Count number of repeated character in a given string - sql

Question

How do I count the number of occurrences of repeated $ character in the given strings.
For ex:
String = '$$$$ABC$$$DE$$$' --> Answer is 4,3,3
String = '###$$%%ANE$$$$$' --> Answer is 2,5
I have no idea how to do it so did not do any attempts.
Thanks for your help.
For Reproducing:
DDL and Inserts:
Create table xyz(text varchar(200));
Insert into xyz values('$$$$ABC$$$DE$$$');
Insert into xyz values('###$$%%ANE$$$$$');
What I need to do: Count the repeated number of '$'
Desired output, based on the sample data in #1 above.
text = '$$$$ABC$$$DE$$$' --> Answer is 4,3,3
text = '###$$%%ANE$$$$$' --> Answer is 2,5
SQL Server version: Microsoft SQL Server 2019 (RTM) - 15.0.2000.5

Answer 1

Please try the following solution. It will work starting from SQL Server 2017 onwards.
It is based on use of the TRANSLATE() function, and XML and XQuery.
SQL
-- DDL and sample data population, start
DECLARE #tbl TABLE (ID INT IDENTITY PRIMARY KEY, tokens VARCHAR(30));
INSERT INTO #tbl (tokens) VALUES
('$$$$ABC$$$DE$$$'), --> Answer is 4,3,3
('###$$%%ANE$$$$$'); --> Answer is 2,5
-- DDL and sample data population, end
DECLARE #separator CHAR(1) = SPACE(1);
;WITH cte AS
(
SELECT *
, REPLACE(TRANSLATE(tokens, '$', SPACE(1)),' ','') AS JunkCharacters
FROM #tbl
)
SELECT *
, REPLACE(TRY_CAST('<root><r><![CDATA[' +
REPLACE(TRANSLATE(tokens, TRIM(JunkCharacters), SPACE(LEN(TRIM(JunkCharacters)))), #separator, ']]></r><r><![CDATA[') +
']]></r></root>' AS XML)
.query('
for $x in /root/r[text()]
return data(string-length($x))
').value('.', 'VARCHAR(20)'), SPACE(1), ',') AS CleansedTokensCounter
FROM cte;
Output
+----+-----------------+----------------+-----------------------+
| ID | tokens | JunkCharacters | CleansedTokensCounter |
+----+-----------------+----------------+-----------------------+
| 1 | $$$$ABC$$$DE$$$ | ABCDE | 4,3,3 |
| 2 | ###$$%%ANE$$$$$ | ###%%ANE | 2,5 |
+----+-----------------+----------------+-----------------------+

Answer 2

We can do this with a number of steps:
We use a tally/numbers table to shred the string into individual characters. The tally is calculated on the fly with a couple of cross-joins and ROW_NUMBER
We then calculate a grouping ID for each group of characters, using a standard gaps-and-islands technique: a windowed sum of each starting row
Filter down to the character we want, group it by ID and return a count of rows in each group.
This returns a new row for every group of $ characters
Create table xyz(text varchar(200));
Insert into xyz values('$$$$ABC$$$DE$$$');
Insert into xyz values('###$$%%ANE$$$$$');
WITH
L0 AS ( SELECT 1 AS c
FROM (VALUES(1),(1),(1),(1),(1),(1),(1),(1),
(1),(1),(1),(1),(1),(1),(1),(1)) AS D(c) ),
L1 AS ( SELECT 1 AS c FROM L0 AS A CROSS JOIN L0 AS B ),
-- you can allow for larger strings with more cross-joins
Nums AS ( SELECT ROW_NUMBER() OVER(ORDER BY (SELECT NULL)) AS rownum
FROM L1 )
SELECT
xyz.[text],
r.numRepetitions
FROM xyz
CROSS APPLY (
SELECT numRepetitions = COUNT(*)
FROM (
SELECT TOP(LEN(xyz.[text]))
thisChar = SUBSTRING(xyz.[text], rownum, 1),
groupId = SUM(CASE WHEN rownum = 1 OR SUBSTRING(xyz.[text], rownum, 1) <> SUBSTRING(xyz.[text], rownum - 1, 1) THEN 1 ELSE 0 END)
OVER (ORDER BY rownum ROWS UNBOUNDED PRECEDING)
FROM Nums
ORDER BY rownum
) AS chars
WHERE thisChar = '$'
GROUP BY groupId
) AS r;
If you want a single comma-separated list of row-counts, you need to subquery again
CROSS APPLY (
SELECT numRepetitions = STRING_AGG(CAST(numRepetitions AS varchar(10)), ',')
FROM (
SELECT numRepetitions = COUNT(*)
FROM (
SELECT TOP(LEN(xyz.[text]))
thisChar = SUBSTRING(xyz.[text], rownum, 1),
groupId = SUM(CASE WHEN rownum = 1 OR SUBSTRING(xyz.[text], rownum, 1) <> SUBSTRING(xyz.[text], rownum - 1, 1) THEN 1 ELSE 0 END)
OVER (ORDER BY rownum ROWS UNBOUNDED PRECEDING)
FROM Nums
ORDER BY rownum
) AS chars
WHERE thisChar = '$'
GROUP BY groupId
) AS groups
) AS r;