Ohh I like that a lot, that's a nice workaround! Definitely not something I have ever considered before.  ... View more

Well unfortunately this isn't showing me the types and formats of all of the data items like I was hoping it would, so let's try a different approach. If you go to create a new calculated item, in the Data Items pane it should group all Dates or Date types together, can you show me what it has? Like this:     Also did you try the list table with just the two data items dummy_date and "Predicted: result (Difference from previous period) 1"? Did that work or do you see the same problem/warning? ... View more

That warning means your table has more than one Date with different formats referenced, and the periodic calculation will not work that way. I would expect it to work if your list table just had dummy_date and the "Predicted: result (Difference from previous period)" calculation in it.  But if any of those other data items are dates of another format (for instance is var1 a datetype with format of Year?) then you'll have this problem.  If you're not sure, one we can check for you is if you click the list table object and press Ctrl+Alt+Q on your keyboard. A new window will appear for Query Diagnostics. Save what you see in the Query tab and attach that to this post, and we'll see what's going on in it.  ... View more

As I mentioned, I think your best course of action is to get a technical support track opened. This might require a bit more investigation than can be done over the SAS communities.  I also didn't see answers to the other questions I had asked which I'll summarize here: Do you see this happen for every report you try to open all of the time? Does network connectivity matter, for instance if you're connecting over VPN or from an office location?  Are you aware of any other users in your organization that have the same issue? A screenshot of the prompt you're getting might be useful too, it's still not clear to me if you're seeing the actual SASLogon prompt or something else.  ... View more

Ah ok so this is SAS 9.4. In that case the rest APIs I mentioned won't be relevant.  What do you do with it after you export it, or what are you wanting to do? Generally speaking, manipulating the XML definitions is not something we could officially provide support for (since the schemas aren't publicly documented as mentioned), but that's not to say you can't do it or changes you make might not work.  ... View more

I think the problem you're going to find here is once you filter out these categorical values for Current_Product that have the Months you're looking for, then it won't matter what you do unfortunately with Hidden measures or anything else.  This does remind me though, another way to get the same type of crossing for other objects is to use the Lattice roles. I made up some data similar to what you're doing, so here in a bar chart with Lattice Row = Current_Product:   Between the two values for "Current_Product", all months will be displayed and you'll see no bars for months where there is no data. But again once you get rid of one of these Products, the months for those will go away too.  ... View more

Echoing what @Sam_SAS said, what you are wanting to do is not currently possible. Visual Analytics cannot make up data points that do not exist in the data. The only exception is crosstabs where we have to make up data crossings that may not exist in the data.   There is an open feature request for the line chart scenario and I have added this thread to it. The only way around it for the time being will be to add rows for the non-existent months into your data.  ... View more

Can you clarify what exactly it is you are trying to do? The XML schemes are not publicly documented for any VA release so I'm not certain what you mean by built-in later versions, so I'm assuming we are talking about two different things maybe.  There are rest APIs where you can retrieve the report content (in JSON format) for both Viya 3.5 and Viya 4: Viya 3.5 => https://developer.sas.com/apis/rest/v3.5/Visualization/#reports Viya 4 => https://developer.sas.com/apis/rest/Visualization/#reports   I don't know if that's helpful but if you can provide more information about what you're doing we can give better advice I expect.  ... View more

Assuming I haven't misunderstood what you are trying to do (and there is a decent chance I have), I don't think this is going to work as long as you have to have both Event ID and Task ID in the same list table. With both columns present, there is no way to aggregate the way you're wanting, at least that I can find.   Borrowing on Bruehl's suggestion, I made up some sample data that looks like this: So let us say you want to be able to filter to Event ID 2 because it has zero active tasks. I then created a calculated item to say if Task Activity = Active Task, return 1, else return 0:   Then I added a new List Table with just Event ID. I then added a local filter to this List Table like this: where I'm filtering on my new Task Activity Number = 0 AND it's filtering on aggregated values. This gives me a List Table that is filtered to just Event IDs that have zero active tasks: I believe this is more or less what Bruehl was getting at.  I could then use this list table and create an action to another list table that has Task Details for example, so that clicking a row in my Event IDs table filters my Task table. Again once you add Task ID to this first Event table, the whole thing breaks but if you keep them in separate table objects, it might be doable.   Another option I have not fully explored myself yet is making use of Aggregated Datasources to potentially create the relational tables you mentioned earlier. Assuming all of the relevant columns exist in the single table you currently have you, you could go to the Data Options menu and select "New data from aggregation of <table name>":     Then you could pick only the event columns. Create table. Repeat again and pick only the task columns plus the event ID column. Create that table. Then maybe you could join these together in the way you're wanting, or work with them separately in separate report objects?  One downside to this approach, there is a performance hit when creating tables like this. Each time you open the report this way, these tables are generated in the user's personal CAS library so there might be a delay opening the report each time, and these tables take up space on the CAS server too that you may or may not have. I haven't fully thought this approach out yet though so this may not be feasible either, but just something to possibly explore.       ... View more