Validator

Allows you to define custom logic to determine whether your local data is still valid or if it needs to be refreshed from the remote source using the Fetcher.

Purpose of the Validator

Data Freshness: Ensures the data served to your application is up-to-date and meets your application’s specific validity criteria.
Optimizing Network Usage: Prevents unnecessary network calls by using valid cached data when appropriate.
Consistency Control: Gives you fine-grained control over when to refresh data, enhancing consistency between the client and server.

APIs

Validator

Validator has the following structure:

kotlin

interface Validator<Output : Any> {
    suspend fun isValid(item: Output): Boolean
}

Parameter: Output
Type: Any
Required: Required
Description: The type representing your domain data model. For example, if you have a Store<Int, Post>, the Output is Post.
Parameter: isValid(item: Output)
Type: Boolean
Required: Optional
Description: A suspending function that determines whether the given item is still valid.

Data Flow

Data Retrieval Request
1
Data Retrieval Request
When your application requests cached data from the Store, it first checks the Memory Cache and the Source of Truth to see if the data is available.
Note
The Validator only operates on data from the Source of Truth and does not validate data coming directly from the network.
Validation Check
2
Validation Check
The Validator's isValid method is called with the cached data as the parameter.
Validity Determination
3
Validity Determination
If isValid returns true, the Store considers the cached data valid and servies it to the application.

If isValid returns false, the Store deems the cached data invalid and proceeds to fetch fresh data from the remote source using the Fetcher.
Data Update
4
Data Update
If new data is fetched, it's stored in the Memory Cache and the Source of Truth for future requests.

Implementing a Validator

You can create a Validator using the Validator.by factory method:

kotlin

val validator = Validator.by { item ->
    // Your custom validation logic here
}

Examples

Time-Based Validation

If your application has data that should be refreshed every 24 hours:

kotlin

val validator = Validator.by { item ->
    Clock.System.now() < item.expiresAt
}

Versioning

If your application data model changes and you need to invalidate old cached data:

kotlin

val validator = Validator.by { item ->
    item.version == CURRENT_VERSION
}

User Authentication

In cases where authentication tokens expire:

kotlin

val validator = Validator.by { item ->
    !item.token.isExpired()
}

Best Practices

Keep Validation Logic Lightweight: The isValid function should execute quickly to avoid slowing down data retrieval. Complex computations or I/O operations should be avoided.
No Side Effects: The Validator should not modify the data or state. It should only assess the validity of the provided item.
Decide on Validity Criteria: Clearly define what makes data valid or invalid in your application’s context. This could be based on timestamps, data content, user preferences, or other domain-specific factors.