SQL Server: How to query within a string? - sql-server

Question

I have a requirement where my client wants me to retrieve specific information from a text column
Following is the sample of the same
the student scored following result: class: 6 subject: result: english 80 math 23
science 45
The expected outcome needs to be like -
English Maths Science
80 23 45
I tried using string_split
select value from STRING_SPLIT( (select value from mytable where [student roll number] = 'SCH-01097') , ' ' )
but that only split the value into multiple rows that can't be queried.
I also tried using LTRIM with CHARINDEX approach, but the column have different text and not always organized. the initial text is different most of the time.
can this be done?
edit - I am close but just not there yet
So far I have reached here
SELECT VALUE FROM STRING_SPLIT ((select
substring(value, charindex('Block',value),1000)
from mytable where [rollnumber ] = 'SCH-01097'),' ') WHERE VALUE <> ' '
this gives me everything I need but in a single column
class6:
Subject
result
english
80
math
23
science
45
now how to make it in desired table form?

Answer 1

To maintain the order of the split values this answer uses DelimitedSplit8K. Something like this works.
[Edit] Instead of having specific strings in a CTE, the query now uses 'stems' to map multiple strings to the same class. For example, if English is entered as En it will still be mapped to English.
Table and data
drop table if exists #tTest;
go
create table #tTest(
string Varchar(256));
insert #tTest(string) values
('the student scored following result: class: 6 subject: result: english 80 math 23');
DelimitedSplit8k
CREATE FUNCTION dbo.DelimitedSplit8K
--===== Define I/O parameters
(#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
;
Query
;with
stems_cte(stem, word) as (
select 'English', 'English' union all
select 'En', 'English' union all
select 'Math', 'Math' union all
select 'Maths', 'Math' union all
select 'Science', 'Science'),
splt_cte(string, str_val, ndx, lead_ndx, lead_len, rn) as (
select t.string, ds.Item, charndx.ndx,
lead(charndx.ndx) over (order by ds.ItemNumber),
lead(len(ds.[Item])) over (order by ds.ItemNumber),
ItemNumber
from #tTest t
cross apply dbo.DelimitedSplit8K(t.string, ' ') ds
cross apply (select charindex(ds.Item, t.string, 1) ndx) charndx
where Item <> ' '),
spec_rows_cte(word, ndx, lead_ndx, lead_len, rn) as (
select sp.word, sc.ndx, sc.lead_ndx, sc.lead_len, sc.rn
from splt_cte sc
join stems_cte sp on sc.str_val=sp.stem)
select max(case when src.word='English' then substring(sc.string, src.lead_ndx, src.lead_len) else null end) English,
max(case when src.word='Math' then substring(sc.string, src.lead_ndx, src.lead_len) else null end) Math,
max(case when src.word='Science' then substring(sc.string, src.lead_ndx, src.lead_len) else null end) Science
from splt_cte sc
join spec_rows_cte src on sc.rn=src.rn;
Output
English Math Science
80 23 NULL

Answer 2

You can insert the results from your query into a table variable with an Identity column, then get the next row for each required subset:
declare #tmp table (Id int identity, Value varchar(20))
Insert into #tmp (VALUE)
SELECT VALUE FROM STRING_SPLIT ((select
substring(value, charindex('Block',value),1000)
from mytable where [rollnumber ] = 'SCH-01097'),' ') WHERE VALUE <> ' '
select
English = (Select top 1 Value From #tmp where Id = (Select Id + 1 From #tmp where Value = 'english')),
Math = (Select top 1 Value From #tmp where Id = (Select Id + 1 From #tmp where Value = 'math')),
Science = (Select top 1 Value From #tmp where Id = (Select Id + 1 From #tmp where Value = 'science'))
Output: