Spaces:
Sleeping
Sleeping
File size: 9,262 Bytes
e5b884f | 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 | import uuid
from typing import Dict, Optional
from datetime import datetime, timedelta
from app.services.RAG_service import RAGService
from app.database.database import SessionDatabase
from app.schemas.request_models import DocumentTypeSchema
class Session:
def __init__(self, session_id:str):
self.session_id = session_id
self.created_at = datetime.now()
self.last_activity = datetime.now()
self.rag_service : Optional[RAGService] = None
self.document_uploaded = False
self.vector_store_created = False
self.document_info = {}
self.username = None
def update_activity(self):
self.last_activity = datetime.now()
def is_expired(self, timeout_minutes: int = 60) -> bool:
return datetime.now() - self.last_activity > timedelta(minutes=timeout_minutes)
class SessionManager:
def __init__(self):
self.sessions: Dict[str, Session] = {}
self.db = SessionDatabase()
def create_session(self, username: str = "anonymous") -> str:
session_id = str(uuid.uuid4())
session = Session(session_id)
session.username = username
self.sessions[session_id] = session
# Save to database
self.db.create_session(session_id, username)
return session_id
def get_session(self, session_id: str) -> Optional[Session]:
# First check in-memory sessions
if session_id in self.sessions:
session = self.sessions[session_id]
if not session.is_expired():
session.update_activity()
return session
else:
del self.sessions[session_id]
# If not in memory, try to restore from database
db_session = self.db.get_session(session_id)
if db_session and db_session['is_active']:
# Restore session to memory
session = Session(session_id)
session.username = db_session['username']
session.document_uploaded = db_session['chunks_count'] > 0
session.vector_store_created = db_session['pinecone_index'] is not None
session.document_info = {
'filename': db_session['document_name'],
'type': db_session['document_type'],
'chunks_count': db_session['chunks_count']
}
# Initialize RAG service if document exists (same as restore_session)
if session.vector_store_created:
print(f"[SessionManager] Restoring RAG service for session {session_id}")
session.rag_service = RAGService()
# Set the basic attributes
session.rag_service.index = db_session['pinecone_index']
session.rag_service.namespace = db_session['pinecone_namespace']
session.rag_service.Document_path = db_session['document_path']
session.rag_service.url = db_session['document_url']
# Create a mock splitter with the keywords file path for restored sessions
from app.ingestion.text_splitter import splitting_text
from app.utils.metadata_utils import MetadataService
# Recreate the document type information
metadataservice = MetadataService()
mock_scheme = DocumentTypeSchema(document_types="Insurance") # Default for restored sessions
document_type = metadataservice.Return_document_model(mock_scheme)
session.rag_service.DocumentTypeScheme = mock_scheme
session.rag_service.Document_Type = document_type
# Create splitter instance to maintain the keywords file path
session.rag_service.splitter = splitting_text(documentTypeSchema=document_type, llm=session.rag_service.llm)
# Generate the expected keywords file path based on document name
import os
document_name = db_session['document_name'] or 'unknown'
keywords_filename = document_name.replace(".", "").replace("\\", "").replace("/", "") + ".json"
session.rag_service.splitter.Keywordsfile_path = os.path.join("app/data/", keywords_filename)
print(f"[SessionManager] RAG service restored with index: {session.rag_service.index}")
self.sessions[session_id] = session
return session
return None
def update_session_document(self, session_id: str, document_name: str,
document_type: str, chunks_count: int,
pinecone_index: str = None, pinecone_namespace: str = None,
document_path: str = None, document_url: str = None):
"""Update session with document information"""
self.db.update_session(
session_id,
document_name=document_name,
document_type=document_type,
chunks_count=chunks_count,
pinecone_index=pinecone_index,
pinecone_namespace=pinecone_namespace,
document_path=document_path,
document_url=document_url
)
# Update in-memory session if exists
if session_id in self.sessions:
session = self.sessions[session_id]
session.document_uploaded = True
session.vector_store_created = pinecone_index is not None
session.document_info = {
'filename': document_name,
'type': document_type,
'chunks_count': chunks_count
}
def get_user_sessions(self, username: str):
"""Get all sessions for a user"""
return self.db.get_user_sessions(username)
def restore_session(self, session_id: str) -> bool:
"""Restore a session from database"""
db_session = self.db.get_session(session_id)
if db_session and db_session['is_active']:
session = Session(session_id)
session.username = db_session['username']
session.document_uploaded = db_session['chunks_count'] > 0
session.vector_store_created = db_session['pinecone_index'] is not None
session.document_info = {
'filename': db_session['document_name'],
'type': db_session['document_type'],
'chunks_count': db_session['chunks_count']
}
# Initialize RAG service if document exists
if session.vector_store_created:
print(f"[SessionManager] Restoring RAG service for session {session_id}")
session.rag_service = RAGService()
# Set the basic attributes
session.rag_service.index = db_session['pinecone_index']
session.rag_service.namespace = db_session['pinecone_namespace']
session.rag_service.Document_path = db_session['document_path']
session.rag_service.url = db_session['document_url']
# Create a mock splitter with the keywords file path for restored sessions
from app.ingestion.text_splitter import splitting_text
from app.utils.metadata_utils import MetadataService
# Recreate the document type information
metadataservice = MetadataService()
mock_scheme = DocumentTypeSchema(document_types="Insurance") # Default for restored sessions
document_type = metadataservice.Return_document_model(mock_scheme)
session.rag_service.DocumentTypeScheme = mock_scheme
session.rag_service.Document_Type = document_type
# Create splitter instance to maintain the keywords file path
session.rag_service.splitter = splitting_text(documentTypeSchema=document_type, llm=session.rag_service.llm)
# Generate the expected keywords file path based on document name
import os
document_name = db_session['document_name'] or 'unknown'
keywords_filename = document_name.replace(".", "").replace("\\", "").replace("/", "") + ".json"
session.rag_service.splitter.Keywordsfile_path = os.path.join("app/data/", keywords_filename)
print(f"[SessionManager] RAG service restored with index: {session.rag_service.index}")
self.sessions[session_id] = session
return True
return False
def delete_session(self, session_id:str):
if session_id in self.sessions:
del self.sessions[session_id]
self.db.deactivate_session(session_id)
def cleanup_expired_sessions(self):
expired_sessions = [
sid for sid, session in self.sessions.items()
if session.is_expired()
]
for sid in expired_sessions:
del self.sessions[sid]
session_manager = SessionManager()
|