Skip to content

Conversation

@jsonbailey
Copy link
Contributor

@jsonbailey jsonbailey commented Dec 9, 2025

Note

Adopt FDv1 data system in client

  • Replace LDClient’s direct data source/store wiring with Impl::DataSystem::FDv1; FDv1 wraps the original feature_store once (avoids nested wrappers), injects data_source_update_sink, exposes store, flag_change_broadcaster, and status providers, and cleans up via stop (including store_wrapper.stop).
  • Delegate data_store_status_provider and data_source_status_provider from LDClient to the data system; add flush delegator to the event processor; refine SDK key nil validation for offline/LDD/custom sources.
  • Update readiness and data access: initialized? now compares data_availability vs target_availability; all reads switched to data_system.store in evaluations and all_flags_state.
  • Default data source creation moved into FDv1 (streaming by default; logs when disabling streaming and using polling).

Tests

  • New spec/impl/data_system/fdv1_spec.rb covering store wrapping (no nested wrappers), processor selection (streaming/polling/offline/LDD), custom data source factory/instance handling, start/stop idempotency, availability semantics, diagnostic accumulator integration, and provider/broadcaster access.

Written by Cursor Bugbot for commit 59ae3c0. This will update automatically on new commits. Configure here.

@jsonbailey jsonbailey requested a review from a team as a code owner December 9, 2025 18:50
# @return [Array<EvaluationDetail, [LaunchDarkly::Impl::Model::FeatureFlag, nil], [String, nil]>]
#
def variation_with_flag(key, context, default)
private def variation_with_flag(key, context, default)
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The file used a mix of per method private and all methods after the declaration on line 717 of the original file. Moving to defining per method.

# so we're not connecting to any LD services).
if !config.offline? && sdk_key.nil?
# SDK key can be nil only if using LDD or custom data source with events disabled
if config.send_events || (!config.use_ldd? && config.data_source.nil?)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This conditional doesn't make sense to me.

The SDK key is required if:

  1. We are sending events, or
  2. We aren't in daemon mode AND we don't have a data source?

If we aren't in daemon mode, then we should have a data source, and that's when an SDK key is required. If there isn't a data source, what is the key for?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'll update the comments to clarify what is happening here. Essentially if we are using a custom data source (like FileData) we don't require an SDK key. If the datasource is nil, then we create a default data source (either polling or streaming) which require an sdk key.

#
def initialized?
@config.offline? || @config.use_ldd? || @data_source.initialized?
@data_system.data_availability == @data_system.target_availability
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Are we sure these lines are equivalent given our previous discussions about this? I can't remember exactly where we landed on that.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

They should be equivalent for FDv1. This will need to be altered a bit when we introduce FDv2.

  • Offline returns and expects DEFAULTS
  • Use_Ldd returns and expects REFRESHED (because it uses null processor) ✅
  • If the data_source.Initialized is true the data_system returns REFRESHED and CACHED if false, but we expect REFRESHED

# @return [LaunchDarkly::Interfaces::DataStore::StatusProvider]
#
attr_reader :data_store_status_provider
def data_store_status_provider
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You could also make these delegators if you wanted to tighten the syntax up some. It would be something like:

require 'forwardable'

# ...

extend Forwardable
def_delegators :@data_system, :data_store_status_provider, :data_source_status_provider

@jsonbailey jsonbailey requested a review from keelerm84 December 24, 2025 15:20
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants