I'm having some difficulty with pivoting rows into columns as I also want to name the columns. Here is my current code, modified:
SELECT Message, value
FROM Table1
CROSS APPLY
(
SELECT value FROM STRING_SPLIT(Message,'"')
WHERE value LIKE '%.%'
)
AS SourceTable
And my current output:
Message value
------------ -----
longmessage1 hello
longmessage1 hi
longmessage1 hey
longmessage1 hola
Just for the sake of shortness, I replaced the actual Message with longmessage1 above. My desired output:
Message greeting1 greeting2 greeting3 greeting4
------------ --------- --------- --------- ---------
longmessage1 hello hi hey hola
The maximum amount of greetings is six, and if a Message doesn't have six, I'm fine with the value of, say greeting 4 and 5 to be NULL.
FYI- I am using SQL Server. I think I could somehow use PIVOT to do this but I'm stuck on the custom column name part and if CROSS APPLY was even the right idea. If anyone could offer some suggestions, that'd be terrific. Thank you!
You can use row_number() and conditional aggregation:
SELECT t1.Message, a.*
FROM Table1 t1 CROSS APPLY
(SELECT MAX(CASE WHEN seqnum = 1 THEN value END) as greeting1,
MAX(CASE WHEN seqnum = 2 THEN value END) as greeting2,
MAX(CASE WHEN seqnum = 3 THEN value END) as greeting3,
MAX(CASE WHEN seqnum = 4 THEN value END) as greeting4,
MAX(CASE WHEN seqnum = 5 THEN value END) as greeting5,
MAX(CASE WHEN seqnum = 6 THEN value END) as greeting6
FROM (SELECT s.value,
ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) as seqnum
FROM STRING_SPLIT(t1.Message,'"')
WHERE value LIKE '%.%'
) s
) s;
Note: In practice, this will probably preserve the ordering of the values. However that is not guaranteed -- based on the documentation.
i am trying to alias the pivot column to scenario1, scenario2, scenario3 instead of 1,2,3. I am getting error.
select *
from (select *
from (select s.campaign_id campaign_id, s.scenario_index scenario_index
from scenario s, campaign c where s.campaign_id = c.campaign_id)
pivot (max(scenario_index)for scenario_index in (1,2,3))
)a
thank you, aggregation gives the result with alias now. The requirement i have is to combine these columns with another query which is
select CASE WHEN AWARD_TYPE = 0 THEN award_rate||' points'
when AWARD_TYPE = 1 then Award_rate||' %'
when award_type=2 then RATIO_POINTS||' points per '||RATIO_MON_UNIT||' AED' End
from points_rule p
where c.pt_basic_rule_id = p.point_rule_id ) as pool_awards,
this query comes as a column and then the scenario1, 2,3 should come as 3 columns with the value of the pool_award based on the campaign_id
Just use conditional aggregation:
select s.campaign_id,
max(case when scenario_index = 1 then 1 end) as scenario1,
max(case when scenario_index = 2 then 1 end) as scenario2,
max(case when scenario_index = 3 then 1 end) as scenario3
from scenario s join
campaign c
on s.campaign_id = c.campaign_id
group by campaign_id;
You can use an alias in IN clause of the PIVOT as follows:
select *
from (select *
from (select s.campaign_id campaign_id, s.scenario_index scenario_index
from scenario s, campaign c where s.campaign_id = c.campaign_id)
pivot (max(scenario_index)for scenario_index in (1 as scenario1,2 as scenario2,3 as scenario3))
)a
I'm stuck trying to transpose a set of rows into a table. In my stored procedure, I take a delimited string as input, and need to transpose it.
SELECT *
FROM string_split('123,4,1,0,0,5|324,2,0,0,0,4','|')
CROSS APPLY string_split(value,',')
From which I receive:
value value
123,4,1,0,0,5 123
123,4,1,0,0,5 4
123,4,1,0,0,5 1
123,4,1,0,0,5 0
123,4,1,0,0,5 0
123,4,1,0,0,5 5
324,2,0,0,0,4 324
324,2,0,0,0,4 2
324,2,0,0,0,4 0
324,2,0,0,0,4 0
324,2,0,0,0,4 0
324,2,0,0,0,4 4
The values delimited by | are client details. And within each client, there are six attributes, delimited by ,. I would like an output table of:
ClientId ClientTypeId AttrA AttrB AttrC AttrD
------------------------------------------------
123 4 0 0 0 5
324 2 0 0 0 4
What's the best way to go about this? I've been looking at PIVOT but can't make it work because it seems like I need row numbers, at least.
This answer assumes that row number function will "follow the order" of the string. If it does not you will need to write your own split that includes a row number in the resulting table. (This is asked on the official documentation page but there is no official answer given).
SELECT
MAX(CASE WHEN col = 1 THEN item ELSE null END) as ClientId,
MAX(CASE WHEN col = 2 THEN item ELSE null END) as ClientTypeId,
MAX(CASE WHEN col = 3 THEN item ELSE null END) as AttrA,
MAX(CASE WHEN col = 4 THEN item ELSE null END) as AttrB,
MAX(CASE WHEN col = 5 THEN item ELSE null END) as AttrC,
MAX(CASE WHEN col = 6 THEN item ELSE null END) as AttrD
FROM (
SELECT A.value as org, B.value as item,
ROW_NUMBER() OVER (partition by A.value) as col
FROM string_split('123,4,1,0,0,5|324,2,0,0,0,4','|') as A
CROSS APPLY string_split(A.value,',') as B
) X
GROUP BY org
You might get a message about nulls in aggregate function ignored. (I always forget which platforms care and which don't.) If you do you can replace the null with 0.
Note, this is not as fast and using a CTE to find the 5 commas in the string with CHARINDEX and then using SUBSTRING to extract the values. But I'm to lazy to write up that solution which I would need to test to get all the off by 1 issues right. Still, I suggest you do it that way if you have a big data set.
I know you already got it pretty much answered, but here you can find a PIVOT solution
select [ClientID],[ClientTypeId],[AttrA],[AttrB],[AttrC],[AttrD]
FROM
(
select case when ColumnRow = 1 then 'ClientID'
when ColumnRow = 2 then 'ClientTypeId'
when ColumnRow = 3 then 'AttrA'
when ColumnRow = 4 then 'AttrB'
when ColumnRow = 5 then 'AttrC'
when ColumnRow = 6 then 'AttrD' else null end as
ColumnRow,t.value,ColumnID from (
select ColumnID,z.value as stringsplit,b.value, cast(Row_number()
over(partition by z.value order by z.value) as
varchar(50)) as ColumnRow from (SELECT cast(Row_number() over(order by
a.value) as
varchar(50)) as ColumnID,
a.value
FROM string_split('123,4,1,0,0,5|324,2,0,0,0,4','|') a
)z
CROSS APPLY string_split(value,',') b
)t
) AS SOURCETABLE
PIVOT
(
MAX(value)
FOR ColumnRow IN ([ClientID],[ClientTypeId],[AttrA],[AttrB],[AttrC],
[AttrD])
)
AS
PivotTable
Here is my table in my Microsoft Access database:
Terr(Prim) Terr(Sec) Qty
---------- --------- -------
A A 0.5
A B 0.5
I want to change it to like this through SQL:
Type Terr Qty
----------- ---- ----
Original A 0.5
Original A 0.5
50-50 give A -0.5
50-50 take B 0.5
Rules:
Create two columns, [Type] and [Terr]
if [Terr(Prim)] = [Terr (Sec)], then keep the row and [Type] = "Original"
if [Terr (Prim)] <> [Terr (Sec)], then:
Row 1: [Type] = "Original", [Terr] = [Terr(Prim)], the other columns remain the same
Row 2: [Type] = "50-50 Give", [Terr] = [Terr (Prim)], [Qty] turns into negative
Row 3: [Type] = "50-50 take", [Terr] = [Terr (Sec)], the other columns remain the same
delete [Terr(Prim)] and [Terr (Sec)]
I think you can use a query like this:
SELECT "Original" As [Type], [TerrPrim] AS [Terr], Qty
FROM t
UNION ALL
SELECT "50-50 Give" As [Type], [TerrPrim] AS [Terr], -Qty
FROM t
WHERE [TerrPrim] <> [TerrSec]
UNION ALL
SELECT "50-50 take" As [Type], [TerrSec] AS Terr, Qty
FROM t
WHERE [TerrPrim] <> [TerrSec];
I find it hard to word what I am trying to achieve. I have a table that looks like this:
user char
---------
a | x
a | y
a | z
b | x
b | x
b | y
c | y
c | y
c | z
How do I write a query that would return me the following result?
user x y z
-------
a |1|1|1|
b |2|1|0|
c |0|2|1|
the numbers represent the no of occurences of chars in the original table
EDIT:
The chars values are unknown hence the solution cannot be restricted to these values. Sorry for not mentioning it sooner. I am using Oracle DB but planning to use JPQL to construct the query.
select user,
sum(case when char='x' then 1 else 0 end) as x,
sum(case when char='y' then 1 else 0 end) as y,
sum(case when char='z' then 1 else 0 end) as z
from thetable
group by user
Or, if you don't mind stacking vertically, this solution will give you a solution that works even with unknown sets of characters:
select user, char, count(*) as count
from thetable
group by user, char
This will give you:
user char count
a x 1
a y 1
a z 1
b x 2
If you want to string an unknown set of values out horizontally (as in your demo output), you're going to need to get into dynamic queries... the SQL standard is not designed to generate output with an unknown number of columns... Hope this is helpful!
Another option, using T-SQL PIVOT (SQL SERVER 2005+)
select *
from userchar as t
pivot
(
count([char]) for [char] in ([x],[y],[z])
) as p
Result:
user x y z
----------- ----------- ----------- -----------
a 1 1 1
b 2 1 0
c 0 2 1
(3 row(s) affected)
Edit ORACLE:
You can build a similar PIVOT table using ORACLE.
The tricky part is that you need the right column names in the IN ([x],[y],[z],...) statement. It shouldn't be too hard to construct the SQL query in code, getting a (SELECT DISTINCT [char] from table) and appending it to your base query.
Pivoting rows into columns dynamically in Oracle
If you don't know the exact values on which to PIVOT, you'll either need to do something procedural or mess with dynamic sql (inside an anonymous block), or use XML (in 11g).
If you want the XML approach, it would be something like:
with x as (
select 'a' as usr, 'x' as val from dual
union all
select 'a' as usr, 'y' as val from dual
union all
select 'b' as usr, 'x' as val from dual
union all
select 'b' as usr, 'x' as val from dual
union all
select 'c' as usr, 'z' as val from dual
)
select * from x
pivot XML (count(val) as val_cnt for val in (ANY))
;
Output:
USR VAL_XML
a <PivotSet><item><column name = "VAL">x</column><column name = "VAL_CNT">1</column></item><item><column name = "VAL">y</column><column name = "VAL_CNT">1</column></item></PivotSet>
b <PivotSet><item><column name = "VAL">x</column><column name = "VAL_CNT">2</column></item></PivotSet>
c <PivotSet><item><column name = "VAL">z</column><column name = "VAL_CNT">1</column></item></PivotSet>
Hope that helps