DATE_SUB() Getting the recent one month or year records from MySQL table

Syntax of DATE_SUB()

DATE_SUB(date, INTERVAL, expression, UNIT)

Example

SELECT DATE_SUB(  '2016-12-25', INTERVAL 3 DAY )

Output is 2016-12-22

We have subtracted three days form the given date by using DATE_SUB() function. In place of DAY we can use Month, year, hour, minute, second etc , here is the list.

unit Value	Expected expr Format	Example
DAY	DAYS	DATE_SUB( '2016-11-29', INTERVAL 10 DAY )
MONTH	MONTHS	DATE_SUB( '2015-11-20', INTERVAL 5 MONTH )
WEEK	WEEKS	DATE_SUB( '2016-08-20', INTERVAL 5 WEEK )
QUARTER	QUARTERS	DATE_SUB( '2016-08-20', INTERVAL 2 QUARTER )
YEAR	YEARS	DATE_SUB( '2016-02-23', INTERVAL 2 YEAR )
YEAR_MONTH	'YEARS-MONTHS'	DATE_SUB( '2016-02-23', INTERVAL '2-5' YEAR_MONTH )

To use the above Examples add SELECT at left and run the query.

SELECT DATE_SUB( '2016-02-23', INTERVAL 2 YEAR ); // 2014-02-23 
SELECT DATE_SUB( CURDATE(), INTERVAL 2 YEAR ); // 2018-02-23

The second query depends on the todays date, so your result will be different.
Some time we have to collect last 7 or 15 days or X days (or month, year or week) data from MySQL table.

For example let us find out who are the new members joined in our forum in last week. One shop may be interested in knowing new products added in last one month. What are the books arrived in last one year. Here irrespective of the date values we want the records of last X days from today, or we can say that the records between today and last X days ( month , year or week) are required.

We will use the MySQL function CURDATE() to get the today's date.

To get the difference in today date and previous day or month we have to use the MySQL function DATE_SUB

DATE_SUB is a MySQL function which takes date expression, the interval and the constant to return the date value for further calculation.

Here are some sample queries on how to get the records as per requirements . �

Last 10 days records

select * from dt_table where  `date` >= DATE_SUB(CURDATE(), INTERVAL 10 DAY)

The above query will return last 10 days records. Note that this query will return all future dates also. To exclude future dates we have to modify the above command a little by using between query to get records. Here is the modified one.

SELECT * FROM dt_table WHERE `date` BETWEEN DATE_SUB( CURDATE( ) ,INTERVAL 10 DAY ) AND CURDATE( )

Present Month Records

Starting from 1st day of the current month till now.

SELECT * FROM `dt_table` WHERE  date between  DATE_FORMAT(CURDATE() ,'%Y-%m-01') AND CURDATE()

Present Year Records

Starting from 1st Jan of the current Year till now.

SELECT * FROM `dt_table` WHERE  date between  DATE_FORMAT(CURDATE() ,'%Y-01-01') AND CURDATE()

Last one month records

Let us try to get records added in last one month

SELECT * FROM dt_table where  `date` >= DATE_SUB(CURDATE(), INTERVAL 1 MONTH)

Here also future records will be returned so we can take care of that by using BETWEEN commands if required.

SELECT * FROM dt_table WHERE date BETWEEN DATE_SUB(CURDATE(), INTERVAL 1 MONTH)  AND CURDATE();

Using Year

select * from dt_table WHERE `date` >= DATE_SUB(CURDATE(), INTERVAL 1 YEAR)

Using year with BETWEEN

SELECT * FROM dt_table WHERE date BETWEEN DATE_SUB(CURDATE(), INTERVAL 1 YEAR)  AND CURDATE();

Records of previous month of any year

SELECT * FROM  dt_table WHERE MONTH( DATE ) = MONTH( DATE_SUB(CURDATE(),INTERVAL 1 MONTH ))

Records of previous month of same year

SELECT * FROM  dt_table WHERE MONTH( DATE ) = MONTH( DATE_SUB(CURDATE(),INTERVAL 1 MONTH )) 
AND 
YEAR( DATE ) = YEAR( DATE_SUB(CURDATE( ),INTERVAL 1 MONTH ))

Note the difference between Last one month record and Previous month record

Records of two date ranges

We can collect records between a particular date ranges by using between command and DATE_SUB. Here are some queries to generate records between two date ranges.

SELECT * FROM dt_table WHERE `date` BETWEEN DATE_SUB( CURDATE( ) ,INTERVAL 3 MONTH ) AND DATE_SUB( CURDATE( ) ,INTERVAL 0 MONTH )

This query will return records between last three months. This query again we will modify to get the records between three moths and six months.

SELECT * FROM dt_table WHERE `date` BETWEEN DATE_SUB( CURDATE( ) ,INTERVAL 6 MONTH ) AND DATE_SUB( CURDATE( ) ,INTERVAL 3 MONTH )

Now let us change this to get records between 6 month and 12 month.

SELECT * FROM dt_table WHERE `date` BETWEEN DATE_SUB( CURDATE( ) ,INTERVAL 12  MONTH ) AND DATE_SUB( CURDATE( ) ,INTERVAL 6 MONTH )

With this you can understand how the records between a month range or a year range can be collected from a table. Note that the months ranges are calculated starting from current day. So if we are collecting records of last three months and we are in 15th day of 9th month then records of 15th day of 6th month we will get but the records of 14th day of 6th month will be returning on next query that is between 3 months and 6 months.

Records of present week

SELECT * FROM `dt_table` WHERE WEEKOFYEAR(date)=WEEKOFYEAR(CURDATE())

Records of previous week

SELECT * FROM `dt_table` WHERE WEEKOFYEAR(date)=WEEKOFYEAR(CURDATE())-1

Records of next week

SELECT * FROM `dt_table` WHERE WEEKOFYEAR(date)=WEEKOFYEAR(CURDATE())+1

Records of present week all working days ( Mon - Fri )

SELECT * FROM `dt_table` WHERE WEEKOFYEAR(date)=WEEKOFYEAR(CURDATE()) 
AND 
WEEKDAY(date)  BETWEEN 1 AND 5

Records of present week all working days till today

SELECT * FROM `dt_table` 
WHERE WEEKOFYEAR( DATE ) = WEEKOFYEAR( CURDATE( ) )
AND WEEKDAY( DATE ) BETWEEN 1 AND WEEKDAY(CURDATE())

Using Date and time Queries

Now let us calculate with Time

unit Value	Expected expr Format
Example
MICROSECOND	MICROSECONDS
DATE_SUB( '2016-02-23 20:55:58', INTERVAL 225 MICROSECOND )
SECOND	SECONDS
DATE_SUB( '2016-02-23 20:55:58', INTERVAL 2 SECOND )
MINUTE	MINUTES
DATE_SUB( '2016-02-23 20:55:58', INTERVAL 2 MINUTE )
HOUR	HOURS
DATE_SUB( '2016-02-23 20:55:58', INTERVAL 5 HOUR )
SECOND_MICROSECOND	'SECONDS. MICROSECONDS'
DATE_SUB( '2016-02-23 20:55:58', INTERVAL '1.543' SECOND_MICROSECOND )
MINUTE_MICROSECOND	'MINUTES: SECONDS. MICROSECONDS'
DATE_SUB ( '2016-02-23 20:55:58', INTERVAL '5:2.743' MINUTE_MICROSECOND )
MINUTE_SECOND	'MINUTES: SECONDS'
DATE_SUB( '2016-02-23 20:55:58', INTERVAL '5:2' MINUTE_SECOND )
HOUR_MICROSECOND	'HOURS:MINUTES: SECONDS. MICROSECONDS'
DATE_SUB( '2016-02-23 20:55:58', INTERVAL '5:2:1.249' HOUR_MICROSECOND)
HOUR_SECOND	'HOURS:MINUTES: SECONDS'
DATE_SUB( '2016-02-23 20:55:58', INTERVAL '5:2:1' HOUR_SECOND )
HOUR_MINUTE	'HOURS:MINUTES'
DATE_SUB( '2016-02-23 20:55:52', INTERVAL '5:2' HOUR_MINUTE )
DAY_MICROSECOND	'DAYS HOURS: MINUTES: SECONDS. MICROSECONDS'
DATE_SUB( '2016-02-23 20:55:52', INTERVAL '2 5:2:24.879' DAY_MICROSECOND
DAY_SECOND	'DAYS HOURS: MINUTES:SECONDS'
DATE_SUB( '2016-02-23 20:55:52', INTERVAL '2 5:2:24' DAY_SECOND)
DAY_MINUTE	'DAYS HOURS: MINUTES'
DATE_SUB( '2016-02-23 20:55:52', INTERVAL '2 5:2' DAY_MINUTE)
DAY_HOUR	'DAYS HOURS'
DATE_SUB( '2016-02-23 20:55:52', INTERVAL '2 5' DAY_HOUR)

Our sample table dt_table_tm stores login date with time in a field along with one more column showing event_id.

All records of Last 5 Hours

SELECT * FROM `dt_table_tm` WHERE tm>=DATE_SUB(NOW(), INTERVAL 5 HOUR)

All records of Last 48 Hours

SELECT * FROM `dt_table_tm` WHERE tm>=DATE_SUB(NOW(), INTERVAL 48 HOUR)

All records of 15 hours 12 minutes

SELECT * FROM `dt_table_tm` WHERE tm>=DATE_SUB(NOW() , INTERVAL '15:12' HOUR_MINUTE)

In place of NOW() we can use specific date and time ( timestamp ) Note : While using date use the format YYYY-mm-dd ( YEAR - Month - date) and while using time use HH:MM:SS ( Hour : Minutes : Seconds )

SELECT * FROM  `dt_table_tm` WHERE tm >= DATE_SUB( '2018-08-10 11:50:00', INTERVAL '2:18:33' HOUR_SECOND )

All logins between 9 and 10 hours ( from 9 and 10 both inclusive till it is not 11 )

SELECT * FROM `dt_table_tm` WHERE HOUR(tm) between 9 and 10

All logins betwee 9 and 10 for a perticular month and year

SELECT * FROM `dt_table_tm` WHERE HOUR(tm) between 9 and 10 and MONTH(tm)=08 and year(tm)=2018

By Using date_format()

SELECT * FROM `dt_table_tm` WHERE HOUR(tm) between 9 and 10 and date_format(tm,'%Y-%m')='2018-08'

All logins between 11 and 18 hours on a prticular day

SELECT * FROM `dt_table_tm` WHERE HOUR(tm) between 11 and 18 and date(tm)='2018-08-09'

All logins between two input times in hour : minutes : seconds on a perticular day

SELECT * FROM `dt_table_tm` WHERE date(tm)='2018-08-09' AND  TIME(tm) BETWEEN TIME('9:00:00') AND TIME('10:15:00')

Counting logins at different hours in all days ( using GROUP BY )

SELECT hour(tm) , count(event_id)   FROM `dt_table_tm` group by HOUR(tm)

First Login ( tm ) of all days. We will use GROUP BY and MIN()

SELECT DATE(tm), MIN(tm),DATE_FORMAT(MIN(tm),'%H : %i :%s') time 
FROM `dt_table_tm` GROUP BY  DATE(tm)

First Login of all days after a particular hour. ( HOUR(tm) >16 )

SELECT DATE(tm), MIN(tm),DATE_FORMAT(MIN(tm),'%H : %i :%s') time
 FROM `dt_table_tm`  WHERE HOUR(tm) > 16 GROUP BY  DATE(tm)

First login of all days after a particular HOUR Minutes and seconds.

SELECT DATE(tm), MIN(tm),DATE_FORMAT(MIN(tm),'%H : %i :%s') time 
 FROM `dt_table_tm`  WHERE TIME(tm) > '16:01:00'  GROUP BY  DATE(tm)

Credit & Source: https://www.plus2net.com/sql_tutorial/date-lastweek.php

Web Development

Thursday, March 25, 2021

Getting the recent week, one month or year records from MySQL table