Cloud Firestore + Android is easy

A while ago, Google has released Cloud Firestore. Cloud Firestore is a cloud-based NoSQL database, which Google positions as a replacement for the Realtime Database. In this article I want to tell how to start using it.

Opportunities

Cloud Firestore allows you to store data on a remote server, easily get access to them and monitor changes in real time. The documentation has an excellent comparison of Cloud Firestore and Realtime Database.

Creation and connection to the project

In the Firebase console, select Database and click on Create database. Next, select the access settings. It’s enough test mode for acquaintance, but on the production it is better to approach this issue more seriously. You can read more about access modes here.

To configure the project, we have to make the following steps:

1. Add Firebase to the project according to the instructions from here.
2. Add dependency to app/build.gradle

implementation ‘com.google.firebase:firebase-firestore:21.0.0’

Now everything is ready.

To get acquainted with the basic techniques of working with Cloud Firestore, I made a simple application. The source code for the app can be found here:

LukyanovAnatoliy/CloudFirestoreExample

To make the application work you need to create a project in the Firebase console and add the google-services.json file to the project in Android Studio.

Data storage structure

Firestore uses collections and documents to store data. A document is an entry that contains any fields. Documents are combined into collections. Also, the document may contain nested collections, but on Android it is not supported. If we draw an analogy with the SQL database, the collection is a table, and the document is an entry in this table. One collection may contain documents with a different set of fields.

Receive and write data

In order to get all the documents of a collection, the following code is enough

Here we request all documents from the Tasks collection.

The library allows you to create queries with parameters. The following code shows how to get documents from the collection by condition

Here we request all documents from the Tasks collection, in which the title field corresponds to the value of Task1.

While getting the documents, they can be immediately converted into our data classes

For writing, you have to create a Hashmap with data (where the name of the field acts as a key, and the value of this field as a value) and transfer it to the library. You can see that in the following code

In this example, a new document will be created and the Firestore will generate an id for it. To set your own id you need to do the following

In this case, if there is no document with id equal to New task, then it will be created, and if it is, then the specified fields will be updated.

Another option to create/update a document

Subscribe to changes

Firestore allows you to subscribe to data changes. You can subscribe to changes in the collection as well as changes to a specific document

querySnapshot.documents — contains an updated list of all documents
querySnapshot.documentChanges — contains a list of changes. Each object contains a modified document and a type of change. 3 types of changes are possible:
ADDED — documents added
MODIFIED — documents updated,
REMOVED — documents deleted

Loading large amounts of data

Realtime Database provides more or less convenient mechanism for loading large amounts of data, which based on manually editing a json file and loading it. Firestore does not provide anything. It was very inconvenient to add new documents until I found a way to download a large amount of information as easily as possible. So that you do not have such issues as me, I will attach the instructions below on how to quickly and easily load a large amount of data. The instruction was found on the Internet.

1. Install Node.js and npm
2. Install the firebase-admin package by running the command

npm install firebase-admin — save

3. Generate a json file with collection data. An example can be found in the Tasks.json file.

4. For loading we need an access key. How to get it is well described in this article.

5. In the export.js file fill with your data
require(‘./firestore_key.json’) — file with access key. I put it in the folder with the script
<YOU_DATABASE> — the name of your firestore database
“./json/Tasks.json” — the path to the file in which the data lie
[‘created’] — list of field names with type Timestamp

6. Run script

node export.js

The script uses the practices of dalenguyen.

Summary

I used Cloud Firestore in one of my projects and did not have any serious problems. One of my collections contains about 15,000 documents and requests for it pass fairly quickly and this is without the use of indexes. Using Cloud Firestore together with Room and Remote Config, you can significantly reduce the number of calls to the database and not exceed the free limits. By using a free rate you can read 50,000 documents, write 20,000 and remove 20,000 per a day.

Example of working application