phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-07-29 15:23:51 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

minor fixes

Browse source 
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
This commit is contained in:
Francisco Javier Arceo 2025-07-02 17:04:35 -04:00
parent febae72092
commit fee32f2248
3 changed files with 13 additions and 21 deletions
Download patch file Download diff file Expand all files Collapse all files

30
llama_stack/providers/remote/vector_io/milvus/milvus.py
View file

@ -41,7 +41,6 @@ VECTOR_INDEX_PREFIX = f"vector_index:milvus:{VERSION}::"
OPENAI_VECTOR_STORES_PREFIX = f"openai_vector_stores:milvus:{VERSION}::" OPENAI_VECTOR_STORES_PREFIX = f"openai_vector_stores:milvus:{VERSION}::"
OPENAI_VECTOR_STORES_FILES_PREFIX = f"openai_vector_stores_files:milvus:{VERSION}::" OPENAI_VECTOR_STORES_FILES_PREFIX = f"openai_vector_stores_files:milvus:{VERSION}::"
OPENAI_VECTOR_STORES_FILES_CONTENTS_PREFIX = f"openai_vector_stores_files_contents:milvus:{VERSION}::" OPENAI_VECTOR_STORES_FILES_CONTENTS_PREFIX = f"openai_vector_stores_files_contents:milvus:{VERSION}::"
from llama_stack.providers.utils.vector_io.chunk_utils import generate_chunk_id
class MilvusIndex(EmbeddingIndex): class MilvusIndex(EmbeddingIndex):
@ -252,16 +251,6 @@ class MilvusVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
key = f"{OPENAI_VECTOR_STORES_PREFIX}{store_id}" key = f"{OPENAI_VECTOR_STORES_PREFIX}{store_id}"
await self.kvstore.delete(key) await self.kvstore.delete(key)
async def _save_openai_vector_store_file(
self, store_id: str, file_id: str, file_info: dict[str, Any], file_contents: list[dict[str, Any]]
) -> None:
"""Save vector store file metadata to Milvus database."""
assert self.kvstore is not None
key = f"{OPENAI_VECTOR_STORES_FILES_PREFIX}{store_id}:{file_id}"
await self.kvstore.set(key=key, value=json.dumps(file_info))
content_key = f"{OPENAI_VECTOR_STORES_FILES_CONTENTS_PREFIX}{store_id}:{file_id}"
await self.kvstore.set(key=content_key, value=json.dumps(file_contents))
async def _load_openai_vector_stores(self) -> dict[str, dict[str, Any]]: async def _load_openai_vector_stores(self) -> dict[str, dict[str, Any]]:
"""Load all vector store metadata from persistent storage.""" """Load all vector store metadata from persistent storage."""
assert self.kvstore is not None assert self.kvstore is not None
@ -274,6 +263,12 @@ class MilvusVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
self, store_id: str, file_id: str, file_info: dict[str, Any], file_contents: list[dict[str, Any]] self, store_id: str, file_id: str, file_info: dict[str, Any], file_contents: list[dict[str, Any]]
) -> None: ) -> None:
"""Save vector store file metadata to Milvus database.""" """Save vector store file metadata to Milvus database."""
if store_id not in self.openai_vector_stores:
store_info = await self._load_openai_vector_stores(store_id)
if not store_info:
logger.error(f"OpenAI vector store {store_id} not found")
raise ValueError(f"No vector store found with id {store_id}")
try: try:
if not await asyncio.to_thread(self.client.has_collection, "openai_vector_store_files"): if not await asyncio.to_thread(self.client.has_collection, "openai_vector_store_files"):
file_schema = MilvusClient.create_schema( file_schema = MilvusClient.create_schema(
@ -314,7 +309,6 @@ class MilvusVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
schema=content_schema, schema=content_schema,
) )
# Save file metadata
file_data = [ file_data = [
{ {
"store_file_id": f"{store_id}_{file_id}", "store_file_id": f"{store_id}_{file_id}",
@ -332,7 +326,7 @@ class MilvusVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
# Save file contents # Save file contents
contents_data = [ contents_data = [
{ {
"chunk_id": generate_chunk_id(file_id, content.get("chunk_id", None)), "chunk_id": content.get("chunk_metadata").get("chunk_id"),
"store_file_id": f"{store_id}_{file_id}", "store_file_id": f"{store_id}_{file_id}",
"store_id": store_id, "store_id": store_id,
"file_id": file_id, "file_id": file_id,
@ -355,7 +349,7 @@ class MilvusVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
if not await asyncio.to_thread(self.client.has_collection, "openai_vector_store_files"): if not await asyncio.to_thread(self.client.has_collection, "openai_vector_store_files"):
return {} return {}
query_filter = f"store_id == '{store_id}' AND file_id == '{file_id}'" query_filter = f"store_file_id == '{store_id}_{file_id}'"
results = await asyncio.to_thread( results = await asyncio.to_thread(
self.client.query, self.client.query,
collection_name="openai_vector_store_files", collection_name="openai_vector_store_files",
@ -380,14 +374,15 @@ class MilvusVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
if not await asyncio.to_thread(self.client.has_collection, "openai_vector_store_files_contents"): if not await asyncio.to_thread(self.client.has_collection, "openai_vector_store_files_contents"):
return [] return []
query_filter = f"store_id == '{store_id}' AND file_id == '{file_id}'" query_filter = (
f"store_id == '{store_id}' AND file_id == '{file_id}' AND store_file_id == '{store_id}_{file_id}'"
)
results = await asyncio.to_thread( results = await asyncio.to_thread(
self.client.query, self.client.query,
collection_name="openai_vector_store_files_contents", collection_name="openai_vector_store_files_contents",
filter=query_filter, filter=query_filter,
output_fields=["chunk_id", "store_id", "file_id", "content"], output_fields=["chunk_id", "store_id", "file_id", "content"],
) )
print(f"\nresults from milvus = {results}\n")
contents = [] contents = []
for result in results: for result in results:
@ -398,8 +393,6 @@ class MilvusVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
logger.error(f"Failed to decode content for store {store_id}, file {file_id}: {e}") logger.error(f"Failed to decode content for store {store_id}, file {file_id}: {e}")
return contents return contents
except Exception as e: except Exception as e:
print(f"failed {e}")
logger.error(f"Error loading openai vector store file contents for {file_id} in store {store_id}: {e}") logger.error(f"Error loading openai vector store file contents for {file_id} in store {store_id}: {e}")
return [] return []
@ -428,7 +421,6 @@ class MilvusVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
async def _delete_openai_vector_store_file_from_storage(self, store_id: str, file_id: str) -> None: async def _delete_openai_vector_store_file_from_storage(self, store_id: str, file_id: str) -> None:
"""Delete vector store file metadata from Milvus database.""" """Delete vector store file metadata from Milvus database."""
print("milvus is trying to delete stuff")
try: try:
if not await asyncio.to_thread(self.client.has_collection, "openai_vector_store_files"): if not await asyncio.to_thread(self.client.has_collection, "openai_vector_store_files"):
return return

3
tests/integration/vector_io/test_openai_vector_stores.py
View file

@ -31,7 +31,7 @@ def skip_if_provider_doesnt_support_openai_vector_stores(client_with_models):
def skip_if_provider_doesnt_support_openai_vector_store_files_api(client_with_models): def skip_if_provider_doesnt_support_openai_vector_store_files_api(client_with_models):
vector_io_providers = [p for p in client_with_models.providers.list() if p.api == "vector_io"] vector_io_providers = [p for p in client_with_models.providers.list() if p.api == "vector_io"]
for p in vector_io_providers: for p in vector_io_providers:
if p.provider_type in ["inline::faiss", "inline::sqlite-vec", "inlihne::milvus"]: if p.provider_type in ["inline::faiss", "inline::sqlite-vec", "inline::milvus"]:
return return
pytest.skip("OpenAI vector stores are not supported by any provider") pytest.skip("OpenAI vector stores are not supported by any provider")
@ -524,7 +524,6 @@ def test_openai_vector_store_attach_files_on_creation(compat_client_with_empty_s
file_ids = valid_file_ids + [failed_file_id] file_ids = valid_file_ids + [failed_file_id]
num_failed = len(file_ids) - len(valid_file_ids) num_failed = len(file_ids) - len(valid_file_ids)
# Create a vector store
vector_store = compat_client.vector_stores.create( vector_store = compat_client.vector_stores.create(
name="test_store", name="test_store",
file_ids=file_ids, file_ids=file_ids,

1
tests/verifications/openai_api/test_responses.py
View file

@ -296,6 +296,7 @@ def test_response_non_streaming_file_search(
if isinstance(openai_client, LlamaStackAsLibraryClient): if isinstance(openai_client, LlamaStackAsLibraryClient):
pytest.skip("Responses API file search is not yet supported in library client.") pytest.skip("Responses API file search is not yet supported in library client.")
print(f"list models = {openai_client.models.list()}")
test_name_base = get_base_test_name(request) test_name_base = get_base_test_name(request)
if should_skip_test(verification_config, provider, model, test_name_base): if should_skip_test(verification_config, provider, model, test_name_base):
pytest.skip(f"Skipping {test_name_base} for model {model} on provider {provider} based on config.") pytest.skip(f"Skipping {test_name_base} for model {model} on provider {provider} based on config.")