The Firebase Realtime Database integration lets you sync personal information between Rownd and a single Firebase Realtime Database (RTDB) instance. When configured, PII stored inside of an RTDB instance will be discoverable and manageable by the data owners. All of this is done by leveraging Google Cloud Functions and the Rownd platform. Check out the Google Cloud prerequisites to ensure your GCP account and Project are ready for the Rownd Integration.

​
Creating the integration

  1. From the Integrations tab in the sidebar, click on the Add Integration button.
  2. Choose the Firebase - Realtime Database connector from the Connector Catalog
  3. Enter a name for your new integration. After you’ve entered a descriptive name, click Next
  4. Click the Begin authentication button and sign in to your Google Account to grant Rownd access to your Firebase project.
    Firebase Realtime Database runs on Google Cloud. Rownd leverages the Google Cloud Platform to receive Realtime Database updates. Rownd will need permission to read Google Cloud Projects, read and write Firebase resources, and create Cloud Functions. For a full list of access that Rownd needs, click here.
    Grant Rownd the access to your Google Account, and then click the Next button when authentication is complete.
  5. Select the Firebase Project, Application, and Database that you’d like to link with Rownd. When finished, click Next
  6. In the Database Settings step, you will provide Rownd with information about where and how personal information is stored within your Firebase Realtime Database. The RTDB structure is essentially a large JSON object. You will specify the JSON path to where the personal information exists. If PII is stored in multiple places within the database, you can add multiple paths to the configuration.

​
Defining User Data Locations

Each entry in the User Data Locations list has three fields; Path, Foreign Key, and Reference.
  • Path (required) - The database location where PII is stored. The value is a JSON path like /path/to/data
For each unique Path that you provide, Rownd will create a Google Cloud Function configured to send updates to Rownd whenever data under that path changes. As you construct this list of paths, consider all database locations where PII is stored.
The Path can include the wildcard character '*' if the PII is stored nested beneath an arbitrary or unique field (like an ID).For instance, PII could be stored at /users/user123/profile or /users/user456/profile. In such a case, you could define the path with the wildcard character to match both paths: /users/*/profile.
The wildcard character '*' can only be used once within a path.
  • Foreign Key - The field name within the database document, or path variable, whose value references another location in the database
This value can have two forms: Object Field The Foreign Key can reference a field within the database object. For instance, if the document at /addresses looks like: 
{
  "userId": "user123",
  "street": "1983 Wolfpack Ln",
  "city": "Raleigh",
  "state": "NC",
  "zip": "27609"
}
A valid Foreign Key would be "userId". The resolved value would be "user123". Path Variable The Foreign Key can reference a path variable. For instance, if the Path is /addresses/{userId}, the Foreign Key can be {userId}. Rownd will interpret the value of the foreign key as its matched value in a resolved path. (e.g. a resolved path of /addresses/user123 would yield a foreign key value of "user123") By convention, Foreign Keys defined with surrounding curly braces {} will be interpreted as path variables. Foreign Keys without curly braces will be assumed document field names.
  • Reference - A database location where the Foreign Key’s resolved value can be used to retrieve additional PII. Reference is only used in combination with a Foreign Key.
This value will always be a JSON path and include the Foreign Key within it surrounded by curly braces {}. For instance, if Foreign Key is "userId", then the path could look something like "/users/{userId}/profile" For more examples of defining user data locations, see the examples section.
  1. Create the Integration
Once you’re satisfied with the database configuration, click on Create to create your Integration. Creation will take a couple of minutes. Rownd is busy creating and configuring resources within your GCP Project. While you wait, feel free to track the progress in the Integrations table by hovering over the Integration’s status. After Creation Database Indexes Depending on the configured paths in step #5 (Database Settings), Rownd may require certain indexes within your Realtime Database Rules. Rownd will show you the expected database indexes in the confirmation screen after you create your new Integration. Transfer those indexes into your database rules via the Firebase UI Find out more about how to create Realtime Database rules here. Attach your Integration to a Rownd Application After creating your Integration, you must attach it to an existing Rownd application.
  1. From the Integrations table, click the overflow icon on your new Integration and select Attach to application
  2. Choose the application from the selector and click Next
  3. Map data between the Rownd application and your new Integration
You will see a list of all fields that exist in your chosen application’s schema. Enter the name of the corresponding field within your Firebase Realtime Database. For instance, you could have a field in your Rownd application field called first_name and a corresponding field in your Realtime Database called firstName. Enter firstName into the input box next to first_name. A mapping for the email field is required. Rownd uses this field to identify and search for users. Rownd can only manage fields for which you have provided a mapping. Therefore, define as many mappings as possible to get the greatest value from Rownd. Once you finish mapping fields, click Save