VWO Feature Management and Experimentation SDK for Ruby

The VWO Feature Management and Experimentation SDK (VWO FME Ruby SDK) enables Ruby developers to integrate feature flagging and experimentation into their applications. This SDK provides full control over feature rollout, A/B testing, and event tracking, allowing teams to manage features dynamically and gain insights into user behavior.

• Ruby 2.6 or later

Add this line to your application's Gemfile:

gem 'vwo-fme-ruby-sdk'

Or install it directly:

gem install vwo-fme-ruby-sdk

The following example demonstrates initializing the SDK with a VWO account ID and SDK key, setting a user context, checking if a feature flag is enabled, and tracking a custom event.

require 'vwo'

# Initialize VWO client
vwo_client = VWO.init({
    sdk_key: '32-alpha-numeric-sdk-key',
    account_id: '123456'
})

# Check if feature is enabled for user
user_context = { id: 'unique_user_id' }
flag = vwo_client.get_flag('feature_key', user_context)

if flag.is_enabled
  puts 'Feature is enabled!'

  # Get feature variable
  value = flag.get_variable('feature_variable', 'default_value')
  puts "Variable value: #{value}"
end

# Track an event
vwo_client.track_event('event_name', user_context)

# Set attribute(s)
vwo_client.set_attribute({ attribute_key: 'attribute_value' }, user_context)

To customize the SDK further, additional parameters can be passed to the init method. Here's a table describing each option:

Refer to the official VWO documentation for additional parameter details.

The context object uniquely identifies users and is crucial for consistent feature rollouts. A typical context includes an id for identifying the user. It can also include other attributes that can be used for targeting and segmentation, such as customVariables, userAgent and ipAddress.

The following table explains all the parameters in the context hash:

user_context = {
  id: 'unique_user_id',
  customVariables: { age: 25, location: 'US' },
  userAgent: 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/130.0.0.0 Safari/537.36',
  ipAddress: '1.1.1.1'
}

Feature Flags serve as the foundation for all testing, personalization, and rollout rules within FME. To implement a feature flag, first use the get_flag API to retrieve the flag configuration. The get_flag API provides a simple way to check if a feature is enabled for a specific user and access its variables. It returns a feature flag object that contains methods for checking the feature's status and retrieving any associated variables.

Example usage:

flag = vwo_client.get_flag('feature_key', user_context)
is_enabled = flag.is_enabled

if is_enabled
  puts 'Feature is enabled!'

  # Get and use feature variable with type safety
  variable_value = flag.get_variable('feature_variable', 'default_value')
  puts "Variable value: #{variable_value}"
else
  puts 'Feature is not enabled!'
end

Feature flags can be enhanced with connected metrics to track key performance indicators (KPIs) for your features. These metrics help measure the effectiveness of your testing rules by comparing control versus variation performance, and evaluate the impact of personalization and rollout campaigns. Use the track_event API to track custom events like conversions, user interactions, and other important metrics:

Example usage:

vwo_client.track_event('event_name', user_context, { amount: 49.99 })

See Tracking Conversions documentation for more information.

User attributes provide rich contextual information about users, enabling powerful personalization. The set_attribute method provides a simple way to associate these attributes with users in VWO for advanced segmentation. Here's what you need to know about the method parameters:

vwo_client.set_attribute({ userType: 'paid' }, user_context)

See Pushing Attributes documentation for additional information.

The poll_interval is an optional parameter that allows the SDK to automatically fetch and update settings from the VWO server at specified intervals. Setting this parameter ensures your application always uses the latest configuration. The polling interval can be configured in three ways:

1. Set via SDK options: If poll_interval is specified in the initialization options (must be >= 1000 milliseconds), that interval will be used
2. VWO Application Settings: If configured in your VWO application settings, that interval will be used
3. Default Fallback: If neither of the above is set, a 10 minute (600,000 milliseconds) polling interval is used

Setting this parameter ensures your application always uses the latest configuration by periodically checking for and applying any updates.

# poll_interval is in milliseconds
vwo_client = VWO.init({ account_id: '123456', sdk_key: '32-alpha-numeric-sdk-key', poll_interval: 60000 })

The VWO FME Gateway Service is an optional but powerful component that enhances VWO's Feature Management and Experimentation (FME) SDKs. It acts as a critical intermediary for pre-segmentation capabilities based on user location and user agent (UA). By deploying this service within your infrastructure, you benefit from minimal latency and strengthened security for all FME operations.

The Gateway Service is required in the following scenarios:

• When using pre-segmentation features based on user location or user agent.
• For applications requiring advanced targeting capabilities.
• It's mandatory when using any thin-client SDK (e.g., Go).

The gateway can be customized by passing the gateway_service parameter in the init configuration.

vwo_client = VWO.init({
  account_id: '123456',
  sdk_key: '32-alpha-numeric-sdk-key',
  gateway_service: {
    url: 'http://custom.gateway.com',
  },
});

Refer to the Gateway Documentation for further details.

The SDK operates in a stateless mode by default, meaning each get_flag call triggers a fresh evaluation of the flag against the current user context.

To optimize performance and maintain consistency, you can implement a custom storage mechanism by passing a storage parameter during initialization. This allows you to persist feature flag decisions in your preferred database system (like Redis, MongoDB, or any other data store).

Key benefits of implementing storage:

• Improved performance by caching decisions
• Consistent user experience across sessions
• Reduced load on your application

The storage mechanism ensures that once a decision is made for a user, it remains consistent even if campaign settings are modified in the VWO Application. This is particularly useful for maintaining a stable user experience during A/B tests and feature rollouts.

class StorageConnector
  def get(feature_key, user_id)
    # Return stored data based on feature_key and user_id
  end

  def set(data)
    # Store data using data[:feature_key] and data[:user_id]
  end
end

vwo_client = VWO.init({
    account_id: '123456',
    sdk_key: '32-alpha-numeric-sdk-key',
    storage: StorageConnector.new
})

VWO by default logs all ERROR level messages to your server console. To gain more control over VWO's logging behaviour, you can use the logger parameter in the init configuration.

# Set log level
vwo_client = VWO.init({
    account_id: '123456',
    sdk_key: '32-alpha-numeric-sdk-key',
    logger: {
        level: 'DEBUG'
    }
})

# Set log level
vwo_client = VWO.init({
    account_id: '123456',
    sdk_key: '32-alpha-numeric-sdk-key',
    logger: {
        level: 'DEBUG',
        prefix: 'CUSTOM LOG PREFIX'
    }
})

VWO FME SDKs help you integrate with several third-party tools, be it analytics, monitoring, customer data platforms, messaging, etc., by implementing a very basic and generic callback capable of receiving VWO-specific properties that can then be pushed to any third-party tool.

def callback(data)
    puts "Integration data: #{data}"
end

vwo_client = VWO.init({
    account_id: '123456',
    sdk_key: '32-alpha-numeric-sdk-key',
    integrations: {
        callback: method(:callback)
    }
})

The SDK leverages threading to efficiently manage concurrent operations. Threading is enabled by default, but can be disabled by configuring the threading parameter during initialization. This gives you control over the SDK's concurrency behavior based on your application's needs.

About max_pool_size

The max_pool_size parameter sets the upper limit for the number of background threads the SDK can use to process events concurrently.

• Short summary: It controls how many events can be processed in parallel by background threads.
• Detailed explanation:
  • When threading is enabled, the SDK uses a thread pool to process events (such as tracking calls) in the background.
  • max_pool_size determines the maximum number of threads that can run at the same time.
  • If all threads are busy, new events are placed in the queue (up to max_queue_size).
  • Increasing max_pool_size can improve throughput for high-traffic applications, but may increase resource usage (CPU, memory).
  • Setting it too high may overwhelm your system; setting it too low may cause event processing delays.
  • Adjust this value based on your application's concurrency needs and available resources.

About max_queue_size

The max_queue_size parameter controls how many events can be waiting in the background processing queue managed by the SDK's thread pool.

• Short summary: It sets the upper limit for how many events can be queued for background processing.
• Detailed explanation:
  • When you use threading (enabled by default), events such as tracking calls are placed in a queue to be processed by background threads.
  • max_queue_size determines the maximum number of events that can wait in this queue.
  • If the queue is full (all threads are busy and the queue has reached its limit), any new event will be processed immediately in the main thread (blocking it until done) until space becomes available in the queue.
  • This mechanism helps balance throughput and resource usage, ensuring you don't lose events but also don't overload your system.
  • You can adjust this value based on your application's expected event volume and performance needs.

When threading is disabled, all tracking calls will block the main execution thread until they complete. This means your application will wait for each VWO operation before continuing.

Example showing blocking behavior:

# By disabling threading, the SDK will wait for the response from the server for each tracking call.
vwo_client = VWO.init({
    account_id: '123456',
    sdk_key: '32-alpha-numeric-sdk-key',
    threading: {
        enabled: false
    },
})

Threading in the VWO SDK provides several important benefits:

1. Asynchronous Event Tracking: When enabled, all tracking calls are processed asynchronously in the background. This prevents these network calls from blocking your application's main execution flow.

2. Improved Performance: By processing tracking and network operations in separate threads, your application remains responsive and can continue serving user requests without waiting for VWO operations to complete.

Example of how threading improves performance:

• Without threading: Each tracking call blocks until the server responds
• With threading: Tracking calls return immediately while processing happens in background

The SDK uses a thread pool to manage these concurrent operations efficiently. The default pool size of 5 threads is suitable for most applications, but you can adjust it based on your needs:

# By default, threading is enabled and the max_pool_size is set to 5.
# you can customize the max_pool_size by passing the max_pool_size parameter in the threading configuration.
vwo_client = VWO.init({
    account_id: '123456',
    sdk_key: '32-alpha-numeric-sdk-key',
    threading: {
        enabled: true,
        max_pool_size: 10
    },
})

The batch_event_ configuration allows you to optimize network requests by batching multiple events together. This is particularly useful for high-traffic applications where you want to reduce the number of API calls.

Example usage:

require 'vwo'

# flushCallBack method
def call(error, data)
    # custom implementation here
end

# Initialize VWO client
vwo_client = VWO.init({
    sdk_key: '32-alpha-numeric-sdk-key',
    account_id: '123456',
    batch_event_data: {
        events_per_request: 50, # Optional: 50 events per request (default is 100)
        request_time_interval: 60 # Optional: send events every 60 seconds (default is 600 seconds)
        flush_callback: method(:call) # Optional: callback to execute after flush 
    },
})

• You can also manually flush events using the flush_events() method:

vwo_client.flush_events()

The version history tracks changes, improvements, and bug fixes in each version. For a full history, see the CHANGELOG.md.

chmod +x ./start-dev.sh
bash start-dev.sh
bundle install

ruby tests/e2e/run_all_tests.rb

We welcome contributions to improve this SDK! Please read our contributing guidelines before submitting a PR.

Our Code of Conduct outlines expectations for all contributors and maintainers.

Apache License, Version 2.0

Copyright 2025 Wingify Software Pvt. Ltd.