MsSql 自定义分数段，按分数段统计考试人次

--分数段分布 
DECLARE @levels VARCHAR(100) = '10,20,30,40,50,60,70,80,90,100'; --自定义分数段
DECLARE @paperId VARCHAR(100)='0000000000001019'--试卷编号
WITH tbTemp AS (--处理分数段
         SELECT L.level
         FROM   (
                    SELECT [value] = CONVERT(XML, '<v>' + REPLACE(@levels, ',', '</v><v>') + '</v>')
                ) T
                OUTER APPLY (
             SELECT LEVEL = CONVERT(INT, N.v.value('.', 'varchar(100)'))
             FROM   T.[value].nodes('/v') N(v)
         ) L
     ), 
     tbLevel AS (--生成分数段区间
         SELECT (
                    SELECT ISNULL(MAX(t.level), 0)
                    FROM   tbTemp AS t
                    WHERE  t.LEVEL < tt.level
                )         AS level0,
                tt.level  AS level1
         FROM   tbTemp    AS tt
     ),
     tbScore AS(--处理成绩为null或0分，为了分配到0-？区间
         SELECT (CASE WHEN tp.GetScore > 0 THEN tp.GetScore ELSE 0.1 END) AS 
                pScore
         FROM   成绩表 AS tp
         WHERE  tp.PaperID = @paperId
     )
--统计区间人次
SELECT tl.level0,
       tl.level1,
       (
           SELECT COUNT(0)
           FROM   tbScore AS ts
           WHERE  ts.pScore > level0
                  AND ts.pScore <= level1
       )        AS ExamCount
FROM   tbLevel  AS tl;

levels = '10,20,30,40,50,60,70,80,90,100'; 

levels = '40,60,80,100';