Hey there
I suppose this is more a question around implementation rather than use. As part of a project I have built an OData API for use by our company and customers, with the plan to integrate our data into Excel and PowerBI for reporting. We have tested extensively and I do believe we meet OData minimal conformance, but we have implemented this endpoint in TypeScript rather than the Microsoft C# libraries recommended in the documentation. Our entire tech stack uses TypeScript so we thought it simpler / more maintainable to build our own OData feed than try integrate a new language into the mix.
The issue is that an error gets thrown when trying to connect to Excel / PowerBI with out endpoint - "[DataFormat.Error] OData: We were unable to detect required payload information". This is difficult to debug as we are unsure what payload information it is trying to detect / what exactly is failing. Does anyone have a list of steps for an endpoint to pass as an OData feed? We only allow for version 4.0 of OData and the necessary headers are returned alongside the format being xml. Does it need to be an atom service document rather?
Using Web.Contents we can achieve what we are trying to do, but it would be nice to get it recognised as a feed for full functionality. Any help / guidance on how to best tackle this would be greatly appreciated. With the documentation expecting C# and no real way to debug the endpoint we are kind of stuck at the minute. Thanks!
