Hi. New user on GD platform here. How do you reco...
# gooddata-platform
i
Hi. New user on GD platform here. How do you recommend to handle dates on the DW? We use the traditional date dimension, but we are seeing some odd things. Also the date field appears magically on top of the fields list and I don’t know where it comes from. It comes from the date dimension? So every insight that is linked to the date dimension is filtered by this date? lets say we have this:
Copy code
table_surgeries_fact
patient_sk -> dim_patient -> date_birth_sk -> dim_date
surgery_date_sk -> dim_date
….
If the date is applied to all fields, then is also applied to the date_birth and it could explain some numbers we are seeing. What’s your recommendation? Having date/datetime fields instead of dim_date will work better with GD engine? Can we fix this on the Logical Model?
👀 1
m
Hi Ivan, I assume you are using the GoodData platform. Dates can be loaded in either of the following formats:
Copy code
dd-mm-yyyy
mm-dd-yyyy
yyyy-mm-dd
dd-MMM-yyyy
MMM-dd-yyyy
yyyy-MMM-dd
MMM dd, yyyy
If you are importing a table/view that contains dates, not one but two datasets will be added to your LDM: one Date dataset for the dates, and the other one with the rest of information from the table/view. These two datasets will be automatically connected with a relationship, and the Date dataset will become a reference in that other dataset. I recommend checking the resources below: Dates in Logical Data Models | GoodData Free Using Date Dimensions in GoodData | The GoodData Community
j
Hi, each date should be mapped to separate date dimension dataset If there are multiple dates in a fact dataset. The Date in the data catalog is somewhat special, this is why it is listed as the first. Role of date is selected when you add Date to insight in Analyze. The selection of role ("as" parameter with selection of date dataset) is pre-selected based on common use in other insights, metrics etc. and can be changed manually. For example metric "Number of births" will have probably "Birth date" dimension dataset pre-selected. Surgery date will be recommended If you slice "Number of surgeries" by date. You may need to do the selection manually for the first time when used with a fact/metric. Later it will be recommended right based on frequency of use together with date datasets. The main advantage comes on dashboard. You may apply single filter (e.g. This month) and each widget may use different role of date depending on which metric it displays. The role of date for the dashboard filter can be changed in widget configuration. There is also recommendation based on frequency but it prioritises selection in insight over general usage.
i
Sorry, our version is CN, not platform
same thing applies?
j
My answer is relevant for all platforms 🙂
🙌 3