Forum Discussion

JavierLopezFALP's avatar
JavierLopezFALP
Frequent Visitor
1 year ago
Solved

CLOCK Table Filtering

Greetings.   I've got a growing appointments table (950k+, since 2023, well over 2 milllion if I up to 2021), for which I need numerous flags that depend on the appointment time. In order to reduce...
Need Help
Tips and Tricks
  • JavierLopezFALP's avatar
    JavierLopezFALP
    1 year ago

    I figured it out. Turns out that the date_time had miliseconds precision. But because  extracted the data as datetime, the format within PBI gave no clue about it. And the decimals remained even after "splitting" the time value.

     

    Whereas my Clock_Table had "integer values behind" each time value.

     

    I solved i with

    cast (date_time as datetime2(0))

     

    Thanks for your input.

     

JavierLopezFALP's avatar
JavierLopezFALP
Frequent Visitor
1 year ago

I'm sorry, I don't follow you. I did not want to bin the time values in the appointments table, with its growing 950k+ rows.

 

Which is why I made the 86.400 time values table. And it works, partially

 

 

The column on the left is form the CLOCK table, and the column on the right is from APP table. And some second-sensitive values are caught in the relationship.

 

Now, out of 913k appointments (after other filters, unrelated to this subject) few are left out of the relationship. But, those 913k share 48.807 distinct time values, and 48.369 are left inrelated.

So, what is the problem?

 

Thanks beforehand.