@featurevisor/roku

BrightScript SDK for Roku is meant to be used with kopytko-framework.

However, if you don't use it, you can simply copy all SDK files and their dependencies to your project (a version will be prepared in the future if anyone is interested).

Visit https://featurevisor.com/docs/sdks/roku for more information.

• Installation
• Usage
• Options
  • bucketKeySeparator
  • configureAndInterceptStaticContext
  • configureBucketKey
  • configureBucketValue
  • context
  • datafile
  • datafileUrl
  • initialFeatures
  • interceptContext
  • logLevel
  • logger
  • onActivation
  • onReady
  • onRefresh
  • onUpdate
  • refreshInterval
  • stickyFeatures
• API
  • f.isEnabled
  • f.getVariation
  • f.getVariable
  • f.activate
  • f.getAllEvaluations
  • f.evaluateFlag
  • f.evaluateVariation
  • f.evaluateVariable
  • f.getFeature
  • f.getContext
  • f.setContext
  • f.onActivation
  • f.onReady
  • f.onRefresh
  • f.onUpdate
  • f.clear
  • f.getRevision
  • f.isReady
  • f.refresh
  • f.setDatafile
  • f.setSticky
  • f.setStickyFeatures (deprecated)
  • f.startRefreshing
  • f.stopRefreshing
• Evaluation object
• Differences from the JS SDK

Installation

npm i -P @featurevisor/roku

Usage

Initialize the SDK (creates FeaturevisorInstance node). For example, in the new MyFeaturevisorNode created:

' @import /components/libs/featurevisor/FeaturevisorSDK.brs from @featurevisor/roku

sub init()
  f = FeaturevisorSDK()
  f.createInstance({
    datafileUrl: "<featurevisor-datafile-url>",
  })
end sub

You can also pass an existing instance to SDK, to not create a new instance, but to use an existing one to use FeaturevisorSDK methods that are invoked on it.

' @import /components/libs/featurevisor/FeaturevisorSDK.brs from @featurevisor/roku

sub init()
  f = FeaturevisorSDK()
  f.createInstance({
    ' options from an existing instance are kept but could be overridden
  }, existingInstance)
end sub

Options

Options you can pass when creating Featurevisor SDK instance:

bucketKeySeparator

• Type: string
• Required: no
• Defaults to: .

configureAndInterceptStaticContext

• Type: associativeArray
• Required: no

The context for configureBucketKey, configureBucketValue, and interceptContext functions, this object will be accessible via m in those functions.

configureBucketKey

• Type: function
• Required: no

Use it to take over bucketing key generation process.

' @import /components/libs/featurevisor/FeaturevisorSDK.brs from @featurevisor/roku

sub init()
  f = FeaturevisorSDK()
  f.createInstance({
    configureBucketKey: function (feature as Dynamic, context as Object, bucketKey as String) as String
      return bucketKey
    end function,
  })
end sub

configureBucketValue

• Type: function
• Required: no

Use it to take over bucketing process.

' @import /components/libs/featurevisor/FeaturevisorSDK.brs from @featurevisor/roku

sub init()
  f = FeaturevisorSDK()
  f.createInstance({
    configureBucketValue: function (feature as Dynamic, context as Object, bucketValue as String) as Integer
      return bucketValue ' 0 to 100000
    end function,
  })
end sub

context

• Type: associativeArray
• Required: no

Set a persistent instance-level context that is automatically merged with per-call context in all evaluation functions. Per-call context values take precedence over instance context values.

' @import /components/libs/featurevisor/FeaturevisorSDK.brs from @featurevisor/roku

sub init()
  f = FeaturevisorSDK()
  f.createInstance({
    datafileUrl: "<featurevisor-datafile-url>",
    context: {
      userId: "user-123",
      country: "nl",
    },
  })
end sub

datafile

• Type: associativeArray
• Required: either datafile or datafileUrl is required

Use it to pass the datafile object directly.

datafileUrl

• Type: string
• Required: either datafile or datafileUrl is required

Use it to pass the URL to fetch the datafile from.

initialFeatures

• Type: associativeArray
• Required: no

Pass set of initial features with their variation and (optional) variables that you want the SDK to return until the datafile is fetched and parsed:

' @import /components/libs/featurevisor/FeaturevisorSDK.brs from @featurevisor/roku

sub init()
  f = FeaturevisorSDK()
  f.createInstance({
    initialFeatures: {
      myFeatureKey: {
        enabled: true,

        ' optional
        variation: "treatment",
        variables: {
          myVariableKey: "my-variable-value",
        },
      },
    },
  })
end sub

interceptContext

• Type: function
• Required: no

Intercept given context before they are used to bucket the user:

' @import /components/libs/featurevisor/FeaturevisorSDK.brs from @featurevisor/roku

sub init()
  defaultContext = {
    platform: "roku",
    locale: "en_US",
    country: "US",
    timezone: "America/New_York",
  }
  f = FeaturevisorSDK()
  f.createInstance({
    configureAndInterceptStaticContext: defaultContext,
    interceptContext: function (context as Object) as Object
      joinedContext = {}
      joinedContext.append(m)
      joinedContext.append(context)

      return joinedContext
    end function,
  })
end sub

logLevel

• Type: string
• Required: no
• Default: "info"
• Allowed values: "debug" | "info" | "warn" | "error" | "fatal"

Controls the minimum severity of log messages printed by the SDK.

f.createInstance({
  datafileUrl: "<featurevisor-datafile-url>",
  logLevel: "debug",
})

logger

• Type: function
• Required: no

Custom log handler. Receives level (string), message (string), and details (associativeArray). When provided, replaces the default print-based output.

f.createInstance({
  datafileUrl: "<featurevisor-datafile-url>",
  logger: function (level as String, message as String, details as Object)
    print "[MyApp] [";level;"] ";message
  end function,
})

onActivation

• Type: associativeArray
• Required: no
• Structure: { callback: function, context?: associativeArray }

Capture activated features along with their evaluated variation:

' @import /components/libs/featurevisor/FeaturevisorSDK.brs from @featurevisor/roku

sub init()
  f = FeaturevisorSDK()
  f.createInstance({
    onActivation: {
      callback: sub (data as Object)
        ' feature has been activated
      end sub,
      context: {}, ' optional context for the callback
    },
  })
end sub

data Object fields:

• captureContext
• feature
• context
• variationValue

captureContext will only contain attributes that are marked as capture: true in the Attributes' YAML files.

onReady

• Type: associativeArray
• Required: no
• Structure: { callback: function, context?: associativeArray }

Triggered maximum once when the SDK is ready to be used.

' @import /components/libs/featurevisor/FeaturevisorSDK.brs from @featurevisor/roku

sub init()
  f = FeaturevisorSDK()
  f.createInstance({
    onReady: {
      callback: sub ()
        ' agent has been createInstanced and it is ready
      end sub,
      context: {}, ' optional context for the callback
    },
  })
end sub

onRefresh

• Type: associativeArray
• Required: no
• Structure: { callback: function, context?: associativeArray }

Triggered every time the datafile is refreshed.

Works only when datafileUrl and refreshInterval are set.

' @import /components/libs/featurevisor/FeaturevisorSDK.brs from @featurevisor/roku

sub init()
  f = FeaturevisorSDK()
  f.createInstance({
    onRefresh: {
      callback: sub ()
        ' datafile has been refreshed
      end sub,
      context: {}, ' optional context for the callback
    },
  })
end sub

onUpdate

• Type: associativeArray
• Required: no
• Structure: { callback: function, context?: associativeArray }

Triggered every time the datafile is refreshed, and the newly fetched datafile is detected to have different content than last fetched one.

Works only when datafileUrl and refreshInterval are set.

' @import /components/libs/featurevisor/FeaturevisorSDK.brs from @featurevisor/roku

sub init()
  f = FeaturevisorSDK()
  f.createInstance({
    onUpdate: {
      callback: sub ()
        ' datafile has been updated (the revision has changed)
      end sub,
      context: {}, ' optional context for the callback
    },
  })
end sub

refreshInterval

• Type: integer (in seconds)
• Required: no

Set the interval grater than zero to refresh the datafile.

' @import /components/libs/featurevisor/FeaturevisorSDK.brs from @featurevisor/roku

sub init()
  f = FeaturevisorSDK()
  f.createInstance({
    datafileUrl: "<featurevisor-datafile-url>",
    refreshInterval: 60 * 5, ' every 5 minutes
  })
end sub

stickyFeatures

• Type: associativeArray
• Required: no

If set, the SDK will skip evaluating the datafile and return variation and variable results from this object instead.

If a feature key is not present in this object, the SDK will continue to evaluate the datafile.

' @import /components/libs/featurevisor/FeaturevisorSDK.brs from @featurevisor/roku

sub init()
  f = FeaturevisorSDK()
  f.createInstance({
    stickyFeatures: {
      myFeatureKey: {
        enabled: true,

        ' optional
        variation: "treatment",
        variables: {
          myVariableKey: "my-variable-value",
        },
      },
    },
  })
end sub

API

f.isEnabled

Check if a feature is enabled or not.

f.isEnabled(featureKey as String, context = {} as Object) as Boolean

f.getVariation

Get feature variation.

f.getVariation(feature as Dynamic, context = {} as Object) as Dynamic

f.getVariable

Get feature variable.

f.getVariable(feature as Dynamic, variableKey as String, context = {} as Object) as Dynamic

Also supports additional type specific methods, returns the value of the desired type, or Invalid if the value does not exist or it does not have a desired type:

• f.getVariableBoolean(feature as Dynamic, variableKey as String, context = {} as Object) as Dynamic
• f.getVariableString(feature as Dynamic, variableKey as String, context = {} as Object) as Dynamic
• f.getVariableInteger(feature as Dynamic, variableKey as String, context = {} as Object) as Dynamic
• f.getVariableDouble(feature as Dynamic, variableKey as String, context = {} as Object) as Dynamic
• f.getVariableArray(feature as Dynamic, variableKey as String, context = {} as Object) as Dynamic
• f.getVariableObject(feature as Dynamic, variableKey as String, context = {} as Object) as Dynamic
• f.getVariableJSON(feature as Dynamic, variableKey as String, context = {} as Object) as Dynamic

f.activate

Same as getVariation, but also calls the onActivation callback.

This is a convenience method meant to be called when you know the User has been exposed to your Feature, and you also want to track the activation.

f.activate(feature as Dynamic, context = {} as Object) as Object

f.getAllEvaluations

Evaluate all features (or a specified subset) at once and return an associativeArray keyed by feature key.

f.getAllEvaluations(context = {} as Object, featureKeys = [] as Object) as Object

Each entry contains:

• enabled (Boolean) — whether the feature is enabled
• variation (Dynamic) — variation value, if the feature has variations
• variables (associativeArray) — map of all variable values, if the feature has variables

evaluations = f.getAllEvaluations({ userId: "user-123" })
' evaluations = {
'   myFeature: { enabled: true, variation: "control", variables: { color: "blue" } },
'   anotherFeature: { enabled: false },
' }

Pass a list of feature keys to evaluate only a subset:

evaluations = f.getAllEvaluations({ userId: "user-123" }, ["myFeature", "anotherFeature"])

f.evaluateFlag

Returns the full evaluation object for a feature flag, including the reason why the result was produced. Useful for debugging or when you need more than just the boolean result.

f.evaluateFlag(featureKey as String, context = {} as Object) as Object

evaluation = f.evaluateFlag("myFeature", { userId: "user-123" })
' evaluation.enabled => true/false
' evaluation.reason  => "allocated", "forced", "sticky", etc.

f.evaluateVariation

Returns the full evaluation object for a feature variation.

f.evaluateVariation(feature as Dynamic, context = {} as Object) as Object

evaluation = f.evaluateVariation("myFeature", { userId: "user-123" })
' evaluation.variationValue => "control" / "treatment" / Invalid
' evaluation.reason         => "allocated", "forced", "no_match", etc.

f.evaluateVariable

Returns the full evaluation object for a feature variable.

f.evaluateVariable(feature as Dynamic, variableKey as String, context = {} as Object) as Object

evaluation = f.evaluateVariable("myFeature", "color", { userId: "user-123" })
' evaluation.variableValue  => "blue"
' evaluation.reason         => "allocated", "defaulted", "variable_not_found", etc.

f.getFeature

Returns the raw feature definition object from the datafile, or Invalid if the feature is not found.

f.getFeature(feature as Dynamic) as Object

f.getContext

Returns the merged context (instance-level context + per-call context).

f.getContext(context = {} as Object) as Object

mergedContext = f.getContext({ sessionId: "session-456" })

f.setContext

Sets or merges the instance-level context. This context is automatically merged into every evaluation call.

f.setContext(context as Object, replace = false as Boolean)

• replace = false (default): merges context into the existing instance context
• replace = true: replaces the instance context entirely

f.setContext({ userId: "user-123", country: "nl" })

' Later, replace entirely
f.setContext({ userId: "user-456" }, true)

f.onActivation

Adds on activation callback which will be called after an feature activation.

f.onActivation(func as Function, context = Invalid as Object)

f.onReady

Adds on ready callback which will be called after an instance is ready (datafile is saved).

It should be called before createInstance

f.onReady(func as Function, context = Invalid as Object)

f.onRefresh

Adds on refresh callback which will be called after a successful datafile refresh. But the file doesn't need to change.

f.onRefresh(func as Function, context = Invalid as Object)

f.onUpdate

Adds on update callback which will be called after a successful datafile refresh when it has been changed compared to the previous one.

f.onUpdate(func as Function, context = Invalid as Object)

f.clear

Stop refreshing and clear the whole instance. It needs to be initialized once again.

f.clear()

f.getRevision

Get the datafile revision.

f.getRevision() as String

f.isReady

Check if the instance is ready to be used (the datafile is set).

f.isReady() as Boolean

f.refresh

Manually refresh datafile.

f.refresh()

f.setDatafile

Set datafile manually.

f.setDatafile(datafile as Object)

f.setSticky

Set or merge sticky features. When replace is false (default), the provided features are merged with the existing ones. When replace is true, the entire sticky map is replaced.

f.setSticky(stickyFeatures as Object, replace = false as Boolean)

' Merge sticky features
f.setSticky({
  myFeature: { enabled: true, variation: "control" },
})

' Replace all sticky features at once
f.setSticky({
  myFeature: { enabled: false },
}, true)

f.setStickyFeatures

Deprecated — use f.setSticky instead.

Set sticky features, replacing the entire map.

f.setStickyFeatures(stickyFeatures as Object)

f.startRefreshing

Resume or start refreshing if refreshInterval was provided.

f.startRefreshing()

f.stopRefreshing

Stop refreshing.

f.stopRefreshing()

Evaluation object

The evaluateFlag, evaluateVariation, and evaluateVariable methods return an associativeArray with the following shape:

| Field | Type | Description | | --------------- | --------------- | ------------------------------------------------------ | | featureKey | String | The evaluated feature key | | reason | String | Why this result was produced (see table below) | | enabled | Boolean | Whether the feature is enabled (flag evaluations) | | variation | associativeArray | Matched variation object (variation evaluations) | | variationValue| Dynamic | The variation's value | | variableKey | String | The evaluated variable key (variable evaluations) | | variableValue | Dynamic | The resolved variable value | | variableSchema| associativeArray | The variable schema definition | | bucketValue | Integer | The bucket value used (0–100000) | | ruleKey | String | The matched traffic rule key | | error | associativeArray | Error details if reason is "error" |

Evaluation reasons

| Reason | Description | | ---------------------------- | ------------------------------------------------------------------ | | "allocated" | Regular bucketing allocation matched | | "defaulted" | Variable default value used (e.g. useDefaultWhenDisabled: true) | | "disabled" | Feature is disabled | | "error" | An unexpected error occurred during evaluation | | "forced" | Matched a forced rule | | "initial" | Instance not yet ready; using initialFeatures value | | "no_match" | No traffic rule matched | | "no_variations" | Feature has no variations defined | | "feature_not_found" | Feature not found in the datafile | | "variable_not_found" | Variable key not found in the feature's schema | | "out_of_range" | Feature is in a mutually-exclusive group but bucket is out of range| | "override" | Variable overridden by a condition inside a variation (v1 format) | | "required" | A required feature is not enabled | | "rule" | Matched a traffic rule | | "sticky" | Using a sticky feature override | | "variable_default" | Variable falls back to its defaultValue | | "variable_disabled" | Feature is disabled; variable disabledValue is returned | | "variable_override_rule" | Variable overridden directly by a traffic rule | | "variable_override_variation" | Variable overridden by a condition inside a variation | | "variation_disabled" | Feature is disabled; disabledVariationValue is returned |

Differences from the JS SDK

The Roku SDK intentionally diverges from the JavaScript SDK in the following ways:

| Feature | Why | |---------|-----| | datafileUrl + HTTP fetch | Roku apps fetch the datafile themselves; there is no built-in fetch API outside the SDK | | refreshInterval + startRefreshing() / stopRefreshing() | TV apps need periodic polling; no background threads available | | refresh() manual method | Caller-driven refresh on app resume | | onReady / onRefresh / onUpdate / onDatafileChanged / onContextChanged / onStickyChanged callbacks | Roku uses SGNode field observers, not Promises or EventEmitter | | activate() method | Experiment activation tracking for analytics | | interceptContext / configureBucketKey / configureBucketValue callbacks | Legacy callback API; kept alongside the hooks system | | initialFeatures option | Return known defaults before the datafile loads | | spawn() / child instances | No concrete Roku use case identified; deferred — no equivalent implementation planned |

License

MIT © Błażej Chełkowski