cleveragents-core/features/langchain_chat_provider_coverage.feature

Feature: LangChain chat provider coverage
  As a maintainer focused on provider reliability
  I want Behave scenarios that exercise the LangChain chat provider
  So that uncovered lines in langchain_chat_provider.py gain coverage

  @coverage @langchain
  Scenario: LangChain provider surfaces validation failures as user-facing error
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider generates changes with a validation failure response
    Then the provider response should contain the generated change data and validation error
    And the progress callback should record the LangChain workflow milestones
    And the graph should be invoked with the supplied project context

  @coverage @langchain
  Scenario: LangChain provider returns an error response when the graph invocation fails
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider generates changes and the graph raises an exception
    Then the provider response should capture the graph failure
    And the progress callback should end at 100 percent even on failure

  @coverage @langchain @streaming
  Scenario: LangChain provider streams node events for progress updates
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider streams changes with incremental graph events
    Then the provider response should contain the streamed change payload
    And the progress callback should include streaming milestones
    And the LangChain graph stream should receive the thread-aware configuration
    And the provider should report the estimated token usage

  @coverage @langchain @streaming @api
  Scenario: LangChain provider stream API emits workflow events with fallback token estimation
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider stream API emits workflow events with fallback token estimation
    Then the stream API should emit workflow events with the provider response
    And the progress callback should include streaming milestones
    And the LangChain graph stream should receive the thread-aware configuration
    And the provider response should contain the streamed change payload
    And the provider should report the estimated token usage

  @coverage @langchain @retry
  Scenario: LangChain provider unwraps nested retry failures
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider generates changes and retries exhaust with nested errors
    Then the provider response should capture the nested retry failure
    And the progress callback should end at 100 percent even on failure

  @coverage @langchain @no_progress
  Scenario: LangChain provider handles invalid state data without a progress callback
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider generates changes without a progress callback and the graph returns invalid response data
    Then the provider response should surface the state error without changes
    And the progress callback should not receive updates

  @coverage @langchain @streaming @api @errors
  Scenario: LangChain provider stream API propagates midstream errors with cleanup
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider stream API raises an error after partial events
    Then the stream API should propagate the streaming error to the caller
    And the stream API progress should include the final failure signal

  @coverage @langchain @streaming @errors
  Scenario: LangChain provider records partial streaming progress when the stream fails
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider streams changes and the stream raises an exception after unknown node events
    Then the provider response should capture the graph failure
    And the progress callback should capture partial streaming progress before failure



  @coverage @langchain @streaming @fallback
  Scenario: LangChain provider replays workflow milestones when streaming is disabled
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider streams changes without streaming support
    Then the fallback stream should replay workflow milestones with the response

  @coverage @langchain @openai
  Scenario: LangChain provider records OpenAI callback token usage
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider generates changes with an OpenAI usage callback
    Then the provider should report the callback-derived usage metrics

  @coverage @langchain @usage
  Scenario: LangChain provider skips usage logging when no metrics exist
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider logs usage without token or cost data
    Then the provider should skip emitting usage metrics

  @coverage @langchain @streaming @no_progress
  Scenario: LangChain provider stream API handles non-dict payloads without progress callbacks
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider stream API emits non-dict payloads without a progress callback
    Then the stream API should emit workflow events with the provider response
    And the stream API should preserve non-dict payloads in events
    And the progress callback should not receive updates
    And the LangChain graph stream should receive the thread-aware configuration
    And the provider should report the estimated token usage

  @coverage @langchain @streaming @no_progress @errors
  Scenario: LangChain provider stream API raises errors without progress callbacks
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider stream API raises an error without a progress callback
    Then the stream API should propagate the streaming error to the caller
    And the progress callback should not receive updates

  @coverage @langchain @errors
  Scenario: LangChain provider unwraps opaque nested error containers
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider generates changes with opaque nested error containers
    Then the provider response should expose the opaque nested error message

  @coverage @langchain @errors @depth
  Scenario: LangChain provider truncates overly deep error chains
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider generates changes with overly deep error containers
    Then the provider response should include the truncated deep error context

  @coverage @langchain @errors @exceptions
  Scenario: LangChain provider surfaces exception based errors
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider generates changes with exception based errors
    Then the provider response should capture the graph failure
    And the progress callback should end at 100 percent even on failure

  @coverage @langchain @errors @fallback
  Scenario: LangChain provider stringifies opaque error mappings
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider generates changes with opaque error mappings
    Then the provider response should stringify the opaque error mapping

  @coverage @langchain @errors @fallback
  Scenario: LangChain provider stringifies numeric error payloads
    Given a LangChain chat provider is configured with a fake LangChain graph
    When the provider generates changes with numeric error payloads
    Then the provider response should capture the graph failure