Spaces:
Sleeping
Sleeping
File size: 12,099 Bytes
8bf4d58 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 |
"""Main orchestrator for coordinating all components."""
import logging
from typing import Dict, Any, Optional
from enum import Enum
from src.core.config import get_settings
from src.retrieval.vector_store import get_vector_store
from src.agents.local_data_agent import LocalDataAgent
from src.agents.search_agent import SearchAgent
from src.agents.cloud_agent import CloudAgent
from src.agents.aggregator_agent import AggregatorAgent
from src.agents.snowflake_agent import SnowflakeAgent
from src.tools.calculator import get_calculator
from src.tools.web_search import get_web_search
from src.tools.database_query import get_database_query
from openai import OpenAI
logger = logging.getLogger(__name__)
class Tier(Enum):
"""System tiers."""
BASIC_RAG = "basic"
AGENT_WITH_TOOLS = "agent"
ADVANCED_AGENTIC = "advanced"
class Orchestrator:
"""Main orchestrator for the RAG system."""
def __init__(self):
"""Initialize orchestrator."""
self.settings = get_settings()
self.client = OpenAI(**self.settings.get_openai_client_kwargs())
self.model = self.settings.openai_model
# Initialize components
self.vector_store = get_vector_store()
# Initialize agents (lazy loading)
self._local_agent: Optional[LocalDataAgent] = None
self._search_agent: Optional[SearchAgent] = None
self._cloud_agent: Optional[CloudAgent] = None
self._snowflake_agent: Optional[SnowflakeAgent] = None
self._aggregator_agent: Optional[AggregatorAgent] = None
# Initialize tools
self.calculator = get_calculator()
self.web_search = get_web_search()
self.database_query = get_database_query()
async def process_query(
self,
query: str,
tier: str = "basic",
session_id: Optional[str] = None,
) -> Dict[str, Any]:
"""
Process a query using the specified tier.
Args:
query: User query
tier: System tier ("basic", "agent", or "advanced")
session_id: Optional session ID for memory
Returns:
Response dictionary
"""
try:
tier_enum = Tier(tier.lower())
if tier_enum == Tier.BASIC_RAG:
return await self._process_basic_rag(query, session_id)
elif tier_enum == Tier.AGENT_WITH_TOOLS:
return await self._process_agent_with_tools(query, session_id)
elif tier_enum == Tier.ADVANCED_AGENTIC:
return await self._process_advanced_agentic(query, session_id)
else:
raise ValueError(f"Unknown tier: {tier}")
except ValueError as e:
logger.error(f"Invalid tier: {e}")
return {
"success": False,
"error": f"Invalid tier: {tier}",
}
except Exception as e:
logger.error(f"Error processing query: {e}")
return {
"success": False,
"error": str(e),
}
async def _process_basic_rag(
self,
query: str,
session_id: Optional[str],
) -> Dict[str, Any]:
"""Process query using basic RAG (retrieval + generation)."""
try:
# Check if OpenAI API key is configured
if not self.settings.openai_api_key:
return {
"success": False,
"error": "OpenAI API key not configured. Please set OPENAI_API_KEY in your .env file.",
"tier": "basic",
}
# Retrieve relevant documents
results = self.vector_store.search(query=query, n_results=5)
# Build context - use retrieved documents if available, otherwise use empty context
if results["documents"]:
context_parts = ["Retrieved documents:"]
for i, (doc, metadata) in enumerate(
zip(results["documents"], results["metadatas"]), 1
):
source = metadata.get("source", "Unknown")
context_parts.append(f"\n[{i}] Source: {source}")
# Ensure doc is a string
doc_str = str(doc) if doc else ""
context_parts.append(f"Content: {doc_str[:500]}...")
context = "\n".join(context_parts)
sources = [
{"id": id, "metadata": meta}
for id, meta in zip(results["ids"], results["metadatas"])
]
else:
context = "No relevant documents found in the knowledge base."
sources = []
# Generate response using LLM
messages = [
{
"role": "system",
"content": "You are a helpful assistant that answers questions based on the provided context.",
},
{
"role": "user",
"content": f"Context:\n{context}\n\nQuestion: {query}",
},
]
try:
response = self.client.chat.completions.create(
model=self.model,
messages=messages,
temperature=0.7,
)
answer = response.choices[0].message.content
except Exception as api_error:
error_msg = str(api_error)
if "quota" in error_msg.lower() or "429" in error_msg:
raise Exception("OpenAI API quota exceeded. Please check your billing and plan details.")
elif "api key" in error_msg.lower() or "401" in error_msg:
raise Exception("Invalid OpenAI API key. Please check your .env file.")
else:
raise Exception(f"OpenAI API error: {error_msg}")
return {
"success": True,
"answer": answer,
"tier": "basic",
"sources": sources,
"model": self.model,
}
except Exception as e:
logger.error(f"Error in basic RAG: {e}", exc_info=True)
return {
"success": False,
"error": f"Error processing query: {str(e)}",
"tier": "basic",
}
async def _process_agent_with_tools(
self,
query: str,
session_id: Optional[str],
) -> Dict[str, Any]:
"""Process query using agent with tools."""
try:
# Check if OpenAI API key is configured
if not self.settings.openai_api_key:
return {
"success": False,
"error": "OpenAI API key not configured. Please set OPENAI_API_KEY in your .env file.",
"tier": "agent",
}
# Use local agent with tools enabled
if not self._local_agent:
self._local_agent = LocalDataAgent(use_planning=True)
# Add tools to agent
self._local_agent.add_tool(
tool=self.calculator.get_tool_schema(),
tool_function=lambda expression: self.calculator.calculate(expression),
)
if self.settings.has_web_search():
async def web_search_tool(query: str, max_results: int = 5):
return await self.web_search.search(query, max_results)
self._local_agent.add_tool(
tool=self.web_search.get_tool_schema(),
tool_function=web_search_tool,
)
if self.settings.database_url:
def db_query_tool(sql: str, limit: int = 100):
return self.database_query.query(sql, limit)
self._local_agent.add_tool(
tool=self.database_query.get_tool_schema(),
tool_function=db_query_tool,
)
# Process query
response = await self._local_agent.process(query, session_id)
return {
**response,
"tier": "agent",
}
except Exception as e:
logger.error(f"Error in agent with tools: {e}", exc_info=True)
return {
"success": False,
"error": f"Error processing query: {str(e)}",
"tier": "agent",
}
async def _process_advanced_agentic(
self,
query: str,
session_id: Optional[str],
) -> Dict[str, Any]:
"""Process query using advanced agentic RAG with multiple agents."""
try:
# Check if OpenAI API key is configured
if not self.settings.openai_api_key:
return {
"success": False,
"error": "OpenAI API key not configured. Please set OPENAI_API_KEY in your .env file.",
"tier": "advanced",
}
# Use aggregator agent
if not self._aggregator_agent:
self._aggregator_agent = AggregatorAgent(use_planning=True)
# Add Snowflake agent if configured
if self.settings.has_snowflake() and not self._snowflake_agent:
snowflake_config = self.settings.get_snowflake_config()
self._snowflake_agent = SnowflakeAgent(
snowflake_config=snowflake_config,
use_planning=False
)
# Note: AggregatorAgent will automatically discover SnowflakeAgent
# through its agent selection logic
# Process query
response = await self._aggregator_agent.process(query, session_id)
return {
**response,
"tier": "advanced",
}
except Exception as e:
logger.error(f"Error in advanced agentic: {e}", exc_info=True)
return {
"success": False,
"error": f"Error processing query: {str(e)}",
"tier": "advanced",
}
def get_agent_status(self) -> Dict[str, Any]:
"""Get status of all agents."""
status = {
"tiers_available": ["basic", "agent", "advanced"],
"agents": {},
}
if self._local_agent:
status["agents"]["local"] = self._local_agent.get_status()
if self._search_agent:
status["agents"]["search"] = self._search_agent.get_status()
if self._cloud_agent:
status["agents"]["cloud"] = self._cloud_agent.get_status()
if self._snowflake_agent:
status["agents"]["snowflake"] = self._snowflake_agent.get_status()
if self._aggregator_agent:
status["agents"]["aggregator"] = self._aggregator_agent.get_status()
return status
def get_system_info(self) -> Dict[str, Any]:
"""Get system information."""
return {
"vector_store": {
"document_count": self.vector_store.count(),
"collection_name": self.settings.chroma_collection_name,
},
"tools": {
"calculator": True,
"web_search": self.settings.has_web_search(),
"database": bool(self.settings.database_url),
"snowflake": self.settings.has_snowflake(),
},
"memory": {
"short_term_enabled": True,
"long_term_enabled": self.settings.long_term_memory_enabled,
},
"model": self.model,
}
# Global instance
_orchestrator: Optional[Orchestrator] = None
def get_orchestrator() -> Orchestrator:
"""Get or create the global orchestrator instance."""
global _orchestrator
if _orchestrator is None:
_orchestrator = Orchestrator()
return _orchestrator
|