I've got the following table:
booking_id
user_id
11
1
12
76
13
932
14
1
15
626
16
1
17
3232
I want to access the 2nd maximum booking_id for user 1.
The expected result is user_id = 1, booking_id = 14.
I've been working over these hellish flames for way too long, this doesn't do any good:
select booking.user_id, b1.booking_id from booking
left join(select
user_id,
booking_id
from booking
where booking_id = (select
max(booking_id)
from booking
where booking_id <> (select
max(booking_id)
from booking))
group by user_id)
as b1 on b1.user_id = booking.user_id
where booking.user_id = '1'
Please note I've managed to do it as a calculated column but that's useless, I need the derived table.
If you are using MySQL, you can avoid the (rather messy) double sub-query by using LIMIT & OFFSET
Just add order by booking_id desc LIMIT 1 OFFSET 1 and you will get the second highest booking_id. For example ...
select * from booking where user_id = 1 order by booking_id desc OFFSET 1 LIMIT 1
I tested this on one of my tables & it worked fine. If you have an index on booking_id it should be really fast.
If you want the second highest booking for the user who holds the highest booking, then this should work
SELECT * FROM booking
WHERE user_id in
(select user_id from booking order by booking_id desc limit 1)
ORDER BY booking_id DESC LIMIT 1 OFFSET 1
The sub-query finds the user_id of the user with the highest booking, then the main query finds their second highest booking
A simple way to do it is using LIMIT OFFSET:
SELECT *
FROM booking
WHERE user_id = 1
ORDER BY booking_id DESC
LIMIT 1 OFFSET 1
Demo here
By using the answer in this question What is the simplest SQL Query to find the second largest value? https://stackoverflow.com/a/7362165/14491685
you can integrate with your query to get it like this:
select * from booking
where booking_id =
(select max(booking_id) from booking
where user_id =1
and booking_id not in (SELECT MAX(booking_id ) FROM booking ))
Related
booking table
id status
1 booked
1 booked
3 cancelled
2 cancelled
2 booked
1 cancelled
1 cancelled
1 booked
select id, count(status) as tot_cancel
from tbl_booking
where id=1 and status='cancelled';
result will be
id tot_cancel
1 2
BUT I NEED last 5 records of user id = 1 and cancelled count not total table record cancelled
I understand that you want the count of cancelled records over the last 5 records of user 1. It is possibe, but you need a column that defines the ordering of the rows, so it is unambiguous which records are last.
Assuming that such column exists in your table and is called ordering_id, you can do:
select sum(status = 'cancelled') no_cancelled
from (select status from mytable where id = 1 order by ordering_id desc limit 5) t
You can do conditional aggregation with limit clause :
select 1 as id, sum(status = 'cancelled')
from table t
where id = 1
order by ? -- use ordering column instead ?
limit 5;
You can use correlated subquery for all ids :
select t.id, sum(t.status = 'cancelled')
from table t
where t.? in (select t1.?
from table t1
where t1.id = t.id
order by t1.?
limit 5
)
group by t.id;
The following does what it supposed to
it returns 200+ distinct records w/o the "limit 2"
What I want is to return 2 distinct records, but it stops after the 1st 2 records, meaning I only get 2 records
select distinct LEFT(`name`, LOCATE("(", `name`)-1), user_id, id
from ppbv79_listings
where user_id = 3798 and category_id = 30
group by LEFT(`name`, LOCATE("(", `name`)-1)
limit 2
Name user_id id
Germany 1213 Used Carl Sonnenschein 3798 2160555
Germany 1213 Used Carl Sonnenschein 3798 2160556
Try this:
select A.`trimmedName`, A.user_id, A.id
from
(select LEFT(`name`, LOCATE("(", `name`)-1)
`trimmedName`, user_id, id,count(category_id) `count`
from ppbv79_listings
where user_id = 3798 and category_id = 30
group by LEFT(`name`, LOCATE("(", `name`)-1), user_id, id
order by `count` desc) A
limit 2;
I assume there are some repetitions you would want to removed, and fetch the just the top 2 rows of data that repeats most.
I have four tables with the following structure.
Table 1:
Project - have unique project names (prj_name)
Table 2:
my_records - have the following fields:
record_id,prj_name,my_dept,record_submit_date,record_state
Table 3:
record_states have multiple states where 'Completed' is one.
Table 4:custom_dept_list
dept_name
I need to get the percentage of (records have state as completed) and (Total records) grouped by my_project where my_dept in custom_dept_list and record_submit_date is greater than "some date"
I have tried the following:
Query:
select prj_name,count(record_id) as total,((select count(record_id) from
my_records where record_state='Completed')/(count(record_id)))*100 as
percent from my_records,custom_dept_list where record_state='Completed'
and record_submit_date >= ( CURDATE() - INTERVAL 15 DAY ) and
my_dept=dept_name group by prj_name order by percent desc;
Total records for project A = 50
Total records for project A with record_state='Completed' = 30
Ratio is not coming - (30/50)*100 = 60
It is giving some very big value.
Below is the data from my_records, i have removed record_submit date to make it simple:
|1|prj1|dept1|Completed
|2|prj1|dept1|XYZ
|3|prj1|dept1|Completed
|4|prj1|dept2|XYZ
|5|prj1|dept2|Completed
|6|prj1|dept1|XYZ
|7|prj1|dept1|XYZ
|8|prj1|dept1|XYZ
|9|prj1|dept2|XYZ
|10|prj1|dept2|XYZ
|11|prj1|dept2|Completed
|12|prj1|dept2|Completed
|13|prj1|dept2|Completed
|14|prj1|dept3|XYZ
|15|prj1|dept4|Completed
|16|prj1|dept4|XYZ
|17|prj1|dept5|Completed
|18|prj1|dept6|XYZ
|19|prj1|dept7|XYZ
|20|prj1|dept8|XYZ
|21|prj1|dept10|XYZ
|22|prj1|dept2|XYZ
|23|prj1|dept2|Completed
|24|prj1|dept2|Completed
|25|prj1|dept2|Completed
Data From Custom_dept_List:
dept_name
dept1
dept3
dept4
dept5
dept6
dept8
dept10
I have tried the following queries :
Query 1
select count(record_id) as count,prj_name from my_records,custom_dept_list where my_dept=dept_name group by prj_name order by count desc;
Ouput -- 13
Query 2
select count(record_id) as count,prj_name from my_records,custom_dept_list where my_dept=dept_name and record_state='Completed' group by prj_name order by count desc;
Output -- 4
Query 3
select prj_name,count(record_id) as total,count(case when record_state='Completed' then record_id end) /count(record_id) *100 as percent from my_records join custom_dept_list on my_dept = dept_name where record_state = 'Completed' group by prj_name order by percent desc;
Output :
prj_name total percent
prj1 4 100.0000
First of all, please use proper join instead of multiple tables in your from clause.
Then, you don't need that inner query to get the count with a specific record_state, you can use a case inside the count:
select prj_name,
count(record_id) as total,
count(case when record_state='Completed' then record_id end) /
count(record_id) * 100 as percent
from my_records
join custom_dept_list
on my_dept = dept_name
where record_submit_date >= ( CURDATE() - INTERVAL 15 DAY )
group by prj_name
order by percent desc;
Your problem was probably caused by that inner query, that was not counting each project's completed records, but all the completed records instead.
you do not need this record_state = 'Completed' condition because of this you get only completed record as total recoded. so try without it.
select prj_name,
count(record_id) as total,
count(case when record_state='Completed' then record_id end) /
count(record_id) * 100 as percent
from my_records
join custom_dept_list
on my_dept = dept_name
where record_submit_date >= ( CURDATE() - INTERVAL 15 DAY )
group by prj_name
I have a table which looks like this
courseid session_date title published
1 2012-07-01 Training Course A 0
1 2012-07-02 Training Course A 0
2 2012-07-04 Training Course B 1
2 2012-07-07 Training Course B 1
3 2012-07-05 Training Course C 1
3 2012-07-06 Training Course C 1
4 2012-07-07 Training Course D 1
4 2012-07-10 Training Course D 1
The table has two entries for each ID and Title because the session_date column shows the start date and the end date of the course.
I am trying to create a query that will pull the next five courses without showing any courses in the past.
I have gotten this far
SELECT session_date, title, courseid
FROM table
WHERE published = 1 AND session_date > DATE(NOW())
ORDER BY session_date ASC LIMIT 0,5
This pulls rows from the table for the next five session-dates but it includes both start dates and finish dates whereas I need the next five courses ordered by start date.
I need to create a query that will pull the earliest session_date for each courseid but ignore the row with the latest session_date for that same courseid but I am at a complete loss of how to do this.
Any help or advice would be most gratefully received.
If you group your results by course and select the MAX(session_date), you will get the latest of the dates associated with each course (i.e. the finish date):
SELECT courseid, MIN(session_date) AS start_date
FROM `table`
WHERE published = 1
GROUP BY courseid
HAVING start_date > CURRENT_DATE
ORDER BY start_date ASC
LIMIT 5
See it on sqlfiddle.
What you need to do is retrieve only the rows with the minimum session_date per courseid group and order by that resulting set:
SELECT
b.*
FROM
(
SELECT courseid, MIN(session_date) AS mindate
FROM tbl
GROUP BY courseid
) a
INNER JOIN
tbl b ON a.courseid = b.courseid AND a.mindate = b.session_date
WHERE
b.session_date > NOW() AND
b.published = 1
ORDER BY
b.session_date
LIMIT 5
But a much better design would be to only have one row per courseid and have two columns specifying start and end dates:
tbl
------------------
courseid [PK]
start_date
end_date
title
published
Then you can simply do:
SELECT *
FROM tbl
WHERE start_date > NOW() AND published = 1
ORDER BY start_date
LIMIT 5
Since values of all the columns in your SELECT clause are repeating, just use DISTINCT
SELECT distinct session_date, title, courseid
FROM table
WHERE published = 1 AND session_date > DATE(NOW())
ORDER BY session_date ASC LIMIT 0,5
Is it possible to limit the results a mysql select query returns with a condition?
For example I have a reviews table:
review_id, member_id, text, date
And I'd like to get the latest 10 reviews but member_id = 123 should only be taken once
Can this be achieved with a single query?
My interpretation of the problem:
the 10 most recent reviews
including at most 1 review with member_id = 123
I'm going to solve this by:
starting with the full reviews result set
removing all reviews that have member_id = 123 except for the most recent one
from the modified result set, take the 10 most recent
Here's the query:
create view newest123 as ( -- this gets the newest review for member_id 123
select *
from reviews
where member_id = 123
order by date desc limit 1
)
select *
from (
select * from newest123
union
select * -- all the reviews that aren't for member_id 123
from reviews
where member_id != 123) filtered
order by date desc limit 10 -- sort 'em and take the top 10