feat!: Restructure provider factory, add create methods, migrate to uv by jsonbailey · Pull Request #102 · launchdarkly/python-server-sdk-ai

jsonbailey · 2026-03-10T22:55:37Z

Requirements

I have added test coverage for new or changed functionality
I have followed the repository's pull request submission guidelines
I have validated my changes against all supported platform versions

Related issues

Provide links to any issues in this repository or elsewhere relating to this pull request.

Describe the solution you've provided

This PR contains two major changes:

1. Provider factory restructuring (breaking)

feat!: Rename OpenAIProvider → OpenAIRunnerFactory; rename openai_provider.py → openai_runner_factory.py
feat!: Rename LangChainProvider → LangChainRunnerFactory; rename langchain_provider.py → langchain_runner_factory.py
feat!: Remove abstract static create(ai_config) from AIProvider ABC; see new create_[model|agent|agent_graph]
feat: Add create_model(), create_agent(), create_agent_graph() instance methods to AIProvider ABC (non-abstract, default warns)
feat!: Replace AIProviderFactory with RunnerFactory; delete ai_provider_factory.py
feat: RunnerFactory.create_model/agent/agent_graph() replace async AIProviderFactory.create()
fix: LDClient now calls RunnerFactory.create_model() synchronously instead of awaiting AIProviderFactory.create()

2. Poetry → uv migration

Migrate all packages from Poetry to uv as the package manager
Convert root pyproject.toml to a uv workspace configuration
Switch build backend from poetry-core to hatchling (langchain and openai provider packages)
Update all Makefiles to use uv run / uv sync / uv build
Update CI workflows (.github/workflows/ci.yml, release-please.yml) to use astral-sh/setup-uv instead of actions/setup-python + abatilo/actions-poetry
Update CI composite action (.github/actions/ci/action.yml) accordingly
Update CONTRIBUTING.md with uv-based setup instructions
Update .gitignore (poetry.lock → uv.lock, add .venv/)

Describe alternatives you've considered

Provide a clear and concise description of any alternative solutions or features you've considered.

Additional context

Items for reviewer attention

OpenAI import name inconsistency — In runner_factory.py, the LangChain connector is imported as LangChainRunnerFactory, but the OpenAI connector is still imported as OpenAIProvider. Verify this is intentional and matches the actual export from ldai_openai/__init__.py.
CI action uv sync --all-groups without workspace_path — The old CI action ran poetry install scoped to inputs.workspace_path. The new action runs uv sync --all-groups at the repo root (workspace-level). Confirm this correctly installs deps for the specific package being tested, and doesn't mask missing dependency declarations.
Release workflow changes — release-please.yml removes all explicit actions/setup-python and abatilo/actions-poetry steps, relying on the composite CI action to set up uv. Verify the build/publish steps still produce correct distribution artifacts with hatchling instead of poetry-core.
uv.lock not committed — .gitignore excludes uv.lock, which is appropriate for libraries but means CI won't have reproducible installs. This is a deliberate tradeoff documented in the gitignore comment.

Note

Medium Risk
Introduces a breaking API rename and changes provider instantiation/creation flow, which can affect downstream integrations and runtime provider selection. Also adds new factory entry points (create_agent, create_agent_graph) that are currently provider-dependent and may return None if unsupported.

Overview
Replaces AIProviderFactory with a new RunnerFactory that centralizes provider loading, fallback selection, and capability construction via create_model, create_agent, and create_agent_graph.

Refactors the provider interface: AIProvider now acts as a per-provider factory with non-abstract create_* methods (defaulting to warnings/None), and the SDK client now uses RunnerFactory.create_model when creating chats and judges.

Updates the OpenAI and LangChain provider packages to expose OpenAIRunnerFactory/LangChainRunnerFactory and implement instance create_model (removing the async static create), including OpenAI client reuse/auto-creation and corresponding test updates.

^{Written by Cursor Bugbot for commit 22f67de. This will update automatically on new commits. Configure here.}

Link to Devin session: https://app.devin.ai/sessions/7d5bda4d9dbe4ae0b950b30a50485e60
Requested by: @keelerm84

packages/sdk/server-ai/src/ldai/providers/runner_factory.py

Keep uv-based CI setup (drop poetry steps added in main), accept version bump to 0.16.1 from main while preserving x-release-please-version comment. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

feat!: Rename AIProviderFactory → RunnerFactory feat!: Rename OpenAIProvider → OpenAIRunnerFactory import from ldai_openai.openai_runner_factory feat!: Rename LangChainProvider to LangChainRunnerFactory import from ldai_langchain.langchain_runner_factory feat: Add create_model(), create_agent(), create_agent_graph() to AIProvider ABC (non-abstract, default warns)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Resolve conflicts: - Makefiles (langchain, openai, server-ai): accept main's `uv sync --all-groups` over branch's `--group dev` to pick up the tooling update from main - client.py: keep branch's `RunnerFactory` import (the PR 1 rename) over main's stale `AIProviderFactory` import Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

packages/sdk/server-ai/src/ldai/providers/runner_factory.py

- Remove module-level docstring strings from all modified files (not idiomatic Python) - Replace all "connector" terminology with "provider" in class docstrings, method docstrings, and log warning messages - Fix stale "ExecutorFactory" references in provider class docstrings → RunnerFactory - Add assert guards for Optional[BaseChatModel] before invocation in LangChainRunnerFactory to satisfy mypy - Fix trailing blank lines (W391) across modified files Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

packages/ai-providers/server-ai-openai/src/ldai_openai/openai_runner_factory.py

Completes the rename that was done for the file (openai_provider.py → openai_runner_factory.py) but missed the class itself. Updates all references in __init__.py, runner_factory.py, and tests. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Drop # --- AIProvider factory methods ---, # --- Model invocation ---, # --- Convenience accessors ---, and # --- Public API --- from the factory and provider classes. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

- Remove stale PR 2 migration note from AIProvider docstring - Remove 'two ways' prose from runner factory class and __init__ docstrings - Replace log.warn with log.warning in ai_provider.py Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

cursor

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, have a team admin enable autofix in the Cursor dashboard.}

packages/ai-providers/server-ai-langchain/src/ldai_langchain/langchain_runner_factory.py

🤖 I have created a release *beep* *boop* --- <details><summary>launchdarkly-server-sdk-ai: 0.17.0</summary> ## [0.17.0](launchdarkly-server-sdk-ai-0.16.1...launchdarkly-server-sdk-ai-0.17.0) (2026-04-02) ### ⚠ BREAKING CHANGES * Bump minimum LangChain version to 1.0.0 * Split track_metrics_of into sync and async variants ([#112](#112)) * Remove node-scoped methods from AIGraphTracker (track_node_invocation, track_tool_call, track_node_judge_response), use related AIConfigTracker methods instead * Extract shared utilities to openai_helper * Extract shared utilities to langchain_helper * Restructure provider factory and support additional create methods ([#102](#102)) ### Features * Add get_ai_usage_from_response to openai_helper ([4fab18f](4fab18f)) * Add get_tool_calls_from_response and sum_token_usage_from_messages to langchain_helper ([4fab18f](4fab18f)) * Add LangGraphAgentGraphRunner ([56ce0fd](56ce0fd)) * Add LDAIClient.create_agent() returning ManagedAgent ([53fd95e](53fd95e)) * Add ManagedAgentGraph support ([#111](#111)) ([56ce0fd](56ce0fd)) * Add ModelRunner ABC with invoke_model() and ([453c71c](453c71c)) * Add OpenAIAgentGraphRunner ([56ce0fd](56ce0fd)) * Add OpenAIAgentRunner with agentic tool-calling loop ([53fd95e](53fd95e)) * add optimization package stub ([872e81e](872e81e)) * Add optional graph_key to all LDAIConfigTracker track_* methods for graph correlation ([4fab18f](4fab18f)) * Add track_tool_call/track_tool_calls to LDAIConfigTracker ([4fab18f](4fab18f)) * Adds optimization package stub ([58b7731](58b7731)) * Bump minimum LangChain version to 1.0.0 ([dc592c5](dc592c5)) * Deprecated Chat object in favor of ManagedModel ([453c71c](453c71c)) * Deprecated create_chat(), use create_model() on the LDAIClient ([453c71c](453c71c)) * Drop support for python 3.9 ([#114](#114)) ([dc592c5](dc592c5)) * Extract shared utilities to langchain_helper ([453c71c](453c71c)) * Extract shared utilities to openai_helper ([453c71c](453c71c)) * Introduce ManagedAgent and AgentRunner implementations ([#110](#110)) ([53fd95e](53fd95e)) * Introduce ManagedModel and ModelRunner (PR-3) ([#104](#104)) ([453c71c](453c71c)) * Restructure provider factory and support additional create methods ([#102](#102)) ([e6e4907](e6e4907)) * Split track_metrics_of into sync and async variants ([#112](#112)) ([4fab18f](4fab18f)) ### Bug Fixes * make AIGraphTracker.track_total_tokens accept Optional[TokenUsage], skip when None or total <= 0 ([4fab18f](4fab18f)) * Remove node-scoped methods from AIGraphTracker (track_node_invocation, track_tool_call, track_node_judge_response), use related AIConfigTracker methods instead ([4fab18f](4fab18f)) * use time.perf_counter_ns() for sub-millisecond precision in duration calculations ([4fab18f](4fab18f)) </details> <details><summary>launchdarkly-server-sdk-ai-langchain: 0.4.0</summary> ## [0.4.0](launchdarkly-server-sdk-ai-langchain-0.3.2...launchdarkly-server-sdk-ai-langchain-0.4.0) (2026-04-02) ### ⚠ BREAKING CHANGES * Bump minimum LangChain version to 1.0.0 * Split track_metrics_of into sync and async variants ([#112](#112)) * Remove node-scoped methods from AIGraphTracker (track_node_invocation, track_tool_call, track_node_judge_response), use related AIConfigTracker methods instead * Extract shared utilities to openai_helper * Extract shared utilities to langchain_helper * Restructure provider factory and support additional create methods ([#102](#102)) ### Features * Add get_ai_usage_from_response to openai_helper ([4fab18f](4fab18f)) * Add get_tool_calls_from_response and sum_token_usage_from_messages to langchain_helper ([4fab18f](4fab18f)) * Add LangGraphAgentGraphRunner ([56ce0fd](56ce0fd)) * Add LDAIClient.create_agent() returning ManagedAgent ([53fd95e](53fd95e)) * Add ManagedAgentGraph support ([#111](#111)) ([56ce0fd](56ce0fd)) * Add ModelRunner ABC with invoke_model() and ([453c71c](453c71c)) * Add OpenAIAgentGraphRunner ([56ce0fd](56ce0fd)) * Add OpenAIAgentRunner with agentic tool-calling loop ([53fd95e](53fd95e)) * add optimization package stub ([872e81e](872e81e)) * Add optional graph_key to all LDAIConfigTracker track_* methods for graph correlation ([4fab18f](4fab18f)) * Add track_tool_call/track_tool_calls to LDAIConfigTracker ([4fab18f](4fab18f)) * Adds optimization package stub ([58b7731](58b7731)) * Bump minimum LangChain version to 1.0.0 ([dc592c5](dc592c5)) * Deprecated Chat object in favor of ManagedModel ([453c71c](453c71c)) * Deprecated create_chat(), use create_model() on the LDAIClient ([453c71c](453c71c)) * Drop support for python 3.9 ([#114](#114)) ([dc592c5](dc592c5)) * Extract shared utilities to langchain_helper ([453c71c](453c71c)) * Extract shared utilities to openai_helper ([453c71c](453c71c)) * Introduce ManagedAgent and AgentRunner implementations ([#110](#110)) ([53fd95e](53fd95e)) * Introduce ManagedModel and ModelRunner (PR-3) ([#104](#104)) ([453c71c](453c71c)) * Restructure provider factory and support additional create methods ([#102](#102)) ([e6e4907](e6e4907)) * Split track_metrics_of into sync and async variants ([#112](#112)) ([4fab18f](4fab18f)) ### Bug Fixes * make AIGraphTracker.track_total_tokens accept Optional[TokenUsage], skip when None or total <= 0 ([4fab18f](4fab18f)) * Remove node-scoped methods from AIGraphTracker (track_node_invocation, track_tool_call, track_node_judge_response), use related AIConfigTracker methods instead ([4fab18f](4fab18f)) * use time.perf_counter_ns() for sub-millisecond precision in duration calculations ([4fab18f](4fab18f)) </details> <details><summary>launchdarkly-server-sdk-ai-openai: 0.3.0</summary> ## [0.3.0](launchdarkly-server-sdk-ai-openai-0.2.1...launchdarkly-server-sdk-ai-openai-0.3.0) (2026-04-02) ### ⚠ BREAKING CHANGES * Bump minimum LangChain version to 1.0.0 * Extract shared utilities to openai_helper * Extract shared utilities to langchain_helper * Restructure provider factory and support additional create methods ([#102](#102)) ### Features * Add LangGraphAgentGraphRunner ([56ce0fd](56ce0fd)) * Add LDAIClient.create_agent() returning ManagedAgent ([53fd95e](53fd95e)) * Add ManagedAgentGraph support ([#111](#111)) ([56ce0fd](56ce0fd)) * Add ModelRunner ABC with invoke_model() and ([453c71c](453c71c)) * Add OpenAIAgentGraphRunner ([56ce0fd](56ce0fd)) * Add OpenAIAgentRunner with agentic tool-calling loop ([53fd95e](53fd95e)) * add optimization package stub ([872e81e](872e81e)) * Adds optimization package stub ([58b7731](58b7731)) * Bump minimum LangChain version to 1.0.0 ([dc592c5](dc592c5)) * Deprecated Chat object in favor of ManagedModel ([453c71c](453c71c)) * Deprecated create_chat(), use create_model() on the LDAIClient ([453c71c](453c71c)) * Drop support for python 3.9 ([#114](#114)) ([dc592c5](dc592c5)) * Extract shared utilities to langchain_helper ([453c71c](453c71c)) * Extract shared utilities to openai_helper ([453c71c](453c71c)) * Introduce ManagedAgent and AgentRunner implementations ([#110](#110)) ([53fd95e](53fd95e)) * Introduce ManagedModel and ModelRunner (PR-3) ([#104](#104)) ([453c71c](453c71c)) * Restructure provider factory and support additional create methods ([#102](#102)) ([e6e4907](e6e4907)) </details> <details><summary>launchdarkly-server-sdk-ai-optimization: 0.1.0</summary> ## 0.1.0 (2026-04-02) ### ⚠ BREAKING CHANGES * Bump minimum LangChain version to 1.0.0 ### Features * Add optimization package stub ([#109](#109)) ([ebd5166](ebd5166)) * Adds optimization package stub ([58b7731](58b7731)) * Adds optimization package stub ([cc85a05](cc85a05)) * Bump minimum LangChain version to 1.0.0 ([dc592c5](dc592c5)) * Drop support for python 3.9 ([#114](#114)) ([dc592c5](dc592c5)) ### Bug Fixes * consistency with other makefiles ([b9a5601](b9a5601)) </details> --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please).  --- > [!NOTE] > **Low Risk** > Low risk because this PR only updates release metadata (versions, manifests, provenance docs, and changelogs) without changing runtime code paths. > > **Overview** > Bumps package versions for the release: `launchdarkly-server-sdk-ai` to `0.17.0`, LangChain provider to `0.4.0`, OpenAI provider to `0.3.0`, and introduces/bumps `launchdarkly-server-sdk-ai-optimization` to `0.1.0` (via `.release-please-manifest.json` and each package’s `pyproject.toml`/`__init__.__version__`). > > Updates `CHANGELOG.md` entries and `PROVENANCE.md` version references to match the new release numbers. > > <sup>Written by [Cursor Bugbot](https://cursor.com/dashboard?tab=bugbot) for commit ae3e183. This will update automatically on new commits. Configure [here](https://cursor.com/dashboard?tab=bugbot).</sup>  --------- Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: jsonbailey <jbailey@launchdarkly.com>

jsonbailey added 4 commits March 10, 2026 15:49

chore: Migrate from Poetry to uv workspace

f54d5cf

chore: Update GitHub Actions workflows to use uv

3ab2baf

fix: Specify out-dir in mono repo

1ea080a

address code review feedback

397d2d4

jsonbailey marked this pull request as ready for review March 10, 2026 22:59

jsonbailey requested a review from a team as a code owner March 10, 2026 22:59

jsonbailey changed the title ~~feat: Support addtional create methods for agent and agent_graph~~ feat: Support additional create methods for agent and agent_graph Mar 10, 2026

cursor bot reviewed Mar 10, 2026

View reviewed changes

packages/sdk/server-ai/src/ldai/providers/runner_factory.py Outdated Show resolved Hide resolved

jsonbailey force-pushed the jb/aic-1664/adapter-executor-factory branch from dd194a8 to fd13f66 Compare March 11, 2026 17:25

jsonbailey marked this pull request as draft March 11, 2026 17:26

cursor bot reviewed Mar 11, 2026

View reviewed changes

packages/sdk/server-ai/src/ldai/providers/runner_factory.py Show resolved Hide resolved

jsonbailey force-pushed the jb/aic-1664/adapter-executor-factory branch from fd13f66 to 4bef65c Compare March 11, 2026 17:55

jsonbailey changed the title ~~feat: Support additional create methods for agent and agent_graph~~ feat: Support additional create methods for agent and agent_graph (PR-1) Mar 11, 2026

jsonbailey changed the title ~~feat: Support additional create methods for agent and agent_graph (PR-1)~~ feat: Support additional create methods for agent and agent_graph (PR-2) Mar 11, 2026

jsonbailey and others added 3 commits March 25, 2026 08:54

chore: Merge main and resolve conflicts

6ec32fc

Keep uv-based CI setup (drop poetry steps added in main), accept version bump to 0.16.1 from main while preserving x-release-please-version comment. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

always install

dee9255

jsonbailey force-pushed the jb/aic-1664/adapter-executor-factory branch from 4bef65c to e10be52 Compare March 25, 2026 14:33

fix: Remove unnecessary async from RunnerFactory public methods

f5ac6cc

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

jsonbailey marked this pull request as ready for review March 25, 2026 14:56

keelerm84 approved these changes Mar 25, 2026

View reviewed changes

Base automatically changed from jb/aic-1664/uv-conversion to main March 25, 2026 15:51

cursor bot reviewed Mar 25, 2026

View reviewed changes

packages/sdk/server-ai/src/ldai/providers/runner_factory.py Outdated Show resolved Hide resolved

cursor bot reviewed Mar 25, 2026

View reviewed changes

packages/ai-providers/server-ai-openai/src/ldai_openai/openai_runner_factory.py Outdated Show resolved Hide resolved

jsonbailey and others added 3 commits March 25, 2026 11:47

Remove non-idiomatic section header comments

cb64cd3

Drop # --- AIProvider factory methods ---, # --- Model invocation ---, # --- Convenience accessors ---, and # --- Public API --- from the factory and provider classes. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

cursor bot reviewed Mar 25, 2026

View reviewed changes

packages/ai-providers/server-ai-langchain/src/ldai_langchain/langchain_runner_factory.py Show resolved Hide resolved

jsonbailey changed the title ~~feat: Support additional create methods for agent and agent_graph (PR-2)~~ feat!: Restructure provider factory and support additional create methods (PR-2) Mar 25, 2026

jsonbailey merged commit e6e4907 into main Mar 25, 2026
38 checks passed

jsonbailey deleted the jb/aic-1664/adapter-executor-factory branch March 25, 2026 17:20

github-actions bot mentioned this pull request Mar 26, 2026

chore: release main #113

Merged

devin-ai-integration bot changed the title ~~feat!: Restructure provider factory and support additional create methods (PR-2)~~ feat!: Restructure provider factory, add create methods, migrate to uv Apr 1, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat!: Restructure provider factory, add create methods, migrate to uv#102

feat!: Restructure provider factory, add create methods, migrate to uv#102
jsonbailey merged 13 commits intomainfrom
jb/aic-1664/adapter-executor-factory

jsonbailey commented Mar 10, 2026 •

edited by devin-ai-integration bot

Loading

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

cursor bot left a comment

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

jsonbailey commented Mar 10, 2026 • edited by devin-ai-integration bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

1. Provider factory restructuring (breaking)

2. Poetry → uv migration

Items for reviewer attention

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

cursor bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

jsonbailey commented Mar 10, 2026 •

edited by devin-ai-integration bot

Loading