split comma seprate value from table in sql server

问题:

i have one SQL table in which many records, i want to know how many names are in it and how much time one name in it.

Table NameMst

Name

john,smith,alax,rock
smith,alax,sira
john,rock
rock,sira

I want to find how much name are there and count of its.

expected output should be like this

Name           Count

john           2
smith          2
alax           2
rock           3
sira           2

help me to resolved it.

回答1:

SELECT y.Name, count(*) Count
FROM
(VALUES
('john,smith,alax,rock'),
('smith,alax,sira'),
('john,rock'),
('rock,sira')) x(names)
CROSS APPLY
(
SELECT t.c.value('.', 'VARCHAR(2000)') Name
     FROM (
         SELECT x = CAST('<t>' + 
               REPLACE(x.names, ',', '</t><t>') + '</t>' AS XML)
     ) a
     CROSS APPLY x.nodes('/t') t(c)
     ) y
GROUP BY y.Name

Result:

Name   Count
alax   2
john   2
rock   3
sira   2
smith  2

回答2:

You can extract the names using a recursive CTE and some string parsing. The rest is just aggregation:

with cte as (
      select (case when names like '%,%'
                   then left(names, charindex(',', names) - 1)
                   else names
              end) as name,
             (case when names like '%,%'
                   then substring(names, charindex(',', names) + 1, len(names))
              end) as names
      from names
      union all
      select (case when names like '%,%'
                   then left(names, charindex(',', names) - 1)
                   else names
              end) as name,
             (case when names like '%,%'
                   then substring(names, charindex(',', names) + 1, len(names))
              end)
      from cte
      where names is not null
     )
select name, count(*)
from cte
group by name;

As you have probably figured out, storing comma delimited lists in SQL Server is a bad idea. You should have an association/junction table with one row per name (and other columns describing the list it is in).

回答3:

DECLARE @table1 TABLE ( id VARCHAR(50) )
DECLARE @table TABLE ( id1 INT,id VARCHAR(50) )
INSERT INTO @table (id1,id) values (1, 'JOHN,rom')
INSERT INTO @table (id1,id) values (2,'Micky,Raju')
INSERT INTO @table (id1,id) values (2,'Micky,Raju')
INSERT INTO @table (id1,id) values (2,'Micky,Raju')
DECLARE @Min INT,@Max INT ,@str1 VARCHAR(100),@str2 VARCHAR(100)

DECLARE @x INT = 0
DECLARE @firstcomma INT = 0
DECLARE @nextcomma INT = 0

SELECT @x =   LEN(id) - LEN(REPLACE(id, ',', '')) + 1 from @table -- number of ids in id_list

WHILE @x > 0
    BEGIN
        SELECT @nextcomma =  CASE WHEN CHARINDEX(',', id, @firstcomma + 1) = 0
                              THEN LEN(id) + 1
                              ELSE CHARINDEX(',', id, @firstcomma - 1)
                         END FROM @table
                         --select @nextcomma
        INSERT  INTO @table1
        SELECT  ( SUBSTRING(id, @firstcomma + 1, (@nextcomma - @firstcomma) - 1) ) FROM @table 
        SELECT @firstcomma = CHARINDEX(',', id, @firstcomma + 1)FROM @table
        SET @x = @x - 1
    END

SELECT  DISTINCT id,COUNT(id)
FROM    @table1
GROUP BY id