In OAC when I am creating a dataset, How can I pass a Filter from the "Dashboard filters to the SQL? I attached a image like it was don in BIP Publisher
Hi @User_SYYDG
Curious on the use case for this one. Are you trying to restrict the dataset itself or just the output?
Generally, when creating a dataset you would load the whole set then add Workbook filters or dashboard prompts to restrict the output.
If you are trying to apply security to the dataset I don't believe this is possible.
One thing you could try is to create a Parameter on a workbook under the (x) tab of the data panel and then use that parameter in the sql so it would be something like WHERE A.SAAR_ID= @parameter("SAAR_ID")('').
Dan
A dataset is meant to be the same object for everybody, therefore there isn't a real concept of "passing a filter to the SQL query defining the dataset".
There is an exception that is meant to bring a feature a bit like row-level security in the semantic model: you can set dataset filters by application role. https://docs.oracle.com/en/middleware/bi/analytics-server/user-oas/use-role-based-filters-dataset.html
It is probably not what you have in mind anyway. Therefore your closest option is what Daniel mentioned above: you have a filter applied to the dataset when you use it. Of course this mean that your user can potentially get rid of the filter and have access to the whole dataset (and here is where the filter by application role plays a role as those can't be removed).
Depending on what you try to achieve, if it's for security reasons to not show the whole data to everybody or a functional/performance reason, your choice on how to implement what you need will be different.
I'm running into a similar issue and can provide you with a use case. I'm making a workbook to summarize data for a single customer so our Customer Service reps can work with them while on a call. I have millions of customers, all with multiple bills, multiple services we offer and multiple other data points that need included in said workbook. Pulling all the data for all the customers is extremely slow and caching is not even an option as it's too much data. The Semantic modeler seems to be the answer, but it's a lot more work to set it up than a dataset…
This sounds like more of a filtering situation than a security situation. The filters you set on a canvas or chart will be passed down to the queries. I would recommend having a required parameter such as customer or account and when it is not set, you will just see blank charts.
has a walk through.
Hi @Henrikues
With that much data I would suggest building it into the semantic model. This will enable you to pre-calculate the results rather than it being done at runtime.
Personally, I have always considered the dataset capability to be more for self service exploration and proof of concept work to prove the value of adding data before creating a governed model in the semantic modeller.
That's what I need to do... I built in Classic / BIP for now as it's a lot simpler to build even though it's rather limited for the users as far as functionality goes. When I load all of the data in dataset, I get the error below when trying to make visuals: State: HY000. Code: 43121. [nQSError: 43121] Max Row Limit Exceeded (HY000) Thanks for the reply, I'll proceed with the Semantic Model!
State: HY000. Code: 43121. [nQSError: 43121] Max Row Limit Exceeded (HY000)
Explore how to build agents using code in AI Data Platform. We'll create an agent that examines and analyzes the contents of a Github repository and reports its findings. The session demonstrates how to use different developer tools to create agents including the AIDP extension for VS Code and Codex. Using these tools we…
What is wrong with this code
Am I missing something or is there no way to search through previous posts to make sure I don't duplicate a question? All I see on the filter options are tags or status… a general search for a word or phrase would be helpful.
We were under the impression that when exporting a visualization in DV to CSV format hidden columns would be include from the grammar panel. This is not happening. See below for a screenshot showing 3 hidden columns that do not appear when exporting the pivot table to CSV. This seems to have worked a month or so back but…
Migrated RPD from obiee12c to OAS 2024. post migration when we try to access dashboards on OAS we are getting below error: State: HY000. Code: 10058. [NQODBC] [SQL_STATE: HY000] [nQSError: 10058] A general error has occurred. (HY000) State: HY000. Code: 43275. [nQSError: 43275] Message returned from OBIS…
There is no option for legend title in many visualization types.
