File size: 1,751 Bytes
bde2f3a | 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 | """
DataBus RAG Provider — wire the world-class RAG engine into DataBus.
"""
import logging
logger = logging.getLogger("databus.rag_provider")
async def _rag_search_provider(**kwargs) -> dict | None:
"""DataBus provider for hybrid RAG search across all collections."""
from dotenv import load_dotenv
load_dotenv("/root/backend/.env", override=True)
query = kwargs.get("query", kwargs.get("q", ""))
collections = kwargs.get("collections", [])
top_k = int(kwargs.get("limit", 10))
enrich = kwargs.get("enrich", False)
if not query:
return {
"error": "query required",
"collections": [
"rmi_news",
"rmi_scams",
"rmi_research",
"rmi_entities",
"rmi_security",
],
}
try:
from app.databus.rag_engine import ai_enrich, hybrid_search
if isinstance(collections, str):
collections = [c.strip() for c in collections.split(",") if c.strip()]
if not collections:
collections = ["rmi_news"]
results = hybrid_search(query, collections, top_k)
if enrich and results.get("results"):
results["ai_summary"] = ai_enrich(query, results["results"])
return {
"query": query,
"results": results.get("results", []),
"ai_summary": results.get("ai_summary", ""),
"collections_searched": collections,
"total_collections": 5,
"source": "RMI RAG Engine (Qdrant + Ollama embeddings + Redis cache)",
"model": "nomic-embed-text (768d)",
}
except Exception as e:
return {"error": str(e), "source": "RMI RAG Engine"}
|