CODEWITHSUNDEEP
androidfirebasegoogle-cloud-firestore
Jéwôm'
Jéwôm'

Reputation: 3971

Giving the UID for the document's name

I would like to know what is the best practice to save users preferences in my firestore database. I would try to explain with an example...

Case 1

I have this kind of Document in my "users" Collection (the name is random generated by Firebase) with 3 fields :

In my Android project, when I want to search the Document of the user "DFDDE45554SDC", I search where user_uid = "DFDDE45554SDC".

Case 2

I have this kind of Document in my "users" Collection (the name is created with the UID of the user) with 2 fields :

In my Android project, when I want to search the Document of the user "DFDDE45554SDC", I just search the Document "DFDDE45554SDC".

I specify : I don't want duplicate users. So, what is the best practice (security, optimisation,...) ? Why ?

Upvotes: 3

Views: 1700

Answers (1)

Grimthorr
Grimthorr

Reputation: 6926

I would suggest that Case 2 is more effective, for a few reasons:

  • We already know the user's ID, so don't need to use a different ID here.
  • Using usersCollection.document(userId) is simple to construct and is a direct DocumentReference, rather than a Query, therefore:
    • A DocumentReference can be stored in the Firestore database, whereas a Query cannot.
    • A DocumentReference would likely scale better than instructing the Firestore database to perform a filter query using whereEqualTo("user_uid", userId) (although with indexing, the performance difference is likely negligible at this point).
    • A Query will always return a collection of results (even if there is only 1), rather than the exact document.
  • There isn't currently a need for a different randomly-generated ID for each document within the users collection because the user ID is already unique.
  • You only need 1 document for each user, so this is a sure-fire way to ensure there won't be any duplicates.

The only real incentive I can think of to use Case 1 would be to standardise your document naming scheme with other collections in your database, but this doesn't really matter so much with Firestore.

For a quick example of the two in Android:

Case 1

db.collection("users")
    .whereEqualTo("user_uid", "DFDDE45554SDC")
    .limit(1)
    .get().addOnCompleteListener(new OnCompleteListener<QuerySnapshot>() {
        @Override
        public void onComplete(@NonNull Task<QuerySnapshot> task) {
            if (task.isSuccessful()) {
                for (DocumentSnapshot document : task.getResult()) {
                    // Even with limit(1), we still receive a collection
                    // so iterate this to obtain the desired document
                }
            }
        }
    });

Case 2

db.collection("users")
    .document("DFDDE45554SDC")
    .get().addOnCompleteListener(new OnCompleteListener<DocumentSnapshot>() {
        @Override
        public void onComplete(@NonNull Task<DocumentSnapshot> task) {
            if (task.isSuccessful() && task.getResult() != null) {
                // We have access to the single desired document directly
                DocumentSnapshot document = task.getResult();
            }
        }
    });

Upvotes: 7

Related Questions