Merge branch 'main' into fix-knowledgestorage-default-instantiation

src/crewai/knowledge/knowledge.py

@@ -14,13 +14,13 @@ class Knowledge(BaseModel):
     Knowledge is a collection of sources and setup for the vector store to save and query relevant context.
     Args:
         sources: List[BaseKnowledgeSource] = Field(default_factory=list)
-        storage: KnowledgeStorage = Field(default_factory=KnowledgeStorage)
+        storage: Optional[KnowledgeStorage] = Field(default=None)
         embedder_config: Optional[Dict[str, Any]] = None
     """
 
     sources: List[BaseKnowledgeSource] = Field(default_factory=list)
     model_config = ConfigDict(arbitrary_types_allowed=True)
-    storage: KnowledgeStorage = Field(default_factory=KnowledgeStorage)
+    storage: Optional[KnowledgeStorage] = Field(default=None)
     embedder_config: Optional[Dict[str, Any]] = None
     collection_name: Optional[str] = None
 

src/crewai/knowledge/source/base_file_knowledge_source.py

@@ -22,7 +22,7 @@ class BaseFileKnowledgeSource(BaseKnowledgeSource, ABC):
         default_factory=list, description="The path to the file"
     )
     content: Dict[Path, str] = Field(init=False, default_factory=dict)
-    storage: KnowledgeStorage = Field(default_factory=KnowledgeStorage)
+    storage: Optional[KnowledgeStorage] = Field(default=None)
     safe_file_paths: List[Path] = Field(default_factory=list)
 
     @field_validator("file_path", "file_paths", mode="before")
@@ -62,7 +62,10 @@ class BaseFileKnowledgeSource(BaseKnowledgeSource, ABC):
 
     def _save_documents(self):
         """Save the documents to the storage."""
-        self.storage.save(self.chunks)
+        if self.storage:
+            self.storage.save(self.chunks)
+        else:
+            raise ValueError("No storage found to save documents.")
 
     def convert_to_path(self, path: Union[Path, str]) -> Path:
         """Convert a path to a Path object."""

src/crewai/knowledge/source/base_knowledge_source.py

@@ -16,7 +16,7 @@ class BaseKnowledgeSource(BaseModel, ABC):
     chunk_embeddings: List[np.ndarray] = Field(default_factory=list)
 
     model_config = ConfigDict(arbitrary_types_allowed=True)
-    storage: KnowledgeStorage = Field(default_factory=KnowledgeStorage)
+    storage: Optional[KnowledgeStorage] = Field(default=None)
     metadata: Dict[str, Any] = Field(default_factory=dict)  # Currently unused
     collection_name: Optional[str] = Field(default=None)
 
@@ -46,4 +46,7 @@ class BaseKnowledgeSource(BaseModel, ABC):
         Save the documents to the storage.
         This method should be called after the chunks and embeddings are generated.
         """
-        self.storage.save(self.chunks)
+        if self.storage:
+            self.storage.save(self.chunks)
+        else:
+            raise ValueError("No storage found to save documents.")

src/crewai/project/crew_base.py

@@ -1,6 +1,6 @@
 import inspect
 from pathlib import Path
-from typing import Any, Callable, Dict, List, TypeVar, Union, cast
+from typing import Any, Callable, Dict, TypeVar, cast
 
 import yaml
 from dotenv import load_dotenv
@@ -116,33 +116,13 @@ def CrewBase(cls: T) -> T:
         def _map_agent_variables(
             self,
             agent_name: str,
-            agent_info: Union[Dict[str, Any], List[Dict[str, Any]]],
+            agent_info: Dict[str, Any],
             agents: Dict[str, Callable],
             llms: Dict[str, Callable],
             tool_functions: Dict[str, Callable],
             cache_handler_functions: Dict[str, Callable],
             callbacks: Dict[str, Callable],
         ) -> None:
-            """Maps agent variables from configuration to internal state.
-            
-            Args:
-                agent_name: Name of the agent.
-                agent_info: Configuration as a dictionary or list of configurations.
-                agents: Dictionary of agent functions.
-                llms: Dictionary of LLM functions.
-                tool_functions: Dictionary of tool functions.
-                cache_handler_functions: Dictionary of cache handler functions.
-                callbacks: Dictionary of callback functions.
-            
-            Raises:
-                ValueError: When an empty list is provided as agent_info.
-            """
-            # If agent_info is a list, use the first item as the configuration
-            if isinstance(agent_info, list):
-                if not agent_info:
-                    raise ValueError(f"Empty agent configuration list for agent {agent_name}")
-                agent_info = agent_info[0]
-                
             if llm := agent_info.get("llm"):
                 try:
                     self.agents_config[agent_name]["llm"] = llms[llm]()

tests/project/test_yaml_config.py

@@ -1,115 +0,0 @@
-import os
-import sys
-import tempfile
-from pathlib import Path
-
-import pytest
-import yaml
-
-
-class TestYamlConfig:
-    """Tests for YAML configuration handling."""
-    
-    def test_list_format_in_yaml(self):
-        """Test that list format in YAML is handled correctly."""
-        # Create a test YAML content with list format
-        yaml_content = """
-        test_agent:
-          - name: test_agent
-            role: Test Agent
-            goal: Test Goal
-        """
-        
-        # Parse the YAML content
-        data = yaml.safe_load(yaml_content)
-        
-        # Get the agent_info which should be a list
-        agent_name = "test_agent"
-        agent_info = data[agent_name]
-        
-        # Verify it's a list
-        assert isinstance(agent_info, list)
-        
-        # Create a function that simulates the behavior of _map_agent_variables
-        # with our fix applied
-        def map_agent_variables(agent_name, agent_info):
-            # This is the fix we implemented
-            if isinstance(agent_info, list):
-                if not agent_info:
-                    raise ValueError(f"Empty agent configuration list for agent {agent_name}")
-                agent_info = agent_info[0]
-                
-            # Try to access a dictionary method on agent_info
-            # This would fail with AttributeError if agent_info is still a list
-            value = agent_info.get("name")
-            return value
-        
-        # Call the function - this would raise AttributeError before the fix
-        result = map_agent_variables(agent_name, agent_info)
-        
-    def test_empty_list_in_yaml(self):
-        """Test that empty list in YAML raises appropriate error."""
-        # Create a test YAML content with empty list
-        yaml_content = """
-        test_agent: []
-        """
-        
-        # Parse the YAML content
-        data = yaml.safe_load(yaml_content)
-        
-        # Get the agent_info which should be an empty list
-        agent_name = "test_agent"
-        agent_info = data[agent_name]
-        
-        # Verify it's a list
-        assert isinstance(agent_info, list)
-        assert len(agent_info) == 0
-        
-        # Create a function that simulates the behavior of _map_agent_variables
-        def map_agent_variables(agent_name, agent_info):
-            if isinstance(agent_info, list):
-                if not agent_info:
-                    raise ValueError(f"Empty agent configuration list for agent {agent_name}")
-                agent_info = agent_info[0]
-            return agent_info
-        
-        # Call the function - should raise ValueError
-        with pytest.raises(ValueError, match=f"Empty agent configuration list for agent {agent_name}"):
-            map_agent_variables(agent_name, agent_info)
-    def test_multiple_items_in_list(self):
-        """Test that when multiple items are in the list, the first one is used."""
-        # Create a test YAML content with multiple items in the list
-        yaml_content = """
-        test_agent:
-          - name: first_agent
-            role: First Agent
-            goal: First Goal
-          - name: second_agent
-            role: Second Agent
-            goal: Second Goal
-        """
-        
-        # Parse the YAML content
-        data = yaml.safe_load(yaml_content)
-        
-        # Get the agent_info which should be a list
-        agent_name = "test_agent"
-        agent_info = data[agent_name]
-        
-        # Verify it's a list with multiple items
-        assert isinstance(agent_info, list)
-        assert len(agent_info) > 1
-        
-        # Create a function that simulates the behavior of _map_agent_variables
-        def map_agent_variables(agent_name, agent_info):
-            if isinstance(agent_info, list):
-                if not agent_info:
-                    raise ValueError(f"Empty agent configuration list for agent {agent_name}")
-                agent_info = agent_info[0]
-            return agent_info.get("name")
-        
-        # Call the function - should return name from the first item
-        result = map_agent_variables(agent_name, agent_info)
-        
-        # Verify only the first item was used
-        assert result == "first_agent"