Hi Team, When I pull any attribute from the subfie...
# gooddata-cloud
v
Hi Team, When I pull any attribute from the subfield dataset, the facts from the other datasets are no longer visible. Similarly, when I pull facts, the attributes from other datasets also get hidden. https://wizehive-zengine-dev.cloud.gooddata.com/analyze/#/VPCx4wTuv3awL2SRV_SNk9h8-03tfp67PL3-/reportId/edit From what I understand, this behavior seems to be related to how the datasets are connected—particularly the grain and reference relationships between them. If the issue is the same, could you please explain how these datasets should be connected? I’m finding the grain and reference relationships a bit confusing. Thanks CC: @Anjali Mandowara @Abhijeet Gupta
j
Hi Vivek, What you’re seeing is expected behavior in Analytical Designer. GoodData can only show objects together if they are connected through a single valid relationship path and share a compatible grain (primary key level). When you add an attribute from the Subfield dataset, the tool checks whether the selected facts can be safely joined to it. If there isn’t a clear and unambiguous relationship path — or if the datasets are at different grains (for example, one is per Applicant and another is per Applicant–Subfield) — GoodData hides the unrelated objects to prevent incorrect results. In most cases, this happens when: • The datasets are connected at different grains, or • There is a many-to-many relationship without a proper bridge dataset, or • There are multiple possible join paths between datasets. To resolve this, the model usually needs to ensure: • Each dataset has a clearly defined grain (primary key), • Fact datasets are not connected directly to other fact datasets, • Relationships allow a single, clear path between the objects you want to analyze together. This is very similar case to what we've already discussed here Sending some further documentation which might help: https://help.gooddata.com/classic/en/data-integration/data-modeling-in-gooddata/logic[…]a-model-components-in-gooddata/facts-in-logical-data-models/ https://www.gooddata.com/docs/cloud/model-data/concepts/primary-key/ https://www.gooddata.com/docs/cloud/model-data/concepts/dataset/ https://www.gooddata.com/docs/cloud/model-data/concepts/primary-key/
I would also recommend to go through our GoodData University Modeling section. The doc is bit older but same principles still apply: https://university.gooddata.com/tutorials/data-modeling/
v
Ok @Julius Kos