rebild project

Dockerfile

@@ -1,39 +1,33 @@
-FROM python:3.11-slim
+FROM python:3.9-slim
 
 WORKDIR /app
 
 # Install system dependencies
 RUN apt-get update && apt-get install -y \
     build-essential \
-    curl \
-    software-properties-common \
     && rm -rf /var/lib/apt/lists/*
 
-# Create directories and set permissions
-RUN mkdir -p /app/vector_store /app/chat_history /app/.cache /app/logs && \
-    chmod 777 /app/vector_store /app/chat_history /app/.cache /app/logs
-
-# Set environment variables
-ENV TRANSFORMERS_CACHE=/app/.cache/huggingface
-ENV HF_HOME=/app/.cache/huggingface
-ENV XDG_CACHE_HOME=/app/.cache
-ENV PYTHONUNBUFFERED=1
-
-# Create cache directories with proper permissions
-RUN mkdir -p /app/.cache/huggingface && \
-    chmod -R 777 /app/.cache
-
+# Copy requirements file
 COPY requirements.txt .
+
+# Install Python dependencies
 RUN pip install --no-cache-dir -r requirements.txt
 
+# Copy the application
 COPY . .
 
-RUN chown -R 1000:1000 /app && \
-    chmod -R 755 /app
+# Create directories for persistent storage
+RUN mkdir -p vector_store
+RUN mkdir -p chat_history
 
-EXPOSE 8000
+# Make sure the static directory exists
+RUN mkdir -p static
+
+# Copy the frontend to the static directory
+COPY index.html static/
 
-USER 1000
+# Expose the port the app runs on
+EXPOSE 8000
 
-# Изменяем команду запуска для сохранения логов
-CMD ["sh", "-c", "uvicorn app:app --host 0.0.0.0 --port 8000 --log-level debug 2>&1 | tee /app/logs/app.log"]
+# Command to run the application
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "8000"]

app - Copy.py

@@ -0,0 +1,417 @@
+import os
+import time
+from dotenv import load_dotenv
+from langchain_groq import ChatGroq
+from langchain_huggingface import HuggingFaceEmbeddings
+from langchain_community.vectorstores import FAISS
+from langchain_text_splitters import RecursiveCharacterTextSplitter
+from langchain_community.document_loaders import WebBaseLoader
+from langchain_core.prompts import PromptTemplate
+from langchain_core.output_parsers import StrOutputParser
+from datetime import datetime
+import json
+import traceback
+from fastapi import FastAPI, HTTPException, Request
+from fastapi.responses import JSONResponse
+from pydantic import BaseModel
+from api import router as analysis_router
+from utils import ChatAnalyzer, setup_chat_analysis
+import requests.exceptions
+import aiohttp
+from typing import Union
+import uvicorn
+import logging
+from rich import print as rprint
+from rich.console import Console
+from rich.panel import Panel
+from rich.table import Table
+
+console = Console()
+
+# Базовая настройка логирования
+logging.basicConfig(level=logging.DEBUG)
+logger = logging.getLogger(__name__)
+
+# Определение путей
+VECTOR_STORE_PATH = os.path.join(os.getcwd(), "vector_store")
+CHAT_HISTORY_PATH = os.path.join(os.getcwd(), "chat_history")
+
+app = FastAPI(title="Status Law Assistant API")
+
+class ChatRequest(BaseModel):
+    message: str
+
+class ChatResponse(BaseModel):
+    response: str
+
+def check_vector_store():
+    """Проверка наличия векторной базы"""
+    index_path = os.path.join(VECTOR_STORE_PATH, "index.faiss")
+    return os.path.exists(index_path)
+
+@app.get("/")
+async def root():
+    """Базовый эндпоинт с информацией о состоянии"""
+    return {
+        "status": "ok",
+        "vector_store_ready": check_vector_store(),
+        "timestamp": datetime.now().isoformat()
+    }
+
+@app.get("/status")
+async def get_status():
+    """Получение статуса векторной базы"""
+    return {
+        "vector_store_exists": check_vector_store(),
+        "can_chat": check_vector_store(),
+        "vector_store_path": VECTOR_STORE_PATH
+    }
+
+@app.post("/build-knowledge-base")
+async def build_kb():
+    """Эндпоинт для построения базы знаний"""
+    try:
+        if check_vector_store():
+            return {
+                "status": "exists",
+                "message": "Knowledge base already exists"
+            }
+        
+        # Инициализируем embeddings только когда нужно построить базу
+        embeddings = HuggingFaceEmbeddings(
+            model_name="sentence-transformers/all-MiniLM-L6-v2"
+        )
+        vector_store = build_knowledge_base(embeddings)
+        
+        return {
+            "status": "success",
+            "message": "Knowledge base built successfully"
+        }
+    except Exception as e:
+        logger.error(f"Failed to build knowledge base: {str(e)}")
+        raise HTTPException(
+            status_code=500,
+            detail=f"Failed to build knowledge base: {str(e)}"
+        )
+
+@app.post("/chat", response_model=ChatResponse)
+async def chat_endpoint(request: ChatRequest):
+    """Эндпоинт чата"""
+    if not check_vector_store():
+        raise HTTPException(
+            status_code=400,
+            detail="Knowledge base not found. Please build it first using /build-knowledge-base endpoint"
+        )
+    
+    try:
+        # Инициализируем компоненты только при необходимости
+        llm = ChatGroq(
+            model_name="llama-3.3-70b-versatile",
+            temperature=0.6,
+            api_key=os.getenv("GROQ_API_KEY")
+        )
+        
+        embeddings = HuggingFaceEmbeddings(
+            model_name="sentence-transformers/all-MiniLM-L6-v2"
+        )
+        
+        vector_store = FAISS.load_local(
+            VECTOR_STORE_PATH,
+            embeddings,
+            allow_dangerous_deserialization=True
+        )
+
+        # Остальная логика чата...
+        context_docs = vector_store.similarity_search(request.message)
+        context_text = "\n".join([d.page_content for d in context_docs])
+        
+        prompt_template = PromptTemplate.from_template('''
+            You are a helpful and polite legal assistant at Status Law.
+            Answer the question based on the context provided.
+            Context: {context}
+            Question: {question}
+        ''')
+        
+        chain = prompt_template | llm | StrOutputParser()
+        response = chain.invoke({
+            "context": context_text,
+            "question": request.message
+        })
+        
+        return ChatResponse(response=response)
+        
+    except Exception as e:
+        logger.error(f"Chat error: {str(e)}")
+        raise HTTPException(
+            status_code=500,
+            detail=f"Chat error: {str(e)}"
+        )
+
+# --------------- Knowledge Base Management ---------------
+URLS = [
+    "https://status.law",
+    "https://status.law/about",
+    "https://status.law/careers",  
+    "https://status.law/tariffs-for-services-against-extradition-en",
+    "https://status.law/challenging-sanctions",
+    "https://status.law/law-firm-contact-legal-protection"
+    "https://status.law/cross-border-banking-legal-issues", 
+    "https://status.law/extradition-defense", 
+    "https://status.law/international-prosecution-protection", 
+    "https://status.law/interpol-red-notice-removal",  
+    "https://status.law/practice-areas",  
+    "https://status.law/reputation-protection",
+    "https://status.law/faq"
+]
+
+def build_knowledge_base(_embeddings):
+    """Build or update the knowledge base"""
+    try:
+        start_time = time.time()
+        documents = []
+        
+        # Ensure vector store directory exists
+        if not os.path.exists(VECTOR_STORE_PATH):
+            os.makedirs(VECTOR_STORE_PATH, exist_ok=True)
+        
+        for url in URLS:
+            try:
+                loader = WebBaseLoader(url)
+                docs = loader.load()
+                documents.extend(docs)
+            except Exception as e:
+                print(f"Failed to load {url}: {str(e)}")
+                continue
+
+        if not documents:
+            raise HTTPException(status_code=500, detail="No documents loaded")
+
+        text_splitter = RecursiveCharacterTextSplitter(
+            chunk_size=500,
+            chunk_overlap=100
+        )
+        chunks = text_splitter.split_documents(documents)
+        
+        vector_store = FAISS.from_documents(chunks, _embeddings)
+        vector_store.save_local(
+            folder_path=VECTOR_STORE_PATH,
+            index_name="index"
+        )
+        
+        if not os.path.exists(os.path.join(VECTOR_STORE_PATH, "index.faiss")):
+            raise HTTPException(status_code=500, detail="FAISS index file not created")
+            
+        return vector_store
+            
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Knowledge base creation failed: {str(e)}")
+
+# --------------- API Models ---------------
+class ChatRequest(BaseModel):
+    message: str
+
+class ChatResponse(BaseModel):
+    response: str
+
+# --------------- API Routes ---------------
+@app.post("/chat", response_model=ChatResponse)
+async def chat_endpoint(request: ChatRequest):
+    try:
+        llm, embeddings = init_models()
+        
+        if not os.path.exists(VECTOR_STORE_PATH):
+            vector_store = build_knowledge_base(embeddings)
+        else:
+            vector_store = FAISS.load_local(
+                VECTOR_STORE_PATH,
+                embeddings,
+                allow_dangerous_deserialization=True
+            )
+
+        # Add retry logic for network operations
+        max_retries = 3
+        retry_count = 0
+        
+        while retry_count < max_retries:
+            try:
+                context_docs = vector_store.similarity_search(request.message)
+                context_text = "\n".join([d.page_content for d in context_docs])
+                
+                prompt_template = PromptTemplate.from_template('''
+                    You are a helpful and polite legal assistant at Status Law.
+                    You answer in the language in which the question was asked.
+                    Answer the question based on the context provided.
+                    
+                    # ... остальной текст промпта ...
+
+                    Context: {context}
+                    Question: {question}
+                    
+                    Response Guidelines:
+                    1. Answer in the user's language
+                    2. Cite sources when possible
+                    3. Offer contact options if unsure
+                    ''')
+                
+                chain = prompt_template | llm | StrOutputParser()
+                response = chain.invoke({
+                    "context": context_text,
+                    "question": request.message
+                })
+                
+                log_interaction(request.message, response, context_text)
+                return ChatResponse(response=response)
+                
+            except (requests.exceptions.RequestException, aiohttp.ClientError) as e:
+                retry_count += 1
+                if retry_count == max_retries:
+                    raise HTTPException(
+                        status_code=503,
+                        detail={
+                            "error": "Network error after maximum retries",
+                            "detail": str(e),
+                            "type": "network_error"
+                        }
+                    )
+                await asyncio.sleep(1 * retry_count)  # Exponential backoff
+                
+    except Exception as e:
+        if isinstance(e, (requests.exceptions.RequestException, aiohttp.ClientError)):
+            raise HTTPException(
+                status_code=503,
+                detail={
+                    "error": "Network error occurred",
+                    "detail": str(e),
+                    "type": "network_error"
+                }
+            )
+        raise HTTPException(status_code=500, detail=str(e))
+
+# --------------- Logging ---------------
+def log_interaction(user_input: str, bot_response: str, context: str):
+    try:
+        log_entry = {
+            "timestamp": datetime.now().isoformat(),
+            "user_input": user_input,
+            "bot_response": bot_response,
+            "context": context[:500],
+            "kb_version": datetime.now().strftime("%Y%m%d-%H%M%S")
+        }
+        
+        os.makedirs("chat_history", exist_ok=True)
+        log_path = os.path.join("chat_history", "chat_logs.json")
+        
+        with open(log_path, "a", encoding="utf-8") as f:
+            f.write(json.dumps(log_entry, ensure_ascii=False) + "\n")
+            
+    except Exception as e:
+        print(f"Logging error: {str(e)}")
+        print(traceback.format_exc())
+
+# Add health check endpoint
+@app.get("/health")
+async def health_check():
+    try:
+        # Check if models can be initialized
+        llm, embeddings = init_models()
+        
+        # Check if vector store is accessible
+        if os.path.exists(VECTOR_STORE_PATH):
+            vector_store = FAISS.load_local(
+                VECTOR_STORE_PATH,
+                embeddings,
+                allow_dangerous_deserialization=True
+            )
+        
+        return {
+            "status": "healthy",
+            "vector_store": "available" if os.path.exists(VECTOR_STORE_PATH) else "not_found"
+        }
+        
+    except Exception as e:
+        return JSONResponse(
+            status_code=503,
+            content={
+                "status": "unhealthy",
+                "error": str(e)
+            }
+        )
+
+# Add diagnostic endpoint
+@app.get("/directory-status")
+async def check_directory_status():
+    """Check status of required directories"""
+    return {
+        "vector_store": {
+            "exists": os.path.exists(VECTOR_STORE_PATH),
+            "path": os.path.abspath(VECTOR_STORE_PATH),
+            "contents": os.listdir(VECTOR_STORE_PATH) if os.path.exists(VECTOR_STORE_PATH) else []
+        },
+        "chat_history": {
+            "exists": os.path.exists(CHAT_HISTORY_PATH),
+            "path": os.path.abspath(CHAT_HISTORY_PATH),
+            "contents": os.listdir(CHAT_HISTORY_PATH) if os.path.exists(CHAT_HISTORY_PATH) else []
+        }
+    }
+
+# Добавим функцию для вывода статуса
+def print_startup_status():
+    """Print application startup status with rich formatting"""
+    try:
+        # Create status table
+        table = Table(show_header=True, header_style="bold magenta")
+        table.add_column("Component", style="cyan")
+        table.add_column("Status", style="green")
+        
+        # Check directories
+        vector_store_exists = os.path.exists(VECTOR_STORE_PATH)
+        chat_history_exists = os.path.exists(CHAT_HISTORY_PATH)
+        
+        table.add_row(
+            "Vector Store Directory",
+            "✅ Created" if vector_store_exists else "❌ Missing"
+        )
+        table.add_row(
+            "Chat History Directory",
+            "✅ Created" if chat_history_exists else "❌ Missing"
+        )
+        
+        # Check environment variables
+        table.add_row(
+            "GROQ API Key",
+            "✅ Set" if os.getenv("GROQ_API_KEY") else "❌ Missing"
+        )
+        
+        # Create status panel
+        status_panel = Panel(
+            table,
+            title="[bold blue]Status Law Assistant API Status[/bold blue]",
+            border_style="blue"
+        )
+        
+        # Print startup message and status
+        console.print("\n")
+        console.print("[bold green]🚀 Server started successfully![/bold green]")
+        console.print(status_panel)
+        console.print("\n[bold yellow]API Documentation:[/bold yellow]")
+        console.print("📚 Swagger UI: http://0.0.0.0:8000/docs")
+        console.print("📘 ReDoc: http://0.0.0.0:8000/redoc\n")
+        
+    except Exception as e:
+        console.print(f"[bold red]Error printing status: {str(e)}[/bold red]")
+
+if __name__ == "__main__":
+    import uvicorn
+    
+    port = int(os.getenv("PORT", 8000))
+    logger.info(f"Starting server on port {port}")
+    
+    config = uvicorn.Config(
+        app,
+        host="0.0.0.0",
+        port=port,
+        log_level="debug"
+    )
+    
+    server = uvicorn.Server(config)
+    server.run()

app.py

@@ -1,5 +1,11 @@
 import os
 import time
+import uvicorn
+from fastapi import FastAPI, HTTPException, Request
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import HTMLResponse
+from fastapi.staticfiles import StaticFiles
+from fastapi.templating import Jinja2Templates
 from dotenv import load_dotenv
 from langchain_groq import ChatGroq
 from langchain_huggingface import HuggingFaceEmbeddings
@@ -11,407 +17,391 @@ from langchain_core.output_parsers import StrOutputParser
 from datetime import datetime
 import json
 import traceback
-from fastapi import FastAPI, HTTPException, Request
-from fastapi.responses import JSONResponse
+from typing import Dict, List, Optional
 from pydantic import BaseModel
-from api import router as analysis_router
-from utils import ChatAnalyzer, setup_chat_analysis
-import requests.exceptions
-import aiohttp
-from typing import Union
-import uvicorn
-import logging
-from rich import print as rprint
-from rich.console import Console
-from rich.panel import Panel
-from rich.table import Table
+from huggingface_hub import Repository, snapshot_download
 
-console = Console()
+# Initialize environment variables
+load_dotenv()
 
-# Базовая настройка логирования
-logging.basicConfig(level=logging.DEBUG)
-logger = logging.getLogger(__name__)
+# Constants for paths and URLs
+VECTOR_STORE_PATH = "vector_store"
+HF_DATASET_REPO = "Rulga/LS_chat"
+LOCAL_CHAT_HISTORY_PATH = "chat_history"
 
-# Определение путей
-VECTOR_STORE_PATH = os.path.join(os.getcwd(), "vector_store")
-CHAT_HISTORY_PATH = os.path.join(os.getcwd(), "chat_history")
+URLS = [
+    "https://status.law",
+    "https://status.law/about",
+    "https://status.law/careers",  
+    "https://status.law/tariffs-for-services-of-protection-against-extradition",
+    "https://status.law/challenging-sanctions",
+    "https://status.law/law-firm-contact-legal-protection", 
+    "https://status.law/cross-border-banking-legal-issues", 
+    "https://status.law/extradition-defense", 
+    "https://status.law/international-prosecution-protection", 
+    "https://status.law/interpol-red-notice-removal",  
+    "https://status.law/practice-areas",  
+    "https://status.law/reputation-protection",
+    "https://status.law/faq"
+]
 
+# Initialize the FastAPI app
 app = FastAPI(title="Status Law Assistant API")
 
+# Add CORS middleware
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+# Define request and response models
 class ChatRequest(BaseModel):
     message: str
-
+    conversation_id: Optional[str] = None
+    
 class ChatResponse(BaseModel):
     response: str
+    conversation_id: str
+    
+class BuildKnowledgeBaseResponse(BaseModel):
+    status: str
+    message: str
+    details: Optional[Dict] = None
 
-def check_vector_store():
-    """Проверка наличия векторной базы"""
-    index_path = os.path.join(VECTOR_STORE_PATH, "index.faiss")
-    return os.path.exists(index_path)
-
-@app.get("/")
-async def root():
-    """Базовый эндпоинт с информацией о состоянии"""
-    return {
-        "status": "ok",
-        "vector_store_ready": check_vector_store(),
-        "timestamp": datetime.now().isoformat()
-    }
-
-@app.get("/status")
-async def get_status():
-    """Получение статуса векторной базы"""
-    return {
-        "vector_store_exists": check_vector_store(),
-        "can_chat": check_vector_store(),
-        "vector_store_path": VECTOR_STORE_PATH
-    }
+# Global variables for models and knowledge base
+llm = None
+embeddings = None
+vector_store = None
+kb_info = {
+    'build_time': None,
+    'size': None,
+    'version': '1.1'
+}
 
-@app.post("/build-knowledge-base")
-async def build_kb():
-    """Эндпоинт для построения базы знаний"""
+# --------------- Hugging Face Dataset Integration ---------------
+def init_hf_dataset_integration():
+    """Initialize integration with Hugging Face dataset for persistence"""
     try:
-        if check_vector_store():
-            return {
-                "status": "exists",
-                "message": "Knowledge base already exists"
-            }
-        
-        # Инициализируем embeddings только когда нужно построить базу
-        embeddings = HuggingFaceEmbeddings(
-            model_name="sentence-transformers/all-MiniLM-L6-v2"
-        )
-        vector_store = build_knowledge_base(embeddings)
+        # Download the latest snapshot of the dataset if it exists
+        if os.getenv("HF_TOKEN"):
+            # With authentication if token provided
+            snapshot_download(
+                repo_id=HF_DATASET_REPO,
+                repo_type="dataset",
+                local_dir="./data_snapshot",
+                token=os.getenv("HF_TOKEN")
+            )
+        else:
+            # Try without authentication for public datasets
+            snapshot_download(
+                repo_id=HF_DATASET_REPO,
+                repo_type="dataset",
+                local_dir="./data_snapshot"
+            )
         
-        return {
-            "status": "success",
-            "message": "Knowledge base built successfully"
-        }
+        # Check if vector store exists in the downloaded data
+        if os.path.exists("./data_snapshot/vector_store/index.faiss"):
+            # Copy to the local vector store path
+            os.makedirs(VECTOR_STORE_PATH, exist_ok=True)
+            os.system(f"cp -r ./data_snapshot/vector_store/* {VECTOR_STORE_PATH}/")
+            return True
     except Exception as e:
-        logger.error(f"Failed to build knowledge base: {str(e)}")
-        raise HTTPException(
-            status_code=500,
-            detail=f"Failed to build knowledge base: {str(e)}"
-        )
+        print(f"Error downloading dataset: {e}")
+    
+    return False
 
-@app.post("/chat", response_model=ChatResponse)
-async def chat_endpoint(request: ChatRequest):
-    """Эндпоинт чата"""
-    if not check_vector_store():
-        raise HTTPException(
-            status_code=400,
-            detail="Knowledge base not found. Please build it first using /build-knowledge-base endpoint"
-        )
+def upload_to_hf_dataset():
+    """Upload the vector store and chat history to the Hugging Face dataset"""
+    if not os.getenv("HF_TOKEN"):
+        print("HF_TOKEN not set, cannot upload to Hugging Face")
+        return False
     
     try:
-        # Инициализируем компоненты только при необходимости
-        llm = ChatGroq(
-            model_name="llama-3.3-70b-versatile",
-            temperature=0.6,
-            api_key=os.getenv("GROQ_API_KEY")
-        )
-        
-        embeddings = HuggingFaceEmbeddings(
-            model_name="sentence-transformers/all-MiniLM-L6-v2"
+        # Clone the repository
+        repo = Repository(
+            local_dir="./data_upload",
+            clone_from=HF_DATASET_REPO,
+            repo_type="dataset",
+            token=os.getenv("HF_TOKEN")
         )
         
-        vector_store = FAISS.load_local(
-            VECTOR_STORE_PATH,
-            embeddings,
-            allow_dangerous_deserialization=True
-        )
-
-        # Остальная логика чата...
-        context_docs = vector_store.similarity_search(request.message)
-        context_text = "\n".join([d.page_content for d in context_docs])
+        # Copy the vector store files
+        if os.path.exists(f"{VECTOR_STORE_PATH}/index.faiss"):
+            os.makedirs("./data_upload/vector_store", exist_ok=True)
+            os.system(f"cp -r {VECTOR_STORE_PATH}/* ./data_upload/vector_store/")
         
-        prompt_template = PromptTemplate.from_template('''
-            You are a helpful and polite legal assistant at Status Law.
-            Answer the question based on the context provided.
-            Context: {context}
-            Question: {question}
-        ''')
+        # Copy the chat history
+        if os.path.exists(f"{LOCAL_CHAT_HISTORY_PATH}/chat_logs.json"):
+            os.makedirs("./data_upload/chat_history", exist_ok=True)
+            os.system(f"cp -r {LOCAL_CHAT_HISTORY_PATH}/* ./data_upload/chat_history/")
         
-        chain = prompt_template | llm | StrOutputParser()
-        response = chain.invoke({
-            "context": context_text,
-            "question": request.message
-        })
+        # Push to Hugging Face
+        repo.push_to_hub(commit_message="Update vector store and chat history")
+        return True
+    except Exception as e:
+        print(f"Error uploading to dataset: {e}")
+        return False
+
+# --------------- Enhanced Logging ---------------
+def log_interaction(user_input: str, bot_response: str, context: str, conversation_id: str):
+    """Log interactions with error handling"""
+    try:
+        log_entry = {
+            "timestamp": datetime.now().isoformat(),
+            "conversation_id": conversation_id,
+            "user_input": user_input,
+            "bot_response": bot_response,
+            "context": context[:500] if context else "",
+            "kb_version": kb_info['version']
+        }
         
-        return ChatResponse(response=response)
+        os.makedirs(LOCAL_CHAT_HISTORY_PATH, exist_ok=True)
+        log_path = os.path.join(LOCAL_CHAT_HISTORY_PATH, "chat_logs.json")
         
+        with open(log_path, "a", encoding="utf-8") as f:
+            f.write(json.dumps(log_entry, ensure_ascii=False) + "\n")
+            
+        # Upload to Hugging Face after logging
+        upload_to_hf_dataset()
+            
     except Exception as e:
-        logger.error(f"Chat error: {str(e)}")
-        raise HTTPException(
-            status_code=500,
-            detail=f"Chat error: {str(e)}"
-        )
+        print(f"Logging error: {str(e)}")
+        print(traceback.format_exc())
 
-# --------------- Knowledge Base Management ---------------
-URLS = [
-    "https://status.law",
-    "https://status.law/about",
-    "https://status.law/careers",  
-    "https://status.law/tariffs-for-services-against-extradition-en",
-    "https://status.law/challenging-sanctions",
-    "https://status.law/law-firm-contact-legal-protection"
-    "https://status.law/cross-border-banking-legal-issues", 
-    "https://status.law/extradition-defense", 
-    "https://status.law/international-prosecution-protection", 
-    "https://status.law/interpol-red-notice-removal",  
-    "https://status.law/practice-areas",  
-    "https://status.law/reputation-protection",
-    "https://status.law/faq"
-]
+# --------------- Model Initialization ---------------
+def init_models():
+    """Initialize AI models"""
+    global llm, embeddings
+    
+    if not llm:
+        try:
+            llm = ChatGroq(
+                model_name="llama-3.3-70b-versatile",
+                temperature=0.6,
+                api_key=os.getenv("GROQ_API_KEY")
+            )
+        except Exception as e:
+            print(f"LLM initialization failed: {str(e)}")
+            raise HTTPException(status_code=500, detail=f"LLM initialization failed: {str(e)}")
+    
+    if not embeddings:
+        try:
+            embeddings = HuggingFaceEmbeddings(
+                model_name="intfloat/multilingual-e5-large-instruct"
+            )
+        except Exception as e:
+            print(f"Embeddings initialization failed: {str(e)}")
+            raise HTTPException(status_code=500, detail=f"Embeddings initialization failed: {str(e)}")
+    
+    return llm, embeddings
 
-def build_knowledge_base(_embeddings):
+# --------------- Knowledge Base Management ---------------
+def build_knowledge_base():
     """Build or update the knowledge base"""
+    global vector_store, kb_info
+    
+    _, _embeddings = init_models()
+    
     try:
         start_time = time.time()
         documents = []
         
-        # Ensure vector store directory exists
-        if not os.path.exists(VECTOR_STORE_PATH):
-            os.makedirs(VECTOR_STORE_PATH, exist_ok=True)
+        # Create folder in advance
+        os.makedirs(VECTOR_STORE_PATH, exist_ok=True)
         
+        # Load documents
         for url in URLS:
             try:
                 loader = WebBaseLoader(url)
                 docs = loader.load()
                 documents.extend(docs)
+                print(f"Loaded {url}")
             except Exception as e:
                 print(f"Failed to load {url}: {str(e)}")
                 continue
-
+                
         if not documents:
-            raise HTTPException(status_code=500, detail="No documents loaded")
+            raise HTTPException(status_code=500, detail="No documents loaded!")
 
+        # Split into chunks
         text_splitter = RecursiveCharacterTextSplitter(
             chunk_size=500,
             chunk_overlap=100
         )
         chunks = text_splitter.split_documents(documents)
         
+        # Create vector store
         vector_store = FAISS.from_documents(chunks, _embeddings)
         vector_store.save_local(
             folder_path=VECTOR_STORE_PATH,
             index_name="index"
         )
         
+        # Verify file creation
         if not os.path.exists(os.path.join(VECTOR_STORE_PATH, "index.faiss")):
-            raise HTTPException(status_code=500, detail="FAISS index file not created")
+            raise HTTPException(status_code=500, detail="FAISS index file not created!")
             
-        return vector_store
+        # Update info
+        kb_info.update({
+            'build_time': time.time() - start_time,
+            'size': sum(
+                os.path.getsize(os.path.join(VECTOR_STORE_PATH, f)) 
+                for f in ["index.faiss", "index.pkl"]
+            ) / (1024 ** 2),
+            'version': datetime.now().strftime("%Y%m%d-%H%M%S")
+        })
+        
+        # Upload to Hugging Face
+        upload_to_hf_dataset()
+        
+        return {
+            "status": "success",
+            "message": "Knowledge base successfully created!",
+            "details": kb_info
+        }
             
     except Exception as e:
-        raise HTTPException(status_code=500, detail=f"Knowledge base creation failed: {str(e)}")
+        error_msg = f"Knowledge base creation failed: {str(e)}"
+        print(error_msg)
+        print(traceback.format_exc())
+        raise HTTPException(status_code=500, detail=error_msg)
 
-# --------------- API Models ---------------
-class ChatRequest(BaseModel):
-    message: str
+def load_knowledge_base():
+    """Load the knowledge base from disk"""
+    global vector_store
+    
+    if vector_store:
+        return vector_store
+        
+    _, _embeddings = init_models()
+    
+    try:
+        vector_store = FAISS.load_local(
+            VECTOR_STORE_PATH,
+            _embeddings,
+            allow_dangerous_deserialization=True
+        )
+        return vector_store
+    except Exception as e:
+        error_msg = f"Failed to load knowledge base: {str(e)}"
+        print(error_msg)
+        print(traceback.format_exc())
+        return None
 
-class ChatResponse(BaseModel):
-    response: str
+# --------------- API Endpoints ---------------
+@app.get("/")
+async def root():
+    """Root endpoint that shows app status"""
+    vector_store_exists = os.path.exists(os.path.join(VECTOR_STORE_PATH, "index.faiss"))
+    
+    return {
+        "status": "running",
+        "knowledge_base_exists": vector_store_exists,
+        "kb_info": kb_info if vector_store_exists else None
+    }
+
+@app.post("/build-kb", response_model=BuildKnowledgeBaseResponse)
+async def build_kb_endpoint():
+    """Endpoint to build/rebuild the knowledge base"""
+    return build_knowledge_base()
 
-# --------------- API Routes ---------------
 @app.post("/chat", response_model=ChatResponse)
 async def chat_endpoint(request: ChatRequest):
+    """Endpoint to chat with the assistant"""
+    # Check if knowledge base exists
+    if not os.path.exists(os.path.join(VECTOR_STORE_PATH, "index.faiss")):
+        raise HTTPException(
+            status_code=400, 
+            detail="Knowledge base not found. Please build it first with /build-kb"
+        )
+    
+    # Use provided conversation ID or generate a new one
+    conversation_id = request.conversation_id or f"conv_{datetime.now().strftime('%Y%m%d%H%M%S')}"
+    
     try:
-        llm, embeddings = init_models()
+        # Load models and knowledge base
+        _llm, _ = init_models()
+        _vector_store = load_knowledge_base()
         
-        if not os.path.exists(VECTOR_STORE_PATH):
-            vector_store = build_knowledge_base(embeddings)
-        else:
-            vector_store = FAISS.load_local(
-                VECTOR_STORE_PATH,
-                embeddings,
-                allow_dangerous_deserialization=True
-            )
-
-        # Add retry logic for network operations
-        max_retries = 3
-        retry_count = 0
-        
-        while retry_count < max_retries:
-            try:
-                context_docs = vector_store.similarity_search(request.message)
-                context_text = "\n".join([d.page_content for d in context_docs])
-                
-                prompt_template = PromptTemplate.from_template('''
-                    You are a helpful and polite legal assistant at Status Law.
-                    You answer in the language in which the question was asked.
-                    Answer the question based on the context provided.
-                    
-                    # ... остальной текст промпта ...
-
-                    Context: {context}
-                    Question: {question}
-                    
-                    Response Guidelines:
-                    1. Answer in the user's language
-                    2. Cite sources when possible
-                    3. Offer contact options if unsure
-                    ''')
-                
-                chain = prompt_template | llm | StrOutputParser()
-                response = chain.invoke({
-                    "context": context_text,
-                    "question": request.message
-                })
-                
-                log_interaction(request.message, response, context_text)
-                return ChatResponse(response=response)
-                
-            except (requests.exceptions.RequestException, aiohttp.ClientError) as e:
-                retry_count += 1
-                if retry_count == max_retries:
-                    raise HTTPException(
-                        status_code=503,
-                        detail={
-                            "error": "Network error after maximum retries",
-                            "detail": str(e),
-                            "type": "network_error"
-                        }
-                    )
-                await asyncio.sleep(1 * retry_count)  # Exponential backoff
-                
-    except Exception as e:
-        if isinstance(e, (requests.exceptions.RequestException, aiohttp.ClientError)):
+        if not _vector_store:
             raise HTTPException(
-                status_code=503,
-                detail={
-                    "error": "Network error occurred",
-                    "detail": str(e),
-                    "type": "network_error"
-                }
+                status_code=500, 
+                detail="Failed to load knowledge base"
             )
-        raise HTTPException(status_code=500, detail=str(e))
-
-# --------------- Logging ---------------
-def log_interaction(user_input: str, bot_response: str, context: str):
-    try:
-        log_entry = {
-            "timestamp": datetime.now().isoformat(),
-            "user_input": user_input,
-            "bot_response": bot_response,
-            "context": context[:500],
-            "kb_version": datetime.now().strftime("%Y%m%d-%H%M%S")
-        }
         
-        os.makedirs("chat_history", exist_ok=True)
-        log_path = os.path.join("chat_history", "chat_logs.json")
+        # Retrieve context
+        context_docs = _vector_store.similarity_search(request.message)
+        context_text = "\n".join([d.page_content for d in context_docs])
         
-        with open(log_path, "a", encoding="utf-8") as f:
-            f.write(json.dumps(log_entry, ensure_ascii=False) + "\n")
-            
-    except Exception as e:
-        print(f"Logging error: {str(e)}")
-        print(traceback.format_exc())
+        # Generate response
+        prompt_template = PromptTemplate.from_template('''
+            You are a helpful and polite legal assistant at Status Law.
+            You answer in the language in which the question was asked.
+            Answer the question based on the context provided.
+            If you cannot answer based on the context, say so politely and offer to contact Status Law directly via the following channels:
+            - For all users: +32465594521 (landline phone).
+            - For English and Swedish speakers only: +46728495129 (available on WhatsApp, Telegram, Signal, IMO).
+            - Provide a link to the contact form: [Contact Form](https://status.law/law-firm-contact-legal-protection/).
+            If the user has questions about specific services and their costs, suggest they visit the page https://status.law/tariffs-for-services-of-protection-against-extradition-and-international-prosecution/ for detailed information.
 
-# Add health check endpoint
-@app.get("/health")
-async def health_check():
-    try:
-        # Check if models can be initialized
-        llm, embeddings = init_models()
+            Ask the user additional questions to understand which service to recommend and provide an estimated cost. For example, clarify their situation and needs to suggest the most appropriate options.
+
+            Also, offer free consultations if they are available and suitable for the user's request.
+            Answer professionally but in a friendly manner.
+
+            Example:
+            Q: How can I challenge the sanctions?
+            A: To challenge the sanctions, you should consult with our legal team, who specialize in this area. Please contact us directly for detailed advice. You can fill out our contact form here: [Contact Form](https://status.law/law-firm-contact-legal-protection/).
+
+            Context: {context}
+            Question: {question}
+            
+            Response Guidelines:
+            1. Answer in the user's language
+            2. Cite sources when possible
+            3. Offer contact options if unsure
+            ''')
         
-        # Check if vector store is accessible
-        if os.path.exists(VECTOR_STORE_PATH):
-            vector_store = FAISS.load_local(
-                VECTOR_STORE_PATH,
-                embeddings,
-                allow_dangerous_deserialization=True
-            )
+        chain = prompt_template | _llm | StrOutputParser()
+        response = chain.invoke({
+            "context": context_text,
+            "question": request.message
+        })
+        
+        # Log the interaction
+        log_interaction(request.message, response, context_text, conversation_id)
         
         return {
-            "status": "healthy",
-            "vector_store": "available" if os.path.exists(VECTOR_STORE_PATH) else "not_found"
+            "response": response,
+            "conversation_id": conversation_id
         }
-        
+                
     except Exception as e:
-        return JSONResponse(
-            status_code=503,
-            content={
-                "status": "unhealthy",
-                "error": str(e)
-            }
-        )
-
-# Add diagnostic endpoint
-@app.get("/directory-status")
-async def check_directory_status():
-    """Check status of required directories"""
-    return {
-        "vector_store": {
-            "exists": os.path.exists(VECTOR_STORE_PATH),
-            "path": os.path.abspath(VECTOR_STORE_PATH),
-            "contents": os.listdir(VECTOR_STORE_PATH) if os.path.exists(VECTOR_STORE_PATH) else []
-        },
-        "chat_history": {
-            "exists": os.path.exists(CHAT_HISTORY_PATH),
-            "path": os.path.abspath(CHAT_HISTORY_PATH),
-            "contents": os.listdir(CHAT_HISTORY_PATH) if os.path.exists(CHAT_HISTORY_PATH) else []
-        }
-    }
+        error_msg = f"Error generating response: {str(e)}"
+        print(error_msg)
+        print(traceback.format_exc())
+        raise HTTPException(status_code=500, detail=error_msg)
 
-# Добавим функцию для вывода статуса
-def print_startup_status():
-    """Print application startup status with rich formatting"""
+# Initialize dataset integration at startup
+@app.on_event("startup")
+async def startup_event():
+    """Initialize on startup"""
+    # Try to load existing knowledge base from Hugging Face
+    init_hf_dataset_integration()
+    
+    # Preload embeddings model to reduce first-request latency
     try:
-        # Create status table
-        table = Table(show_header=True, header_style="bold magenta")
-        table.add_column("Component", style="cyan")
-        table.add_column("Status", style="green")
-        
-        # Check directories
-        vector_store_exists = os.path.exists(VECTOR_STORE_PATH)
-        chat_history_exists = os.path.exists(CHAT_HISTORY_PATH)
-        
-        table.add_row(
-            "Vector Store Directory",
-            "✅ Created" if vector_store_exists else "❌ Missing"
-        )
-        table.add_row(
-            "Chat History Directory",
-            "✅ Created" if chat_history_exists else "❌ Missing"
-        )
-        
-        # Check environment variables
-        table.add_row(
-            "GROQ API Key",
-            "✅ Set" if os.getenv("GROQ_API_KEY") else "❌ Missing"
-        )
-        
-        # Create status panel
-        status_panel = Panel(
-            table,
-            title="[bold blue]Status Law Assistant API Status[/bold blue]",
-            border_style="blue"
-        )
-        
-        # Print startup message and status
-        console.print("\n")
-        console.print("[bold green]🚀 Server started successfully![/bold green]")
-        console.print(status_panel)
-        console.print("\n[bold yellow]API Documentation:[/bold yellow]")
-        console.print("📚 Swagger UI: http://0.0.0.0:8000/docs")
-        console.print("📘 ReDoc: http://0.0.0.0:8000/redoc\n")
-        
+        global embeddings
+        if not embeddings:
+            embeddings = HuggingFaceEmbeddings(
+                model_name="intfloat/multilingual-e5-large-instruct"
+            )
     except Exception as e:
-        console.print(f"[bold red]Error printing status: {str(e)}[/bold red]")
+        print(f"Warning: Failed to preload embeddings: {e}")
 
+# Run the application
 if __name__ == "__main__":
-    import uvicorn
-    
-    port = int(os.getenv("PORT", 8000))
-    logger.info(f"Starting server on port {port}")
-    
-    config = uvicorn.Config(
-        app,
-        host="0.0.0.0",
-        port=port,
-        log_level="debug"
-    )
-    
-    server = uvicorn.Server(config)
-    server.run()
+    uvicorn.run("app:app", host="0.0.0.0", port=8000)

index.html

@@ -0,0 +1,276 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>Status Law Assistant</title>
+    <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0/dist/css/bootstrap.min.css" rel="stylesheet">
+    <style>
+        body {
+            background-color: #f8f9fa;
+        }
+        .chat-container {
+            max-width: 800px;
+            margin: 30px auto;
+            background: white;
+            border-radius: 10px;
+            box-shadow: 0 4px 20px rgba(0, 0, 0, 0.1);
+            overflow: hidden;
+        }
+        .chat-header {
+            padding: 15px 20px;
+            background: linear-gradient(135deg, #2B5876 0%, #4E4376 100%);
+            color: white;
+            border-bottom: 1px solid #e6e6e6;
+        }
+        .chat-area {
+            height: 400px;
+            overflow-y: auto;
+            padding: 20px;
+            background-color: #f8f9fa;
+        }
+        .user-message, .bot-message {
+            padding: 8px 15px;
+            margin-bottom: 10px;
+            border-radius: 18px;
+            max-width: 75%;
+            word-wrap: break-word;
+        }
+        .user-message {
+            background-color: #e2f0ff;
+            margin-left: auto;
+            border-bottom-right-radius: 5px;
+        }
+        .bot-message {
+            background-color: #f0f0f0;
+            margin-right: auto;
+            border-bottom-left-radius: 5px;
+        }
+        .input-area {
+            padding: 15px;
+            background-color: white;
+            border-top: 1px solid #e6e6e6;
+        }
+        .kb-status {
+            padding: 10px 20px;
+            background-color: #f8f9fa;
+            border-top: 1px solid #e6e6e6;
+            font-size: 0.9em;
+            color: #6c757d;
+        }
+        .kb-badge {
+            font-size: 0.8em;
+            padding: 5px 10px;
+            border-radius: 15px;
+        }
+        .loading {
+            display: inline-block;
+            width: 20px;
+            height: 20px;
+            border: 3px solid rgba(0,0,0,.1);
+            border-radius: 50%;
+            border-top-color: #2B5876;
+            animation: spin 1s ease-in-out infinite;
+        }
+        @keyframes spin {
+            to { transform: rotate(360deg); }
+        }
+    </style>
+</head>
+<body>
+    <div class="container">
+        <div class="chat-container">
+            <div class="chat-header d-flex justify-content-between align-items-center">
+                <h1 class="h4 mb-0">⚖️ Status Law Assistant</h1>
+                <span id="kb-status-badge" class="kb-badge bg-warning">Checking...</span>
+            </div>
+            
+            <div id="kb-action-area" class="p-3 bg-light d-none">
+                <div class="alert alert-warning">
+                    Knowledge base not found. You need to build it before chatting.
+                </div>
+                <button id="build-kb-btn" class="btn btn-primary">Build Knowledge Base</button>
+            </div>
+            
+            <div id="chat-area" class="chat-area">
+                <div class="bot-message">
+                    Hello! I'm the Status Law assistant. How can I help you with your legal questions?
+                </div>
+            </div>
+            
+            <div class="input-area">
+                <div class="input-group">
+                    <input 
+                        type="text" 
+                        id="user-input" 
+                        class="form-control" 
+                        placeholder="Type your message here..." 
+                        aria-label="Message"
+                    >
+                    <button id="send-btn" class="btn btn-primary">Send</button>
+                </div>
+            </div>
+            
+            <div class="kb-status">
+                <small id="kb-info">Loading knowledge base info...</small>
+            </div>
+        </div>
+    </div>
+
+    <script>
+        // Global variables
+        let conversationId = null;
+        
+        // DOM elements
+        const userInput = document.getElementById('user-input');
+        const sendBtn = document.getElementById('send-btn');
+        const chatArea = document.getElementById('chat-area');
+        const kbInfo = document.getElementById('kb-info');
+        const kbStatusBadge = document.getElementById('kb-status-badge');
+        const kbActionArea = document.getElementById('kb-action-area');
+        const buildKbBtn = document.getElementById('build-kb-btn');
+        
+        // Check knowledge base status on load
+        checkKnowledgeBaseStatus();
+        
+        // Event listeners
+        sendBtn.addEventListener('click', sendMessage);
+        userInput.addEventListener('keypress', function(e) {
+            if (e.key === 'Enter') {
+                sendMessage();
+            }
+        });
+        
+        buildKbBtn.addEventListener('click', buildKnowledgeBase);
+        
+        // Functions
+        async function checkKnowledgeBaseStatus() {
+            try {
+                const response = await fetch('/');
+                const data = await response.json();
+                
+                if (data.knowledge_base_exists) {
+                    kbStatusBadge.className = 'kb-badge bg-success';
+                    kbStatusBadge.textContent = 'Ready';
+                    kbActionArea.classList.add('d-none');
+                    
+                    if (data.kb_info) {
+                        const date = new Date(data.kb_info.build_time * 1000);
+                        const formattedDate = date.toLocaleString();
+                        kbInfo.textContent = `Knowledge base version: ${data.kb_info.version || 'Unknown'}, Size: ${data.kb_info.size ? data.kb_info.size.toFixed(2) + ' MB' : 'Unknown'}`;
+                    }
+                } else {
+                    kbStatusBadge.className = 'kb-badge bg-danger';
+                    kbStatusBadge.textContent = 'Not Ready';
+                    kbActionArea.classList.remove('d-none');
+                    kbInfo.textContent = 'Knowledge base not found. Please build it first.';
+                }
+            } catch (error) {
+                console.error('Error checking KB status:', error);
+                kbStatusBadge.className = 'kb-badge bg-danger';
+                kbStatusBadge.textContent = 'Error';
+                kbInfo.textContent = 'Error checking knowledge base status.';
+            }
+        }
+        
+        async function buildKnowledgeBase() {
+            try {
+                kbStatusBadge.className = 'kb-badge bg-warning';
+                kbStatusBadge.textContent = 'Building...';
+                buildKbBtn.disabled = true;
+                buildKbBtn.innerHTML = '<span class="loading me-2"></span> Building...';
+                
+                const response = await fetch('/build-kb', {
+                    method: 'POST'
+                });
+                
+                const data = await response.json();
+                
+                if (response.ok) {
+                    kbStatusBadge.className = 'kb-badge bg-success';
+                    kbStatusBadge.textContent = 'Ready';
+                    kbActionArea.classList.add('d-none');
+                    
+                    if (data.details) {
+                        kbInfo.textContent = `Knowledge base version: ${data.details.version || 'Unknown'}, Size: ${data.details.size ? data.details.size.toFixed(2) + ' MB' : 'Unknown'}`;
+                    }
+                    
+                    // Add a system message
+                    addBotMessage("Knowledge base built successfully! You can now ask questions.");
+                } else {
+                    throw new Error(data.detail || 'Failed to build knowledge base');
+                }
+            } catch (error) {
+                console.error('Error building KB:', error);
+                kbStatusBadge.className = 'kb-badge bg-danger';
+                kbStatusBadge.textContent = 'Error';
+                kbInfo.textContent = 'Error building knowledge base.';
+                addBotMessage("There was an error building the knowledge base. Please try again later.");
+            } finally {
+                buildKbBtn.disabled = false;
+                buildKbBtn.textContent = 'Build Knowledge Base';
+            }
+        }
+        
+        async function sendMessage() {
+            const message = userInput.value.trim();
+            if (!message) return;
+            
+            // Add user message to chat
+            addUserMessage(message);
+            userInput.value = '';
+            
+            // Add a temporary bot message with loading indicator
+            const loadingMsgElement = addBotMessage('<span class="loading me-2"></span> Thinking...');
+            
+            try {
+                const response = await fetch('/chat', {
+                    method: 'POST',
+                    headers: {
+                        'Content-Type': 'application/json',
+                    },
+                    body: JSON.stringify({
+                        message: message,
+                        conversation_id: conversationId
+                    }),
+                });
+                
+                const data = await response.json();
+                
+                if (response.ok) {
+                    // Update conversation ID for future messages
+                    conversationId = data.conversation_id;
+                    
+                    // Replace loading message with actual response
+                    loadingMsgElement.innerHTML = data.response;
+                } else {
+                    throw new Error(data.detail || 'Failed to get response');
+                }
+            } catch (error) {
+                console.error('Error sending message:', error);
+                loadingMsgElement.innerHTML = "Sorry, I encountered an error processing your request. Please try again later.";
+            }
+            
+            // Scroll to bottom
+            chatArea.scrollTop = chatArea.scrollHeight;
+        }
+        
+        function addUserMessage(text) {
+            const div = document.createElement('div');
+            div.className = 'user-message';
+            div.textContent = text;
+            chatArea.appendChild(div);
+            chatArea.scrollTop = chatArea.scrollHeight;
+        }
+        
+        function addBotMessage(html) {
+            const div = document.createElement('div');
+            div.className = 'bot-message';
+            div.innerHTML = html;
+            chatArea.appendChild(div);
+            chatArea.scrollTop = chatArea.scrollHeight;
+            return div;
+        }
+    </script>
+</body>
+</html>

requirements.txt

@@ -1,24 +1,14 @@
-langchain-community
-langchain-core
-langchain-huggingface
-langchain-groq
-python-dotenv
-beautifulsoup4
-faiss-cpu
-requests
-langgraph
-langchain-anthropic
-fastapi
-uvicorn[standard]
-pydantic
-python-multipart
-pandas
-langchain
-plotly
-pytest
-httpx
-pytest-asyncio
-aiohttp
-requests
-tenacity
-rich>=10.0.0
+fastapi==0.109.2
+uvicorn==0.27.1
+langchain>=0.1.0
+langchain_groq>=0.1.0
+langchain_huggingface>=0.0.2
+langchain_community>=0.0.13
+langchain_text_splitters>=0.0.1
+langchain_core>=0.1.10
+faiss-cpu>=1.7.4
+python-dotenv>=1.0.0
+huggingface_hub>=0.19.0
+jinja2>=3.0.0
+aiofiles>=0.8.0
+python-multipart>=0.0.6

space.yaml

@@ -0,0 +1,9 @@
+sdk: docker
+title: Status Law Assistant
+emoji: ⚖️
+colorFrom: blue
+colorTo: indigo
+sdk_version: 0.0.1
+app_port: 8000
+pinned: false
+license: mit