This implementation based on the now deprecated Microsoft.Azure.Cosmos.Table SDK. It has been replaced with the Azure.Data.Tables SDK which contains some improvements, such as a generelly simplified usage and operations for upserting and merging of entities.
Migration Guide: https://github.com/Azure/azure-sdk-for-net/blob/main/sdk/tables/Azure.Data.Tables/MigrationGuide.md
To consider: The limitation that only a very limited set of property types can be used also applies to the new SDK. Unlike the old SDK, the new SDK unfortunately does not yet provide a serious way to implement custom serialization (such as the solution presented in this snippet, which automatically serializes non-compliant types as JSON strings). This issue is discussed here: https://github.com/Azure/azure-sdk-for-net/issues/27413
Tables in Storage Account or Cosmos DB provides a very easy to use solution to store simple and not very critical data. As Storage Tables are extremly cheap, can store an almost unlimited amount of entries and are easily connectable from within Power BI, this sample shows a logging scenario as a typical use case.
The properties PartitionKey and RowKey must be provided and the combination of them has to be unique. Use them appropriately. In the base class EntityBase there is a pattern and helper function which simplifies the concatenation and splitting to/from a single identifier.
Supported data types are very much limited. This includes e.g. enum or collection properties. Therefore the attribute EntityJsonPropertyConverter is provided. This just serializes those properties to and from JSON.
For performance reasons it is recommended to not instantiate the client for the table connection on each request. The FunctionHelper class provides a possible way to handle that.