Address PR review: Add constants, IPv4 validation, error handling, and expanded tests

Co-Authored-By: Joe Moura <joao@crewai.com>

src/crewai/agent.py

@@ -142,8 +142,13 @@ class Agent(BaseAgent):
                 self.embedder = crew_embedder
 
             if self.knowledge_sources:
-                from crewai.utilities import sanitize_collection_name
-                knowledge_agent_name = sanitize_collection_name(self.role)
+                try:
+                    from crewai.utilities import sanitize_collection_name
+                    knowledge_agent_name = sanitize_collection_name(self.role)
+                except Exception as e:
+                    self._logger.warning(f"Error sanitizing collection name: {e}")
+                    knowledge_agent_name = "default_agent"
+                    
                 if isinstance(self.knowledge_sources, list) and all(
                     isinstance(k, BaseKnowledgeSource) for k in self.knowledge_sources
                 ):

src/crewai/utilities/__init__.py

@@ -7,7 +7,7 @@ from .parser import YamlParser
 from .printer import Printer
 from .prompts import Prompts
 from .rpm_controller import RPMController
-from .string_utils import sanitize_collection_name
+from .string_utils import sanitize_collection_name, is_ipv4_pattern
 from .exceptions.context_window_exceeding_exception import (
     LLMContextLengthExceededException,
 )
@@ -27,4 +27,5 @@ __all__ = [
     "LLMContextLengthExceededException",
     "EmbeddingConfigurator",
     "sanitize_collection_name",
+    "is_ipv4_pattern",
 ]

src/crewai/utilities/string_utils.py

@@ -84,6 +84,28 @@ def interpolate_only(
 
 from typing import Optional
 
+# Constants for ChromaDB collection name requirements
+MIN_LENGTH = 3
+MAX_LENGTH = 63
+DEFAULT_COLLECTION = "default_collection"
+
+# Compiled regex patterns for better performance
+INVALID_CHARS_PATTERN = re.compile(r"[^a-zA-Z0-9_-]")
+IPV4_PATTERN = re.compile(r"^(\d{1,3}\.){3}\d{1,3}$")
+
+
+def is_ipv4_pattern(name: str) -> bool:
+    """
+    Check if a string matches an IPv4 address pattern.
+
+    Args:
+        name: The string to check
+
+    Returns:
+        True if the string matches an IPv4 pattern, False otherwise
+    """
+    return bool(IPV4_PATTERN.match(name))
+
 
 def sanitize_collection_name(name: Optional[str]) -> str:
     """
@@ -101,10 +123,14 @@ def sanitize_collection_name(name: Optional[str]) -> str:
         A sanitized collection name that meets ChromaDB requirements
     """
     if not name:
-        return "default_collection"
+        return DEFAULT_COLLECTION
+
+    # Handle IPv4 pattern
+    if is_ipv4_pattern(name):
+        name = f"ip_{name}"
 
     # Replace spaces and invalid characters with underscores
-    sanitized = re.sub(r"[^a-zA-Z0-9_-]", "_", name)
+    sanitized = INVALID_CHARS_PATTERN.sub("_", name)
 
     # Ensure it starts with alphanumeric
     if not sanitized[0].isalnum():
@@ -114,12 +140,12 @@ def sanitize_collection_name(name: Optional[str]) -> str:
     if not sanitized[-1].isalnum():
         sanitized = sanitized[:-1] + "z"
 
-    # Ensure length is between 3-63 characters
-    if len(sanitized) < 3:
+    # Ensure length is between MIN_LENGTH-MAX_LENGTH characters
+    if len(sanitized) < MIN_LENGTH:
         # Add padding with alphanumeric character at the end
-        sanitized = sanitized + "x" * (3 - len(sanitized))
-    if len(sanitized) > 63:
-        sanitized = sanitized[:63]
+        sanitized = sanitized + "x" * (MIN_LENGTH - len(sanitized))
+    if len(sanitized) > MAX_LENGTH:
+        sanitized = sanitized[:MAX_LENGTH]
         # Ensure it still ends with alphanumeric after truncation
         if not sanitized[-1].isalnum():
             sanitized = sanitized[:-1] + "z"