I need to get max date for each row over other ids. Of course I can do this with CROSS JOIN and JOIN .
Like this
WITH t AS (
SELECT 1 AS id, rep_date FROM UNNEST(GENERATE_DATE_ARRAY('2021-09-01','2021-09-09', INTERVAL 1 DAY)) rep_date
UNION ALL
SELECT 2 AS id, rep_date FROM UNNEST(GENERATE_DATE_ARRAY('2021-08-20','2021-09-03', INTERVAL 1 DAY)) rep_date
UNION ALL
SELECT 3 AS id, rep_date FROM UNNEST(GENERATE_DATE_ARRAY('2021-08-25','2021-09-05', INTERVAL 1 DAY)) rep_date
)
SELECT id, rep_date, MAX(rep_date) OVER (PARTITION BY id) max_date, max_date_over_others FROM t
JOIN (
SELECT t.id, MAX(max_date) max_date_over_others FROM t
CROSS JOIN (
SELECT id, MAX(rep_date) max_date FROM t
GROUP BY 1
) t1
WHERE t1.id <> t.id
GROUP BY 1
) USING (id)
But it's too wired for huge tables. So I'm looking for the some simpler way to do this. Any ideas?
Your version is good enough I think. But if you want to try other options - consider below approach. It might looks more verbose from first look - but should be more optimal and cheaper to compare with your version with cross join
temp as (
select id,
greatest(
ifnull(max(max_date_for_id) over preceding_ids, '1970-01-01'),
ifnull(max(max_date_for_id) over following_ids, '1970-01-01')
) as max_date_for_rest_ids
from (
select id, max(rep_date) max_date_for_id
from t
group by id
)
window
preceding_ids as (order by id rows between unbounded preceding and 1 preceding),
following_ids as (order by id rows between 1 following and unbounded following)
)
select *
from t
join temp
using (id)
Assuming your original table data just has columns id and dt - wouldn't this solve it? I'm using the fact that if an id has the max dt of everything, then it gets the second-highest over the other id values.
WITH max_dates AS
(
SELECT
id,
MAX(dt) AS max_dt
FROM
data
GROUP BY
id
),
with_top1_value AS
(
SELECT
*,
MAX(dt) OVER () AS max_overall_dt_1,
MIN(dt) OVER () AS min_overall_dt
FROM
max_dates
),
with_top2_values AS
(
SELECT
*,
MAX(CASE WHEN dt = max_overall_dt_1 THEN min_overall_dt ELSE dt END) AS max_overall_dt2
FROM
with_top1_value
),
SELECT
*,
CASE WHEN dt = max_overall_dt1 THEN max_overall_dt2 ELSE max_overall_dt1 END AS max_dt_of_others
FROM
with_top2_values
Related
I am using redshift sql and would like to group users who has overlapping voucher period into a single row instead (showing the minimum start date and max end date)
For E.g if i have these records,
I would like to achieve this result using redshift
Explanation is tat since row 1 and row 2 has overlapping dates, I would like to just combine them together and get the min(Start_date) and max(End_Date)
I do not really know where to start. Tried using row_number to partition them but does not seem to work well. This is what I tried.
select
id,
start_date,
end_date,
lag(end_date, 1) over (partition by id order by start_date) as prev_end_date,
row_number() over (partition by id, (case when prev_end_date >= start_date then 1 else 0) order by start_date) as rn
from users
Are there any suggestions out there? Thank you kind sirs.
This is a type of gaps-and-islands problem. Because the dates are arbitrary, let me suggest the following approach:
Use a cumulative max to get the maximum end_date before the current date.
Use logic to determine when there is no overall (i.e. a new period starts).
A cumulative sum of the starts provides an identifier for the group.
Then aggregate.
As SQL:
select id, min(start_date), max(end_date)
from (select u.*,
sum(case when prev_end_date >= start_date then 0 else 1
end) over (partition by id
order by start_date, voucher_code
rows between unbounded preceding and current row
) as grp
from (select u.*,
max(end_date) over (partition by id
order by start_date, voucher_code
rows between unbounded preceding and 1 preceding
) as prev_end_date
from users u
) u
) u
group by id, grp;
Another approach would be using recursive CTE:
Divide all rows into numbered partitions grouped by id and ordered by start_date and end_date
Iterate over them calculating group_start_date for each row (rows which have to be merged in final result would have the same group_start_date)
Finally you need to group the CTE by id and group_start_date taking max end_date from each group.
Here is corresponding sqlfiddle: http://sqlfiddle.com/#!18/7059b/2
And the SQL, just in case:
WITH cteSequencing AS (
-- Get Values Order
SELECT *, start_date AS group_start_date,
ROW_NUMBER() OVER (PARTITION BY id ORDER BY start_date, end_date) AS iSequence
FROM users),
Recursion AS (
-- Anchor - the first value in groups
SELECT *
FROM cteSequencing
WHERE iSequence = 1
UNION ALL
-- Remaining items
SELECT b.id, b.start_date, b.end_date,
CASE WHEN a.end_date > b.start_date THEN a.group_start_date
ELSE b.start_date
END
AS groupStartDate,
b.iSequence
FROM Recursion AS a
INNER JOIN cteSequencing AS b ON a.iSequence + 1 = b.iSequence AND a.id = b.id)
SELECT id, group_start_date as start_date, MAX(end_date) as end_date FROM Recursion group by id, group_start_date ORDER BY id, group_start_date
Suppose this example query:
select
id
, date
, sum(var) over (partition by id order by date rows 30 preceding) as roll_sum
from tab
When some dates are not present on date column the window will not consider the unexistent dates. How could i make this windowns aggregation including these unexistent dates?
Many thanks!
You can join a sequence containing all dates from a desired interval.
select
*
from (
select
d.date,
q.id,
q.roll_sum
from unnest(sequence(date '2000-01-01', date '2030-12-31')) d
left join ( your_query ) q on q.date = d.date
) v
where v.date > (select min(my_date) from tab2)
and v.date < (select max(my_date) from tab2)
In standard SQL, you would typically use a window range specification, like:
select
id,
date,
sum(var) over (
partition by id
order by date
range interval '30' day preceding
) as roll_sum
from tab
However I am unsure that Presto supports this syntax. You can resort a correlated subquery instead:
select
id,
date,
(
select sum(var)
from tab t1
where
t1.id = t.id
and t1.date >= t.date - interval '30' day
and t1.date <= t.date
) roll_sum
from tab t
I don't think Presto support window functions with interval ranges. Alas. There is an old fashioned way to doing this, by counting "ins" and "outs" of values:
with t as (
select id, date, var, 1 as is_orig
from t
union all
select id, date + interval '30 day', -var, 0
from t
)
select id.*
from (select id, date, sum(var) over (partition by id order by date) as running_30,
sum(is_org) as is_orig
from t
group by id, date
) id
where is_orig > 0
I'm working with a table that has an id and date column. For each id, there's a 90-day window where multiple transactions can be made. The 90-day window starts when the first transaction is made and the clock resets once the 90 days are over. When the new 90-day window begins triggered by a new transaction I want to start the count from the beginning at one. I would like to generate something like this with the two additional columns (window and count) in SQL:
id date window count
name1 7/7/2019 first 1
name1 12/31/2019 second 1
name1 1/23/2020 second 2
name1 1/23/2020 second 3
name1 2/12/2020 second 4
name1 4/1/2020 third 1
name2 6/30/2019 first 1
name2 8/14/2019 first 2
I think getting the rank of the window can be done with a CASE statement and MIN(date) OVER (PARTITION BY id). This is what I have in mind for that:
CASE WHEN MIN(date) OVER (PARTITION BY id) THEN 'first'
WHEN DATEDIFF(day, date, MIN(date) OVER (PARTITION BY id)) <= 90 THEN 'first'
WHEN DATEDIFF(day, date, MIN(date) OVER (PARTITION BY id)) > 90 AND DATEDIFF(day, date, MIN(date) OVER (PARTITION BY id)) <= 180 THEN 'third'
WHEN DATEDIFF(day, date, MIN(date) OVER (PARTITION BY id)) > 180 AND DATEDIFF(day, date, MIN(date) OVER (PARTITION BY id)) <= 270 THEN 'fourth'
ELSE NULL END
And incrementing the counts within the windows would be ROW_NUMBER() OVER (PARTITION BY id, window)?
You cannot solve this problem with window functions only. You need to iterate through the dataset, which can be done with a recursive query:
with
tab as (
select t.*, row_number() over(partition by id order by date) rn
from mytable t
)
cte as (
select id, date, rn, date date0 from tab where rn = 1
union all
select t.id, t.date, t.rn, greatest(t.date, c.date + interval '90' day)
from cte c
inner join tab t on t.id = c.id and t.rn = c.rn + 1
)
select
id,
date,
dense_rank() over(partition by id order by date0) grp,
count(*) over(partition by id order by date0, date) cnt
from cte
The first query in the with clause ranks records having the same id by increasing date; then, the recursive query traverses the data set and computes the starting date of each group. The last step is numbering the groups and computing the window count.
GMB is totally correct that a recursive CTE is needed. I offer this as an alternative form for two reasons. First, because it uses SQL Server syntax, which appears to be the database being used in the question. Second, because it directly calculates window and count without window functions:
with t as (
select t.*, row_number() over (partition by id order by date) as seqnum
from tbl t
),
cte as (
select t.id, t.date, dateadd(day, 90, t.date) as window_end, 1 as window, 1 as count, seqnum
from t
where seqnum = 1
union all
select t.id, t.date,
(case when t.date > cte.window_end then dateadd(day, 90, t.date)
else cte.window_end
end) as window_end,
(case when t.date > cte.window_end then window + 1 else window end) as window,
(case when t.date > cte.window_end then 1 else cte.count + 1 end) as count,
t.seqnum
from cte join
t
on t.id = cte.id and
t.seqnum = cte.seqnum + 1
)
select id, date, window, count
from cte
order by 1, 2;
Here is a db<>fiddle.
We are trying to port a code to run on Amazon Redshift, but Refshift won't run the recursive CTE function. Any good soul that knows how to port this?
with tt as (
select t.*, row_number() over (partition by id order by time) as seqnum
from t
),
recursive cte as (
select t.*, time as grp_start
from tt
where seqnum = 1
union all
select tt.*,
(case when tt.time < cte.grp_start + interval '3 second'
then tt.time
else tt.grp_start
end)
from cte join
tt
on tt.seqnum = cte.seqnum + 1
)
select cte.*,
(case when grp_start = lag(grp_start) over (partition by id order by time)
then 0 else 1
end) as isValid
from cte;
Or, a different code to reproduce the logic below.
It is a binary result that:
it is 1 if it is the first known value of an ID
it is 1 if it is 3 seconds or later than the previous "1" of that ID
It is 0 if it is less than 3 seconds than the previous "1" of that ID
Note 1: this is not the difference in seconds from the previous record
Note 2: there are many IDs in the data set
Note 3: original dataset has ID and Date
Desired output:
https://i.stack.imgur.com/k4KUQ.png
Dataset poc:
http://www.sqlfiddle.com/#!15/41d4b
As of this writing, Redshift does support recursive CTE's: see documentation here
To note when creating a recursive CTE in Redshift:
start the query: with recursive
column names must be declared for all recursive cte's
Consider the following example for creating a list of dates using recursive CTE's:
with recursive
start_dt as (select current_date s_dt)
, end_dt as (select dateadd(day, 1000, current_date) e_dt)
-- the recusive cte, note declaration of the column `dt`
, dates (dt) as (
-- start at the start date
select s_dt dt from start_dt
union all
-- recursive lines
select dateadd(day, 1, dt)::date dt -- converted to date to avoid type mismatch
from dates
where dt <= (select e_dt from end_dt) -- stop at the end date
)
select *
from dates
The below code could help you.
SELECT id, time, CASE WHEN sec_diff is null or prev_sec_diff - sec_diff > 3
then 1
else 0
end FROM (
select id, time, sec_diff, lag(sec_diff) over(
partition by id order by time asc
)
as prev_sec_diff
from (
select id, time, date_part('s', time - lag(time) over(
partition by id order by time asc
)
)
as sec_diff from hon
) x
) y
I have an sql table like that:
Id Date Price
1 21.09.09 25
2 31.08.09 16
1 23.09.09 21
2 03.09.09 12
So what I need is to get min and max date for each id and dif in days between them. It is kind of easy. Using SQLlite syntax:
SELECT id,
min(date),
max(date),
julianday(max(date)) - julianday(min(date)) as dif
from table group by id
Then the tricky one: how can I receive the price per day during this difference period. I mean something like this:
ID Date PricePerDay
1 21.09.09 25
1 22.09.09 0
1 23.09.09 21
2 31.08.09 16
2 01.09.09 0
2 02.09.09 0
2 03.09.09 12
I create a cte as you mentioned with calendar but dont know how to get the desired result:
WITH RECURSIVE
cnt(x) AS (
SELECT 0
UNION ALL
SELECT x+1 FROM cnt
LIMIT (SELECT ((julianday('2015-12-31') - julianday('2015-01-01')) + 1)))
SELECT date(julianday('2015-01-01'), '+' || x || ' days') as date FROM cnt
p.s. If it will be in sqllite syntax-would be awesome!
You can use a recursive CTE to calculate all the days between the min date and max date. The rest is just a left join and some logic:
with recursive cte as (
select t.id, min(date) as thedate, max(date) as maxdate
from t
group by id
union all
select cte.id, date(thedate, '+1 day') as thedate, cte.maxdate
from cte
where cte.thedate < cte.maxdate
)
select cte.id, cte.date,
coalesce(t.price, 0) as PricePerDay
from cte left join
t
on cte.id = t.id and cte.thedate = t.date;
One method is using a tally table.
To build a list of dates and join that with the table.
The date stamps in the DD.MM.YY format are first changed to the YYYY-MM-DD date format.
To make it possible to actually use them as a date in the SQL.
At the final select they are formatted back to the DD.MM.YY format.
First some test data:
create table testtable (Id int, [Date] varchar(8), Price int);
insert into testtable (Id,[Date],Price) values (1,'21.09.09',25);
insert into testtable (Id,[Date],Price) values (1,'23.09.09',21);
insert into testtable (Id,[Date],Price) values (2,'31.08.09',16);
insert into testtable (Id,[Date],Price) values (2,'03.09.09',12);
The SQL:
with Digits as (
select 0 as n
union all select 1
union all select 2
union all select 3
union all select 4
union all select 5
union all select 6
union all select 7
union all select 8
union all select 9
),
t as (
select Id,
('20'||substr([Date],7,2)||'-'||substr([Date],4,2)||'-'||substr([Date],1,2)) as [Date],
Price
from testtable
),
Dates as (
select Id, date(MinDate,'+'||(d2.n*10+d1.n)||' days') as [Date]
from (
select Id, min([Date]) as MinDate, max([Date]) as MaxDate
from t
group by Id
) q
join Digits d1
join Digits d2
where date(MinDate,'+'||(d2.n*10+d1.n)||' days') <= MaxDate
)
select d.Id,
(substr(d.[Date],9,2)||'.'||substr(d.[Date],6,2)||'.'||substr(d.[Date],3,2)) as [Date],
coalesce(t.Price,0) as Price
from Dates d
left join t on (d.Id = t.Id and d.[Date] = t.[Date])
order by d.Id, d.[Date];
The recursive SQL below was totally inspired by the excellent answer from Gordon Linoff.
And a recursive SQL is probably more performant for this anyway.
(He should get the 15 points for the accepted answer).
The difference in this version is that the datestamps are first formatted to YYYY-MM-DD.
with t as (
select Id,
('20'||substr([Date],7,2)||'-'||substr([Date],4,2)||'-'||substr([Date],1,2)) as [Date],
Price
from testtable
),
cte as (
select Id, min([Date]) as [Date], max([Date]) as MaxDate from t
group by Id
union all
select Id, date([Date], '+1 day'), MaxDate from cte
where [Date] < MaxDate
)
select cte.Id,
(substr(cte.[Date],9,2)||'.'||substr(cte.[Date],6,2)||'.'||substr(cte.[Date],3,2)) as [Date],
coalesce(t.Price, 0) as PricePerDay
from cte
left join t
on (cte.Id = t.Id and cte.[Date] = t.[Date])
order by cte.Id, cte.[Date];