Google表格 - 查询以显示给定列表中最早日期的日期和价值
我试图每个月的第一天才能获得价值。
02/06/2000 15:30:00 5.12
05/06/2000 15:30:00 5.2
06/06/2000 15:30:00 5.28
07/06/2000 15:30:00 5.26
08/06/2000 15:30:00 5.33
09/06/2000 15:30:00 5.36
12/06/2000 15:30:00 5.24
13/06/2000 15:30:00 5.33
14/06/2000 15:30:00 5.39
15/06/2000 15:30:00 5.3
16/06/2000 15:30:00 5.29
19/06/2000 15:30:00 5.27
20/06/2000 15:30:00 5.26
21/06/2000 15:30:00 5.33
22/06/2000 15:30:00 5.25
23/06/2000 15:30:00 5.15
26/06/2000 15:30:00 5.21
27/06/2000 15:30:00 5.15
28/06/2000 15:30:00 5.09
29/06/2000 15:30:00 5.11
30/06/2000 15:30:00 4.82
03/07/2000 15:30:00 4.87
04/07/2000 15:30:00 4.92
05/07/2000 15:30:00 4.87
06/07/2000 15:30:00 4.93
07/07/2000 15:30:00 4.85
10/07/2000 15:30:00 5.2
11/07/2000 15:30:00 5.43
12/07/2000 15:30:00 5.54
13/07/2000 15:30:00 5.29
14/07/2000 15:30:00 5.17
17/07/2000 15:30:00 5.29
18/07/2000 15:30:00 5.07
19/07/2000 15:30:00 5.05
20/07/2000 15:30:00 4.92
21/07/2000 15:30:00 5.04
24/07/2000 15:30:00 4.77
25/07/2000 15:30:00 4.91
26/07/2000 15:30:00 4.62
27/07/2000 15:30:00 4.8
28/07/2000 15:30:00 4.59
31/07/2000 15:30:00 4.47
01/08/2000 15:30:00 4.57
02/08/2000 15:30:00 4.51
03/08/2000 15:30:00 4.57
04/08/2000 15:30:00 4.41
07/08/2000 15:30:00 4.3
08/08/2000 15:30:00 4.35
09/08/2000 15:30:00 4.35
10/08/2000 15:30:00 4.33
11/08/2000 15:30:00 4.3
14/08/2000 15:30:00 4.26
16/08/2000 15:30:00 4.26
17/08/2000 15:30:00 4.25
18/08/2000 15:30:00 4.28
21/08/2000 15:30:00 4.36
22/08/2000 15:30:00 4.39
23/08/2000 15:30:00 4.44
24/08/2000 15:30:00 4.45
25/08/2000 15:30:00 4.52
28/08/2000 15:30:00 4.68
29/08/2000 15:30:00 4.71
30/08/2000 15:30:00 4.68
31/08/2000 15:30:00 4.79
04/09/2000 15:30:00 4.8
05/09/2000 15:30:00 4.71
06/09/2000 15:30:00 4.72
07/09/2000 15:30:00 4.73
08/09/2000 15:30:00 4.79
11/09/2000 15:30:00 4.87
12/09/2000 15:30:00 4.83
13/09/2000 15:30:00 4.73
14/09/2000 15:30:00 4.77
15/09/2000 15:30:00 5.13
18/09/2000 15:30:00 5.1
19/09/2000 15:30:00 5.1
20/09/2000 15:30:00 5.13
21/09/2000 15:30:00 5.12
22/09/2000 15:30:00 5.46
25/09/2000 15:30:00 5.57
26/09/2000 15:30:00 5.32
27/09/2000 15:30:00 5.32
28/09/2000 15:30:00 5.48
29/09/2000 15:30:00 5.9
03/10/2000 15:30:00 6.08
04/10/2000 15:30:00 6.07
05/10/2000 15:30:00 6.3
06/10/2000 15:30:00 6.52
09/10/2000 15:30:00 6.39
10/10/2000 15:30:00 6.1
11/10/2000 15:30:00 6.06
12/10/2000 15:30:00 6.12
13/10/2000 15:30:00 6.47
16/10/2000 15:30:00 6.42
17/10/2000 15:30:00 6.23
18/10/2000 15:30:00 6.2
19/10/2000 15:30:00 6.68
20/10/2000 15:30:00 7.18
23/10/2000 15:30:00 6.86
24/10/2000 15:30:00 6.89
25/10/2000 15:30:00 7.29
26/10/2000 15:30:00 7.42
27/10/2000 15:30:00 7.33
30/10/2000 15:30:00 7.29
31/10/2000 15:30:00 7.33
01/11/2000 15:30:00 7.48
02/11/2000 15:30:00 7.57
03/11/2000 15:30:00 7.9
06/11/2000 15:30:00 8.11
07/11/2000 15:30:00 8.36
08/11/2000 15:30:00 8.31
09/11/2000 15:30:00 8.49
10/11/2000 15:30:00 8.58
13/11/2000 15:30:00 8.67
14/11/2000 15:30:00 8.65
15/11/2000 15:30:00 8.76
16/11/2000 15:30:00 8.77
17/11/2000 15:30:00 8.91
20/11/2000 15:30:00 9.03
21/11/2000 15:30:00 9.21
22/11/2000 15:30:00 9.19
23/11/2000 15:30:00 9.35
24/11/2000 15:30:00 9.59
27/11/2000 15:30:00 9.79
28/11/2000 15:30:00 9.98
29/11/2000 15:30:00 10.07
30/11/2000 15:30:00 10.15
01/12/2000 15:30:00 10.19
04/12/2000 15:30:00 10.4
05/12/2000 15:30:00 10.35
06/12/2000 15:30:00 10.95
07/12/2000 15:30:00 11.06
08/12/2000 15:30:00 11.32
11/12/2000 15:30:00 11.55
12/12/2000 15:30:00 11.43
13/12/2000 15:30:00 11.72
14/12/2000 15:30:00 11.71
15/12/2000 15:30:00 11.81
18/12/2000 15:30:00 12.07
19/12/2000 15:30:00 12.09
20/12/2000 15:30:00 12.28
21/12/2000 15:30:00 12.28
我希望查询
02/06/2000 5.12
03/07/2000 4.87
01/08/2000 4.57
04/09/2000 4.8
03/10/2000 6.08
01/11/2000 7.48
01/12/2000 10.19
目前显示,我在不同的列中进行多个分类。任何简化的查询建议都将是可观的。
提前致谢。
此致。
I am trying to fetch value first day of every month.
02/06/2000 15:30:00 5.12
05/06/2000 15:30:00 5.2
06/06/2000 15:30:00 5.28
07/06/2000 15:30:00 5.26
08/06/2000 15:30:00 5.33
09/06/2000 15:30:00 5.36
12/06/2000 15:30:00 5.24
13/06/2000 15:30:00 5.33
14/06/2000 15:30:00 5.39
15/06/2000 15:30:00 5.3
16/06/2000 15:30:00 5.29
19/06/2000 15:30:00 5.27
20/06/2000 15:30:00 5.26
21/06/2000 15:30:00 5.33
22/06/2000 15:30:00 5.25
23/06/2000 15:30:00 5.15
26/06/2000 15:30:00 5.21
27/06/2000 15:30:00 5.15
28/06/2000 15:30:00 5.09
29/06/2000 15:30:00 5.11
30/06/2000 15:30:00 4.82
03/07/2000 15:30:00 4.87
04/07/2000 15:30:00 4.92
05/07/2000 15:30:00 4.87
06/07/2000 15:30:00 4.93
07/07/2000 15:30:00 4.85
10/07/2000 15:30:00 5.2
11/07/2000 15:30:00 5.43
12/07/2000 15:30:00 5.54
13/07/2000 15:30:00 5.29
14/07/2000 15:30:00 5.17
17/07/2000 15:30:00 5.29
18/07/2000 15:30:00 5.07
19/07/2000 15:30:00 5.05
20/07/2000 15:30:00 4.92
21/07/2000 15:30:00 5.04
24/07/2000 15:30:00 4.77
25/07/2000 15:30:00 4.91
26/07/2000 15:30:00 4.62
27/07/2000 15:30:00 4.8
28/07/2000 15:30:00 4.59
31/07/2000 15:30:00 4.47
01/08/2000 15:30:00 4.57
02/08/2000 15:30:00 4.51
03/08/2000 15:30:00 4.57
04/08/2000 15:30:00 4.41
07/08/2000 15:30:00 4.3
08/08/2000 15:30:00 4.35
09/08/2000 15:30:00 4.35
10/08/2000 15:30:00 4.33
11/08/2000 15:30:00 4.3
14/08/2000 15:30:00 4.26
16/08/2000 15:30:00 4.26
17/08/2000 15:30:00 4.25
18/08/2000 15:30:00 4.28
21/08/2000 15:30:00 4.36
22/08/2000 15:30:00 4.39
23/08/2000 15:30:00 4.44
24/08/2000 15:30:00 4.45
25/08/2000 15:30:00 4.52
28/08/2000 15:30:00 4.68
29/08/2000 15:30:00 4.71
30/08/2000 15:30:00 4.68
31/08/2000 15:30:00 4.79
04/09/2000 15:30:00 4.8
05/09/2000 15:30:00 4.71
06/09/2000 15:30:00 4.72
07/09/2000 15:30:00 4.73
08/09/2000 15:30:00 4.79
11/09/2000 15:30:00 4.87
12/09/2000 15:30:00 4.83
13/09/2000 15:30:00 4.73
14/09/2000 15:30:00 4.77
15/09/2000 15:30:00 5.13
18/09/2000 15:30:00 5.1
19/09/2000 15:30:00 5.1
20/09/2000 15:30:00 5.13
21/09/2000 15:30:00 5.12
22/09/2000 15:30:00 5.46
25/09/2000 15:30:00 5.57
26/09/2000 15:30:00 5.32
27/09/2000 15:30:00 5.32
28/09/2000 15:30:00 5.48
29/09/2000 15:30:00 5.9
03/10/2000 15:30:00 6.08
04/10/2000 15:30:00 6.07
05/10/2000 15:30:00 6.3
06/10/2000 15:30:00 6.52
09/10/2000 15:30:00 6.39
10/10/2000 15:30:00 6.1
11/10/2000 15:30:00 6.06
12/10/2000 15:30:00 6.12
13/10/2000 15:30:00 6.47
16/10/2000 15:30:00 6.42
17/10/2000 15:30:00 6.23
18/10/2000 15:30:00 6.2
19/10/2000 15:30:00 6.68
20/10/2000 15:30:00 7.18
23/10/2000 15:30:00 6.86
24/10/2000 15:30:00 6.89
25/10/2000 15:30:00 7.29
26/10/2000 15:30:00 7.42
27/10/2000 15:30:00 7.33
30/10/2000 15:30:00 7.29
31/10/2000 15:30:00 7.33
01/11/2000 15:30:00 7.48
02/11/2000 15:30:00 7.57
03/11/2000 15:30:00 7.9
06/11/2000 15:30:00 8.11
07/11/2000 15:30:00 8.36
08/11/2000 15:30:00 8.31
09/11/2000 15:30:00 8.49
10/11/2000 15:30:00 8.58
13/11/2000 15:30:00 8.67
14/11/2000 15:30:00 8.65
15/11/2000 15:30:00 8.76
16/11/2000 15:30:00 8.77
17/11/2000 15:30:00 8.91
20/11/2000 15:30:00 9.03
21/11/2000 15:30:00 9.21
22/11/2000 15:30:00 9.19
23/11/2000 15:30:00 9.35
24/11/2000 15:30:00 9.59
27/11/2000 15:30:00 9.79
28/11/2000 15:30:00 9.98
29/11/2000 15:30:00 10.07
30/11/2000 15:30:00 10.15
01/12/2000 15:30:00 10.19
04/12/2000 15:30:00 10.4
05/12/2000 15:30:00 10.35
06/12/2000 15:30:00 10.95
07/12/2000 15:30:00 11.06
08/12/2000 15:30:00 11.32
11/12/2000 15:30:00 11.55
12/12/2000 15:30:00 11.43
13/12/2000 15:30:00 11.72
14/12/2000 15:30:00 11.71
15/12/2000 15:30:00 11.81
18/12/2000 15:30:00 12.07
19/12/2000 15:30:00 12.09
20/12/2000 15:30:00 12.28
21/12/2000 15:30:00 12.28
I want a query to display
02/06/2000 5.12
03/07/2000 4.87
01/08/2000 4.57
04/09/2000 4.8
03/10/2000 6.08
01/11/2000 7.48
01/12/2000 10.19
Presently, I am doing multiple sorting in different columns. Any simplified query suggestion will be appreciable.
Thanks in advance.
Best Regards.
如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。

绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱,如果其他用户或者作者回复了您的评论,将不能在第一时间通知您!
发布评论
评论(2)
使用:
更新:
use:
update:
尝试一下,假设您的数据从第2行开始。
结果(日期格式化的DD/mm/yyyy)
使用Google Finance,请尝试
explanation
sequence
sequence(loday()() - 日期(2020,1,(2020,1,) 1),1,日期(2020,1,1),1)
将给出以来所有日期的列表,因为1/1/2020text(unique(eomonth(____)))
将保留独特的值,因为yyyy-mm eomonth> eomonthTry it, assuming your data starts at line 2.
result (dates formatted dd/MM/yyyy)
with google finance, try
explanation
sequence(today()-date(2020,1,1),1,date(2020,1,1),1)
will give a list of all dates since 1/1/2020text(unique(eomonth(___)))
will retain unique values as yyyy-MM EOMONTH