Bug: creating dataset filters via CloudFormation

m0ltar · August 16, 2023, 3:36pm

I added a simple dataset filter in the UI.

Then I exported the dataset definition using describe API call, and the DataTransforms came out like this:

{
    "FilterOperation": {
        "ConditionExpression": "NOT ({path}=\"\")"
    }
},

When I used exactly the same values to push back the filter via CloudFormation, the UI now displays this:

screenshot-20230816T163415-a0wx72dL@2x

And it is not possible to edit or delete it. Choosing options from the three-dot menu does nothing. No errors, no console errors, no network calls. Just simply no action.

Meanwhile, describe API call still shows the same setting:

{
    "FilterOperation": {
        "ConditionExpression": "NOT ({path}=\"\")"
    }
}

m0ltar · August 17, 2023, 2:16pm

Looks like the same report here from someone using Boto. So I think it’s not limited to CloudFormation, but API as well.

github.com/boto/boto3

'Custom filter list' filters for QuickSight Datasets are stuck after update from boto3

opened 10:01AM - 24 May 23 UTC

closed 08:24PM - 06 Jun 23 UTC

manudb9

response-requested service-api closed-for-staleness quicksight

### Describe the bug The boto3 documentation relative to update QuickSight Data…sets only defines a single way to create dataset filters: by defining `FilterOperations` in the `DataTransforms` section of a `LogicalTableMap`. ```python response = client.update_data_set( ..., LogicalTableMap={ 'string': { ... 'DataTransforms': [ { ..., 'FilterOperation': { 'ConditionExpression': 'string' }, .... ) ``` To understand the syntax, we have created a dataset, using AWS dashboard, with a Custom Filter list on a field, and then extracted the dataset JSON definition by using the method `describe_data_set`. ![image](https://github.com/boto/boto3/assets/16865472/db046f4a-730e-4f85-84cc-f14f00b8508b) ![image](https://github.com/boto/boto3/assets/16865472/59862925-35ae-4eb8-a911-20681e0e4a18) This is the part of the filter definition: ``` { ..., 'LogicalTableMap': { 'data-dev-geo-analytics': { 'DataTransforms': [ ..., { 'FilterOperation': { 'ConditionExpression': '{industry}="value1" OR (({industry}="value2") OR (({industry}="value2")))' } } ], ... } }, ... } ``` We have used the same JSON definition to update the dataset from boto3, but the result is different than the one create from AWS UI. The filter appears on UI with a different format and cannot be edited or deleted from the UI: ![image](https://github.com/boto/boto3/assets/16865472/fd4910f8-4aee-436b-8b08-ab64265a8eed) ### Expected Behavior The filter in the dataset is of type "Custom Filter List", and can be edited and deleted from QuickSight UI. ### Current Behavior The filter appears on UI with a different format and cannot be edited or deleted from the UI. Nonetheless, the filter is correctly applied to the data, since rows not matching the condition are discarded. ### Reproduction Steps 1. Create or update a dataset from boto3 QuickSight client, containing a LogicalTableMap with DataTransforms FilterOperations. 2. Go to QuickSight in AWS dashboard. 3. Click on `Edit dataset`. 4. Go to Filters tab. 5. Try deleting or editing the filter. ### Possible Solution Add support for Custom Filter List in boto3 QuickSight Client. ### Additional Information/Context _No response_ ### SDK version used boto3==1.26.101 ### Environment details (OS name and version, etc.) Python 3.7 MacOs 12.2.1 (21D62)

DylanM · November 14, 2023, 4:50pm

Hello @m0ltar, I am sorry this was never responded to! I know it has been a few months, so have you discovered a fix for this or noticed a change in this functionality? If not, I can make sure this topic is marked as a bug for the development team. Thanks for the thorough documentation of the issue you were experiencing, it is super helpful!

m0ltar · November 14, 2023, 5:29pm

I have not tested this again, not sure if the bug is still there.

DylanM · November 16, 2023, 4:28pm

Hello @m0ltar, if you come across this issue again I would recommend filing a case with AWS Support where we can dive into the details so that we can help you further. Here are the steps to open a support case. If your company has someone who manages your AWS account, you might not have direct access to AWS Support and will need to raise an internal ticket to your IT team or whomever manages your AWS account. They should be able to open an AWS Support case on your behalf.

For now, I will go ahead and leave this topic tagged as a bug for our AWS development team and archive the question. Thank you for highlighting this issue!