Topics started by jfhbrook
I have a scenario I’m working through that looks something like this: I have 2 tables (in practice way more than 2 tables) that entail GA site analytics that are both joined with a content items dimensions table and each with a date dimension. One fact table, with aggregated metrics, looks like this: cms_id | date | metric | value 123abc | yesterday | 'pageviews' | 10000 and another fact table, which deals with traffic source metrics specifically, looks like this: cms_id | date | source | medium | campaign | pageviews 123abc | yesterday | twitter | socialmarketing | SF | 1200 Treating cms_id as a consolidated dimension works great because I can specify that both of these tables join with my content items table on that cms_id, and that’s fine. But I don’t have a corresponding dimension table for dates! So right now, in my explore there are 2 date dimensions and you need to know coming in which dimension to use depending on which table you’re working with! What’s the right way to model
Many of our non-Looker tools have the ability to mark incompatible dimensions and measures, which will disallow the user from using them in combination. For example, in DFP, “Creative Size” is incompatible with requests-related metrics, because requests are for one of a range of acceptable creatives rather than for a specific creative. On DFP specifically, the UI “grays out” the incompatible dimensions/metrics and disallows selecting them. A screenshot: We’re trying to pull similar ad metrics into looker, and are wanting to reimplement some version of this graying out. Does LookML has a way of tagging a group of dimensions/measures as incompatible? Is there a way to fake it?
I have a table (representing youtube subscriber metrics) that contains numbers aggregated across dates: channel | date | subscribers my_channel | 2/1 | 1500 my_channel | 1/31 | 1495 my_channel | 1/30 | 1487 my_other_channel | 2/1 | 1200 my_other_channel | 1/31 | 1160 my_other_channel | 1/30 | 1158 I currently have my measure (subscribers) described as a sum, because it might be appropriate to calculate the total amount of subscribers across channels (I’ll leave that one up to my users)…but it’s not appropriate to calculate sums across date, since the numbers are already summed! It seems this problem can be avoided by making sure you select the date field when doing the aggregate, but it would be fantastic if there was a way to enforce this constraint at the lookml level. Is this possible?
Already have an account? Login
Login to the community
No account yet? Create an account
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.