Applications often have the concept of "saved searches", where their users' historical queries are saved. An extension of this feature, is the ability to receive a notification when new "items" are added which match any of these historical, saved searches.
To put it more tangibly, let's say you run a real estate listing application. Your client is on the market for a house in the $500k - $750k range. Whenever they create a new search with different criteria, maybe one search includes a garage and the other has no HOA fees, you will be logging each of these deviations in a "historical queries" collection. Now whenever a new house is added to the homes collection, your system will search this new house through the entire historical queries collection, find any matches, and notify the user.
Using PyMongo's Logger class, we will register our own logger via:
monitoring.register(QueryLogger())So now, whenever a query is sent to the database, we are copying it in a new collection.
watch