计算给定字符串中重复字符的数量

Question

如何计算给定字符串中重复 $ 字符的出现次数。

例如：

1. String = '$$$$ABC$$$DE$$$' --> 答案是 4,3,3
2. String = '###$$%%ANE$$$$$' --> 答案是 2,5

我不知道怎么做所以没有做任何尝试。

感谢您的帮助。

用于复制：

1. DDL 和插入：

Create table xyz(text varchar(200));
Insert into xyz values('$$$$ABC$$$DE$$$');
Insert into xyz values('###$$%%ANE$$$$$');

1. 我需要做的：统计'$'的重复次数

2. 所需的输出，基于上面 #1 中的示例数据。

   text = '$$$$ABC$$$DE$$$' --> 答案是 4,3,3
   text = '###$$%%ANE$$$$$' --> 答案是 2,5

3. SQL Server 版本：Microsoft SQL Server 2019 (RTM) - 15.0.2000.5

Answer 1

请尝试以下解决方案。它将从 SQL Server 2017 开始工作。

它基于 TRANSLATE() 函数以及 XML 和 XQuery 的使用。

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;

输出

+----+-----------------+----------------+-----------------------+
| ID |     tokens      | JunkCharacters | CleansedTokensCounter |
+----+-----------------+----------------+-----------------------+
|  1 | $$$$ABC$$$DE$$$ | ABCDE          |                 4,3,3 |
|  2 | ###$$%%ANE$$$$$ | ###%%ANE       |                   2,5 |
+----+-----------------+----------------+-----------------------+

Answer 2

我们可以通过以下几个步骤来做到这一点：

• 我们使用计数/数字表将字符串分解为单个字符。计数是通过几个交叉连接和ROW_NUMBER 即时计算的
• 然后我们为每组字符计算一个分组 ID，使用标准的间隙和岛技术：每个起始行的窗口总和
• 过滤到我们想要的字符，按 ID 分组并返回每组中的行数。

这将为每组 $ 字符返回一个新行

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;

如果你想要一个逗号分隔的行计数列表，你需要再次子查询

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;