【Transact-SQL】统计某字段中的值第一次出现后的2小时内出现的次数

时间:2022-06-26 07:29:16

原文: 【Transact-SQL】统计某字段中的值第一次出现后的2小时内出现的次数

 

table1
name               createdate
a              2011-03-01 10:00:00
a              2011-03-01 11:00:00
a              2011-03-01 14:00:00
b              2011-03-01 13:00:00
b              2011-03-01 13:20:00
b              2011-03-01 14:00:00
 
查询结果为
name             createdate                    count
a             2011-03-01 10:00:00               2
a             2011-03-01 14:00:00               1
b             2011-03-01 13:00:00               3

就相当于是统计name字段中的值在第一次出现后的2小时内,总共出现了几次?

 

这个是网上的解答:


   
  
  1. declare @table1 table( name nvarchar,createdate smalldatetime)
  2. insert into @table1
  3. select ‘a‘, ‘2011-03-01 10:00:00‘
  4. union all select ‘a‘, ‘2011-03-01 11:00:00‘
  5. union all select ‘a‘, ‘2011-03-01 14:00:00‘
  6. union all select ‘b‘, ‘2011-03-01 13:00:00‘
  7. union all select ‘b‘, ‘2011-03-01 13:20:00‘
  8. union all select ‘b‘, ‘2011-03-01 14:00:00‘
  9. select name,
  10. createdate,
  11. ( select count(createdate)
  12. from @table1 b
  13. where a.name=b.name and
  14. a.createdate<=b.createdate and
  15. dateadd(hh, 2,a.createdate) >= b.createdate
  16. ) as count
  17. from @table1 a
  18. where not exists
  19. ( select 1 from
  20. @table1 b
  21. where a.name=b.name and
  22. a.createdate>b.createdate and
  23. a.createdate< dateadd(hh, 2,b.createdate))
  24. group by name,createdate


但是这个解答其实是有问题的,当把临时表中的第3条数据的createdate改为‘2011-03-01 12:00:00‘,那么显示的结果是:

name        createdate                           count
a                2011-03-01 10:00:00        3
b                2011-03-01 13:00:00        3

在其中没有包括createdate为‘2011-03-01 12:00:00‘的记录,因为这个时间到为‘2011-03-01 10:00:00‘是超过2个小时了,也就是说为‘2011-03-01 10:00:00‘是第一个出现时间,到为‘2011-03-01 11:59:59‘为止,接下来应该是从‘2011-03-01 12:00:00‘开始的下个区间了,而这里显然是有问题的。

以下是我写的解法,虽然效率不是太高,但是能解决这个问题:


   
  
  1. declare @table1 table( name nvarchar,createdate smalldatetime)
  2. insert into @table1
  3. select ‘a‘, ‘2011-03-01 10:00:00‘
  4. union all select ‘a‘, ‘2011-03-01 11:00:00‘
  5. union all select ‘a‘, ‘2011-03-01 12:00:00‘
  6. union all select ‘b‘, ‘2011-03-01 13:10:00‘
  7. union all select ‘b‘, ‘2011-03-01 13:20:00‘
  8. union all select ‘b‘, ‘2011-03-01 14:30:00‘
  9. union all select ‘b‘, ‘2011-03-01 15:15:00‘
  10. union all select ‘b‘, ‘2011-03-01 16:00:00‘
  11. union all select ‘b‘, ‘2011-03-01 17:00:00‘
  12. ;with aa --按照name分区,同时按照createdate排序编号
  13. as
  14. (
  15. select name,
  16. createdate,
  17. ROW_NUMBER() over( partition by name
  18. order by createdate) as k1
  19. from @table1
  20. ),
  21. r
  22. as
  23. (
  24. select v.name,
  25. starts=v.k1, --区间开始的编号
  26. ends= isnull(
  27. min( case when v.k1<a.k1
  28. and DATEADD( hour, 2,v.createdate) <= a.createdate
  29. then a.k1
  30. else null
  31. end) -1,
  32. max( case when v.k1<a.k1
  33. then a.k1
  34. else v.k1
  35. end)
  36. ), --区间结尾的编号
  37. isnull(
  38. min( case when v.k1<a.k1
  39. and DATEADD( hour, 2,v.createdate) <= a.createdate
  40. then a.k1
  41. else null
  42. end) -1,
  43. max( case when v.k1<a.k1
  44. then a.k1
  45. else v.k1
  46. end)
  47. ) - v.k1 as diff --区间结尾编号与区间开始编号之间的差值
  48. from aa v
  49. inner join aa a
  50. on v.name = a.name --只关联name相等的
  51. group by v.name,
  52. v.k1
  53. having isnull(
  54. min( case when v.k1<a.k1
  55. and DATEADD( hour, 2,v.createdate) <= a.createdate
  56. then a.k1
  57. else null
  58. end) -1,
  59. max( case when v.k1<a.k1
  60. then a.k1
  61. else v.k1
  62. end)
  63. ) >=v.k1
  64. and
  65. isnull(
  66. max( case when v.k1>a.k1 and
  67. DATEADD( hour, -2,v.createdate) >= a.createdate
  68. then v.k1 - 1
  69. else null
  70. end) 1,
  71. min( case when v.k1>a.k1
  72. then a.k1
  73. else v.k1
  74. end)
  75. ) = v.k1
  76. )
  77. --select * from r
  78. select aa.name,
  79. aa.createdate,
  80. diff 1
  81. from r
  82. inner join aa
  83. on aa.name = r.name
  84. and aa.k1 =r.starts
  85. where not exists
  86. ( select 1
  87. from r rr
  88. where rr.name = r.name and
  89. rr.starts <> r.starts and
  90. rr.starts < r.starts and
  91. (rr.ends = r.ends or
  92. rr.ends = r.starts)
  93. )

 不过发现我的这个解法也是有问题的,最大的问题在与不能准确的确定上限在那里,还得继续考虑问题的解法。

下面这个也是有问题的:


   
  
  1. declare @table1 table( name nvarchar,createdate smalldatetime)
  2. insert into @table1
  3. select ‘a‘, ‘2011-03-01 10:00:00‘
  4. union all select ‘a‘, ‘2011-03-01 11:00:00‘
  5. union all select ‘a‘, ‘2011-03-01 12:00:00‘
  6. --union all select ‘b‘,‘2011-03-01 13:10:00‘
  7. --union all select ‘b‘,‘2011-03-01 13:20:00‘
  8. --union all select ‘b‘,‘2011-03-01 14:30:00‘
  9. --union all select ‘b‘,‘2011-03-01 15:15:00‘
  10. --union all select ‘b‘,‘2011-03-01 16:00:00‘
  11. --union all select ‘b‘,‘2011-03-01 17:15:00‘
  12. ;with a --按照name分区,同时按照createdate排序编号
  13. as
  14. (
  15. select name,
  16. createdate,
  17. ROW_NUMBER() over( partition by name
  18. order by createdate) as k1
  19. from @table1
  20. ),
  21. c
  22. as
  23. (
  24. select a1.name,
  25. a1.createdate,
  26. a1.k1,
  27. MIN(a2.createdate) as nextCreatedate,
  28. MIN(a2.k1) as nextK1
  29. from a a1
  30. inner join a a2
  31. on a1.name = a2.name
  32. and a2.createdate < DATEADD( hour, 2,a1.createdate)
  33. and a2.createdate >= a1.createdate
  34. and a1.k1 <= a2.k1
  35. group by a1.name,
  36. a1.createdate,
  37. a1.k1
  38. ),
  39. w
  40. as
  41. (
  42. select name,
  43. createdate,
  44. k1
  45. --null,
  46. --null,
  47. --null
  48. from a
  49. where k1 = 1
  50. union all
  51. select c.name,
  52. c.nextCreatedate,
  53. c.nextK1
  54. from W
  55. inner join a
  56. on a.name = w.name
  57. and dateadd( hour, 2,w.createdate) <= a.createdate
  58. and w.k1 <= a.k1
  59. and w.createdate <> ‘2011-03-01 12:00:00‘
  60. inner join c
  61. on w.name = c.name
  62. and w.createdate = c.createdate
  63. and w.k1 = c.k1
  64. where w.k1 <= 3
  65. )
  66. SELECT *
  67. FROM w


下面的解法是正确的,不过用的是T-SQL,不是纯sql了:


   
  
  1. declare @table1 table( name nvarchar( 100),createdate smalldatetime)
  2. declare @table2 table( name nvarchar( 100),createdate smalldatetime,rnum bigint)
  3. declare @temp table( name nvarchar( 100),createdate smalldatetime,rnum bigint)
  4. declare @i int = 1;
  5. insert into @table1
  6. select ‘a‘, ‘2011-03-01 10:00:00‘
  7. union all select ‘a‘, ‘2011-03-01 11:00:00‘
  8. union all select ‘a‘, ‘2011-03-01 12:00:00‘
  9. union all select ‘b‘, ‘2011-03-01 13:10:00‘
  10. union all select ‘b‘, ‘2011-03-01 13:20:00‘
  11. union all select ‘b‘, ‘2011-03-01 14:30:00‘
  12. union all select ‘b‘, ‘2011-03-01 15:15:00‘
  13. union all select ‘b‘, ‘2011-03-01 16:00:00‘
  14. union all select ‘b‘, ‘2011-03-01 17:16:00‘
  15. union all select ‘b‘, ‘2011-03-01 17:15:00‘
  16. ;with a --按照name分区,同时按照createdate排序编号
  17. as
  18. (
  19. select name,
  20. createdate,
  21. ROW_NUMBER() over( partition by name
  22. order by createdate) as k1
  23. from @table1
  24. )
  25. insert into @table2
  26. select * from a
  27. insert into @temp
  28. select name,
  29. createdate,
  30. rnum
  31. from @table2
  32. where rnum = 1
  33. --select * from @temp
  34. while @i <= ( select MAX(rnum) from @table2)
  35. begin
  36. insert into @temp
  37. select t2.name,
  38. min(t2.createdate),
  39. @i 1
  40. from @temp t1
  41. inner join @table2 t2
  42. on t1.name = t2.name
  43. and t2.createdate >= dateadd( hour, 2,t1.createdate)
  44. where t1.rnum = @i
  45. group by t2.name
  46. set @i = @i 1
  47. end
  48. ;with r
  49. as
  50. (
  51. select name,
  52. createdate
  53. from @temp
  54. group by name,
  55. createdate
  56. )
  57. select r.name,
  58. r.createdate,
  59. COUNT( 1)
  60. from r
  61. inner join @table1 t
  62. on t.name = r.name
  63. and t.createdate >= r.createdate
  64. and t.createdate < DATEADD( HOUR, 2,r.createdate)
  65. group by r.name,
  66. r.createdate

 

其实这个问题是个递归问题,由上一个找到下一个,但是得构造一下:


   
  
  1. declare @table1 table( name nvarchar,createdate smalldatetime)
  2. insert into @table1
  3. select ‘a‘, ‘2011-03-01 10:00:00‘
  4. union all select ‘a‘, ‘2011-03-01 11:00:00‘
  5. union all select ‘a‘, ‘2011-03-01 12:00:00‘
  6. union all select ‘a‘, ‘2011-03-01 12:20:00‘
  7. union all select ‘b‘, ‘2011-03-01 13:10:00‘
  8. union all select ‘b‘, ‘2011-03-01 13:20:00‘
  9. union all select ‘b‘, ‘2011-03-01 14:30:00‘
  10. union all select ‘b‘, ‘2011-03-01 15:15:00‘
  11. union all select ‘b‘, ‘2011-03-01 16:00:00‘
  12. union all select ‘b‘, ‘2011-03-01 17:20:00‘
  13. union all select ‘b‘, ‘2011-03-01 17:15:00‘
  14. union all select ‘b‘, ‘2011-03-01 19:16:00‘
  15. union all select ‘b‘, ‘2011-03-01 17:15:00‘
  16. ;with a --按照name分区,同时按照createdate排序编号
  17. as
  18. (
  19. select name,
  20. createdate,
  21. ROW_NUMBER() over( partition by name
  22. order by createdate) as k1
  23. from @table1
  24. ),
  25. c --对于每个时间,找到大于这个时间2小时的时间中最小那个时间
  26. as
  27. (
  28. select a1.name,
  29. a1.createdate,
  30. a1.k1,
  31. MIN(a2.createdate) as nextCreatedate,
  32. MIN(a2.k1) as nextK1
  33. from a a1
  34. inner join a a2
  35. on a1.name = a2.name
  36. and a2.createdate >= DATEADD( hour, 2,a1.createdate)
  37. group by a1.name,
  38. a1.createdate,
  39. a1.k1
  40. union all
  41. select a.name, null, null,a.createdate, 1 --构造递归运行时需要的层级
  42. from a
  43. where k1 = 1
  44. ),
  45. w --递归查询
  46. as
  47. (
  48. select c.name,
  49. c.createdate,
  50. c.k1,
  51. c.nextCreatedate,
  52. c.nextK1,
  53. 1 as lev
  54. from c
  55. where createdate is null
  56. and k1 is null
  57. union all
  58. select c.name,
  59. c.createdate,
  60. c.k1,
  61. c.nextCreatedate,
  62. c.nextK1,
  63. lev 1
  64. from W
  65. inner join c
  66. on w.name = c.name
  67. and w.nextCreatedate = c.createdate
  68. )
  69. SELECT distinct name,
  70. nextCreatedate,
  71. nextK1
  72. FROM w



 

 

 

【Transact-SQL】统计某字段中的值第一次出现后的2小时内出现的次数【Transact-SQL】统计某字段中的值第一次出现后的2小时内出现的次数 不想长大啊 发布了416 篇原创文章 · 获赞 135 · 访问量 94万 他的留言板 关注