jarvis-ai-assistant 0.1.138__py3-none-any.whl → 0.1.141__py3-none-any.whl

jarvis/jarvis_platform/oyi.py

@@ -1,307 +0,0 @@
-import mimetypes
-import os
-from typing import Dict, List, Tuple
-from jarvis.jarvis_platform.base import BasePlatform
-import requests
-import json
-
-from jarvis.jarvis_utils.output import OutputType, PrettyOutput
-
-class OyiModel(BasePlatform):
-    """Oyi model implementation"""
-
-    platform_name = "oyi"
-    BASE_URL = "https://api-10086.rcouyi.com"
-
-    def get_model_list(self) -> List[Tuple[str, str]]:
-        """Get model list"""
-        self.get_available_models()
-        return [(name,info['desc']) for name,info in self.models.items()]
-
-    def __init__(self):
-        """Initialize model"""
-        super().__init__()
-        self.models = {}
-        self.messages = []
-        self.system_message = ""
-        self.conversation = None
-        self.first_chat = True
-
-        self.token = os.getenv("OYI_API_KEY")
-        if not self.token:
-            PrettyOutput.print("OYI_API_KEY 未设置", OutputType.WARNING)
-
-        self.model_name = os.getenv("JARVIS_MODEL") or "deepseek-chat"
-        if self.model_name not in [m.split()[0] for m in self.get_available_models()]:
-            PrettyOutput.print(f"警告: 选择的模型 {self.model_name} 不在可用列表中", OutputType.WARNING)
-
-
-    def set_model_name(self, model_name: str):
-        """Set model name"""
-
-        self.model_name = model_name
-
-
-    def create_conversation(self) -> bool:
-        """Create a new conversation"""
-        try:
-            headers = {
-                'Authorization': f'Bearer {self.token}',
-                'Content-Type': 'application/json',
-                'Accept': 'application/json',
-                'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36'
-            }
-
-            payload = {
-                "id": 0,
-                "roleId": 0,
-                "title": "New conversation",
-                "isLock": False,
-                "systemMessage": "",
-                "params": json.dumps({
-                    "model": self.model_name,
-                    "is_webSearch": True,
-                    "message": [],
-                    "systemMessage": None,
-                    "requestMsgCount": 65536,
-                    "temperature": 0.8,
-                    "speechVoice": "Alloy",
-                    "max_tokens": 8192,
-                    "chatPluginIds": []
-                })
-            }
-
-            response = requests.post(
-                f"{self.BASE_URL}/chatapi/chat/save",
-                headers=headers,
-                json=payload
-            )
-
-            if response.status_code == 200:
-                data = response.json()
-                if data['code'] == 200 and data['type'] == 'success':
-                    self.conversation = data
-                    return True
-                else:
-                    PrettyOutput.print(f"创建会话失败: {data['message']}", OutputType.WARNING)
-                    return False
-            else:
-                PrettyOutput.print(f"创建会话失败: {response.status_code}", OutputType.WARNING)
-                return False
-
-        except Exception as e:
-            PrettyOutput.print(f"创建会话失败: {str(e)}", OutputType.ERROR)
-            return False
-
-    def set_system_message(self, message: str):
-        """Set system message"""
-        self.system_message = message
-
-    def chat(self, message: str) -> str:
-        """Execute chat with the model
-
-        Args:
-            message: User input message
-
-        Returns:
-            str: Model response
-        """
-        try:
-            # 确保有会话ID
-            if not self.conversation:
-                if not self.create_conversation():
-                    raise Exception("Failed to create conversation")
-
-            # 1. 发送消息
-            headers = {
-                'Authorization': f'Bearer {self.token}',
-                'Content-Type': 'application/json',
-                'Accept': 'application/json, text/plain, */*',
-                'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36',
-                'Origin': 'https://ai.rcouyi.com',
-                'Referer': 'https://ai.rcouyi.com/'
-            }
-
-            payload = {
-                "topicId": self.conversation['result']['id'] if self.conversation else None,
-                "messages": self.messages,
-                "content": message,
-                "contentFiles": []
-            }
-
-            # 如果有上传的文件，添加到请求中
-            if self.first_chat:
-                message = self.system_message + "\n" + message
-                payload["content"] = message
-                self.first_chat = False
-
-            self.messages.append({"role": "user", "content": message})
-
-            # 发送消息
-            response = requests.post(
-                f"{self.BASE_URL}/chatapi/chat/message",
-                headers=headers,
-                json=payload
-            )
-
-            if response.status_code != 200:
-                error_msg = f"聊天请求失败: {response.status_code}"
-                PrettyOutput.print(error_msg, OutputType.WARNING)
-                raise Exception(error_msg)
-
-            data = response.json()
-            if data['code'] != 200 or data['type'] != 'success':
-                error_msg = f"聊天失败: {data.get('message', '未知错误')}"
-                PrettyOutput.print(error_msg, OutputType.WARNING)
-                raise Exception(error_msg)
-
-            message_id = data['result'][-1]
-
-            # 获取响应内容
-            response = requests.post(
-                f"{self.BASE_URL}/chatapi/chat/message/{message_id}",
-                headers=headers,
-                stream=True
-            )
-
-            if response.status_code == 200:
-                full_response = ""
-                bin = b""
-                for chunk in response.iter_content(decode_unicode=True):
-                    if chunk:
-                        bin += chunk
-                        try:
-                            text = bin.decode('utf-8')
-                        except UnicodeDecodeError:
-                            continue
-                        if not self.suppress_output:
-                            PrettyOutput.print_stream(text)
-                        full_response += text
-                        bin = b""
-
-                PrettyOutput.print_stream_end()
-
-                self.messages.append({"role": "assistant", "content": full_response})
-                return full_response
-            else:
-                error_msg = f"获取响应失败: {response.status_code}"
-                PrettyOutput.print(error_msg, OutputType.WARNING)
-                raise Exception(error_msg)
-        except Exception as e:
-            PrettyOutput.print(f"聊天失败: {str(e)}", OutputType.ERROR)
-            raise e
-
-    def name(self) -> str:
-        """Return model name"""
-        return self.model_name
-
-
-    def delete_chat(self) -> bool:
-        """Delete current chat session"""
-        try:
-            if not self.conversation:
-                return True
-
-            headers = {
-                'Authorization': f'Bearer {self.token}',
-                'Content-Type': 'application/json',
-                'Accept': 'application/json, text/plain, */*',
-                'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36',
-                'Origin': 'https://ai.rcouyi.com',
-                'Referer': 'https://ai.rcouyi.com/'
-            }
-
-            response = requests.post(
-                f"{self.BASE_URL}/chatapi/chat/{self.conversation['result']['id']}",
-                headers=headers,
-                json={}
-            )
-
-            if response.status_code == 200:
-                data = response.json()
-                if data['code'] == 200 and data['type'] == 'success':
-                    self.messages = []
-                    self.conversation = None
-                    self.first_chat = True
-                    return True
-                else:
-                    error_msg = f"删除会话失败: {data.get('message', '未知错误')}"
-                    PrettyOutput.print(error_msg, OutputType.WARNING)
-                    return False
-            else:
-                error_msg = f"删除会话请求失败: {response.status_code}"
-                PrettyOutput.print(error_msg, OutputType.WARNING)
-                return False
-
-
-        except Exception as e:
-            PrettyOutput.print(f"删除会话失败: {str(e)}", OutputType.ERROR)
-            return False
-
-    def get_available_models(self) -> List[str]:
-        """Get available model list
-
-        Returns:
-            List[str]: Available model name list
-        """
-        try:
-            if self.models:
-                return list(self.models.keys())
-
-            headers = {
-                'Content-Type': 'application/json',
-                'Accept': 'application/json, text/plain, */*',
-                'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36',
-                'Origin': 'https://ai.rcouyi.com',
-                'Referer': 'https://ai.rcouyi.com/'
-            }
-
-            response = requests.get(
-                "https://ai.rcouyi.com/config/system.json",
-                headers=headers
-            )
-
-            if response.status_code != 200:
-                PrettyOutput.print(f"获取模型列表失败: {response.status_code}", OutputType.WARNING)
-                return []
-
-            data = response.json()
-
-            # 保存模型信息
-            self.models = {
-                model['value']: model
-                for model in data.get('model', [])
-                if model.get('enable', False)  # 只保存启用的模型
-            }
-
-            # 格式化显示
-            models = []
-            for model in self.models.values():
-                # 基本信息
-                model_name = model['value']
-                model_str = model['label']
-
-                # 添加后缀标签
-                suffix = model.get('suffix', [])
-                if suffix:
-                    # 处理新格式的suffix (字典列表)
-                    if suffix and isinstance(suffix[0], dict):
-                        suffix_str = ', '.join(s.get('tag', '') for s in suffix)
-                    # 处理旧格式的suffix (字符串列表)
-                    else:
-                        suffix_str = ', '.join(str(s) for s in suffix)
-                    model_str += f" ({suffix_str})"
-
-                # 添加描述或提示
-                info = model.get('tooltip') or model.get('description', '')
-                if info:
-                    model_str += f" - {info}"
-
-                model['desc'] = model_str
-                models.append(model_name)
-
-            return sorted(models)
-
-        except Exception as e:
-            PrettyOutput.print(f"获取模型列表失败: {str(e)}", OutputType.WARNING)
-            return []

jarvis/jarvis_rag/file_processors.py

@@ -1,138 +0,0 @@
-from pathlib import Path
-import fitz  # PyMuPDF for PDF files
-from docx import Document as DocxDocument  # python-docx for DOCX files
-from pptx import Presentation
-import pandas as pd
-import unicodedata
-
-class FileProcessor:
-    """Base class for file processor"""
-    @staticmethod
-    def can_handle(file_path: str) -> bool:
-        """Determine if the file can be processed"""
-        raise NotImplementedError
-
-    @staticmethod
-    def extract_text(file_path: str) -> str:
-        """Extract file text content"""
-        raise NotImplementedError
-
-class TextFileProcessor(FileProcessor):
-    """Text file processor"""
-    ENCODINGS = ['utf-8', 'gbk', 'gb2312', 'latin1']
-    SAMPLE_SIZE = 8192  # Read the first 8KB to detect encoding
-
-    @staticmethod
-    def can_handle(file_path: str) -> bool:
-        """Determine if the file is a text file by trying to decode it"""
-        try:
-            # Read the first part of the file to detect encoding
-            with open(file_path, 'rb') as f:
-                sample = f.read(TextFileProcessor.SAMPLE_SIZE)
-
-            # Check if it contains null bytes (usually represents a binary file)
-            if b'\x00' in sample:
-                return False
-
-            # Check if it contains too many non-printable characters (usually represents a binary file)
-            non_printable = sum(1 for byte in sample if byte < 32 and byte not in (9, 10, 13))  # tab, newline, carriage return
-            if non_printable / len(sample) > 0.3:  # If non-printable characters exceed 30%, it is considered a binary file
-                return False
-
-            # Try to decode with different encodings
-            for encoding in TextFileProcessor.ENCODINGS:
-                try:
-                    sample.decode(encoding)
-                    return True
-                except UnicodeDecodeError:
-                    continue
-
-            return False
-
-        except Exception:
-            return False
-
-    @staticmethod
-    def extract_text(file_path: str) -> str:
-        """Extract text content, using the detected correct encoding"""
-        detected_encoding = None
-        try:
-            # First try to detect encoding
-            with open(file_path, 'rb') as f:
-                raw_data = f.read()
-
-            # Try different encodings
-            for encoding in TextFileProcessor.ENCODINGS:
-                try:
-                    raw_data.decode(encoding)
-                    detected_encoding = encoding
-                    break
-                except UnicodeDecodeError:
-                    continue
-
-            if not detected_encoding:
-                raise UnicodeDecodeError(f"Failed to decode file with supported encodings: {file_path}") # type: ignore
-
-            # Use the detected encoding to read the file
-            with open(file_path, 'r', encoding=detected_encoding, errors='ignore') as f:
-                content = f.read()
-
-            # Normalize Unicode characters
-            content = unicodedata.normalize('NFKC', content)
-
-            return content
-
-        except Exception as e:
-            raise Exception(f"Failed to read file: {str(e)}")
-
-class PDFProcessor(FileProcessor):
-    """PDF file processor"""
-    @staticmethod
-    def can_handle(file_path: str) -> bool:
-        return Path(file_path).suffix.lower() == '.pdf'
-
-    @staticmethod
-    def extract_text(file_path: str) -> str:
-        text_parts = []
-        with fitz.open(file_path) as doc: # type: ignore
-            for page in doc:
-                text_parts.append(page.get_text()) # type: ignore
-        return "\n".join(text_parts)
-
-class DocxProcessor(FileProcessor):
-    """DOCX file processor"""
-    @staticmethod
-    def can_handle(file_path: str) -> bool:
-        return Path(file_path).suffix.lower() == '.docx'
-
-    @staticmethod
-    def extract_text(file_path: str) -> str:
-        doc = DocxDocument(file_path)
-        return "\n".join([paragraph.text for paragraph in doc.paragraphs])
-
-class PPTProcessor(FileProcessor):
-    """PPT file processor"""
-    @staticmethod
-    def can_handle(file_path: str) -> bool:
-        return Path(file_path).suffix.lower() in ['.ppt', '.pptx']
-
-    @staticmethod
-    def extract_text(file_path: str) -> str:
-        prs = Presentation(file_path)
-        text = []
-        for slide in prs.slides:
-            for shape in slide.shapes:
-                if hasattr(shape, "text"):
-                    text.append(shape.text)  # type: ignore
-        return "\n".join(text)
-
-class ExcelProcessor(FileProcessor):
-    """Excel file processor"""
-    @staticmethod
-    def can_handle(file_path: str) -> bool:
-        return Path(file_path).suffix.lower() in ['.xls', '.xlsx']
-
-    @staticmethod
-    def extract_text(file_path: str) -> str:
-        df = pd.read_excel(file_path)
-        return df.to_string()