mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-12-15 23:32:38 +00:00
update resolver to only pass vector_stores section of run config
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com> Using Router only from VectorDBs Signed-off-by: Francisco Javier Arceo <farceo@redhat.com> removing model_api from vector store providers Signed-off-by: Francisco Javier Arceo <farceo@redhat.com> fix test Signed-off-by: Francisco Javier Arceo <farceo@redhat.com> updating integration tests Signed-off-by: Francisco Javier Arceo <farceo@redhat.com> special handling for replay mode for available providers Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
This commit is contained in:
Francisco Javier Arceo
parent
24a1430c8b
commit
accc4c437e
46 changed files with 397 additions and 702 deletions
|
|
@ -4,27 +4,14 @@
|
|||
# This source code is licensed under the terms described in the LICENSE file in
|
||||
# the root directory of this source tree.
|
||||
|
||||
from llama_stack.core.datatypes import StackRunConfig
|
||||
from llama_stack.providers.datatypes import Api, ProviderSpec
|
||||
|
||||
from .config import QdrantVectorIOConfig
|
||||
|
||||
|
||||
async def get_adapter_impl(
|
||||
config: QdrantVectorIOConfig, deps: dict[Api, ProviderSpec], run_config: StackRunConfig | None = None
|
||||
):
|
||||
async def get_adapter_impl(config: QdrantVectorIOConfig, deps: dict[Api, ProviderSpec]):
|
||||
from .qdrant import QdrantVectorIOAdapter
|
||||
|
||||
vector_stores_config = None
|
||||
if run_config and run_config.vector_stores:
|
||||
vector_stores_config = run_config.vector_stores
|
||||
|
||||
impl = QdrantVectorIOAdapter(
|
||||
config,
|
||||
deps[Api.inference],
|
||||
deps[Api.models],
|
||||
deps.get(Api.files),
|
||||
vector_stores_config,
|
||||
)
|
||||
impl = QdrantVectorIOAdapter(config, deps[Api.inference], deps.get(Api.files))
|
||||
await impl.initialize()
|
||||
return impl
|
||||
|
|
|
|||
|
|
@ -8,10 +8,7 @@ from typing import Any
|
|||
|
||||
from pydantic import BaseModel
|
||||
|
||||
from llama_stack.providers.utils.kvstore.config import (
|
||||
KVStoreConfig,
|
||||
SqliteKVStoreConfig,
|
||||
)
|
||||
from llama_stack.providers.utils.kvstore.config import KVStoreConfig, SqliteKVStoreConfig
|
||||
from llama_stack.schema_utils import json_schema_type
|
||||
|
||||
|
||||
|
|
@ -34,7 +31,6 @@ class QdrantVectorIOConfig(BaseModel):
|
|||
return {
|
||||
"api_key": "${env.QDRANT_API_KEY:=}",
|
||||
"kvstore": SqliteKVStoreConfig.sample_run_config(
|
||||
__distro_dir__=__distro_dir__,
|
||||
db_name="qdrant_registry.db",
|
||||
__distro_dir__=__distro_dir__, db_name="qdrant_registry.db"
|
||||
),
|
||||
}
|
||||
|
|
|
|||
|
|
@ -16,7 +16,6 @@ from qdrant_client.models import PointStruct
|
|||
from llama_stack.apis.common.errors import VectorStoreNotFoundError
|
||||
from llama_stack.apis.files import Files
|
||||
from llama_stack.apis.inference import Inference, InterleavedContent
|
||||
from llama_stack.apis.models import Models
|
||||
from llama_stack.apis.vector_dbs import VectorDB
|
||||
from llama_stack.apis.vector_io import (
|
||||
Chunk,
|
||||
|
|
@ -25,17 +24,12 @@ from llama_stack.apis.vector_io import (
|
|||
VectorStoreChunkingStrategy,
|
||||
VectorStoreFileObject,
|
||||
)
|
||||
from llama_stack.core.datatypes import VectorStoresConfig
|
||||
from llama_stack.log import get_logger
|
||||
from llama_stack.providers.datatypes import VectorDBsProtocolPrivate
|
||||
from llama_stack.providers.inline.vector_io.qdrant import QdrantVectorIOConfig as InlineQdrantVectorIOConfig
|
||||
from llama_stack.providers.utils.kvstore import kvstore_impl
|
||||
from llama_stack.providers.utils.memory.openai_vector_store_mixin import OpenAIVectorStoreMixin
|
||||
from llama_stack.providers.utils.memory.vector_store import (
|
||||
ChunkForDeletion,
|
||||
EmbeddingIndex,
|
||||
VectorDBWithIndex,
|
||||
)
|
||||
from llama_stack.providers.utils.memory.vector_store import ChunkForDeletion, EmbeddingIndex, VectorDBWithIndex
|
||||
|
||||
from .config import QdrantVectorIOConfig as RemoteQdrantVectorIOConfig
|
||||
|
||||
|
|
@ -100,8 +94,7 @@ class QdrantIndex(EmbeddingIndex):
|
|||
chunk_ids = [convert_id(c.chunk_id) for c in chunks_for_deletion]
|
||||
try:
|
||||
await self.client.delete(
|
||||
collection_name=self.collection_name,
|
||||
points_selector=models.PointIdsList(points=chunk_ids),
|
||||
collection_name=self.collection_name, points_selector=models.PointIdsList(points=chunk_ids)
|
||||
)
|
||||
except Exception as e:
|
||||
log.error(f"Error deleting chunks from Qdrant collection {self.collection_name}: {e}")
|
||||
|
|
@ -134,12 +127,7 @@ class QdrantIndex(EmbeddingIndex):
|
|||
|
||||
return QueryChunksResponse(chunks=chunks, scores=scores)
|
||||
|
||||
async def query_keyword(
|
||||
self,
|
||||
query_string: str,
|
||||
k: int,
|
||||
score_threshold: float,
|
||||
) -> QueryChunksResponse:
|
||||
async def query_keyword(self, query_string: str, k: int, score_threshold: float) -> QueryChunksResponse:
|
||||
raise NotImplementedError("Keyword search is not supported in Qdrant")
|
||||
|
||||
async def query_hybrid(
|
||||
|
|
@ -162,17 +150,13 @@ class QdrantVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
|
|||
self,
|
||||
config: RemoteQdrantVectorIOConfig | InlineQdrantVectorIOConfig,
|
||||
inference_api: Inference,
|
||||
models_api: Models,
|
||||
files_api: Files | None = None,
|
||||
vector_stores_config: VectorStoresConfig | None = None,
|
||||
) -> None:
|
||||
super().__init__(files_api=files_api, kvstore=None)
|
||||
self.config = config
|
||||
self.client: AsyncQdrantClient = None
|
||||
self.cache = {}
|
||||
self.inference_api = inference_api
|
||||
self.models_api = models_api
|
||||
self.vector_stores_config = vector_stores_config
|
||||
self.vector_db_store = None
|
||||
self._qdrant_lock = asyncio.Lock()
|
||||
|
||||
|
|
@ -187,11 +171,7 @@ class QdrantVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
|
|||
|
||||
for vector_db_data in stored_vector_dbs:
|
||||
vector_db = VectorDB.model_validate_json(vector_db_data)
|
||||
index = VectorDBWithIndex(
|
||||
vector_db,
|
||||
QdrantIndex(self.client, vector_db.identifier),
|
||||
self.inference_api,
|
||||
)
|
||||
index = VectorDBWithIndex(vector_db, QdrantIndex(self.client, vector_db.identifier), self.inference_api)
|
||||
self.cache[vector_db.identifier] = index
|
||||
self.openai_vector_stores = await self._load_openai_vector_stores()
|
||||
|
||||
|
|
@ -200,18 +180,13 @@ class QdrantVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
|
|||
# Clean up mixin resources (file batch tasks)
|
||||
await super().shutdown()
|
||||
|
||||
async def register_vector_db(
|
||||
self,
|
||||
vector_db: VectorDB,
|
||||
) -> None:
|
||||
async def register_vector_db(self, vector_db: VectorDB) -> None:
|
||||
assert self.kvstore is not None
|
||||
key = f"{VECTOR_DBS_PREFIX}{vector_db.identifier}"
|
||||
await self.kvstore.set(key=key, value=vector_db.model_dump_json())
|
||||
|
||||
index = VectorDBWithIndex(
|
||||
vector_db=vector_db,
|
||||
index=QdrantIndex(self.client, vector_db.identifier),
|
||||
inference_api=self.inference_api,
|
||||
vector_db=vector_db, index=QdrantIndex(self.client, vector_db.identifier), inference_api=self.inference_api
|
||||
)
|
||||
|
||||
self.cache[vector_db.identifier] = index
|
||||
|
|
@ -243,12 +218,7 @@ class QdrantVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
|
|||
self.cache[vector_db_id] = index
|
||||
return index
|
||||
|
||||
async def insert_chunks(
|
||||
self,
|
||||
vector_db_id: str,
|
||||
chunks: list[Chunk],
|
||||
ttl_seconds: int | None = None,
|
||||
) -> None:
|
||||
async def insert_chunks(self, vector_db_id: str, chunks: list[Chunk], ttl_seconds: int | None = None) -> None:
|
||||
index = await self._get_and_cache_vector_db_index(vector_db_id)
|
||||
if not index:
|
||||
raise VectorStoreNotFoundError(vector_db_id)
|
||||
|
|
@ -256,10 +226,7 @@ class QdrantVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
|
|||
await index.insert_chunks(chunks)
|
||||
|
||||
async def query_chunks(
|
||||
self,
|
||||
vector_db_id: str,
|
||||
query: InterleavedContent,
|
||||
params: dict[str, Any] | None = None,
|
||||
self, vector_db_id: str, query: InterleavedContent, params: dict[str, Any] | None = None
|
||||
) -> QueryChunksResponse:
|
||||
index = await self._get_and_cache_vector_db_index(vector_db_id)
|
||||
if not index:
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue