* Add migration

* Delete legacy from bot * Clear old models * Единый http клиент * РАГ полечен
2025-12-24 13:44:52 +03:00
parent 1ce1c23d10
commit 8bdacb4f7a
19 changed files with 200 additions and 565 deletions
@@ -1,172 +0,0 @@
-import json
-from typing import Optional, AsyncIterator
-import httpx
-from tg_bot.config.settings import settings
-
-
-class DeepSeekAPIError(Exception):
-    pass
-
-
-class DeepSeekClient:
-    
-    def __init__(self, api_key: str | None = None, api_url: str | None = None):
-        self.api_key = api_key or settings.DEEPSEEK_API_KEY
-        self.api_url = api_url or settings.DEEPSEEK_API_URL
-        self.timeout = 60.0
-    
-    def _get_headers(self) -> dict[str, str]:
-        if not self.api_key:
-            raise DeepSeekAPIError("API key not set")
-        
-        return {
-            "Content-Type": "application/json",
-            "Authorization": f"Bearer {self.api_key}"
-        }
-    
-    async def chat_completion(
-        self,
-        messages: list[dict[str, str]],
-        model: str = "deepseek-chat",
-        temperature: float = 0.7,
-        max_tokens: Optional[int] = None,
-        stream: bool = False
-    ) -> dict:
-        if not self.api_key:
-            return {
-                "content": "API key not configured",
-                "usage": {
-                    "prompt_tokens": 0,
-                    "completion_tokens": 0,
-                    "total_tokens": 0
-                }
-            }
-        
-        payload = {
-            "model": model,
-            "messages": messages,
-            "temperature": temperature,
-            "stream": stream
-        }
-        
-        if max_tokens is not None:
-            payload["max_tokens"] = max_tokens
-        
-        try:
-            async with httpx.AsyncClient(timeout=self.timeout) as client:
-                response = await client.post(
-                    self.api_url,
-                    headers=self._get_headers(),
-                    json=payload
-                )
-                response.raise_for_status()
-                
-                data = response.json()
-                
-                if "choices" in data and len(data["choices"]) > 0:
-                    content = data["choices"][0]["message"]["content"]
-                else:
-                    raise DeepSeekAPIError("Invalid response format")
-                
-                usage = data.get("usage", {})
-                
-                return {
-                    "content": content,
-                    "usage": {
-                        "prompt_tokens": usage.get("prompt_tokens", 0),
-                        "completion_tokens": usage.get("completion_tokens", 0),
-                        "total_tokens": usage.get("total_tokens", 0)
-                    }
-                }
-                
-        except httpx.HTTPStatusError as e:
-            error_msg = f"API error: {e.response.status_code}"
-            try:
-                error_data = e.response.json()
-                if "error" in error_data:
-                    error_msg = error_data['error'].get('message', error_msg)
-            except:
-                pass
-            raise DeepSeekAPIError(error_msg) from e
-        except httpx.RequestError as e:
-            raise DeepSeekAPIError(f"Connection error: {str(e)}") from e
-        except Exception as e:
-            raise DeepSeekAPIError(str(e)) from e
-    
-    async def stream_chat_completion(
-        self,
-        messages: list[dict[str, str]],
-        model: str = "deepseek-chat",
-        temperature: float = 0.7,
-        max_tokens: Optional[int] = None
-    ) -> AsyncIterator[str]:
-        if not self.api_key:
-            yield "API key not configured"
-            return
-        
-        payload = {
-            "model": model,
-            "messages": messages,
-            "temperature": temperature,
-            "stream": True
-        }
-        
-        if max_tokens is not None:
-            payload["max_tokens"] = max_tokens
-        
-        try:
-            async with httpx.AsyncClient(timeout=self.timeout) as client:
-                async with client.stream(
-                    "POST",
-                    self.api_url,
-                    headers=self._get_headers(),
-                    json=payload
-                ) as response:
-                    response.raise_for_status()
-                    
-                    async for line in response.aiter_lines():
-                        if not line.strip():
-                            continue
-                        
-                        if line.startswith("data: "):
-                            line = line[6:]
-                        
-                        if line.strip() == "[DONE]":
-                            break
-                        
-                        try:
-                            data = json.loads(line)
-                            
-                            if "choices" in data and len(data["choices"]) > 0:
-                                delta = data["choices"][0].get("delta", {})
-                                content = delta.get("content", "")
-                                if content:
-                                    yield content
-                        except json.JSONDecodeError:
-                            continue
-                            
-        except httpx.HTTPStatusError as e:
-            error_msg = f"API error: {e.response.status_code}"
-            try:
-                error_data = e.response.json()
-                if "error" in error_data:
-                    error_msg = error_data['error'].get('message', error_msg)
-            except:
-                pass
-            raise DeepSeekAPIError(error_msg) from e
-        except httpx.RequestError as e:
-            raise DeepSeekAPIError(f"Connection error: {str(e)}") from e
-        except Exception as e:
-            raise DeepSeekAPIError(str(e)) from e
-    
-    async def health_check(self) -> bool:
-        if not self.api_key:
-            return False
-        
-        try:
-            test_messages = [{"role": "user", "content": "test"}]
-            await self.chat_completion(test_messages, max_tokens=1)
-            return True
-        except Exception:
-            return False
-
@@ -1,87 +1,23 @@
-"""HTTP client utilities for making requests to the backend API"""
 import aiohttp
 from typing import Optional
-import ssl
-import os
-
-
-def get_windows_host_ip() -> Optional[str]:
-    """
-    Get the Windows host IP address when running in WSL.
-    In WSL2, the Windows host IP is typically the first nameserver in /etc/resolv.conf.
-    """
-    try:
-        if os.path.exists("/etc/resolv.conf"):
-            with open("/etc/resolv.conf", "r") as f:
-                for line in f:
-                    if line.startswith("nameserver"):
-                        ip = line.split()[1]
-                        if ip not in ["127.0.0.1", "127.0.0.53"] and not ip.startswith("fe80"):
-                            return ip
-    except Exception:
-        pass
-    return None
-
-
-def normalize_backend_url(url: str) -> str:
-    """
-    Normalize backend URL for better compatibility, especially on WSL and Docker.
-    """
-    if not ("localhost" in url or "127.0.0.1" in url):
-        return url
-    if os.path.exists("/.dockerenv"):
-        print(f"Warning: Running in Docker but URL contains localhost: {url}")
-        print("Please set BACKEND_URL environment variable in docker-compose.yml to use Docker service name (e.g., http://backend:8000/api/v1)")
-        return url.replace("localhost", "127.0.0.1")
-    try:
-        if os.path.exists("/proc/version"):
-            with open("/proc/version", "r") as f:
-                version_content = f.read().lower()
-                if "microsoft" in version_content:
-                    windows_ip = get_windows_host_ip()
-                    if windows_ip:
-                        if "localhost" in url or "127.0.0.1" in url:
-                            url = url.replace("localhost", windows_ip).replace("127.0.0.1", windows_ip)
-                            print(f"WSL detected: Using Windows host IP {windows_ip} for backend connection")
-                            return url
-    except Exception as e:
-        print(f"Warning: Could not detect WSL environment: {e}")
-    
-    if url.startswith("http://localhost") or url.startswith("https://localhost"):
-        return url.replace("localhost", "127.0.0.1")
-    return url


 def create_http_session(timeout: Optional[aiohttp.ClientTimeout] = None) -> aiohttp.ClientSession:
    """
-    Create a configured aiohttp ClientSession for backend API requests.
-    
-    Args:
-        timeout: Optional timeout configuration. Defaults to 30 seconds total timeout.
-    
-    Returns:
-        Configured aiohttp.ClientSession
+    Создаем сессию для запросов к бэку
    """
    if timeout is None:
        timeout = aiohttp.ClientTimeout(total=30, connect=10)
    
    connector = aiohttp.TCPConnector(
-        ssl=False,
        limit=100,
-        limit_per_host=30,
-        force_close=True,
-        enable_cleanup_closed=True
+        limit_per_host=30
    )
    
-    ssl_context = ssl.create_default_context()
-    ssl_context.check_hostname = False
-    ssl_context.verify_mode = ssl.CERT_NONE
-    
    return aiohttp.ClientSession(
        connector=connector,
        timeout=timeout,
        headers={
-            "Content-Type": "application/json",
            "Accept": "application/json"
        }
    )
@@ -4,6 +4,7 @@ from aiogram.filters import Command, StateFilter
 from aiogram.fsm.context import FSMContext
 import aiohttp
 from tg_bot.config.settings import settings
+from tg_bot.infrastructure.http_client import create_http_session
 from tg_bot.infrastructure.telegram.states.collection_states import (
    CollectionAccessStates,
    CollectionEditStates
@@ -14,7 +15,7 @@ router = Router()

 async def get_user_collections(telegram_id: str):
    try:
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            async with session.get(
                f"{settings.BACKEND_URL}/collections/",
                headers={"X-Telegram-ID": telegram_id}
@@ -33,7 +34,7 @@ async def get_collection_documents(collection_id: str, telegram_id: str):
        url = f"{settings.BACKEND_URL}/documents/collection/{collection_id}"
        print(f"DEBUG get_collection_documents: URL={url}, collection_id={collection_id}, telegram_id={telegram_id}")
        
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            async with session.get(
                url,
                headers={"X-Telegram-ID": telegram_id}
@@ -57,7 +58,7 @@ async def get_collection_documents(collection_id: str, telegram_id: str):

 async def search_in_collection(collection_id: str, query: str, telegram_id: str):
    try:
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            async with session.get(
                f"{settings.BACKEND_URL}/documents/collection/{collection_id}",
                params={"search": query},
@@ -78,7 +79,7 @@ async def get_collection_info(collection_id: str, telegram_id: str):
        url = f"{settings.BACKEND_URL}/collections/{collection_id}"
        print(f"DEBUG get_collection_info: URL={url}, collection_id={collection_id}, telegram_id={telegram_id}")
        
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            async with session.get(
                url,
                headers={"X-Telegram-ID": telegram_id}
@@ -103,7 +104,7 @@ async def get_collection_info(collection_id: str, telegram_id: str):
 async def get_collection_access_list(collection_id: str, telegram_id: str):
    """Получить список пользователей с доступом к коллекции"""
    try:
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            async with session.get(
                f"{settings.BACKEND_URL}/collections/{collection_id}/access",
                headers={"X-Telegram-ID": telegram_id}
@@ -122,7 +123,7 @@ async def grant_collection_access(collection_id: str, telegram_id: str, owner_te
        url = f"{settings.BACKEND_URL}/collections/{collection_id}/access/telegram/{telegram_id}"
        print(f"DEBUG grant_collection_access: URL={url}, target_telegram_id={telegram_id}, owner_telegram_id={owner_telegram_id}")
        
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            async with session.post(
                url,
                headers={"X-Telegram-ID": owner_telegram_id}
@@ -145,7 +146,7 @@ async def grant_collection_access(collection_id: str, telegram_id: str, owner_te
 async def revoke_collection_access(collection_id: str, telegram_id: str, owner_telegram_id: str):
    """Отозвать доступ к коллекции"""
    try:
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            async with session.delete(
                f"{settings.BACKEND_URL}/collections/{collection_id}/access/telegram/{telegram_id}",
                headers={"X-Telegram-ID": owner_telegram_id}
@@ -281,7 +282,7 @@ async def show_collection_menu(callback: CallbackQuery):
    collection_name = collection_info.get("name", "Коллекция")
    
    try:
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            async with session.get(
                f"{settings.BACKEND_URL}/users/telegram/{telegram_id}"
            ) as response:
@@ -673,7 +674,7 @@ async def process_edit_collection_description(message: Message, state: FSMContex
        if new_description:
            update_data["description"] = new_description
        
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            async with session.put(
                f"{settings.BACKEND_URL}/collections/{collection_id}",
                json=update_data,
@@ -7,6 +7,7 @@ from aiogram.filters import StateFilter
 from aiogram.fsm.context import FSMContext
 import aiohttp
 from tg_bot.config.settings import settings
+from tg_bot.infrastructure.http_client import create_http_session
 from tg_bot.infrastructure.telegram.states.collection_states import (
    DocumentEditStates,
    DocumentUploadStates
@@ -18,7 +19,7 @@ router = Router()
 async def get_document_info(document_id: str, telegram_id: str):
    """Получить информацию о документе"""
    try:
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            async with session.get(
                f"{settings.BACKEND_URL}/documents/{document_id}",
                headers={"X-Telegram-ID": telegram_id}
@@ -34,7 +35,7 @@ async def get_document_info(document_id: str, telegram_id: str):
 async def delete_document(document_id: str, telegram_id: str):
    """Удалить документ"""
    try:
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            async with session.delete(
                f"{settings.BACKEND_URL}/documents/{document_id}",
                headers={"X-Telegram-ID": telegram_id}
@@ -54,7 +55,7 @@ async def update_document(document_id: str, telegram_id: str, title: str = None,
        if content:
            update_data["content"] = content
        
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            async with session.put(
                f"{settings.BACKEND_URL}/documents/{document_id}",
                json=update_data,
@@ -71,7 +72,7 @@ async def update_document(document_id: str, telegram_id: str, title: str = None,
 async def upload_document_to_collection(collection_id: str, file_data: bytes, filename: str, telegram_id: str):
    """Загрузить документ в коллекцию"""
    try:
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            form_data = aiohttp.FormData()
            form_data.add_field('file', file_data, filename=filename, content_type='application/octet-stream')
            
@@ -120,7 +121,7 @@ async def view_document(callback: CallbackQuery):
        response += "\n\n<i>...</i>"
    
    try:
-        async with aiohttp.ClientSession() as session:
+        async with create_http_session() as session:
            async with session.get(
                f"{settings.BACKEND_URL}/collections/{collection_id}",
                headers={"X-Telegram-ID": telegram_id}
@@ -1,7 +1,5 @@
 from aiogram import Router, types
 from aiogram.types import Message
-from datetime import datetime
-import aiohttp
 from tg_bot.config.settings import settings
 from tg_bot.domain.services.user_service import UserService, User
 from tg_bot.application.services.rag_service import RAGService
@@ -60,12 +58,7 @@ async def process_premium_question(message: Message, user: User, question_text:
        answer = rag_result.get("answer", "Извините, не удалось сгенерировать ответ.")
        sources = rag_result.get("sources", [])
        
-        await save_conversation_to_backend(
-            str(message.from_user.id),
-            question_text,
-            answer,
-            sources
-        )
+        # Беседа уже сохранена в бэкенде через API /rag/question
        
        response = (
            f"<b>Ваш вопрос:</b>\n"
@@ -74,18 +67,10 @@ async def process_premium_question(message: Message, user: User, question_text:
        )
        
        if sources:
-            response += f"<b>Источники из коллекций:</b>\n"
-            collections_used = {}
-            for source in sources[:5]:
-                collection_name = source.get('collection', 'Неизвестно')
-                if collection_name not in collections_used:
-                    collections_used[collection_name] = []
-                collections_used[collection_name].append(source.get('title', 'Без названия'))
-            
-            for i, (collection_name, titles) in enumerate(collections_used.items(), 1):
-                response += f"{i}. <b>Коллекция:</b> {collection_name}\n"
-                for title in titles[:2]:
-                    response += f"   • {title}\n"
+            response += f"<b>Источники:</b>\n"
+            for idx, source in enumerate(sources[:5], 1):
+                title = source.get('title', 'Без названия')
+                response += f"{idx}. {title}\n"
            response += "\n<i>Используйте /mycollections для просмотра всех коллекций</i>\n\n"
        
        response += (
@@ -122,12 +107,7 @@ async def process_free_question(message: Message, user: User, question_text: str
        answer = rag_result.get("answer", "Извините, не удалось сгенерировать ответ.")
        sources = rag_result.get("sources", [])
        
-        await save_conversation_to_backend(
-            str(message.from_user.id),
-            question_text,
-            answer,
-            sources
-        )
+        # Уже все сохранили через /rag/question
        
        response = (
            f"<b>Ваш вопрос:</b>\n"
@@ -136,18 +116,10 @@ async def process_free_question(message: Message, user: User, question_text: str
        )
        
        if sources:
-            response += f"<b>Источники из коллекций:</b>\n"
-            collections_used = {}
-            for source in sources[:5]:
-                collection_name = source.get('collection', 'Неизвестно')
-                if collection_name not in collections_used:
-                    collections_used[collection_name] = []
-                collections_used[collection_name].append(source.get('title', 'Без названия'))
-            
-            for i, (collection_name, titles) in enumerate(collections_used.items(), 1):
-                response += f"{i}. <b>Коллекция:</b> {collection_name}\n"
-                for title in titles[:2]:
-                    response += f"   • {title}\n"
+            response += f"<b>Источники:</b>\n"
+            for idx, source in enumerate(sources[:5], 1):
+                title = source.get('title', 'Без названия')
+                response += f"{idx}. {title}\n"
            response += "\n<i>Используйте /mycollections для просмотра всех коллекций</i>\n\n"
        
        response += (
@@ -176,83 +148,7 @@ async def process_free_question(message: Message, user: User, question_text: str
    await message.answer(response, parse_mode="HTML")


-async def save_conversation_to_backend(telegram_id: str, question: str, answer: str, sources: list):
-    try:
-        from tg_bot.config.settings import settings
-        backend_url = settings.BACKEND_URL
-        async with aiohttp.ClientSession() as session:
-            async with session.get(
-                f"{backend_url}/users/telegram/{telegram_id}"
-            ) as user_response:
-                if user_response.status != 200:
-                    return
-                user_data = await user_response.json()
-                user_uuid = user_data.get("user_id")
-            
-            async with session.get(
-                f"{backend_url}/collections/",
-                headers={"X-Telegram-ID": telegram_id}
-            ) as collections_response:
-                collections = []
-                if collections_response.status == 200:
-                    collections = await collections_response.json()
-            
-            collection_id = None
-            if collections:
-                collection_id = collections[0].get("collection_id")
-            else:
-                async with session.post(
-                    f"{backend_url}/collections",
-                    json={
-                        "name": "Основная коллекция",
-                        "description": "Коллекция по умолчанию",
-                        "is_public": False
-                    },
-                    headers={"X-Telegram-ID": telegram_id}
-                ) as create_collection_response:
-                    if create_collection_response.status in [200, 201]:
-                        collection_data = await create_collection_response.json()
-                        collection_id = collection_data.get("collection_id")
-            
-            if not collection_id:
-                return
-            
-            async with session.post(
-                f"{backend_url}/conversations",
-                json={"collection_id": str(collection_id)},
-                headers={"X-Telegram-ID": telegram_id}
-            ) as conversation_response:
-                if conversation_response.status not in [200, 201]:
-                    return
-                conversation_data = await conversation_response.json()
-                conversation_id = conversation_data.get("conversation_id")
-            
-            if not conversation_id:
-                return
-            
-            await session.post(
-                f"{backend_url}/messages",
-                json={
-                    "conversation_id": str(conversation_id),
-                    "content": question,
-                    "role": "user"
-                },
-                headers={"X-Telegram-ID": telegram_id}
-            )
-            
-            await session.post(
-                f"{backend_url}/messages",
-                json={
-                    "conversation_id": str(conversation_id),
-                    "content": answer,
-                    "role": "assistant",
-                    "sources": {"documents": sources}
-                },
-                headers={"X-Telegram-ID": telegram_id}
-            )
-            
-    except Exception as e:
-        print(f"Error saving conversation: {e}")
+#Сново сохраняется в /rag/question


 async def handle_limit_exceeded(message: Message, user: User):