AstrBot 4.3.5__py3-none-any.whl → 4.5.0__py3-none-any.whl

astrbot/core/platform/sources/satori/satori_event.py

@@ -2,7 +2,18 @@ from typing import TYPE_CHECKING
 from astrbot.api import logger
 from astrbot.api.event import AstrMessageEvent, MessageChain
 from astrbot.api.platform import AstrBotMessage, PlatformMetadata
-from astrbot.api.message_components import Plain, Image, At, File, Record, Video, Reply
+from astrbot.api.message_components import (
+    Plain,
+    Image,
+    At,
+    File,
+    Record,
+    Video,
+    Reply,
+    Forward,
+    Node,
+    Nodes,
+)
 
 if TYPE_CHECKING:
     from .satori_adapter import SatoriPlatformAdapter
@@ -48,55 +59,24 @@ class SatoriPlatformEvent(AstrMessageEvent):
             content_parts = []
 
             for component in message.chain:
-                if isinstance(component, Plain):
-                    text = (
-                        component.text.replace("&", "&")
-                        .replace("<", "&lt;")
-                        .replace(">", "&gt;")
-                    )
-                    content_parts.append(text)
-
-                elif isinstance(component, At):
-                    if component.qq:
-                        content_parts.append(f'<at id="{component.qq}"/>')
-                    elif component.name:
-                        content_parts.append(f'<at name="{component.name}"/>')
-
-                elif isinstance(component, Image):
-                    try:
-                        image_base64 = await component.convert_to_base64()
-                        if image_base64:
-                            content_parts.append(
-                                f'<img src="data:image/jpeg;base64,{image_base64}"/>'
-                            )
-                    except Exception as e:
-                        logger.error(f"图片转换为base64失败: {e}")
-
-                elif isinstance(component, File):
-                    content_parts.append(
-                        f'<file src="{component.file}" name="{component.name or "文件"}"/>'
-                    )
-
-                elif isinstance(component, Record):
-                    try:
-                        record_base64 = await component.convert_to_base64()
-                        if record_base64:
-                            content_parts.append(
-                                f'<audio src="data:audio/wav;base64,{record_base64}"/>'
-                            )
-                    except Exception as e:
-                        logger.error(f"语音转换为base64失败: {e}")
-
-                elif isinstance(component, Reply):
-                    content_parts.append(f'<reply id="{component.id}"/>')
-
-                elif isinstance(component, Video):
-                    try:
-                        video_path_url = await component.convert_to_file_path()
-                        if video_path_url:
-                            content_parts.append(f'<video src="{video_path_url}"/>')
-                    except Exception as e:
-                        logger.error(f"视频文件转换失败: {e}")
+                component_content = await cls._convert_component_to_satori_static(
+                    component
+                )
+                if component_content:
+                    content_parts.append(component_content)
+
+                # 特殊处理 Node 和 Nodes 组件
+                if isinstance(component, Node):
+                    # 单个转发节点
+                    node_content = await cls._convert_node_to_satori_static(component)
+                    if node_content:
+                        content_parts.append(node_content)
+
+                elif isinstance(component, Nodes):
+                    # 合并转发消息
+                    node_content = await cls._convert_nodes_to_satori_static(component)
+                    if node_content:
+                        content_parts.append(node_content)
 
             content = "".join(content_parts)
             channel_id = session_id
@@ -138,55 +118,22 @@ class SatoriPlatformEvent(AstrMessageEvent):
             content_parts = []
 
             for component in message.chain:
-                if isinstance(component, Plain):
-                    text = (
-                        component.text.replace("&", "&amp;")
-                        .replace("<", "&lt;")
-                        .replace(">", "&gt;")
-                    )
-                    content_parts.append(text)
-
-                elif isinstance(component, At):
-                    if component.qq:
-                        content_parts.append(f'<at id="{component.qq}"/>')
-                    elif component.name:
-                        content_parts.append(f'<at name="{component.name}"/>')
-
-                elif isinstance(component, Image):
-                    try:
-                        image_base64 = await component.convert_to_base64()
-                        if image_base64:
-                            content_parts.append(
-                                f'<img src="data:image/jpeg;base64,{image_base64}"/>'
-                            )
-                    except Exception as e:
-                        logger.error(f"图片转换为base64失败: {e}")
-
-                elif isinstance(component, File):
-                    content_parts.append(
-                        f'<file src="{component.file}" name="{component.name or "文件"}"/>'
-                    )
-
-                elif isinstance(component, Record):
-                    try:
-                        record_base64 = await component.convert_to_base64()
-                        if record_base64:
-                            content_parts.append(
-                                f'<audio src="data:audio/wav;base64,{record_base64}"/>'
-                            )
-                    except Exception as e:
-                        logger.error(f"语音转换为base64失败: {e}")
-
-                elif isinstance(component, Reply):
-                    content_parts.append(f'<reply id="{component.id}"/>')
-
-                elif isinstance(component, Video):
-                    try:
-                        video_path_url = await component.convert_to_file_path()
-                        if video_path_url:
-                            content_parts.append(f'<video src="{video_path_url}"/>')
-                    except Exception as e:
-                        logger.error(f"视频文件转换失败: {e}")
+                component_content = await self._convert_component_to_satori(component)
+                if component_content:
+                    content_parts.append(component_content)
+
+                # 特殊处理 Node 和 Nodes 组件
+                if isinstance(component, Node):
+                    # 单个转发节点
+                    node_content = await self._convert_node_to_satori(component)
+                    if node_content:
+                        content_parts.append(node_content)
+
+                elif isinstance(component, Nodes):
+                    # 合并转发消息
+                    node_content = await self._convert_nodes_to_satori(component)
+                    if node_content:
+                        content_parts.append(node_content)
 
             content = "".join(content_parts)
             channel_id = self.session_id
@@ -250,3 +197,227 @@ class SatoriPlatformEvent(AstrMessageEvent):
             logger.error(f"Satori 流式消息发送异常: {e}")
 
         return await super().send_streaming(generator, use_fallback)
+
+    async def _convert_component_to_satori(self, component) -> str:
+        """将单个消息组件转换为 Satori 格式"""
+        try:
+            if isinstance(component, Plain):
+                text = (
+                    component.text.replace("&", "&amp;")
+                    .replace("<", "&lt;")
+                    .replace(">", "&gt;")
+                )
+                return text
+
+            elif isinstance(component, At):
+                if component.qq:
+                    return f'<at id="{component.qq}"/>'
+                elif component.name:
+                    return f'<at name="{component.name}"/>'
+
+            elif isinstance(component, Image):
+                try:
+                    image_base64 = await component.convert_to_base64()
+                    if image_base64:
+                        return f'<img src="data:image/jpeg;base64,{image_base64}"/>'
+                except Exception as e:
+                    logger.error(f"图片转换为base64失败: {e}")
+
+            elif isinstance(component, File):
+                return (
+                    f'<file src="{component.file}" name="{component.name or "文件"}"/>'
+                )
+
+            elif isinstance(component, Record):
+                try:
+                    record_base64 = await component.convert_to_base64()
+                    if record_base64:
+                        return f'<audio src="data:audio/wav;base64,{record_base64}"/>'
+                except Exception as e:
+                    logger.error(f"语音转换为base64失败: {e}")
+
+            elif isinstance(component, Reply):
+                return f'<reply id="{component.id}"/>'
+
+            elif isinstance(component, Video):
+                try:
+                    video_path_url = await component.convert_to_file_path()
+                    if video_path_url:
+                        return f'<video src="{video_path_url}"/>'
+                except Exception as e:
+                    logger.error(f"视频文件转换失败: {e}")
+
+            elif isinstance(component, Forward):
+                return f'<message id="{component.id}" forward/>'
+
+            # 对于其他未处理的组件类型，返回空字符串
+            return ""
+
+        except Exception as e:
+            logger.error(f"转换消息组件失败: {e}")
+            return ""
+
+    async def _convert_node_to_satori(self, node: Node) -> str:
+        """将单个转发节点转换为 Satori 格式"""
+        try:
+            content_parts = []
+            if node.content:
+                for content_component in node.content:
+                    component_content = await self._convert_component_to_satori(
+                        content_component
+                    )
+                    if component_content:
+                        content_parts.append(component_content)
+
+            content = "".join(content_parts)
+
+            # 如果内容为空，添加默认内容
+            if not content.strip():
+                content = "[转发消息]"
+
+            # 构建 Satori 格式的转发节点
+            author_attrs = []
+            if node.uin:
+                author_attrs.append(f'id="{node.uin}"')
+            if node.name:
+                author_attrs.append(f'name="{node.name}"')
+
+            author_attr_str = " ".join(author_attrs)
+
+            return f"<message><author {author_attr_str}/>{content}</message>"
+
+        except Exception as e:
+            logger.error(f"转换转发节点失败: {e}")
+            return ""
+
+    @classmethod
+    async def _convert_component_to_satori_static(cls, component) -> str:
+        """将单个消息组件转换为 Satori 格式"""
+        try:
+            if isinstance(component, Plain):
+                text = (
+                    component.text.replace("&", "&amp;")
+                    .replace("<", "&lt;")
+                    .replace(">", "&gt;")
+                )
+                return text
+
+            elif isinstance(component, At):
+                if component.qq:
+                    return f'<at id="{component.qq}"/>'
+                elif component.name:
+                    return f'<at name="{component.name}"/>'
+
+            elif isinstance(component, Image):
+                try:
+                    image_base64 = await component.convert_to_base64()
+                    if image_base64:
+                        return f'<img src="data:image/jpeg;base64,{image_base64}"/>'
+                except Exception as e:
+                    logger.error(f"图片转换为base64失败: {e}")
+
+            elif isinstance(component, File):
+                return (
+                    f'<file src="{component.file}" name="{component.name or "文件"}"/>'
+                )
+
+            elif isinstance(component, Record):
+                try:
+                    record_base64 = await component.convert_to_base64()
+                    if record_base64:
+                        return f'<audio src="data:audio/wav;base64,{record_base64}"/>'
+                except Exception as e:
+                    logger.error(f"语音转换为base64失败: {e}")
+
+            elif isinstance(component, Reply):
+                return f'<reply id="{component.id}"/>'
+
+            elif isinstance(component, Video):
+                try:
+                    video_path_url = await component.convert_to_file_path()
+                    if video_path_url:
+                        return f'<video src="{video_path_url}"/>'
+                except Exception as e:
+                    logger.error(f"视频文件转换失败: {e}")
+
+            elif isinstance(component, Forward):
+                return f'<message id="{component.id}" forward/>'
+
+            # 对于其他未处理的组件类型，返回空字符串
+            return ""
+
+        except Exception as e:
+            logger.error(f"转换消息组件失败: {e}")
+            return ""
+
+    @classmethod
+    async def _convert_node_to_satori_static(cls, node: Node) -> str:
+        """将单个转发节点转换为 Satori 格式"""
+        try:
+            content_parts = []
+            if node.content:
+                for content_component in node.content:
+                    component_content = await cls._convert_component_to_satori_static(
+                        content_component
+                    )
+                    if component_content:
+                        content_parts.append(component_content)
+
+            content = "".join(content_parts)
+
+            # 如果内容为空，添加默认内容
+            if not content.strip():
+                content = "[转发消息]"
+
+            author_attrs = []
+            if node.uin:
+                author_attrs.append(f'id="{node.uin}"')
+            if node.name:
+                author_attrs.append(f'name="{node.name}"')
+
+            author_attr_str = " ".join(author_attrs)
+
+            return f"<message><author {author_attr_str}/>{content}</message>"
+
+        except Exception as e:
+            logger.error(f"转换转发节点失败: {e}")
+            return ""
+
+    async def _convert_nodes_to_satori(self, nodes: Nodes) -> str:
+        """将多个转发节点转换为 Satori 格式的合并转发"""
+        try:
+            node_parts = []
+
+            for node in nodes.nodes:
+                node_content = await self._convert_node_to_satori(node)
+                if node_content:
+                    node_parts.append(node_content)
+
+            if node_parts:
+                return f"<message forward>{''.join(node_parts)}</message>"
+            else:
+                return ""
+
+        except Exception as e:
+            logger.error(f"转换合并转发消息失败: {e}")
+            return ""
+
+    @classmethod
+    async def _convert_nodes_to_satori_static(cls, nodes: Nodes) -> str:
+        """将多个转发节点转换为 Satori 格式的合并转发"""
+        try:
+            node_parts = []
+
+            for node in nodes.nodes:
+                node_content = await cls._convert_node_to_satori_static(node)
+                if node_content:
+                    node_parts.append(node_content)
+
+            if node_parts:
+                return f"<message forward>{''.join(node_parts)}</message>"
+            else:
+                return ""
+
+        except Exception as e:
+            logger.error(f"转换合并转发消息失败: {e}")
+            return ""

astrbot/core/provider/manager.py

@@ -1,6 +1,5 @@
 import asyncio
 import traceback
-from typing import List
 
 from astrbot.core import logger, sp
 from astrbot.core.astrbot_config_mgr import AstrBotConfigManager
@@ -28,7 +27,7 @@ class ProviderManager:
         self.persona_mgr = persona_mgr
         self.acm = acm
         config = acm.confs["default"]
-        self.providers_config: List = config["provider"]
+        self.providers_config: list = config["provider"]
         self.provider_settings: dict = config["provider_settings"]
         self.provider_stt_settings: dict = config.get("provider_stt_settings", {})
         self.provider_tts_settings: dict = config.get("provider_tts_settings", {})
@@ -36,15 +35,15 @@ class ProviderManager:
         # 人格相关属性，v4.0.0 版本后被废弃，推荐使用 PersonaManager
         self.default_persona_name = persona_mgr.default_persona
 
-        self.provider_insts: List[Provider] = []
+        self.provider_insts: list[Provider] = []
         """加载的 Provider 的实例"""
-        self.stt_provider_insts: List[STTProvider] = []
+        self.stt_provider_insts: list[STTProvider] = []
         """加载的 Speech To Text Provider 的实例"""
-        self.tts_provider_insts: List[TTSProvider] = []
+        self.tts_provider_insts: list[TTSProvider] = []
         """加载的 Text To Speech Provider 的实例"""
-        self.embedding_provider_insts: List[EmbeddingProvider] = []
+        self.embedding_provider_insts: list[EmbeddingProvider] = []
         """加载的 Embedding Provider 的实例"""
-        self.rerank_provider_insts: List[RerankProvider] = []
+        self.rerank_provider_insts: list[RerankProvider] = []
         """加载的 Rerank Provider 的实例"""
         self.inst_map: dict[
             str,
@@ -175,7 +174,11 @@ class ProviderManager:
     async def initialize(self):
         # 逐个初始化提供商
         for provider_config in self.providers_config:
-            await self.load_provider(provider_config)
+            try:
+                await self.load_provider(provider_config)
+            except Exception as e:
+                logger.error(traceback.format_exc())
+                logger.error(e)
 
         # 设置默认提供商
         selected_provider_id = sp.get(
@@ -404,10 +407,12 @@ class ProviderManager:
 
             self.inst_map[provider_config["id"]] = inst
         except Exception as e:
-            logger.error(traceback.format_exc())
             logger.error(
                 f"实例化 {provider_config['type']}({provider_config['id']}) 提供商适配器失败：{e}"
             )
+            raise Exception(
+                f"实例化 {provider_config['type']}({provider_config['id']}) 提供商适配器失败：{e}"
+            )
 
     async def reload(self, provider_config: dict):
         await self.terminate_provider(provider_config["id"])

astrbot/core/provider/provider.py

@@ -1,4 +1,5 @@
 import abc
+import asyncio
 from typing import List
 from typing import AsyncGenerator
 from astrbot.core.agent.tool import ToolSet
@@ -203,6 +204,72 @@ class EmbeddingProvider(AbstractProvider):
         """获取向量的维度"""
         ...
 
+    async def get_embeddings_batch(
+        self,
+        texts: list[str],
+        batch_size: int = 16,
+        tasks_limit: int = 3,
+        max_retries: int = 3,
+        progress_callback=None,
+    ) -> list[list[float]]:
+        """批量获取文本的向量，分批处理以节省内存
+
+        Args:
+            texts: 文本列表
+            batch_size: 每批处理的文本数量
+            tasks_limit: 并发任务数量限制
+            max_retries: 失败时的最大重试次数
+            progress_callback: 进度回调函数，接收参数 (current, total)
+
+        Returns:
+            向量列表
+        """
+        semaphore = asyncio.Semaphore(tasks_limit)
+        all_embeddings: list[list[float]] = []
+        failed_batches: list[tuple[int, list[str]]] = []
+        completed_count = 0
+        total_count = len(texts)
+
+        async def process_batch(batch_idx: int, batch_texts: list[str]):
+            nonlocal completed_count
+            async with semaphore:
+                for attempt in range(max_retries):
+                    try:
+                        batch_embeddings = await self.get_embeddings(batch_texts)
+                        all_embeddings.extend(batch_embeddings)
+                        completed_count += len(batch_texts)
+                        if progress_callback:
+                            await progress_callback(completed_count, total_count)
+                        return
+                    except Exception as e:
+                        if attempt == max_retries - 1:
+                            # 最后一次重试失败，记录失败的批次
+                            failed_batches.append((batch_idx, batch_texts))
+                            raise Exception(
+                                f"批次 {batch_idx} 处理失败，已重试 {max_retries} 次: {str(e)}"
+                            )
+                        # 等待一段时间后重试，使用指数退避
+                        await asyncio.sleep(2**attempt)
+
+        tasks = []
+        for i in range(0, len(texts), batch_size):
+            batch_texts = texts[i : i + batch_size]
+            batch_idx = i // batch_size
+            tasks.append(process_batch(batch_idx, batch_texts))
+
+        # 收集所有任务的结果，包括失败的任务
+        results = await asyncio.gather(*tasks, return_exceptions=True)
+
+        # 检查是否有失败的任务
+        errors = [r for r in results if isinstance(r, Exception)]
+        if errors:
+            error_msg = (
+                f"有 {len(errors)} 个批次处理失败: {'; '.join(str(e) for e in errors)}"
+            )
+            raise Exception(error_msg)
+
+        return all_embeddings
+
 
 class RerankProvider(AbstractProvider):
     def __init__(self, provider_config: dict, provider_settings: dict) -> None:

astrbot/core/provider/sources/anthropic_source.py

@@ -10,7 +10,7 @@ from anthropic.types import Message
 from astrbot.core.utils.io import download_image_by_url
 from astrbot.api.provider import Provider
 from astrbot import logger
-from astrbot.core.provider.func_tool_manager import FuncCall
+from astrbot.core.provider.func_tool_manager import ToolSet
 from ..register import register_provider_adapter
 from astrbot.core.provider.entities import LLMResponse
 from typing import AsyncGenerator
@@ -104,7 +104,7 @@ class ProviderAnthropic(Provider):
 
         return system_prompt, new_messages
 
-    async def _query(self, payloads: dict, tools: FuncCall) -> LLMResponse:
+    async def _query(self, payloads: dict, tools: ToolSet | None) -> LLMResponse:
         if tools:
             if tool_list := tools.get_func_desc_anthropic_style():
                 payloads["tools"] = tool_list
@@ -135,7 +135,7 @@ class ProviderAnthropic(Provider):
         return llm_response
 
     async def _query_stream(
-        self, payloads: dict, tools: FuncCall
+        self, payloads: dict, tools: ToolSet | None
     ) -> AsyncGenerator[LLMResponse, None]:
         if tools:
             if tool_list := tools.get_func_desc_anthropic_style():
@@ -326,7 +326,7 @@ class ProviderAnthropic(Provider):
         async for llm_response in self._query_stream(payloads, func_tool):
             yield llm_response
 
-    async def assemble_context(self, text: str, image_urls: List[str] = None):
+    async def assemble_context(self, text: str, image_urls: List[str] | None = None):
         """组装上下文，支持文本和图片"""
         if not image_urls:
             return {"role": "user", "content": text}

astrbot/core/provider/sources/dashscope_source.py

@@ -1,15 +1,14 @@
 import re
 import asyncio
 import functools
-from typing import List
 from .. import Provider, Personality
 from ..entities import LLMResponse
-from ..func_tool_manager import FuncCall
 from ..register import register_provider_adapter
 from astrbot.core.message.message_event_result import MessageChain
 from .openai_source import ProviderOpenAIOfficial
 from astrbot.core import logger, sp
 from dashscope import Application
+from dashscope.app.application_response import ApplicationResponse
 
 
 @register_provider_adapter("dashscope", "Dashscope APP 适配器。")
@@ -62,11 +61,11 @@ class ProviderDashscope(ProviderOpenAIOfficial):
     async def text_chat(
         self,
         prompt: str,
-        session_id: str = None,
-        image_urls: List[str] = [],
-        func_tool: FuncCall = None,
-        contexts: List = None,
-        system_prompt: str = None,
+        session_id=None,
+        image_urls=[],
+        func_tool=None,
+        contexts=None,
+        system_prompt=None,
         model=None,
         **kwargs,
     ) -> LLMResponse:
@@ -122,6 +121,8 @@ class ProviderDashscope(ProviderOpenAIOfficial):
             )
             response = await asyncio.get_event_loop().run_in_executor(None, partial)
 
+        assert isinstance(response, ApplicationResponse)
+
         logger.debug(f"dashscope resp: {response}")
 
         if response.status_code != 200:
@@ -135,12 +136,12 @@ class ProviderDashscope(ProviderOpenAIOfficial):
                 ),
             )
 
-        output_text = response.output.get("text", "")
+        output_text = response.output.get("text", "") or ""
         # RAG 引用脚标格式化
         output_text = re.sub(r"<ref>\[(\d+)\]</ref>", r"[\1]", output_text)
         if self.output_reference and response.output.get("doc_references", None):
             ref_str = ""
-            for ref in response.output.get("doc_references", []):
+            for ref in response.output.get("doc_references", []) or []:
                 ref_title = (
                     ref.get("title", "")
                     if ref.get("title")

astrbot/core/provider/sources/dify_source.py

@@ -1,9 +1,7 @@
 import astrbot.core.message.components as Comp
 import os
-from typing import List
 from .. import Provider
 from ..entities import LLMResponse
-from ..func_tool_manager import FuncCall
 from ..register import register_provider_adapter
 from astrbot.core.utils.dify_api_client import DifyAPIClient
 from astrbot.core.utils.io import download_image_by_url, download_file
@@ -55,11 +53,11 @@ class ProviderDify(Provider):
     async def text_chat(
         self,
         prompt: str,
-        session_id: str = None,
-        image_urls: List[str] = None,
-        func_tool: FuncCall = None,
-        contexts: List = None,
-        system_prompt: str = None,
+        session_id=None,
+        image_urls=None,
+        func_tool=None,
+        contexts=None,
+        system_prompt=None,
         tool_calls_result=None,
         model=None,
         **kwargs,
@@ -223,7 +221,7 @@ class ProviderDify(Provider):
             # Chat
             return MessageChain(chain=[Comp.Plain(chunk)])
 
-        async def parse_file(item: dict) -> Comp:
+        async def parse_file(item: dict):
             match item["type"]:
                 case "image":
                     return Comp.Image(file=item["url"], url=item["url"])

astrbot/core/provider/sources/gemini_embedding_source.py

@@ -32,7 +32,6 @@ class GeminiEmbeddingProvider(EmbeddingProvider):
         self.model = provider_config.get(
             "embedding_model", "gemini-embedding-exp-03-07"
         )
-        self.dimension = provider_config.get("embedding_dimensions", 768)
 
     async def get_embedding(self, text: str) -> list[float]:
         """
@@ -60,4 +59,4 @@ class GeminiEmbeddingProvider(EmbeddingProvider):
 
     def get_dim(self) -> int:
         """获取向量的维度"""
-        return self.dimension
+        return self.provider_config.get("embedding_dimensions", 768)