last_time"  from "MemberOrderLasts"  where id = '1'
//2018-08-16 20:12:16.920642+08

注：对于国内程序员来说，使用“/”容易产生歧义，最好使用“-”做分隔，即 2018-08-16 20:12:16。

2、特殊值

当在 SQL 命令中用作常量时，所有这些值都需要用单引号括起来。

Input String	Valid Types	Description
epoch	date, timestamp	1970-01-01 00:00:00+00 (Unix system time zero)
infinity	date, timestamp	later than all other time stamps
-infinity	date, timestamp	earlier than all other time stamps
now	date, time, timestamp	current transaction's start time
today	date, timestamp	midnight today
tomorrow	date, timestamp	midnight tomorrow
yesterday	date, timestamp	midnight yesterday
allballs	time	00:00:00.00 UTC

INSERT INTO "public"."MemberOrderLasts"("id","mobile","last_time")
VALUES
(2,'13600000000','now');

三、Date/Time TimeZone

---

参考我另一篇：《从 moment -> nodejs -> sequelize -> postgres，你都得设置好时区》。

四、Date/Time Functions and Operators

---

参考文档：https://www.postgresql.org/docs/9.2/static/functions-datetime.html

1、运算符

+ / -（* / / 不赘述，具体看文档）

select date '2018-01-12' + interval '7'
-- 2018-01-12 00:00:07 

select "activatedAt" - "createdAt" AS "diff" from "Members" where id = '373'
-- 19 days 09:07:11.155

2、函数

Function	Return Type	Description	Example	Result
age( timestamp , timestamp )	interval	Subtract arguments, producing a "symbolic" result that uses years and months	age(timestamp '2001-04-10', timestamp '1957-06-13')	43 years 9 mons 27 days
age( timestamp )	interval	Subtract from current_date (at midnight)	age(timestamp '1957-06-13')	43 years 8 mons 3 days
date_part( text , timestamp )	double precision	Get subfield (equivalent to extract); see Section 9.9.1	date_part('hour', timestamp '2001-02-16 20:38:40')	20
date_part( text , interval )	double precision	Get subfield (equivalent to extract); see Section 9.9.1	date_part('month', interval '2 years 3 months')	3
date_trunc( text , timestamp )	timestamp	Truncate to specified precision; see also Section 9.9.2	date_trunc('hour', timestamp '2001-02-16 20:38:40')	2001-02-16 20:00:00
extract(field from timestamp )	double precision	Get subfield; see Section 9.9.1	extract(hour from timestamp '2001-02-16 20:38:40')	20
extract(field from interval )	double precision	Get subfield; see Section 9.9.1	extract(month from interval '2 years 3 months')	3
isfinite( date )	boolean	Test for finite date (not +/-infinity)	isfinite(date '2001-02-16')	true
isfinite( timestamp )	boolean	Test for finite time stamp (not +/-infinity)	isfinite(timestamp '2001-02-16 21:28:30')	true
isfinite( interval )	boolean	Test for finite interval	isfinite(interval '4 hours')	true
justify_days( interval )	interval	Adjust interval so 30-day time periods are represented as months	justify_days(interval '35 days')	1 mon 5 days
justify_hours( interval )	interval	Adjust interval so 24-hour time periods are represented as days	justify_hours(interval '27 hours')	1 day 03:00:00
justify_interval( interval )	interval	Adjust interval using justify_days and justify_hours, with additional sign adjustments	justify_interval(interval '1 mon -1 hour')	29 days 23:00:00

常用：extract / date_part 和 date_trunc 函数

1、extract / date_part
select date_part('hour', timestamp '2001-02-16 20:38:40')
-- 20
select extract('hour' from timestamp '2001-02-16 20:38:40')
-- 20

2、date_trunc
select date_trunc('hour', timestamp '2001-02-16 20:38:40')
-- 2001-02-16 20:00:00

3、查询时间是否重叠

(start1, end1) OVERLAPS (start2, end2)

(start1, length1) OVERLAPS (start2, length2)

// 一、两种用法
// 1、`(start1, end1) OVERLAPS (start2, end2)`
SELECT (DATE '2001-02-16', DATE '2001-12-21') OVERLAPS
       (DATE '2001-10-30', DATE '2002-10-30');
Result: true

// 2、`(start1, length1) OVERLAPS (start2, length2)`
SELECT (DATE '2001-02-16', INTERVAL '100 days') OVERLAPS
       (DATE '2001-10-30', DATE '2002-10-30');
Result: false

// 二、重叠判断 遵循左开右闭的原则
SELECT (DATE '2001-10-29', DATE '2001-10-30') OVERLAPS
       (DATE '2001-10-30', DATE '2001-10-31');
Result: false
SELECT (DATE '2001-10-30', DATE '2001-10-30') OVERLAPS
       (DATE '2001-10-30', DATE '2001-10-31');
Result: true

五、Current Date/Time

---

1、普通

注意：下面的日期/时间在同一个事务里都是不变的。（如果要变化，建议使用下面即将介绍的 statement_timestamp 或者 clock_timestamp ）。

（1）带时区信息

CURRENT_DATE

CURRENT_TIME

CURRENT_TIMESTAMP

CURRENT_TIME(precision)

CURRENT_TIMESTAMP(precision)

还记得上面说的特殊值 now 吗，实际上跟 CURRENT_TIMESTAMP 一样：

//下面三个完全相等
SELECT CURRENT_TIMESTAMP;
SELECT TIMESTAMP 'now';
SELECT now();

（2）不带时区信息

LOCALTIME

LOCALTIMESTAMP

LOCALTIME(precision)

LOCALTIMESTAMP(precision)

仅 Date 没有时区概念。

select CURRENT_TIMESTAMP
// 2018-09-13 15:10:05.639902+08
select LOCALTIMESTAMP
// 2018-09-13 15:10:05.639902

2、事务专用

为了保障同一事务中的多个修改具有相同的时间戳，所以 postgre 提供了针对性的时间函数：

（1） transaction_timestamp() 返回事务开始的时间

其实 transaction_timestamp = CURRENT_TIMESTAMP ，只是针对的场景不同，所以换了个名字。

（2） statement_timestamp() 返回当前语句的开始时间

statement_timestamp() 和 transaction_timestamp() 只在一个事务内的第一条命令返回值相同。

（3） clock_timestamp() 返回实际的当前时间，因此即使在单个 SQL 命令中它的值也会更改。

select clock_timestamp(),clock_timestamp()
-- 2020-03-05 16:59:04.264526+08 | 2020-03-05 16:59:04.264527+08 （不一样）

六、Delaying Execution

---

pg_sleep()让当前的会话进程休眠 seconds 秒以后再执行。

SELECT pg_sleep(1.5);