Customizing Datasets

Now that we've finished modifying our views, let's update the datasets. In my original dataset, I was joining genres and tracks, but I now know that I have a new set of views and I also have a view that I want to use as the base view, so this dataset is gonna need a complete overhaul. Let's just create a new dataset. To add a new dataset click the plus next to datasets and you'll now get a template ready to populate. This is going to include the join template which is really useful. Let's first of all give the state set a name and we can also give it a user friendly label and a description. In the first run, we're going to choose the base view of the set, which is my artist view, and remember to use the name of the view rather than the label. Now we can define the joins. First, I'm going to join the daily tracks view. Let's give our view an alias. This will make my life easier when I'm typing up my join constraints but it won't appear anywhere in the interface. You can also define labels for your views in the joins. This is really useful if you're joining the same view multiple times and you want your end users to have a different label for each iteration of the view. It's also really useful if you're using the same view across multiple datasets for different purposes. In this case I'm not using the same view multiple times and I've already defined labels on my source views so so we'll leave it as is. The constraint is just your join condition. And now for the relationship you can pick any of one to one, one to many, many to many, or many to one. Finally, I'm going to set up a join type. Let's duplicate this view block so I can use it for my other views. Finally, before I commit this dataset, I wanted to show you how to handle errors in the YAML editor. If I make a mistake when I'm typing, the first thing you'll notice is that you get this red squiggly line under the mistake. You'll also notice on the left here that the object I'm working in has been highlighted red. To get more information about the error, simply hover your cursor over the squiggly line and you'll get this useful tooltip that pops up and shows you exactly what's happened. So let's correct that now and the error is resolved. Let's commit the dataset. So while I've got the name of the dataset and the label set up my file name hasn't been changed. So let's give that a easy to remember name. To remove a dataset or view, right click on it and then choose delete. One of these deletes is the delete we just did, the genres. This other delete is actually what happens when you rename a file. It will essentially copy the existing file to the location with the new name and then delete the original version. So now we've customized the dataset, let's head back to the home page and see how everything looks. This is already added to my project, the customer success training lab, so we're actually ready now to go and explore the catalog for real in a canvas.