feat(tests): make inference_recorder into api_recorder (include tool_invoke) (#3403)

Renames `inference_recorder.py` to `api_recorder.py` and extends it to
support recording/replaying tool invocations in addition to inference
calls.

This allows us to record web-search, etc. tool calls and thereafter
apply recordings for `tests/integration/responses`

## Test Plan

```
export OPENAI_API_KEY=...
export TAVILY_SEARCH_API_KEY=...

./scripts/integration-tests.sh --stack-config ci-tests \
   --suite responses --inference-mode record-if-missing
```
This commit is contained in:
Ashwin Bharambe 2025-10-09 14:27:51 -07:00 committed by GitHub
parent 26fd5dbd34
commit f50ce11a3b
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
284 changed files with 296191 additions and 631 deletions

View file

@ -316,13 +316,13 @@ class Stack:
# asked for in the run config.
async def initialize(self):
if "LLAMA_STACK_TEST_INFERENCE_MODE" in os.environ:
from llama_stack.testing.inference_recorder import setup_inference_recording
from llama_stack.testing.api_recorder import setup_api_recording
global TEST_RECORDING_CONTEXT
TEST_RECORDING_CONTEXT = setup_inference_recording()
TEST_RECORDING_CONTEXT = setup_api_recording()
if TEST_RECORDING_CONTEXT:
TEST_RECORDING_CONTEXT.__enter__()
logger.info(f"Inference recording enabled: mode={os.environ.get('LLAMA_STACK_TEST_INFERENCE_MODE')}")
logger.info(f"API recording enabled: mode={os.environ.get('LLAMA_STACK_TEST_INFERENCE_MODE')}")
dist_registry, _ = await create_dist_registry(self.run_config.metadata_store, self.run_config.image_name)
policy = self.run_config.server.auth.access_policy if self.run_config.server.auth else []
@ -381,7 +381,7 @@ class Stack:
try:
TEST_RECORDING_CONTEXT.__exit__(None, None, None)
except Exception as e:
logger.error(f"Error during inference recording cleanup: {e}")
logger.error(f"Error during API recording cleanup: {e}")
global REGISTRY_REFRESH_TASK
if REGISTRY_REFRESH_TASK: