607 lines
20 KiB
Plaintext
607 lines
20 KiB
Plaintext
|
|
import os
|
|||
|
|
import re
|
|||
|
|
import uuid
|
|||
|
|
import shutil
|
|||
|
|
from datetime import datetime
|
|||
|
|
from typing import Optional, Dict, List
|
|||
|
|
import chainlit as cl
|
|||
|
|
import ollama
|
|||
|
|
import fitz # PyMuPDF
|
|||
|
|
from qdrant_client import AsyncQdrantClient
|
|||
|
|
from qdrant_client.models import PointStruct, Distance, VectorParams
|
|||
|
|
from chainlit.data.sql_alchemy import SQLAlchemyDataLayer
|
|||
|
|
from chainlit.data.storage_clients import BaseStorageClient
|
|||
|
|
|
|||
|
|
|
|||
|
|
# === CONFIGURAZIONE ===
|
|||
|
|
DATABASE_URL = os.getenv("DATABASE_URL", "postgresql+asyncpg://ai_user:secure_password_here@postgres:5432/ai_station")
|
|||
|
|
OLLAMA_URL = os.getenv("OLLAMA_URL", "http://192.168.1.243:11434")
|
|||
|
|
QDRANT_URL = os.getenv("QDRANT_URL", "http://qdrant:6333")
|
|||
|
|
WORKSPACES_DIR = "./workspaces"
|
|||
|
|
STORAGE_DIR = "./.files"
|
|||
|
|
|
|||
|
|
os.makedirs(STORAGE_DIR, exist_ok=True)
|
|||
|
|
os.makedirs(WORKSPACES_DIR, exist_ok=True)
|
|||
|
|
|
|||
|
|
|
|||
|
|
# === MAPPING UTENTI E RUOLI ===
|
|||
|
|
USER_PROFILES = {
|
|||
|
|
"giuseppe@defranceschi.pro": {
|
|||
|
|
"role": "admin",
|
|||
|
|
"name": "Giuseppe",
|
|||
|
|
"workspace": "admin_workspace",
|
|||
|
|
"rag_collection": "admin_docs",
|
|||
|
|
"capabilities": ["debug", "system_prompts", "user_management", "all_models"],
|
|||
|
|
"show_code": True
|
|||
|
|
},
|
|||
|
|
"giuseppe.defranceschi@gmail.com": {
|
|||
|
|
"role": "admin",
|
|||
|
|
"name": "Giuseppe",
|
|||
|
|
"workspace": "admin_workspace",
|
|||
|
|
"rag_collection": "admin_docs",
|
|||
|
|
"capabilities": ["debug", "system_prompts", "user_management", "all_models"],
|
|||
|
|
"show_code": True
|
|||
|
|
},
|
|||
|
|
"federica.tecchio@gmail.com": {
|
|||
|
|
"role": "business",
|
|||
|
|
"name": "Federica",
|
|||
|
|
"workspace": "business_workspace",
|
|||
|
|
"rag_collection": "contabilita",
|
|||
|
|
"capabilities": ["pdf_upload", "basic_chat"],
|
|||
|
|
"show_code": False
|
|||
|
|
},
|
|||
|
|
"riccardob545@gmail.com": {
|
|||
|
|
"role": "engineering",
|
|||
|
|
"name": "Riccardo",
|
|||
|
|
"workspace": "engineering_workspace",
|
|||
|
|
"rag_collection": "engineering_docs",
|
|||
|
|
"capabilities": ["code_execution", "data_viz", "advanced_chat"],
|
|||
|
|
"show_code": True
|
|||
|
|
},
|
|||
|
|
"giuliadefranceschi05@gmail.com": {
|
|||
|
|
"role": "architecture",
|
|||
|
|
"name": "Giulia",
|
|||
|
|
"workspace": "architecture_workspace",
|
|||
|
|
"rag_collection": "architecture_manuals",
|
|||
|
|
"capabilities": ["visual_chat", "pdf_upload", "image_gen"],
|
|||
|
|
"show_code": False
|
|||
|
|
}
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
# === CUSTOM LOCAL STORAGE CLIENT ===
|
|||
|
|
class LocalStorageClient(BaseStorageClient):
|
|||
|
|
"""Storage locale su filesystem per file/elementi"""
|
|||
|
|
|
|||
|
|
def __init__(self, storage_path: str):
|
|||
|
|
self.storage_path = storage_path
|
|||
|
|
os.makedirs(storage_path, exist_ok=True)
|
|||
|
|
|
|||
|
|
async def upload_file(
|
|||
|
|
self,
|
|||
|
|
object_key: str,
|
|||
|
|
data: bytes,
|
|||
|
|
mime: str = "application/octet-stream",
|
|||
|
|
overwrite: bool = True,
|
|||
|
|
) -> Dict[str, str]:
|
|||
|
|
"""Salva file localmente"""
|
|||
|
|
file_path = os.path.join(self.storage_path, object_key)
|
|||
|
|
os.makedirs(os.path.dirname(file_path), exist_ok=True)
|
|||
|
|
|
|||
|
|
with open(file_path, "wb") as f:
|
|||
|
|
f.write(data)
|
|||
|
|
|
|||
|
|
return {
|
|||
|
|
"object_key": object_key,
|
|||
|
|
"url": f"/files/{object_key}"
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
# === INIZIALIZZAZIONE DATA LAYER ===
|
|||
|
|
print("🔧 Inizializzazione database...")
|
|||
|
|
storage_client = LocalStorageClient(storage_path=STORAGE_DIR)
|
|||
|
|
|
|||
|
|
try:
|
|||
|
|
data_layer = SQLAlchemyDataLayer(
|
|||
|
|
conninfo=DATABASE_URL,
|
|||
|
|
storage_provider=storage_client,
|
|||
|
|
user_thread_limit=1000,
|
|||
|
|
show_logger=False
|
|||
|
|
)
|
|||
|
|
# ⬇️ QUESTA RIGA È CRUCIALE PER LA PERSISTENZA
|
|||
|
|
cl.data_layer = data_layer
|
|||
|
|
print("✅ SQLAlchemyDataLayer + LocalStorage initialized successfully")
|
|||
|
|
print(f"✅ Data layer set: {cl.data_layer is not None}")
|
|||
|
|
except Exception as e:
|
|||
|
|
print(f"❌ Failed to initialize data layer: {e}")
|
|||
|
|
cl.data_layer = None
|
|||
|
|
|
|||
|
|
|
|||
|
|
# === OAUTH CALLBACK CON RUOLI ===
|
|||
|
|
@cl.oauth_callback
|
|||
|
|
def oauth_callback(
|
|||
|
|
provider_id: str,
|
|||
|
|
token: str,
|
|||
|
|
raw_user_data: Dict[str, str],
|
|||
|
|
default_user: cl.User,
|
|||
|
|
) -> Optional[cl.User]:
|
|||
|
|
"""Validazione e arricchimento dati utente con ruoli"""
|
|||
|
|
if provider_id == "google":
|
|||
|
|
email = raw_user_data.get("email", "").lower()
|
|||
|
|
|
|||
|
|
# Verifica se utente è autorizzato
|
|||
|
|
if email not in USER_PROFILES:
|
|||
|
|
print(f"❌ Utente non autorizzato: {email}")
|
|||
|
|
return None # Nega accesso
|
|||
|
|
|
|||
|
|
# Arricchisci metadata con profilo
|
|||
|
|
profile = USER_PROFILES[email]
|
|||
|
|
default_user.metadata.update({
|
|||
|
|
"picture": raw_user_data.get("picture", ""),
|
|||
|
|
"locale": raw_user_data.get("locale", "en"),
|
|||
|
|
"role": profile["role"],
|
|||
|
|
"workspace": profile["workspace"],
|
|||
|
|
"rag_collection": profile["rag_collection"],
|
|||
|
|
"capabilities": profile["capabilities"],
|
|||
|
|
"show_code": profile["show_code"],
|
|||
|
|
"display_name": profile["name"]
|
|||
|
|
})
|
|||
|
|
|
|||
|
|
print(f"✅ Utente autorizzato: {email} - Ruolo: {profile['role']}")
|
|||
|
|
return default_user
|
|||
|
|
|
|||
|
|
return default_user
|
|||
|
|
|
|||
|
|
|
|||
|
|
# === UTILITY FUNCTIONS ===
|
|||
|
|
def get_user_profile(user_email: str) -> Dict:
|
|||
|
|
"""Recupera profilo utente"""
|
|||
|
|
return USER_PROFILES.get(user_email.lower(), {
|
|||
|
|
"role": "guest",
|
|||
|
|
"name": "Ospite",
|
|||
|
|
"workspace": "guest_workspace",
|
|||
|
|
"rag_collection": "documents",
|
|||
|
|
"capabilities": [],
|
|||
|
|
"show_code": False
|
|||
|
|
})
|
|||
|
|
|
|||
|
|
|
|||
|
|
def create_workspace(workspace_name: str) -> str:
|
|||
|
|
"""Crea directory workspace se non esiste"""
|
|||
|
|
workspace_path = os.path.join(WORKSPACES_DIR, workspace_name)
|
|||
|
|
os.makedirs(workspace_path, exist_ok=True)
|
|||
|
|
return workspace_path
|
|||
|
|
|
|||
|
|
|
|||
|
|
def save_code_to_file(code: str, workspace: str) -> str:
|
|||
|
|
"""Salva blocco codice come file .py"""
|
|||
|
|
timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
|
|||
|
|
file_name = f"code_{timestamp}.py"
|
|||
|
|
file_path = os.path.join(WORKSPACES_DIR, workspace, file_name)
|
|||
|
|
|
|||
|
|
with open(file_path, "w", encoding="utf-8") as f:
|
|||
|
|
f.write(code)
|
|||
|
|
|
|||
|
|
return file_path
|
|||
|
|
|
|||
|
|
|
|||
|
|
def extract_text_from_pdf(pdf_path: str) -> str:
|
|||
|
|
"""Estrae testo da PDF usando PyMuPDF"""
|
|||
|
|
try:
|
|||
|
|
doc = fitz.open(pdf_path)
|
|||
|
|
text_parts = []
|
|||
|
|
|
|||
|
|
for page_num in range(len(doc)):
|
|||
|
|
page = doc[page_num]
|
|||
|
|
text = page.get_text()
|
|||
|
|
text_parts.append(f"--- Pagina {page_num + 1} ---\n{text}\n")
|
|||
|
|
|
|||
|
|
doc.close()
|
|||
|
|
return "\n".join(text_parts)
|
|||
|
|
except Exception as e:
|
|||
|
|
print(f"❌ Errore estrazione PDF: {e}")
|
|||
|
|
return ""
|
|||
|
|
|
|||
|
|
|
|||
|
|
# === QDRANT FUNCTIONS ===
|
|||
|
|
async def get_qdrant_client() -> AsyncQdrantClient:
|
|||
|
|
"""Connessione a Qdrant"""
|
|||
|
|
return AsyncQdrantClient(url=QDRANT_URL)
|
|||
|
|
|
|||
|
|
|
|||
|
|
async def ensure_collection(collection_name: str):
|
|||
|
|
"""Crea collection se non esiste"""
|
|||
|
|
client = await get_qdrant_client()
|
|||
|
|
if not await client.collection_exists(collection_name):
|
|||
|
|
await client.create_collection(
|
|||
|
|
collection_name=collection_name,
|
|||
|
|
vectors_config=VectorParams(size=768, distance=Distance.COSINE)
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
|
|||
|
|
async def get_embeddings(text: str) -> list:
|
|||
|
|
"""Genera embeddings con Ollama"""
|
|||
|
|
max_length = 2000
|
|||
|
|
if len(text) > max_length:
|
|||
|
|
text = text[:max_length]
|
|||
|
|
|
|||
|
|
client = ollama.Client(host=OLLAMA_URL)
|
|||
|
|
|
|||
|
|
try:
|
|||
|
|
response = client.embed(model='nomic-embed-text', input=text)
|
|||
|
|
if 'embeddings' in response:
|
|||
|
|
return response['embeddings'][0]
|
|||
|
|
return response.get('embedding', [])
|
|||
|
|
except Exception as e:
|
|||
|
|
print(f"❌ Errore Embedding: {e}")
|
|||
|
|
return []
|
|||
|
|
|
|||
|
|
|
|||
|
|
def chunk_text(text: str, max_length: int = 1500, overlap: int = 200) -> list:
|
|||
|
|
"""Divide testo in chunks con overlap"""
|
|||
|
|
if len(text) <= max_length:
|
|||
|
|
return [text]
|
|||
|
|
|
|||
|
|
chunks = []
|
|||
|
|
start = 0
|
|||
|
|
|
|||
|
|
while start < len(text):
|
|||
|
|
end = start + max_length
|
|||
|
|
|
|||
|
|
if end < len(text):
|
|||
|
|
last_period = text.rfind('.', start, end)
|
|||
|
|
last_newline = text.rfind('\n', start, end)
|
|||
|
|
split_point = max(last_period, last_newline)
|
|||
|
|
|
|||
|
|
if split_point > start:
|
|||
|
|
end = split_point + 1
|
|||
|
|
|
|||
|
|
chunks.append(text[start:end].strip())
|
|||
|
|
start = end - overlap
|
|||
|
|
|
|||
|
|
return chunks
|
|||
|
|
|
|||
|
|
|
|||
|
|
async def index_document(file_name: str, content: str, collection_name: str) -> bool:
|
|||
|
|
"""Indicizza documento su Qdrant in collection specifica"""
|
|||
|
|
try:
|
|||
|
|
await ensure_collection(collection_name)
|
|||
|
|
chunks = chunk_text(content, max_length=1500)
|
|||
|
|
qdrant_client = await get_qdrant_client()
|
|||
|
|
points = []
|
|||
|
|
|
|||
|
|
for i, chunk in enumerate(chunks):
|
|||
|
|
embeddings = await get_embeddings(chunk)
|
|||
|
|
if not embeddings:
|
|||
|
|
continue
|
|||
|
|
|
|||
|
|
point_id = str(uuid.uuid4())
|
|||
|
|
point = PointStruct(
|
|||
|
|
id=point_id,
|
|||
|
|
vector=embeddings,
|
|||
|
|
payload={
|
|||
|
|
"file_name": file_name,
|
|||
|
|
"content": chunk,
|
|||
|
|
"chunk_index": i,
|
|||
|
|
"total_chunks": len(chunks),
|
|||
|
|
"indexed_at": datetime.now().isoformat()
|
|||
|
|
}
|
|||
|
|
)
|
|||
|
|
points.append(point)
|
|||
|
|
|
|||
|
|
if points:
|
|||
|
|
await qdrant_client.upsert(collection_name=collection_name, points=points)
|
|||
|
|
return True
|
|||
|
|
|
|||
|
|
return False
|
|||
|
|
except Exception as e:
|
|||
|
|
print(f"❌ Errore indicizzazione: {e}")
|
|||
|
|
return False
|
|||
|
|
|
|||
|
|
|
|||
|
|
async def search_qdrant(query_text: str, collection_name: str, limit: int = 5) -> str:
|
|||
|
|
"""Ricerca documenti rilevanti in collection specifica"""
|
|||
|
|
try:
|
|||
|
|
qdrant_client = await get_qdrant_client()
|
|||
|
|
|
|||
|
|
# Verifica se collection esiste
|
|||
|
|
if not await qdrant_client.collection_exists(collection_name):
|
|||
|
|
return ""
|
|||
|
|
|
|||
|
|
query_embedding = await get_embeddings(query_text)
|
|||
|
|
|
|||
|
|
if not query_embedding:
|
|||
|
|
return ""
|
|||
|
|
|
|||
|
|
search_result = await qdrant_client.query_points(
|
|||
|
|
collection_name=collection_name,
|
|||
|
|
query=query_embedding,
|
|||
|
|
limit=limit
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
contexts = []
|
|||
|
|
seen_files = set()
|
|||
|
|
|
|||
|
|
for hit in search_result.points:
|
|||
|
|
if hit.payload:
|
|||
|
|
file_name = hit.payload.get('file_name', 'Unknown')
|
|||
|
|
content = hit.payload.get('content', '')
|
|||
|
|
chunk_idx = hit.payload.get('chunk_index', 0)
|
|||
|
|
score = hit.score if hasattr(hit, 'score') else 0
|
|||
|
|
|
|||
|
|
file_key = f"{file_name}_{chunk_idx}"
|
|||
|
|
if file_key not in seen_files:
|
|||
|
|
seen_files.add(file_key)
|
|||
|
|
# ✅ FIX: Markdown code block corretto
|
|||
|
|
contexts.append(
|
|||
|
|
f"📄 **{file_name}** (chunk {chunk_idx+1}, score: {score:.2f})\n"
|
|||
|
|
f"``````"
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
return "\n\n".join(contexts) if contexts else ""
|
|||
|
|
except Exception as e:
|
|||
|
|
print(f"❌ Errore ricerca Qdrant: {e}")
|
|||
|
|
return ""
|
|||
|
|
|
|||
|
|
|
|||
|
|
# === CHAINLIT HANDLERS ===
|
|||
|
|
@cl.on_chat_start
|
|||
|
|
async def on_chat_start():
|
|||
|
|
"""Inizializzazione chat con profili utente"""
|
|||
|
|
user = cl.user_session.get("user")
|
|||
|
|
|
|||
|
|
if user:
|
|||
|
|
user_email = user.identifier
|
|||
|
|
profile = get_user_profile(user_email)
|
|||
|
|
user_name = profile["name"]
|
|||
|
|
user_role = profile["role"]
|
|||
|
|
workspace = profile["workspace"]
|
|||
|
|
user_picture = user.metadata.get("picture", "")
|
|||
|
|
show_code = profile["show_code"]
|
|||
|
|
capabilities = profile["capabilities"]
|
|||
|
|
else:
|
|||
|
|
user_email = "guest@local"
|
|||
|
|
user_name = "Ospite"
|
|||
|
|
user_role = "guest"
|
|||
|
|
workspace = "guest_workspace"
|
|||
|
|
user_picture = ""
|
|||
|
|
show_code = False
|
|||
|
|
capabilities = []
|
|||
|
|
|
|||
|
|
create_workspace(workspace)
|
|||
|
|
|
|||
|
|
# Salva in sessione
|
|||
|
|
cl.user_session.set("email", user_email)
|
|||
|
|
cl.user_session.set("name", user_name)
|
|||
|
|
cl.user_session.set("role", user_role)
|
|||
|
|
cl.user_session.set("workspace", workspace)
|
|||
|
|
cl.user_session.set("show_code", show_code)
|
|||
|
|
cl.user_session.set("capabilities", capabilities)
|
|||
|
|
cl.user_session.set("rag_collection", profile.get("rag_collection", "documents"))
|
|||
|
|
|
|||
|
|
# Settings basati su ruolo
|
|||
|
|
settings_widgets = [
|
|||
|
|
cl.input_widget.Select(
|
|||
|
|
id="model",
|
|||
|
|
label="🤖 Modello AI",
|
|||
|
|
values=["glm-4.6:cloud", "llama3.2", "mistral", "qwen2.5-coder:32b"],
|
|||
|
|
initial_value="glm-4.6:cloud",
|
|||
|
|
),
|
|||
|
|
cl.input_widget.Slider(
|
|||
|
|
id="temperature",
|
|||
|
|
label="🌡️ Temperatura",
|
|||
|
|
initial=0.7,
|
|||
|
|
min=0,
|
|||
|
|
max=2,
|
|||
|
|
step=0.1,
|
|||
|
|
),
|
|||
|
|
]
|
|||
|
|
|
|||
|
|
# Solo admin può disabilitare RAG
|
|||
|
|
if user_role == "admin":
|
|||
|
|
settings_widgets.append(
|
|||
|
|
cl.input_widget.Switch(
|
|||
|
|
id="rag_enabled",
|
|||
|
|
label="📚 Abilita RAG",
|
|||
|
|
initial=True,
|
|||
|
|
)
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
# ⬇️ INVIA SETTINGS (questo attiva l'icona ⚙️)
|
|||
|
|
await cl.ChatSettings(settings_widgets).send()
|
|||
|
|
|
|||
|
|
# Emoji ruolo
|
|||
|
|
role_emoji = {
|
|||
|
|
"admin": "👑",
|
|||
|
|
"business": "💼",
|
|||
|
|
"engineering": "⚙️",
|
|||
|
|
"architecture": "🏛️",
|
|||
|
|
"guest": "👤"
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
persistence_status = "✅ Attiva" if cl.data_layer else "⚠️ Disattivata"
|
|||
|
|
|
|||
|
|
welcome_msg = f"{role_emoji.get(user_role, '👋')} **Benvenuto, {user_name}!**\n\n"
|
|||
|
|
if user_picture:
|
|||
|
|
welcome_msg += f"\n\n"
|
|||
|
|
|
|||
|
|
welcome_msg += (
|
|||
|
|
f"🎭 **Ruolo**: {user_role.upper()}\n"
|
|||
|
|
f"📁 **Workspace**: `{workspace}`\n"
|
|||
|
|
f"💾 **Persistenza**: {persistence_status}\n"
|
|||
|
|
f"🤖 **Modello**: `glm-4.6:cloud`\n\n"
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
# Capabilities specifiche
|
|||
|
|
if "debug" in capabilities:
|
|||
|
|
welcome_msg += "🔧 **Modalità Debug**: Attiva\n"
|
|||
|
|
if "user_management" in capabilities:
|
|||
|
|
welcome_msg += "👥 **Gestione Utenti**: Disponibile\n"
|
|||
|
|
if not show_code:
|
|||
|
|
welcome_msg += "🎨 **Modalità Visuale**: Codice nascosto\n"
|
|||
|
|
|
|||
|
|
welcome_msg += "\n⚙️ **Usa le Settings (icona ⚙️ in alto a destra) per personalizzare!**"
|
|||
|
|
|
|||
|
|
await cl.Message(content=welcome_msg).send()
|
|||
|
|
|
|||
|
|
|
|||
|
|
@cl.on_settings_update
|
|||
|
|
async def on_settings_update(settings):
|
|||
|
|
"""Gestisce aggiornamento settings utente"""
|
|||
|
|
cl.user_session.set("settings", settings)
|
|||
|
|
|
|||
|
|
model = settings.get("model", "glm-4.6:cloud")
|
|||
|
|
temp = settings.get("temperature", 0.7)
|
|||
|
|
rag = settings.get("rag_enabled", True)
|
|||
|
|
|
|||
|
|
await cl.Message(
|
|||
|
|
content=f"✅ **Settings aggiornati**:\n"
|
|||
|
|
f"- 🤖 Modello: `{model}`\n"
|
|||
|
|
f"- 🌡️ Temperatura: `{temp}`\n"
|
|||
|
|
f"- 📚 RAG: {'✅ Attivo' if rag else '❌ Disattivato'}"
|
|||
|
|
).send()
|
|||
|
|
|
|||
|
|
|
|||
|
|
@cl.on_message
|
|||
|
|
async def on_message(message: cl.Message):
|
|||
|
|
"""Gestione messaggi utente con RAG intelligente"""
|
|||
|
|
user_email = cl.user_session.get("email", "guest")
|
|||
|
|
user_role = cl.user_session.get("role", "guest")
|
|||
|
|
workspace = cl.user_session.get("workspace", "guest_workspace")
|
|||
|
|
show_code = cl.user_session.get("show_code", False)
|
|||
|
|
rag_collection = cl.user_session.get("rag_collection", "documents")
|
|||
|
|
settings = cl.user_session.get("settings", {})
|
|||
|
|
|
|||
|
|
model = settings.get("model", "glm-4.6:cloud")
|
|||
|
|
temperature = settings.get("temperature", 0.7)
|
|||
|
|
|
|||
|
|
# Admin può disabilitare RAG, altri lo hanno sempre attivo
|
|||
|
|
rag_enabled = settings.get("rag_enabled", True) if user_role == "admin" else True
|
|||
|
|
|
|||
|
|
try:
|
|||
|
|
# Gestisci upload file
|
|||
|
|
if message.elements:
|
|||
|
|
await handle_file_uploads(message.elements, workspace, rag_collection)
|
|||
|
|
|
|||
|
|
# RAG Search solo se abilitato
|
|||
|
|
context_text = ""
|
|||
|
|
if rag_enabled:
|
|||
|
|
context_text = await search_qdrant(message.content, rag_collection, limit=5)
|
|||
|
|
|
|||
|
|
# Costruisci prompt con o senza contesto
|
|||
|
|
if context_text:
|
|||
|
|
system_prompt = (
|
|||
|
|
"Sei un assistente AI esperto. "
|
|||
|
|
"Usa il seguente contesto per arricchire la tua risposta, "
|
|||
|
|
"ma puoi anche rispondere usando la tua conoscenza generale se il contesto non è sufficiente."
|
|||
|
|
)
|
|||
|
|
full_prompt = f"{system_prompt}\n\n**CONTESTO DOCUMENTI:**\n{context_text}\n\n**DOMANDA UTENTE:**\n{message.content}"
|
|||
|
|
else:
|
|||
|
|
system_prompt = "Sei un assistente AI esperto e disponibile. Rispondi in modo chiaro e utile."
|
|||
|
|
full_prompt = f"{system_prompt}\n\n**DOMANDA UTENTE:**\n{message.content}"
|
|||
|
|
|
|||
|
|
# Streaming risposta da Ollama
|
|||
|
|
client = ollama.Client(host=OLLAMA_URL)
|
|||
|
|
msg = cl.Message(content="")
|
|||
|
|
await msg.send()
|
|||
|
|
|
|||
|
|
messages = [{"role": "user", "content": full_prompt}]
|
|||
|
|
stream = client.chat(
|
|||
|
|
model=model,
|
|||
|
|
messages=messages,
|
|||
|
|
stream=True,
|
|||
|
|
options={"temperature": temperature}
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
full_response = ""
|
|||
|
|
for chunk in stream:
|
|||
|
|
content = chunk['message']['content']
|
|||
|
|
full_response += content
|
|||
|
|
await msg.stream_token(content)
|
|||
|
|
|
|||
|
|
await msg.update()
|
|||
|
|
|
|||
|
|
# ✅ FIX: Estrai codice Python con regex corretto
|
|||
|
|
code_blocks = re.findall(r"``````", full_response, re.DOTALL)
|
|||
|
|
|
|||
|
|
if code_blocks:
|
|||
|
|
elements = []
|
|||
|
|
|
|||
|
|
# Se show_code è False, nascondi il codice dalla risposta
|
|||
|
|
if not show_code:
|
|||
|
|
cleaned_response = re.sub(
|
|||
|
|
r"``````",
|
|||
|
|
"[💻 Codice eseguito internamente]",
|
|||
|
|
full_response,
|
|||
|
|
flags=re.DOTALL
|
|||
|
|
)
|
|||
|
|
await msg.update(content=cleaned_response)
|
|||
|
|
|
|||
|
|
# Salva codice nel workspace
|
|||
|
|
for code in code_blocks:
|
|||
|
|
file_path = save_code_to_file(code.strip(), workspace)
|
|||
|
|
elements.append(
|
|||
|
|
cl.File(
|
|||
|
|
name=os.path.basename(file_path),
|
|||
|
|
path=file_path,
|
|||
|
|
display="inline" if show_code else "side"
|
|||
|
|
)
|
|||
|
|
)
|
|||
|
|
|
|||
|
|
if show_code:
|
|||
|
|
await cl.Message(
|
|||
|
|
content=f"💾 Codice salvato in workspace `{workspace}`",
|
|||
|
|
elements=elements
|
|||
|
|
).send()
|
|||
|
|
|
|||
|
|
except Exception as e:
|
|||
|
|
await cl.Message(content=f"❌ **Errore:** {str(e)}").send()
|
|||
|
|
|
|||
|
|
|
|||
|
|
async def handle_file_uploads(elements, workspace: str, collection_name: str):
|
|||
|
|
"""Gestisce upload e indicizzazione file in collection specifica"""
|
|||
|
|
for element in elements:
|
|||
|
|
try:
|
|||
|
|
dest_path = os.path.join(WORKSPACES_DIR, workspace, element.name)
|
|||
|
|
shutil.copy(element.path, dest_path)
|
|||
|
|
|
|||
|
|
content = None
|
|||
|
|
|
|||
|
|
if element.name.lower().endswith('.pdf'):
|
|||
|
|
await cl.Message(content=f"📄 Elaborazione PDF **{element.name}**...").send()
|
|||
|
|
content = extract_text_from_pdf(dest_path)
|
|||
|
|
|
|||
|
|
if not content:
|
|||
|
|
await cl.Message(
|
|||
|
|
content=f"⚠️ **{element.name}**: PDF vuoto o non leggibile"
|
|||
|
|
).send()
|
|||
|
|
continue
|
|||
|
|
|
|||
|
|
elif element.name.lower().endswith('.txt'):
|
|||
|
|
with open(dest_path, 'r', encoding='utf-8') as f:
|
|||
|
|
content = f.read()
|
|||
|
|
else:
|
|||
|
|
await cl.Message(
|
|||
|
|
content=f"📁 **{element.name}** salvato in workspace (supportati: .pdf, .txt)"
|
|||
|
|
).send()
|
|||
|
|
continue
|
|||
|
|
|
|||
|
|
if content:
|
|||
|
|
success = await index_document(element.name, content, collection_name)
|
|||
|
|
|
|||
|
|
if success:
|
|||
|
|
word_count = len(content.split())
|
|||
|
|
await cl.Message(
|
|||
|
|
content=f"✅ **{element.name}** indicizzato in `{collection_name}`\n"
|
|||
|
|
f"📊 Parole estratte: {word_count:,}"
|
|||
|
|
).send()
|
|||
|
|
else:
|
|||
|
|
await cl.Message(
|
|||
|
|
content=f"⚠️ Errore indicizzazione **{element.name}**"
|
|||
|
|
).send()
|
|||
|
|
|
|||
|
|
except Exception as e:
|
|||
|
|
await cl.Message(
|
|||
|
|
content=f"❌ Errore con **{element.name}**: {str(e)}"
|
|||
|
|
).send()
|