I need to count the total count of calculated values between -10 and 10.
What I have tried to do is:
WITH routes as (
SELECT
reg,
heading-lag(heading) over (PARTITION BY reg order by time) AS direction
FROM my_table
)
SELECT direction, reg, Count(direction) AS total_count
FROM routes WHERE direction between -10 AND 10
GROUP BY reg, direction;
This counts how many of each value between -10 and 10 each route has. But how to count just one values per route in given range?
Try the following query:
WITH routes as (
SELECT
reg,
heading-lag(heading) over (PARTITION BY reg order by time) AS direction
FROM my_table
)
SELECT COUNT(DISTINCT reg) AS total
FROM routes WHERE direction between -10 AND 10;
Related
For each entry in the loc_of_interest table, I want to find the 15 minute timeslot (from the data in the other cte) with the maximum decrease in count of nearby points. I do not know how to proceed beyond the 'pseudocode' part, and indeed, am uncertain if I am going in the right direction with this existing code as well.
Here is my code:
-- I have two cte's already made
subset_cr -- many rows of data
(device_id, points_geom, time_created)
loc_of_interest -- 2 rows of data
(loc_id, points_geom)
-- here is how I wish to proceed:
with temp as (
SELECT loi.loc_id AS loc_id,
routes.fifteen_min_slot ,
routes.count_of_near_points
FROM loc_of_interest as loi
CROSS JOIN LATERAL (
SELECT date_trunc('hour', routes.time_created) + date_part('minute', routes.time_created)::int / 15 * interval '15 min' as fifteen_min_slot,
count (ST_DWithin(
loi.point_geom::geography,
st_transform(route_points.point_geom,4326)::geography,
100)) as count_of_near_points
FROM subset_cr as routes
) routes
group by 1,2
)
--pseudocode below
for each loc_id
select fifteen_min_slot
from temp
where difference in count_of_near_points is max
Code update:
I have added the following code for the pseudocode I wrote earlier:
tempy as (
select loc_id, fifteen_min_slot, count_of_near_points - lag (count_of_near_points) over (partition by loc_id, order by fifteen_min_slot) as lagging_diff
from temp
)
select loc_id, fifteen_min_slot
from tempy
where lagging_diff = (select max lagging_diff from tempy)
I have the following sample table
And the output should be the first date (for each id) when cum_rev reaches the 100 mark.
I tried the following, because I taught with group bz trick and the where condition i will only get the first occurrence of value higher than 100.
SELECT id
,pd
,cum_rev
FROM (
SELECT id
,pd
,rev
,SUM(rev) OVER (
PARTITION BY id
ORDER BY pd
) AS cum_rev
FROM tab1
)
WHERE cum_rev >= 100
GROUP BY id
But it is not working, and I get the following error. And also when I add an alias is not helping
ERROR: subquery in FROM must have an alias LINE 4: FROM (
^ HINT: For example, FROM (SELECT ...) [AS] foo.
So the desired output is:
2 2015-04-02 135.70
3 2015-07-03 102.36
Do I need another approach? Can anyone help?
Thanks
demo:db<>fiddle
SELECT
id, total
FROM (
SELECT
*,
SUM(rev) OVER (PARTITION BY id ORDER BY pd) - rev as prev_total,
SUM(rev) OVER (PARTITION BY id ORDER BY pd) as total
FROM tab1
) s
WHERE total >= 100 AND prev_total < 100
You can use the cumulative SUM() window function for each id group (partition). To find the first which goes over a threshold you need to check the previous value for being under the threshold while the current one meets it.
PS: You got the error because your subquery is missing an alias. In my example its just s
To keep this simple, let's say I have a table with 100 records that include:
userId
pointsEarned
I would like to group these 100 records (or whatever the total is based on other criteria) into several groups as follows:
Group 1, 15% of total records
Group 2, 25% of total records
Group 3, 10% of total records
Group 4, 10% of total records
Group 5, 40% (remaining of total records, percentage doesn't really matter)
In addition to the above, there will be a minimum of 3 groups and a maximum of 5 groups with varying percentages that always totally 100%. If it makes it easier, the last group will always be the remainder not picked in the other groups.
I'd like to results to be as follows:
groupNbr
userId
pointsEarned
To do this sort of breakup, you need a way to rank the records so that you can decide which group they belong in. If you do not want to randomise the group allocation, and userId is contiguous number, then using userId would be sufficient. However, you probably can't guarantee that, so you need to create some sort of ranking, then use that to split your data into groups. Here is a simple example.
Declare #Total int
Set #Total = Select COUNT(*) from dataTable
Select case
when ranking <= 0.15 * #Total then 1
when ranking <= 0.4 * #Total then 2
when ranking <= 0.5 * #Total then 3
when ranking <= 0.6 * #Total then 4
else 5 end as groupNbr,
userId,
pointsEearned
FROM (Select userId, pointsEarned, ROW_NUMBER() OVER (ORDER BY userId) as ranking From dataTable) A
If you need to randomise which group data end up in, then you need to allocate a random number to each row first, and then rank them by that random number and then split as above.
If you need to make the splits more flexible, you could design a split table that has columns like minPercentage, maxPercentage, groupNbr, fill it with the splits and do something like this
Declare #Total int
Set #Total = Select COUNT(*) from dataTable
Select S.groupNbr
B.userId,
B.pointsEearned
FROM (Select ranking / #Total * 100 as rankPercent, userId, pointsEarned
FROM (Select userId, pointsEarned, ROW_NUMBER() OVER (ORDER BY userId) as ranking From dataTable) A
) B
inner join splitTable S on S.minPercentage <= rankPercent and S.maxPercentage >= rankPercent
I am wondering if there is some easy way, a function, or other method to return data from a query with the following results.
I have a SQL Express DB 2008 R2, a table that contains numerical data in a given column, say col T.
I am given a value X in code and would like to return up to three records. The record where col T equals my value X, and the record before and after, and nothing else. The sort is done on col T. The record before may be beginning of file and therefore not exist, likewise, if X equals the last record then the record after would be non existent, end of file/table.
The value of X may not exist in the table.
This I think is similar to get a range of results in numerical order.
Any help or direction in solving this would be greatly appreciated.
Thanks again,
It might not be the most optimal solution, but:
SELECT T
FROM theTable
WHERE T = X
UNION ALL
SELECT *
FROM
(
SELECT TOP 1 T
FROM theTable
WHERE T > X
ORDER BY T
) blah
UNION ALL
SELECT *
FROM
(
SELECT TOP 1 T
FROM theTable
WHERE T < X
ORDER BY T DESC
) blah2
DECLARE #x int = 100
;WITH t as
(
select ROW_NUMBER() OVER (ORDER BY T ASC) AS row_nm,*
from YourTable
)
, t1 as
(
select *
from t
WHERE T = #x
)
select *
from t
CROSS APPLY t1
WHERE t.row_nm BETWEEN t1.row_nm -1 and t1.row_nm + 1
Is it possible to get the values of just rows 10 through 20? If so how?
If you're using SQL Server 2005 or greater, check out the ROW_NUMBER function: http://msdn.microsoft.com/en-us/library/ms186734.aspx
One way is to do something like the following...
SELECT * FROM (
SELECT TOP x * FROM (
SELECT TOP y fields
FROM table
WHERE conditions
ORDER BY table.field ASC) as foo
ORDER by field DESC) as bar
ORDER by field ASC
x is the number of rows you want returned and y is x+offset.
http://josephlindsay.com/archives/2005/05/27/paging-results-in-ms-sql-server/
Hey, by the asnwer of joelt about row_number(). I did it.
Its like this:
SELECT allianceId, position, points from (select ROW_NUMBER() over (Order by Points DESC) as position, points, allianceId from Alliance) as somethingx where position >= #alliancePosition - 5 and position <= #alliancePosition + 5;