Merge branch 'main' into chore/strong-typing

llama_stack/distribution/build_container.sh

@@ -96,7 +96,7 @@ FROM $container_base
 WORKDIR /app
 
 # We install the Python 3.12 dev headers and build tools so that any
-# C‑extension wheels (e.g. polyleven, faiss‑cpu) can compile successfully.
+# C-extension wheels (e.g. polyleven, faiss-cpu) can compile successfully.
 
 RUN dnf -y update && dnf install -y iputils git net-tools wget \
     vim-minimal python3.12 python3.12-pip python3.12-wheel \
@@ -169,7 +169,7 @@ if [ -n "$run_config" ]; then
     echo "Copying external providers directory: $external_providers_dir"
     cp -r "$external_providers_dir" "$BUILD_CONTEXT_DIR/providers.d"
     add_to_container << EOF
-COPY --chmod=g+w providers.d /.llama/providers.d
+COPY providers.d /.llama/providers.d
 EOF
     fi
 

llama_stack/distribution/configure.py

@@ -17,7 +17,7 @@ from llama_stack.distribution.distribution import (
     builtin_automatically_routed_apis,
     get_provider_registry,
 )
-from llama_stack.distribution.stack import replace_env_vars
+from llama_stack.distribution.stack import cast_image_name_to_string, replace_env_vars
 from llama_stack.distribution.utils.config_dirs import EXTERNAL_PROVIDERS_DIR
 from llama_stack.distribution.utils.dynamic import instantiate_class_type
 from llama_stack.distribution.utils.prompt_for_config import prompt_for_config
@@ -164,7 +164,8 @@ def upgrade_from_routing_table(
 def parse_and_maybe_upgrade_config(config_dict: dict[str, Any]) -> StackRunConfig:
     version = config_dict.get("version", None)
     if version == LLAMA_STACK_RUN_CONFIG_VERSION:
-        return StackRunConfig(**replace_env_vars(config_dict))
+        processed_config_dict = replace_env_vars(config_dict)
+        return StackRunConfig(**cast_image_name_to_string(processed_config_dict))
 
     if "routing_table" in config_dict:
         logger.info("Upgrading config...")
@@ -175,4 +176,5 @@ def parse_and_maybe_upgrade_config(config_dict: dict[str, Any]) -> StackRunConfi
     if not config_dict.get("external_providers_dir", None):
         config_dict["external_providers_dir"] = EXTERNAL_PROVIDERS_DIR
 
-    return StackRunConfig(**replace_env_vars(config_dict))
+    processed_config_dict = replace_env_vars(config_dict)
+    return StackRunConfig(**cast_image_name_to_string(processed_config_dict))

llama_stack/distribution/routers/vector_io.py

@@ -5,6 +5,7 @@
 # the root directory of this source tree.
 
 import asyncio
+import uuid
 from typing import Any
 
 from llama_stack.apis.common.content_types import (
@@ -81,6 +82,7 @@ class VectorIORouter(VectorIO):
         embedding_model: str,
         embedding_dimension: int | None = 384,
         provider_id: str | None = None,
+        vector_db_name: str | None = None,
         provider_vector_db_id: str | None = None,
     ) -> None:
         logger.debug(f"VectorIORouter.register_vector_db: {vector_db_id}, {embedding_model}")
@@ -89,6 +91,7 @@ class VectorIORouter(VectorIO):
             embedding_model,
             embedding_dimension,
             provider_id,
+            vector_db_name,
             provider_vector_db_id,
         )
 
@@ -123,7 +126,6 @@ class VectorIORouter(VectorIO):
         embedding_model: str | None = None,
         embedding_dimension: int | None = None,
         provider_id: str | None = None,
-        provider_vector_db_id: str | None = None,
     ) -> VectorStoreObject:
         logger.debug(f"VectorIORouter.openai_create_vector_store: name={name}, provider_id={provider_id}")
 
@@ -135,17 +137,17 @@ class VectorIORouter(VectorIO):
             embedding_model, embedding_dimension = embedding_model_info
             logger.info(f"No embedding model specified, using first available: {embedding_model}")
 
-        vector_db_id = name
+        vector_db_id = f"vs_{uuid.uuid4()}"
         registered_vector_db = await self.routing_table.register_vector_db(
-            vector_db_id,
-            embedding_model,
-            embedding_dimension,
-            provider_id,
-            provider_vector_db_id,
+            vector_db_id=vector_db_id,
+            embedding_model=embedding_model,
+            embedding_dimension=embedding_dimension,
+            provider_id=provider_id,
+            provider_vector_db_id=vector_db_id,
+            vector_db_name=name,
         )
-
         return await self.routing_table.get_provider_impl(registered_vector_db.identifier).openai_create_vector_store(
-            vector_db_id,
+            name=name,
             file_ids=file_ids,
             expires_after=expires_after,
             chunking_strategy=chunking_strategy,

llama_stack/distribution/routing_tables/vector_dbs.py

@@ -36,6 +36,7 @@ class VectorDBsRoutingTable(CommonRoutingTableImpl, VectorDBs):
         embedding_dimension: int | None = 384,
         provider_id: str | None = None,
         provider_vector_db_id: str | None = None,
+        vector_db_name: str | None = None,
     ) -> VectorDB:
         if provider_vector_db_id is None:
             provider_vector_db_id = vector_db_id
@@ -62,6 +63,7 @@ class VectorDBsRoutingTable(CommonRoutingTableImpl, VectorDBs):
             "provider_resource_id": provider_vector_db_id,
             "embedding_model": embedding_model,
             "embedding_dimension": model.metadata["embedding_dimension"],
+            "vector_db_name": vector_db_name,
         }
         vector_db = TypeAdapter(VectorDBWithOwner).validate_python(vector_db_data)
         await self.register_object(vector_db)

llama_stack/distribution/server/server.py

@@ -47,6 +47,7 @@ from llama_stack.distribution.server.routes import (
     initialize_route_impls,
 )
 from llama_stack.distribution.stack import (
+    cast_image_name_to_string,
     construct_stack,
     replace_env_vars,
     validate_env_pair,
@@ -439,7 +440,7 @@ def main(args: argparse.Namespace | None = None):
                     logger.error(f"Error: {str(e)}")
                     sys.exit(1)
         config = replace_env_vars(config_contents)
-        config = StackRunConfig(**config)
+        config = StackRunConfig(**cast_image_name_to_string(config))
 
     # now that the logger is initialized, print the line about which type of config we are using.
     logger.info(log_line)

llama_stack/distribution/stack.py

@@ -267,6 +267,13 @@ def _convert_string_to_proper_type(value: str) -> Any:
     return value
 
 
+def cast_image_name_to_string(config_dict: dict[str, Any]) -> dict[str, Any]:
+    """Ensure that any value for a key 'image_name' in a config_dict is a string"""
+    if "image_name" in config_dict and config_dict["image_name"] is not None:
+        config_dict["image_name"] = str(config_dict["image_name"])
+    return config_dict
+
+
 def validate_env_pair(env_pair: str) -> tuple[str, str]:
     """Validate and split an environment variable key-value pair."""
     try: