Fix langchain integration test failures + add DBSQL/streamable MCP tests#369

Open

dhruv0811 wants to merge 34 commits intomainfrom

fix-langchain-test-failures

Commits on Mar 13, 2026

Add OBO credential flow integration tests

dhruv0811
and
claude
committed
Fix: use Config(credentials_strategy=...) for ModelServingUserCredentials

dhruv0811
and
claude
committed
Fix: use credentials_strategy (with 's') — the correct WorkspaceClient kwarg

dhruv0811
and
claude
committed
Fix whoami assertions: compare deployer vs end-user SQL results directly

dhruv0811
and
claude
committed
Format test file with ruff

dhruv0811
and
claude
committed
Fix type checker errors: add None guards for SDK optional types

dhruv0811
and
claude
committed
Replace simulated OBO tests with end-to-end agent invocation tests

dhruv0811
and
claude
committed
Add databricks-openai to test dependencies for OBO e2e tests

dhruv0811
and
claude
committed
Add app fixture, serving deploy script, and warm-start for OBO e2e tests

dhruv0811
and
claude
committed
Fix app fixture: add hatch wheel packages config

dhruv0811
and
claude
committed
Add serving agent and deploy script matching working notebook pattern

dhruv0811
and
claude
committed
Fix lint, format, and core_test failures

dhruv0811
and
claude
committed
Fix SP-B identity check: use OBO_TEST_CLIENT_ID directly

dhruv0811
and
claude
committed
Move whoami_serving_agent.py into model_serving_fixture/

dhruv0811
and
claude
committed
Inject warehouse ID at deploy time instead of reading env at import

dhruv0811
and
claude
committed
Fix app whoami tool: return user_name (UUID for SPs) for parity with serving

dhruv0811
and
claude
committed
Fix serving deploy: drop endpoint_name, add input_example

dhruv0811
and
claude
committed
Fix ruff: remove unused imports (shutil, os)

dhruv0811
and
claude
committed
Fix langchain integration test failures

dhruv0811
and
claude
committed
Fix langchain integration test failures

dhruv0811
and
claude
committed
Skip remaining 4 failing tests (streaming usage + mock unit test)

dhruv0811
and
claude
committed
Add DBSQL and raw streamable_http_client MCP integration tests

dhruv0811
and
claude
committed
Add Genie raw streamable_http_client integration test

dhruv0811
and
claude
committed
Refactor raw streamable tests to use pytest.mark.asyncio

dhruv0811
and
claude
committed
Fix langchain integration tests (comprehensive)

dhruv0811
and
claude
committed
Remove RUN_DOGFOOD_TESTS gates — run all tests for evaluation

dhruv0811
and
claude
committed
Remove RUN_ST_ENDPOINT_TESTS gates — run all tests for evaluation

dhruv0811
and
claude
committed
Fix langchain integration tests (clean diff from main)

dhruv0811
and
claude
committed
Fix timeout_and_retries unit test: patch get_openai_client directly

dhruv0811
and
claude
committed
Revert stream_options, fix remaining issues

dhruv0811
and
claude
committed
Fix streaming usage tests: find usage chunk instead of assuming last

dhruv0811
and
claude
committed
Remove OBO file changes that belong to PR #352

dhruv0811
and
claude
committed
Fix ruff formatting

dhruv0811
and
claude
committed
Fix ty errors: add assert for type narrowing in lakebase tests

dhruv0811
and
claude
committed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fix langchain integration test failures + add DBSQL/streamable MCP tests#369

Fix langchain integration test failures + add DBSQL/streamable MCP tests#369
dhruv0811 wants to merge 34 commits intomainfrom
fix-langchain-test-failures

Commits on Mar 13, 2026

Add OBO credential flow integration tests

Fix: use Config(credentials_strategy=...) for ModelServingUserCredentials

Fix: use credentials_strategy (with 's') — the correct WorkspaceClient kwarg

Fix whoami assertions: compare deployer vs end-user SQL results directly

Format test file with ruff

Fix type checker errors: add None guards for SDK optional types

Replace simulated OBO tests with end-to-end agent invocation tests

Add databricks-openai to test dependencies for OBO e2e tests

Add app fixture, serving deploy script, and warm-start for OBO e2e tests

Fix app fixture: add hatch wheel packages config

Add serving agent and deploy script matching working notebook pattern

Fix lint, format, and core_test failures

Fix SP-B identity check: use OBO_TEST_CLIENT_ID directly

Move whoami_serving_agent.py into model_serving_fixture/

Inject warehouse ID at deploy time instead of reading env at import

Fix app whoami tool: return user_name (UUID for SPs) for parity with serving

Fix serving deploy: drop endpoint_name, add input_example

Fix ruff: remove unused imports (shutil, os)

Fix langchain integration test failures

Fix langchain integration test failures

Skip remaining 4 failing tests (streaming usage + mock unit test)

Add DBSQL and raw streamable_http_client MCP integration tests

Add Genie raw streamable_http_client integration test

Refactor raw streamable tests to use pytest.mark.asyncio

Fix langchain integration tests (comprehensive)

Remove RUN_DOGFOOD_TESTS gates — run all tests for evaluation

Remove RUN_ST_ENDPOINT_TESTS gates — run all tests for evaluation

Fix langchain integration tests (clean diff from main)

Fix timeout_and_retries unit test: patch get_openai_client directly

Revert stream_options, fix remaining issues

Fix streaming usage tests: find usage chunk instead of assuming last

Remove OBO file changes that belong to PR #352

Fix ruff formatting

Fix ty errors: add assert for type narrowing in lakebase tests