Reputation: 773
I've tried so hard to understand how to create a pivot table in SQL, but I can't manage it!
I have the following columns:
link_id route_section date_1 StartHour AvJT data_source
....... ............. ....... ........... ...... ............
With 600,000 rows of data.
I need them in the following pivot table;
date_1
StartHour as column headingslink_id
as the row headingAvJT
as the datadata_source
= '1' as the filter.PIVOT TABLE
Link_ID
date_1 StartHour 00001a 000002a 000003a 000004a
20/01/2014 8 456 4657 556 46576
21/01/2014 8 511 4725 601 52154
22/01/2014 8 468 4587 458 47585
23/01/2014 8 456 4657 556 46576
24/01/2014 8 456 4657 556 46576
25/01/2014 8 456 4657 556 46576
26/01/2014 8 456 4657 556 46576
I've managed to get the following code, this works but only gives me date_1 as column heading and not StartHour additionally, or with the filter as date_source = '1'.
Use [C1_20132014]
DECLARE @DynamicPivotQuery AS NVARCHAR(MAX)
DECLARE @ColumnName AS NVARCHAR(MAX)
--Get distinct values of the PIVOT Column
SELECT @ColumnName= ISNULL(@ColumnName + ',','')
+ QUOTENAME(Link_ID)
FROM (SELECT DISTINCT Link_ID FROM C1_May_Routes) AS Link_ID
--Prepare the PIVOT query using the dynamic
SET @DynamicPivotQuery =
N'SELECT Date_1, ' + @ColumnName + '
FROM C1_May_Routes
PIVOT(SUM(AvJT)
FOR Link_ID IN (' + @ColumnName + ')) AS PVTTable'
--Execute the Dynamic Pivot Query
EXEC sp_executesql @DynamicPivotQuery
Thanks for any help,
Henry
Upvotes: 0
Views: 1307
Reputation: 21341
Here you will select the values in a column to show as column in pivot
DECLARE @cols NVARCHAR (MAX)
SELECT @cols = COALESCE (@cols + ',[' + AvJT + ']', '[' + AvJT + ']')
FROM (SELECT DISTINCT AvJT FROM YourTable) PV
ORDER BY AvJT
Now pivot the query
DECLARE @query NVARCHAR(MAX)
SET @query = 'SELECT * FROM
(
SELECT date_1, StartHour,AvJT, data_source
FROM YourTable
) x
PIVOT
(
-- Values in each dynamic column
SUM(data_source)
FOR AvJT IN (' + @cols + ')
) p;'
EXEC SP_EXECUTESQL @query
If you want to do it to where column names are not dynamic, you can do the below query
SELECT DATE_1,STARTHOUR,
MIN(CASE WHEN AvJT='00001a' THEN data_source END) [00001a],
MIN(CASE WHEN AvJT='00002a' THEN data_source END) [00002a],
MIN(CASE WHEN AvJT='00003a' THEN data_source END) [00003a],
MIN(CASE WHEN AvJT='00004a' THEN data_source END) [00004a]
FROM YOURTABLE
GROUP BY DATE_1,STARTHOUR
EDIT :
I am updating for your updated question.
Declare a variable for filtering data_source
DECLARE @DATASOURCE VARCHAR(20) = '1'
Instead of QUOTENAME
, you can use another format to get the columns for pivot
DECLARE @cols NVARCHAR (MAX)
SELECT @cols = COALESCE (@cols + ',[' + Link_ID + ']', '[' + Link_ID + ']')
FROM (SELECT DISTINCT Link_ID FROM C1_May_Routes WHERE data_source=@DATASOURCE) PV
ORDER BY Link_ID
Now pivot
DECLARE @query NVARCHAR(MAX)
SET @query = 'SELECT * FROM
(
-- We will select the data that has to be shown for pivoting
-- with filtered data_source
SELECT date_1, StartHour,AvJT, Link_ID
FROM C1_May_Routes
WHERE data_source = '+@DATASOURCE+'
) x
PIVOT
(
-- Values in each dynamic column
SUM(AvJT)
-- Select columns from @cols
FOR Link_ID IN (' + @cols + ')
) p;'
EXEC SP_EXECUTESQL @query
Upvotes: 1
Reputation: 33581
A crosstab would be something like this. FWIW, I would recommend using better column names than 00001a. Give your column names some meaning so they are easier to work with.
with SortedData as
(
SELECT date_1
, StartHour
, AvJT
, data_source
, ROW_NUMBER() over (partition by date_1 order by AvJT) AS RowNum
FROM YourTable
)
select date_1
, StartHour
, MAX(case when RowNum = 1 then AvJT end) as [00001a]
, MAX(case when RowNum = 2 then AvJT end) as [00002a]
, MAX(case when RowNum = 3 then AvJT end) as [00003a]
, MAX(case when RowNum = 4 then AvJT end) as [00004a]
from SortedData
Upvotes: 0