Table calculations return blank result

Q&A
,
1

Hello, I want to count the number of pipeline names partitioned by pipeline type. I have used a table calculation to obtain my expected result. However, for months where there are no pipeline types, the result is blank.

runningSum(
count(name),
[{created_month} ASC],
[pipeline]
)

image
image738×350 17 KB

For instance, I want above 202103 is 72, not blank.

Has anyone done the same thing before? Please help me out. Thank you!

2

Could you provide a sample dataset in csv to demonstrate your issue?

3

Hi Xing, thank you for reaching out.
The dataset looks like this

image
image2170×328 39.5 KB

created_year,created_month, created_day are calculated field

4

Hi @Xing I found some work around method, I used lag to find the the previous column and will replace month’s value with this value if it’s bank. But it seems can not detect the empty value of the calculation?

ifelse(
{count_type} = NULL,
lag(
     {count_type},
     [{created_month} ASC],
     1,
     ['pipeline']
),
{count_type}
)

I tried isNull() as well

5

What does count_type represent? I do not see that field in your screenshot of the table.

6

Anyway, you should use isNull() as below
ifelse(
isNull({count_type}),
lag(
{count_type},
[{created_month} ASC],
1,
[‘pipeline’]
),
{count_type}
)

7

Hi @Xing .
Thank you, I already tried this, the value is still empty on the new calculated field.
I did some more research and found this post:

It seems a runningSum (or table calculations in general) limitation.

8

At AWS, our roadmap is primarily driven by our customers. Your feedback helps us build a better service. I have tagged this as a feature request.