Recently I was working on this project and I was asked to create a new module to create a new feature. The C# solution, as most technologies, makes use of a database in order to fetch and store data. Basic CRUD operations. As you might have already guessed from the title of this post the project makes use of ADO.NET and Entity Framework to perform such processes. I will have to admit that I am not a fan of ADO.NET or Entity Framework. Personally speaking I prefer to create scripts, stored procedures, views, etc and then open connection strings and retrieve the data as data tables/data sets. Having said that, that was my opinion, and since the project was already in a stable running state I had to adapt and work with Entity Framework, which I’m also comfortable with.

The problem happened when I created a new table via SQL Server Management Studio, let’s call it tbl_weather. This new table would have historical data and store “raw” data that would be processed later on. tbl_weather would have three columns and roughly this type of data.

I then switched to Visual Studio and tried to update my Entity Framework model. Right click > Update model from database > Next > Next > Finish … job done. But wait, where’s tbl_weather?? I tried this step again and again and nothing happened. The connection string was good because my new table was being detected when attempting to update my model from the database. So what is wrong with it?

After a while I tried to delete my table and re-create it but this time round I added another column, ID, and set it as unique primary key. I tried to update my model once more and this time round the table was added to my model straight away. To my understanding it seems that Entity Framework would not accept to add “basic” tables but it would require the new tables to be somewhat more structured. In this case it seemed that not having one of the most basic SQL standard, a primary key, was preventing me from adding my table to my model.

Now, I know that a primary key is a must and that a unique identifier must be present but in this case I was fetching the data from a source and bulk insert it into this table. This data would then be retrieved again, processed and insert in yet another new, better structured table. Therefore, I didn’t really worry the primary key.

Anyways, that’s my two pence that I wanted to share with you guys, until next time, adios!

Bjorn