Best practices for excluding erroring fields from explores

jyau
New Member

I have run into the following situation a bunch of times, and it feels like a best practices recommendation of different approaches would be helpful.

When dealing with the problem described in the post below, and taking the second approach “Exclude the erroring field from explores”, what are the different methods of doing so?

What does this error mean? When you run the LookML validator, you might see an error like this: [image] This error is referring to the field lifetime_orders in the users view. It is saying that users.lifetime_orders cannot access the user_order_facts.lifetime_orders field that it refers to. Why is this error happening? There are a few reasons this error could be occurring: The field you are trying to refer to does not exist. The field is a dimension group without the timeframe appended. Th…

From what I gather, there are a couple of options for doing so:

  1. Exclude the fields at the explore level
    a. Exclude individual fields - fields: [ALL_FIELDS*, -view.field1]
    b. Exclude sets - fields: [ALL_FIELDS*, -set1*]
  2. Exclude the fields at the join level
    a. Include individual fields (excluding everything else) - fields: [field1]
    b. Include sets and exclude individual fields in those sets - fields: [set1*, -field1]

I know some of the syntax above is incorrect, because my LookML still isn’t validating.

  • What is the correct syntax?
  • In which cases does the view need to be scoped? Do sets need to be scoped to the view?
  • Are there any other options I haven’t covered above?

My specific situation is that I have many fields in a single view that cause errors across many explores, so excluding sets at the explore level (1b, above) seems like the best solution, although I can’t get the syntax right.

1 3 2,286
3 REPLIES 3
Top Labels in this Space
Top Solution Authors