Hi, is there any explanation for this kind of viz switcher behavior?

Hi Daniel, I’m happy to help with this issue. First, I need some more information and context. - Are you using the GoodData Cloud platform or an embedded environment? - In case of a embedded environment, how are you embedding the dashboard (please, if is SDK, include versions)? - Is this related to a single visualisation on a dashboard, or is it an entire dashboard? - If it’s a single visualisation, does the issue occur only with that one? - Are multiple users experiencing this issue? If so, could you specify their roles and permissions? - Have you tried accessing it in a different browser? - Could you please provide a TraceID from the developer console? - Did you make recently some significant changes in the workspace/domain? and are you able to see the same data you have in the visualisation in another visualisation, for example table? - Lastly, could you share a direct link to the visualisation or dashboard? Feel free to use direct messaging. Any information you can provide will be important for understand the root of the issue. Thank you!

• this is on the cloud platform ◦ (not embedded) • multiple viz switchers are having issues, sometimes like this • i'm the only person who'd see it • browser change hasn't helped • yes, I'm editting this via declarative YAMLs and the python SDK. i can push an update and get one behavior, refresh the browser and get another behavior, refresh again and get another behavior, so on and on. Just affecting these switchers. • I'll send direct links vai dm

Last question, did you try to edit and send the configuration of the visualisation manually or directly in the GD Cloud UI, without the YAML file and Python SDK? I am assuming in that case will work as expected.

i have not had this issue happen when making visualizations/dbs one by one in the gui

I have impersonate the testing organisation using our Admin user, and I can see correctly the dashboard without any issue. I would like to ask for autorization if I can impersonate your user, as you answered that only you have the issue.

yes, go ahead. let me know what you need from me

Hello Daniel, trying to open the insight "Encs_by_encs_sentiment_Appliances" will result in a message in Analytical Designer stating that "The visualization with this URL does not exist." However, checking for the visualization definition via its corresponding endpoint does prove otherwise: /api/v1/entities/workspaces/Your_Workspace_ID/visualizationObjects/Encs_by_encs_sentiment_Appliances The reason for this is that you have the following ID defined for the ranking filter in it:

"rankingFilter": {
              "measure": {
                "localIdentifier": "bf9274a3f9cf49dd8c961783c7af2530g"
              },
              "operator": "TOP",
              "value": 15
            }
          }

But I am afraid that these ID's don't end with an alpha character. I was able to replicate the same issue in my workspace and have thus determined that you will need to recheck this definition. I understand that you are working with the Python SDK, and it was most likely introduced by it, since adding a ranking filter via the UI will assign the ID in the correct format.

Thanks Moises, this should be enough for me to work with; I'll return if I have more I can't resolve

Hi Daniel, thank you for the update. I would like to clarify, there is no restriction about the length of the string

localdentifier

. The value is used to cross-reference model elements that are part of an execution. The constraint is that the local identifiers of all measures and attributes in an execution definition MUST be unique. And as you wrote, after removing the 'g' works because then the localIdentifier is the proper match. Please let us to know if you need more assistance.