Extract All Instances of String into Concatenated Result - sql

Question

Using SQL Server 2014, I'm wanting to search within a field and return all instances of a string that is found, plus the following word. For example, the text in the column may be:
"exec sproc1 and then some more text here and then maybe execute sproc2 exec storedproc3 and maybe exec sproc1"
I'd like to elegantly return "sproc1, sproc2, storedproc3, sproc1", as each was the word following either exec or execute (as delimited by spaces). As you can see in the example, the leading word may vary, as may the length of the sproc name. I've been able to return the first usage of exec/execute; my issue is that sometimes there are multiple (see below).
REPLACE(REPLACE(CASE
WHEN [sJSTP].[subsystem]='TSQL' AND CHARINDEX('EXECUTE',[sJSTP].[command],1)>0
THEN SUBSTRING([sJSTP].[command],CHARINDEX('EXECUTE',[sJSTP].[command],1)+8,
IIF(
CHARINDEX(' ',[sJSTP].[command],CHARINDEX('EXECUTE',[sJSTP].[command],1)+8)>0,
CHARINDEX(' ',[sJSTP].[command],CHARINDEX('EXECUTE',[sJSTP].[command],1)+8)-CHARINDEX('EXECUTE',[sJSTP].[command],1)-8,
LEN([sJSTP].[command])))
WHEN [sJSTP].[subsystem]='TSQL' AND CHARINDEX('EXEC',[sJSTP].[command],1)>0 AND CHARINDEX('DCEXEC',[sJSTP].[command],1)<=0
THEN SUBSTRING([sJSTP].[command],CHARINDEX('EXEC',[sJSTP].[command],1)+5,
IIF(
CHARINDEX(' ',[sJSTP].[command],CHARINDEX('EXEC',[sJSTP].[command],1)+5)>0,
CHARINDEX(' ',[sJSTP].[command],CHARINDEX('EXEC',[sJSTP].[command],1)+5)-CHARINDEX('EXEC',[sJSTP].[command],1)-5,
LEN([sJSTP].[command])))
END,'[',''),']','') AS sprocname
The ultimate use of this is parsing job commands from the msdb..sysjobsteps table to see what stored procedures are being used.
Edit: Add sample data
Sample 1:
exec quarterly_run 1, 'BW'
exec quarterly_run_2 1, 'QR '
exec quarterly_run 2, 'VAS'
exec quarterly_run 1, 'WR'
exec quarterly_run 3, 'RW'
exec quarterly_run_2 1, 'ASF'
exec quarterly_run_3 1, 'ALL'
Sample 2:
declare #rundate datetime, #rptqtr datetime, #qtr int
set #rundate = getdate()
set #rptqtr = '06/30/2016'
set #qtr = (select datediff(quarter,#rptqtr,#rundate))
exec quarterly_extract #qtr
Sample 3:
exec Daily_Sync_Process
exec Daily_Process

Answer 1

Just another inline option, and not limited to 8K
Example
Declare #YourTable table (ID int,SomeCol varchar(max))
Insert into #YourTable values
(1,'exec quarterly_run 1, ''BW'' exec quarterly_run_2 1, ''QR '' exec quarterly_run 2, ''VAS'' exec quarterly_run 1, ''WR'' exec quarterly_run 3, ''RW'' exec quarterly_run_2 1, ''ASF'' exec quarterly_run_3 1, ''ALL''')
,(2,'declare #rundate datetime, #rptqtr datetime, #qtr int
set #rundate = getdate() set #rptqtr = ''06/30/2016''
set #qtr = (select datediff(quarter,#rptqtr,#rundate))
exec quarterly_extract #qtr
')
,(3,'exec Daily_Sync_Process exec Daily_Process')
;with cte as (
Select A.ID
,C.*
From #YourTable A
Cross Apply (values (replace(replace(SomeCol,char(13),' '),char(10),' '))) B(CleanString)
Cross Apply (
Select RetSeq,RetVal = case when Lag(RetVal,1) over (Order by RetSeq) in ('Exec','Execute') then RetVal else null end
From (
Select RetSeq = Row_Number() over (Order By (Select null))
,RetVal = LTrim(RTrim(B.i.value('(./text())[1]', 'varchar(max)')))
From (Select x = Cast('<x>' + replace((Select replace(CleanString,' ','§§Split§§') as [*] For XML Path('')),'§§Split§§','</x><x>')+'</x>' as xml).query('.')) as A
Cross Apply x.nodes('x') AS B(i)
) C1
) C
)
Select A.ID
,NewString = Stuff((Select ', ' +RetVal From cte Where ID=A.ID Order By RetSeq For XML Path ('')),1,2,'')
From cte A
Group By A.ID
Returns
ID NewString
1 quarterly_run, quarterly_run_2, quarterly_run, quarterly_run, quarterly_run, quarterly_run_2, quarterly_run_3
2 quarterly_extract
3 Daily_Sync_Process, Daily_Process

Answer 2

So if you want to get what is immediately following exec then I'd split on the space, and then use a self join. Here is code using the function below, which is Jeff Moden's splitter.
with cte as(
select
job_id
,step_name
,step_id
,s.ItemNumber
,s.Item
from msdb..sysjobsteps
--split on the space
cross apply dbo.DelimitedSplit8K(command,' ') s)
select
c.job_id
,c.step_id
,c.step_name
,c.Item
,c2.Item
from cte c
--self join to get exec myproc in the same row
full join
cte c2 on
c2.ItemNumber = c.ItemNumber + 1
and c.job_id = c2.job_id
and c.step_id = c2.step_id
--we only care where the base table has exec or execute (not executed, etc)
where c.Item = 'exec' or c.Item = 'execute'
order by
c.job_id, c.step_id, c.ItemNumber
It's important to realize this would fail when, for example, the command was exec someproc which has two spaces. You can fix that with a replace() but you'd have to nest this replace multiple times to account for as many spaces as you want. You'd handle that on the command column in the splitter function
--here we replace two spaces with 1 for the entire command
cross apply dbo.DelimitedSplit8K(replace(command,' ',' '),' ') s)
SPLITER FUNCTION
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE FUNCTION [dbo].[DelimitedSplit8K] (#pString VARCHAR(8000), #pDelimiter CHAR(1))
--WARNING!!! DO NOT USE MAX DATA-TYPES HERE! IT WILL KILL PERFORMANCE!
RETURNS TABLE WITH SCHEMABINDING AS
RETURN
/* "Inline" CTE Driven "Tally Table" produces values from 1 up to 10,000... enough to cover VARCHAR(8000)*/
WITH E1(N) AS (
SELECT 1 UNION ALL SELECT 1 UNION ALL SELECT 1 UNION ALL
SELECT 1 UNION ALL SELECT 1 UNION ALL SELECT 1 UNION ALL
SELECT 1 UNION ALL SELECT 1 UNION ALL SELECT 1 UNION ALL SELECT 1
), --10E+1 or 10 rows
E2(N) AS (SELECT 1 FROM E1 a, E1 b), --10E+2 or 100 rows
E4(N) AS (SELECT 1 FROM E2 a, E2 b), --10E+4 or 10,000 rows max
cteTally(N) AS (--==== This provides the "base" CTE and limits the number of rows right up front
-- for both a performance gain and prevention of accidental "overruns"
SELECT TOP (ISNULL(DATALENGTH(#pString),0)) ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) FROM E4
),
cteStart(N1) AS (--==== This returns N+1 (starting position of each "element" just once for each delimiter)
SELECT 1 UNION ALL
SELECT t.N+1 FROM cteTally t WHERE SUBSTRING(#pString,t.N,1) = #pDelimiter
),
cteLen(N1,L1) AS(--==== Return start and length (for use in substring)
SELECT s.N1,
ISNULL(NULLIF(CHARINDEX(#pDelimiter,#pString,s.N1),0)-s.N1,8000)
FROM cteStart s
)
--===== Do the actual split. The ISNULL/NULLIF combo handles the length for the final element when no delimiter is found.
SELECT ItemNumber = ROW_NUMBER() OVER(ORDER BY l.N1),
Item = SUBSTRING(#pString, l.N1, l.L1)
FROM cteLen l
GO