Not sure what exactly you referring to with “repor...
# gooddata-platform
z
Not sure what exactly you referring to with “reports functionalty”. The FREE edition includes Analytical Designer and Dashboards.
m
I think I'm probably asking the wrong question - I was under the impression reports were somehow a distinct entity separate from dashboards.
I'm trying to work out if it's possible to create views of data with more complicated criteria. So for example I can see how I can filter the available rows to show, say, all orders where the status is "delivered". Is it possible to have a filter expression built up from a logical conjunction of filters e.g. "where orderStatus is 'delivered' OR dateOrdered < '2020-01-01'"? And from what I can see in the interface it isn't possible to add filters for numeric values, or am I missing something? Thanks
j
This is not possible with our UI yet. Workaround solution is to define and store a metric with filter of any complexity a use the metric. True is that if you would like to reuse the filter in different scenarios with different metrics, it is not a convenient solution.
m
Just to confirm my understanding here, let's say I want all data where quantity > 1 OR price > 100, I need to: 1. Create a custom metric like this: SELECT IF {fact/order_lines.quantity} > 1 OR {fact/order_lines.price} > 100 THEN 1 ELSE 0 END 2. Add this to my table 3. Add a filter on this column and show only rows where the value is 1 Is that right? I've tried this and it works, but it seems unfortunate that I have to have this extra column visible on my table.
d
Can't you just put that filter directly into you metric? e.g. something like
Copy code
SELECT {my_metric} WHERE {fact/order_lines.quantity} > 1 OR {fact/order_lines.price} > 100
m
Well I don't really necessarily want a metric - I just want to display a table of data filtered by a simple condition with an OR in it. Broadly how would I go about that?
d
May I ask why you don't want to create a new metric? It's certainly better solution than the extra "IF" metric because: • It doesn't add redundant metric to the insight (as you already said) • It works correctly, contrary to your IF metric. Let me explain. Suppose that the rows in your insight represent Products. Then the version with filter in the metric filters out source data (most probably order_lines - depends on the metric) as expected. On the other hand your "IF" version filters out the whole Product rows, based on the condition whether for given Product product there's at least one line in the data complying to your condition. This is most probably not what you want.