Outlier Pattern

used when some documents are different and needs a different method to interact. example: most of books have a single digit reviews, so embeded method is used. but one or two hot sales have lots of reviews, so reference method needs to be used.

Code Summary: Outlier Pattern

To apply the Outlier pattern to the review documents in our bookstore app, we added the outlier field to the documents in the reviews collection. This field is set to true when the comments array in a review document exceeds the three comments threshold:

{
  "rating": 1,
  "userId": 318,
  "productId": 34538756,
  "review_text": "Lorem ipsum dolor sit amet...",
  "comments": [{...}, {...}, ... ],
  "outlier": true,
  ...
}

To update the existing documents in the reviews collection we used the updateMany method with two arguments. First a filter to match all documents with a comments array larger than 3 elements. The second argument is an aggregation pipeline with $set operator to add the outlier field and set it to true.

db.reviews.updateMany(
    {
        "comments.3": { $exists: true }
    },
    {
        $set: { outlier: true }
    }
);