Spaces:
Runtime error
Runtime error
File size: 6,282 Bytes
1813edc | 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 | """
Qdrant Manager - Vector database client for RAG persistence and retrieval
Manages collections, embeddings, and search operations
"""
from qdrant_client import QdrantClient
from qdrant_client.models import Distance, VectorParams, PointStruct
from typing import List, Dict, Any, Optional
from uuid import uuid4
from backend.config import settings
class QdrantManager:
"""Singleton for managing Qdrant vector database"""
_instance = None
_client = None
def __new__(cls):
if cls._instance is None:
cls._instance = super().__new__(cls)
cls._instance._initialized = False
return cls._instance
def __init__(self):
if not self._initialized:
# Initialize Qdrant client
self.client = QdrantClient(
url=settings.qdrant_url,
api_key=settings.qdrant_api_key
)
self._initialized = True
self._ensure_collections()
def _ensure_collections(self):
"""Create collections if they don't exist"""
collections = [
settings.kb_collection_name,
settings.history_collection_name
]
for collection_name in collections:
try:
# Try to get collection info
self.client.get_collection(collection_name)
except Exception:
# Create if doesn't exist
self.client.create_collection(
collection_name=collection_name,
vectors_config=VectorParams(
size=settings.vector_dimension,
distance=Distance.COSINE
)
)
print(f"✅ Created collection: {collection_name}")
def add_to_kb(self, documents: List[Dict[str, Any]]) -> None:
"""
Add documents to knowledge base collection
Args:
documents: List of dicts with 'id', 'text', 'embedding' keys
"""
from rag.embedding_manager import embedding_manager
points = []
for doc in documents:
embedding = embedding_manager.embed(doc['text'])
point = PointStruct(
id=int(uuid4().int % (10**8)),
vector=embedding,
payload={
"text": doc['text'],
"source": doc.get('source', 'unknown'),
"document_id": doc.get('document_id', str(uuid4()))
}
)
points.append(point)
self.client.upsert(
collection_name=settings.kb_collection_name,
points=points
)
print(f"✅ Added {len(documents)} documents to KB collection")
def search_kb(self, query: str, limit: int = 3) -> List[Dict[str, Any]]:
"""
Search knowledge base
Args:
query: Search query text
limit: Number of results to return
Returns:
List of similar documents
"""
from rag.embedding_manager import embedding_manager
query_embedding = embedding_manager.embed(query)
results = self.client.query_points(
collection_name=settings.kb_collection_name,
query=query_embedding,
limit=limit
).points
return [
{
"text": r.payload['text'],
"source": r.payload.get('source', 'unknown'),
"score": r.score
}
for r in results
]
def search_history(self, query: str, customer_id: str, limit: int = 3) -> List[Dict[str, Any]]:
"""
Search customer history with customer_id filter
Args:
query: Search query text
customer_id: Filter by customer
limit: Number of results
Returns:
List of similar history records for customer
"""
from rag.embedding_manager import embedding_manager
query_embedding = embedding_manager.embed(query)
results = self.client.query_points(
collection_name=settings.history_collection_name,
query=query_embedding,
limit=limit,
query_filter={
"must": [
{
"key": "customer_id",
"match": {"value": customer_id}
}
]
}
).points
return [
{
"text": r.payload['text'],
"customer_id": r.payload.get('customer_id'),
"interaction_type": r.payload.get('interaction_type'),
"score": r.score
}
for r in results
]
def add_to_history(self, customer_id: str, text: str, interaction_type: str) -> None:
"""
Add conversation to customer history
Args:
customer_id: Customer identifier
text: Conversation text
interaction_type: Type of interaction (e.g., 'complaint', 'refund_request')
"""
from rag.embedding_manager import embedding_manager
embedding = embedding_manager.embed(text)
point = PointStruct(
id=int(uuid4().int % (10**8)),
vector=embedding,
payload={
"customer_id": customer_id,
"text": text,
"interaction_type": interaction_type,
"timestamp": str(__import__('datetime').datetime.now())
}
)
self.client.upsert(
collection_name=settings.history_collection_name,
points=[point]
)
def get_collection_info(self, collection_name: str) -> Dict[str, Any]:
"""Get collection statistics"""
info = self.client.get_collection(collection_name)
return {
"name": collection_name,
"points_count": info.points_count,
"vectors_count": info.vectors_count
}
# Global singleton instance
qdrant_manager = QdrantManager()
|