In this lesson we are going to talk about how we can explore schemas. So there are a few ways that we could look at documents to identify our schema. One way would be just to query for the very first document in our collection. This will give you a good overview of this single document, but it's not going to give you an idea of the shape of all of your documents assuming that they might look different. And moreover it's not going to give you much information about the different fields or different data types on those other documents. One way to get around this would be to use the aggregation command. You could use the aggregation framework to tally up different types and number of fields that you have across all your documents. But that's pretty cumbersome, the best way for you to look at your schema in MongoDB is with MongoDB Compass. Let's tale a look at how it works. So this is the full version of MongoDB Compass. There are two versions of Compass, one is free and one is for people with an Enterprise license. However since you're a part of this course and you're using Compass for educational purposes, you are entitled to use the enterprise version of Compass free of charge. So here I've gone ahead and specified the different information necessary to connect to our course cluster. For password I'm going to go ahead and type in ag-password, and I can go down here and click connect. And after a few moments you can now see all the different collections that we have access to on the course cluster. Now, that I'm connected, I've gone ahead and drilled down to the specific collection. And so you can see that Compass allows us to look at different collections and databases on our MongoDB database. But moreover, it allows us to easily view and explore documents. I can go ahead and look at this individual document and open up different array or object fields. I can really easily open up everything by clicking this button, but this is still just letting us look at our data one document at a time. If we go ahead and go over to the schema tab, and click Analyze Schema. By clicking analyze, Compass is retrieving 1,000 random documents to be used to analyze the schema of this collection. As you may have guessed, the sampling is needed in order to provide a timely analysis. This is because if we wanted to read all the documents from the collection, that can be gigabytes or terabytes of data which would not only be slow but could potentially have negative effects on the server. For each field that was found in the sample documents, you can see not only the field's name but when we drill down, we can also see its type. So this is embedded document, this embedded field is of type string, and etc., etc. Moreover, we're able to see the distribution of these types, so for example, this field is sometimes a document. But 17% of the time, at least for this sample of 1,000, it's actually undefined. This means the field does not exist for those documents. This is a great way to help you analyze if you need to clean or transform the data in your database. Moreover, you can actually query your data, here I select this range and it automatically generates a query. And now I can click Analyze and it will only sample documents that meet the query criteria. And now that I've removed the outliers, you can see the distribution of the values from the sample. Additionally, Compass supports geospatial queries, here you can see all the different geospatial points represented in these documents. And you can actually hold Shift, click, and drag to create a geospatial query that will only find documents within that circle. So in summary, Compass makes it really easy for us to visualize schemas and shape of our documents. It allows us to explore the distributions of our different fields. We can also write queries in Compass by either clicking and dragging on the different fields or writing queries ourselves. But I want to point out that I only covered a few of the features that Compass supports, specifically the ones focused on schema exploration. Compass has many, many more features that I highly recommend you dive into by playing around with the tool.
