Shopper or Merchant? Check-out N1QL Querying
N1QL is a next generation query language for Couchbase Server. It goes beyond SQL and the relational model in several ways--most importantly, attributes in N1QL can contain multiple values, and these values can be nested. In this blog, we will go over some N1QL queries that are commonly seen in an e-commerce application. This kind of app would have a variety of rich data related to products, customers, and purchases.
First, lets get a list of products belonging to a particular category (in this case, I am looking for “appliances”)
UNNEST product.categories as categories
WHERE categories = "Appliances"
I’ve now found an appliance that I’m looking to buy, but before I check-out, I need to log into the store. Your app can store user-profiles in Couchbase in the form of JSON documents as shown below :
"phoneNumber": "1-543-962-9861 x534",
Now, lets get a list of customers and their e-mail addresses. To do that, you can use the following N1QL query :
SELECT firstName || " " || lastName as fullName, emailAddress
|| is used to concatenate strings in N1QL. The result of this query will be something like the follows -
"fullName": "Rosella Tremblay"
"fullName": "Erich Toy"
After I login, I add the item to my shopping cart. Before I check-out, I add a few more items that I like into my shopping cart. I then submit my order. Soon later, the dispatch team is notified that an order is placed and would like to review my purchase order to prepare the package to be mailed out.
The following N1QL query, will prepare the shopping purchase order for “purchase0”
SELECT purchases, product, customer
FROM purchases KEY "purchase0" UNNEST purchases.lineItems AS items
JOIN product KEY items.product
JOIN customer KEY purchases.customerId
Here the document that has purchase information is joined with the product and customer documents to get complete purchase order information.
At the end of the year, the store wants to review its annual sales. The following N1QL query calculates the sales month-over-month.
SELECT substr(purchases.purchasedAt, 0, 7) as month, round(sum(product.unitPrice * items.count)/1000000, 3) as revenueMillion
FROM purchases unnest purchases.lineItems as items join product key items.product
GROUP BY substr(purchases.purchasedAt, 0, 7)ORDER BY month
We only skimmed over a few N1QL queries related to e-commerce. To learn more about N1QL, check out http://query.couchbase.com and don’t forget to register for our upcoming N1QL webinar to go over this use-case in more detail.