snow-ai 0.4.15 → 0.4.17

package/dist/api/gemini.js

@@ -1,445 +0,0 @@
-import { getOpenAiConfig, getCustomSystemPrompt, getCustomHeaders, } from '../utils/apiConfig.js';
-import { getSystemPrompt } from './systemPrompt.js';
-import { withRetryGenerator, parseJsonWithFix } from '../utils/retryUtils.js';
-import { addProxyToFetchOptions } from '../utils/proxyUtils.js';
-import { saveUsageToFile } from '../utils/usageLogger.js';
-let geminiConfig = null;
-function getGeminiConfig() {
-    if (!geminiConfig) {
-        const config = getOpenAiConfig();
-        if (!config.apiKey) {
-            throw new Error('Gemini API configuration is incomplete. Please configure API key first.');
-        }
-        const customHeaders = getCustomHeaders();
-        geminiConfig = {
-            apiKey: config.apiKey,
-            baseUrl: config.baseUrl && config.baseUrl !== 'https://api.openai.com/v1'
-                ? config.baseUrl
-                : 'https://generativelanguage.googleapis.com/v1beta',
-            customHeaders,
-            geminiThinking: config.geminiThinking,
-        };
-    }
-    return geminiConfig;
-}
-export function resetGeminiClient() {
-    geminiConfig = null;
-}
-/**
- * Convert OpenAI-style tools to Gemini function declarations
- */
-function convertToolsToGemini(tools) {
-    if (!tools || tools.length === 0) {
-        return undefined;
-    }
-    const functionDeclarations = tools
-        .filter(tool => tool.type === 'function' && 'function' in tool)
-        .map(tool => {
-        if (tool.type === 'function' && 'function' in tool) {
-            // Convert OpenAI parameters schema to Gemini format
-            const params = tool.function.parameters;
-            return {
-                name: tool.function.name,
-                description: tool.function.description || '',
-                parametersJsonSchema: {
-                    type: 'object',
-                    properties: params.properties || {},
-                    required: params.required || [],
-                },
-            };
-        }
-        throw new Error('Invalid tool format');
-    });
-    return [{ functionDeclarations }];
-}
-/**
- * Convert our ChatMessage format to Gemini's format
- * @param messages - The messages to convert
- * @param includeBuiltinSystemPrompt - Whether to include builtin system prompt (default true)
- */
-function convertToGeminiMessages(messages, includeBuiltinSystemPrompt = true) {
-    const customSystemPrompt = getCustomSystemPrompt();
-    let systemInstruction;
-    const contents = [];
-    for (let i = 0; i < messages.length; i++) {
-        const msg = messages[i];
-        if (!msg)
-            continue;
-        // Extract system message as systemInstruction
-        if (msg.role === 'system') {
-            systemInstruction = msg.content;
-            continue;
-        }
-        // Handle tool results
-        if (msg.role === 'tool') {
-            // Find the corresponding function call to get the function name
-            // Look backwards in contents to find the matching tool call
-            let functionName = 'unknown_function';
-            for (let j = contents.length - 1; j >= 0; j--) {
-                const contentMsg = contents[j];
-                if (contentMsg.role === 'model' && contentMsg.parts) {
-                    for (const part of contentMsg.parts) {
-                        if (part.functionCall) {
-                            functionName = part.functionCall.name;
-                            break;
-                        }
-                    }
-                    if (functionName !== 'unknown_function')
-                        break;
-                }
-            }
-            // Tool response must be a valid object for Gemini API
-            // If content is a JSON string, parse it; otherwise wrap it in an object
-            let responseData;
-            const imageParts = [];
-            // Handle images from tool result
-            if (msg.images && msg.images.length > 0) {
-                for (const image of msg.images) {
-                    imageParts.push({
-                        inlineData: {
-                            mimeType: image.mimeType,
-                            data: image.data,
-                        },
-                    });
-                }
-            }
-            if (!msg.content) {
-                responseData = {};
-            }
-            else {
-                let contentToParse = msg.content;
-                // Sometimes the content is double-encoded as JSON
-                // First, try to parse it once
-                const firstParseResult = parseJsonWithFix(contentToParse, {
-                    toolName: 'Gemini tool response (first parse)',
-                    logWarning: false,
-                    logError: false,
-                });
-                if (firstParseResult.success &&
-                    typeof firstParseResult.data === 'string') {
-                    // If it's a string, it might be double-encoded, try parsing again
-                    contentToParse = firstParseResult.data;
-                }
-                // Now parse or wrap the final content
-                const finalParseResult = parseJsonWithFix(contentToParse, {
-                    toolName: 'Gemini tool response (final parse)',
-                    logWarning: false,
-                    logError: false,
-                });
-                if (finalParseResult.success) {
-                    const parsed = finalParseResult.data;
-                    // If parsed result is an object (not array, not null), use it directly
-                    if (typeof parsed === 'object' &&
-                        parsed !== null &&
-                        !Array.isArray(parsed)) {
-                        responseData = parsed;
-                    }
-                    else {
-                        // If it's a primitive, array, or null, wrap it
-                        responseData = { content: parsed };
-                    }
-                }
-                else {
-                    // Not valid JSON, wrap the raw string
-                    responseData = { content: contentToParse };
-                }
-            }
-            // Build parts array with functionResponse and optional images
-            const parts = [
-                {
-                    functionResponse: {
-                        name: functionName,
-                        response: responseData,
-                    },
-                },
-            ];
-            // Add images as inline data parts
-            if (imageParts.length > 0) {
-                parts.push(...imageParts);
-            }
-            contents.push({
-                role: 'user',
-                parts,
-            });
-            continue;
-        }
-        // Handle tool calls in assistant messages
-        if (msg.role === 'assistant' &&
-            msg.tool_calls &&
-            msg.tool_calls.length > 0) {
-            const parts = [];
-            // Add text content if exists
-            if (msg.content) {
-                parts.push({ text: msg.content });
-            }
-            // Add function calls
-            for (const toolCall of msg.tool_calls) {
-                const argsParseResult = parseJsonWithFix(toolCall.function.arguments, {
-                    toolName: `Gemini function call: ${toolCall.function.name}`,
-                    fallbackValue: {},
-                    logWarning: true,
-                    logError: true,
-                });
-                parts.push({
-                    functionCall: {
-                        name: toolCall.function.name,
-                        args: argsParseResult.data,
-                    },
-                });
-            }
-            contents.push({
-                role: 'model',
-                parts,
-            });
-            continue;
-        }
-        // Build message parts
-        const parts = [];
-        // Add text content
-        if (msg.content) {
-            parts.push({ text: msg.content });
-        }
-        // Add images for user messages
-        if (msg.role === 'user' && msg.images && msg.images.length > 0) {
-            for (const image of msg.images) {
-                const base64Match = image.data.match(/^data:([^;]+);base64,(.+)$/);
-                if (base64Match) {
-                    parts.push({
-                        inlineData: {
-                            mimeType: base64Match[1] || image.mimeType,
-                            data: base64Match[2] || '',
-                        },
-                    });
-                }
-            }
-        }
-        // Add to contents
-        const role = msg.role === 'assistant' ? 'model' : 'user';
-        contents.push({ role, parts });
-    }
-    // Handle system instruction
-    // 如果配置了自定义系统提示词（最高优先级，始终添加）
-    if (customSystemPrompt) {
-        systemInstruction = customSystemPrompt;
-        if (includeBuiltinSystemPrompt) {
-            // Prepend default system prompt as first user message
-            contents.unshift({
-                role: 'user',
-                parts: [{ text: getSystemPrompt() }],
-            });
-        }
-    }
-    else if (!systemInstruction && includeBuiltinSystemPrompt) {
-        // 没有自定义系统提示词，但需要添加默认系统提示词
-        systemInstruction = getSystemPrompt();
-    }
-    return { systemInstruction, contents };
-}
-/**
- * Create streaming chat completion using Gemini API
- */
-export async function* createStreamingGeminiCompletion(options, abortSignal, onRetry) {
-    const config = getGeminiConfig();
-    // 使用重试包装生成器
-    yield* withRetryGenerator(async function* () {
-        const { systemInstruction, contents } = convertToGeminiMessages(options.messages, options.includeBuiltinSystemPrompt !== false);
-        // Build request payload
-        const requestBody = {
-            contents,
-            systemInstruction: systemInstruction
-                ? { parts: [{ text: systemInstruction }] }
-                : undefined,
-        };
-        // Add thinking configuration if enabled
-        // Only include generationConfig when thinking is enabled
-        if (config.geminiThinking?.enabled) {
-            requestBody.generationConfig = {
-                thinkingConfig: {
-                    thinkingBudget: config.geminiThinking.budget,
-                },
-            };
-        }
-        // Add tools if provided
-        const geminiTools = convertToolsToGemini(options.tools);
-        if (geminiTools) {
-            requestBody.tools = geminiTools;
-        }
-        // Extract model name from options.model (e.g., "gemini-pro" or "models/gemini-pro")
-        const modelName = options.model.startsWith('models/')
-            ? options.model
-            : `models/${options.model}`;
-        const url = `${config.baseUrl}/${modelName}:streamGenerateContent?key=${config.apiKey}&alt=sse`;
-        const fetchOptions = addProxyToFetchOptions(url, {
-            method: 'POST',
-            headers: {
-                'Content-Type': 'application/json',
-                Authorization: `Bearer ${config.apiKey}`,
-                'x-snow': 'true',
-                ...config.customHeaders,
-            },
-            body: JSON.stringify(requestBody),
-            signal: abortSignal,
-        });
-        const response = await fetch(url, fetchOptions);
-        if (!response.ok) {
-            const errorText = await response.text();
-            throw new Error(`Gemini API error: ${response.status} ${response.statusText} - ${errorText}`);
-        }
-        if (!response.body) {
-            throw new Error('No response body from Gemini API');
-        }
-        let contentBuffer = '';
-        let thinkingTextBuffer = ''; // Accumulate thinking text content
-        let toolCallsBuffer = [];
-        let hasToolCalls = false;
-        let toolCallIndex = 0;
-        let totalTokens = { prompt: 0, completion: 0, total: 0 };
-        // Parse SSE stream
-        const reader = response.body.getReader();
-        const decoder = new TextDecoder();
-        let buffer = '';
-        try {
-            while (true) {
-                const { done, value } = await reader.read();
-                if (done) {
-                    // ✅ 关键修复：检查buffer是否有残留数据
-                    if (buffer.trim()) {
-                        // 连接异常中断，抛出明确错误
-                        throw new Error(`Stream terminated unexpectedly with incomplete data: ${buffer.substring(0, 100)}...`);
-                    }
-                    break; // 正常结束
-                }
-                if (abortSignal?.aborted) {
-                    return;
-                }
-                buffer += decoder.decode(value, { stream: true });
-                const lines = buffer.split('\n');
-                buffer = lines.pop() || '';
-                for (const line of lines) {
-                    const trimmed = line.trim();
-                    if (!trimmed || trimmed.startsWith(':'))
-                        continue;
-                    if (trimmed === 'data: [DONE]' || trimmed === 'data:[DONE]') {
-                        break;
-                    }
-                    // Handle both "event: " and "event:" formats
-                    if (trimmed.startsWith('event:')) {
-                        // Event type, will be followed by data
-                        continue;
-                    }
-                    // Handle both "data: " and "data:" formats
-                    if (trimmed.startsWith('data:')) {
-                        const data = trimmed.startsWith('data: ')
-                            ? trimmed.slice(6)
-                            : trimmed.slice(5);
-                        const parseResult = parseJsonWithFix(data, {
-                            toolName: 'Gemini SSE stream',
-                            logWarning: false,
-                            logError: true,
-                        });
-                        if (parseResult.success) {
-                            const chunk = parseResult.data;
-                            // Process candidates
-                            if (chunk.candidates && chunk.candidates.length > 0) {
-                                const candidate = chunk.candidates[0];
-                                if (candidate.content && candidate.content.parts) {
-                                    for (const part of candidate.content.parts) {
-                                        // Process thought content (Gemini thinking)
-                                        // When part.thought === true, the text field contains thinking content
-                                        if (part.thought === true && part.text) {
-                                            thinkingTextBuffer += part.text;
-                                            yield {
-                                                type: 'reasoning_delta',
-                                                delta: part.text,
-                                            };
-                                        }
-                                        // Process regular text content (when thought is not true)
-                                        else if (part.text) {
-                                            contentBuffer += part.text;
-                                            yield {
-                                                type: 'content',
-                                                content: part.text,
-                                            };
-                                        }
-                                        // Process function calls
-                                        if (part.functionCall) {
-                                            hasToolCalls = true;
-                                            const fc = part.functionCall;
-                                            const toolCall = {
-                                                id: `call_${toolCallIndex++}`,
-                                                type: 'function',
-                                                function: {
-                                                    name: fc.name,
-                                                    arguments: JSON.stringify(fc.args || {}),
-                                                },
-                                            };
-                                            toolCallsBuffer.push(toolCall);
-                                            // Yield delta for token counting
-                                            const deltaText = fc.name + JSON.stringify(fc.args || {});
-                                            yield {
-                                                type: 'tool_call_delta',
-                                                delta: deltaText,
-                                            };
-                                        }
-                                    }
-                                }
-                            }
-                            // Track usage info
-                            if (chunk.usageMetadata) {
-                                totalTokens = {
-                                    prompt: chunk.usageMetadata.promptTokenCount || 0,
-                                    completion: chunk.usageMetadata.candidatesTokenCount || 0,
-                                    total: chunk.usageMetadata.totalTokenCount || 0,
-                                };
-                            }
-                        }
-                    }
-                }
-            }
-        }
-        catch (error) {
-            const { logger } = await import('../utils/logger.js');
-            logger.error('Gemini SSE stream parsing error:', {
-                error: error instanceof Error ? error.message : 'Unknown error',
-                remainingBuffer: buffer.substring(0, 200),
-            });
-            throw error;
-        }
-        // Yield tool calls if any
-        if (hasToolCalls && toolCallsBuffer.length > 0) {
-            yield {
-                type: 'tool_calls',
-                tool_calls: toolCallsBuffer,
-            };
-        }
-        // Yield usage info
-        if (totalTokens.total > 0) {
-            const usageData = {
-                prompt_tokens: totalTokens.prompt,
-                completion_tokens: totalTokens.completion,
-                total_tokens: totalTokens.total,
-            };
-            // Save usage to file system at API layer
-            saveUsageToFile(options.model, usageData);
-            yield {
-                type: 'usage',
-                usage: usageData,
-            };
-        }
-        // Return complete thinking block if thinking content exists
-        const thinkingBlock = thinkingTextBuffer
-            ? {
-                type: 'thinking',
-                thinking: thinkingTextBuffer,
-            }
-            : undefined;
-        // Signal completion
-        yield {
-            type: 'done',
-            thinking: thinkingBlock,
-        };
-    }, {
-        abortSignal,
-        onRetry,
-    });
-}

package/dist/api/models.d.ts

@@ -1,15 +0,0 @@
-export interface Model {
-    id: string;
-    object: string;
-    created: number;
-    owned_by: string;
-}
-export interface ModelsResponse {
-    object: string;
-    data: Model[];
-}
-/**
- * Fetch available models based on configured request method
- */
-export declare function fetchAvailableModels(): Promise<Model[]>;
-export declare function filterModels(models: Model[], searchTerm: string): Model[];

package/dist/api/models.js

@@ -1,139 +0,0 @@
-import { getOpenAiConfig, getCustomHeaders } from '../utils/apiConfig.js';
-/**
- * Fetch models from OpenAI-compatible API
- */
-async function fetchOpenAIModels(baseUrl, apiKey, customHeaders) {
-    const url = `${baseUrl}/models`;
-    const headers = {
-        'Content-Type': 'application/json',
-        ...customHeaders,
-    };
-    if (apiKey) {
-        headers['Authorization'] = `Bearer ${apiKey}`;
-    }
-    const response = await fetch(url, {
-        method: 'GET',
-        headers,
-    });
-    if (!response.ok) {
-        throw new Error(`Failed to fetch models: ${response.status} ${response.statusText}`);
-    }
-    const data = await response.json();
-    return data.data || [];
-}
-/**
- * Fetch models from Gemini API
- */
-async function fetchGeminiModels(baseUrl, apiKey) {
-    // Gemini uses API key as query parameter
-    const url = `${baseUrl}/models?key=${apiKey}`;
-    const response = await fetch(url, {
-        method: 'GET',
-        headers: {
-            'Content-Type': 'application/json',
-        },
-    });
-    if (!response.ok) {
-        throw new Error(`Failed to fetch models: ${response.status} ${response.statusText}`);
-    }
-    const data = await response.json();
-    // Convert Gemini format to standard Model format
-    return (data.models || []).map(model => ({
-        id: model.name.replace('models/', ''), // Remove "models/" prefix
-        object: 'model',
-        created: 0,
-        owned_by: 'google',
-    }));
-}
-/**
- * Fetch models from Anthropic API
- * Supports both Anthropic native format and OpenAI-compatible format for backward compatibility
- */
-async function fetchAnthropicModels(baseUrl, apiKey, customHeaders) {
-    const url = `${baseUrl}/models`;
-    const headers = {
-        'Content-Type': 'application/json',
-        ...customHeaders,
-    };
-    if (apiKey) {
-        headers['x-api-key'] = apiKey;
-        headers['Authorization'] = `Bearer ${apiKey}`;
-    }
-    const response = await fetch(url, {
-        method: 'GET',
-        headers,
-    });
-    if (!response.ok) {
-        throw new Error(`Failed to fetch models: ${response.status} ${response.statusText}`);
-    }
-    const data = await response.json();
-    // Try to parse as Anthropic format first
-    if (data.data && Array.isArray(data.data) && data.data.length > 0) {
-        const firstItem = data.data[0];
-        // Check if it's Anthropic format (has created_at field)
-        if ('created_at' in firstItem && typeof firstItem.created_at === 'string') {
-            // Anthropic native format
-            return data.data.map(model => ({
-                id: model.id,
-                object: 'model',
-                created: new Date(model.created_at).getTime() / 1000,
-                owned_by: 'anthropic',
-            }));
-        }
-        // Fallback to OpenAI format (has created field as number)
-        if ('id' in firstItem && 'object' in firstItem) {
-            // OpenAI-compatible format
-            return data.data;
-        }
-    }
-    // If no data array or empty, return empty array
-    return [];
-}
-/**
- * Fetch available models based on configured request method
- */
-export async function fetchAvailableModels() {
-    const config = getOpenAiConfig();
-    if (!config.baseUrl) {
-        throw new Error('Base URL not configured. Please configure API settings first.');
-    }
-    const customHeaders = getCustomHeaders();
-    try {
-        let models;
-        switch (config.requestMethod) {
-            case 'gemini':
-                if (!config.apiKey) {
-                    throw new Error('API key is required for Gemini API');
-                }
-                models = await fetchGeminiModels(config.baseUrl.replace(/\/$/, ''), config.apiKey);
-                break;
-            case 'anthropic':
-                if (!config.apiKey) {
-                    throw new Error('API key is required for Anthropic API');
-                }
-                models = await fetchAnthropicModels(config.baseUrl.replace(/\/$/, ''), config.apiKey, customHeaders);
-                break;
-            case 'chat':
-            case 'responses':
-            default:
-                // OpenAI-compatible API
-                models = await fetchOpenAIModels(config.baseUrl.replace(/\/$/, ''), config.apiKey, customHeaders);
-                break;
-        }
-        // Sort models alphabetically by id for better UX
-        return models.sort((a, b) => a.id.localeCompare(b.id));
-    }
-    catch (error) {
-        if (error instanceof Error) {
-            throw new Error(`Error fetching models: ${error.message}`);
-        }
-        throw new Error('Unknown error occurred while fetching models');
-    }
-}
-export function filterModels(models, searchTerm) {
-    if (!searchTerm.trim()) {
-        return models;
-    }
-    const lowerSearchTerm = searchTerm.toLowerCase();
-    return models.filter(model => model.id.toLowerCase().includes(lowerSearchTerm));
-}

package/dist/api/responses.d.ts

@@ -1,38 +0,0 @@
-import type { ChatMessage, ToolCall, ChatCompletionTool, UsageInfo } from './types.js';
-export interface ResponseOptions {
-    model: string;
-    messages: ChatMessage[];
-    stream?: boolean;
-    temperature?: number;
-    max_tokens?: number;
-    tools?: ChatCompletionTool[];
-    tool_choice?: 'auto' | 'none' | 'required';
-    reasoning?: {
-        summary?: 'auto' | 'none';
-        effort?: 'low' | 'medium' | 'high';
-    } | null;
-    prompt_cache_key?: string;
-    store?: boolean;
-    include?: string[];
-    includeBuiltinSystemPrompt?: boolean;
-}
-export interface ResponseStreamChunk {
-    type: 'content' | 'tool_calls' | 'tool_call_delta' | 'reasoning_delta' | 'reasoning_started' | 'reasoning_data' | 'done' | 'usage';
-    content?: string;
-    tool_calls?: ToolCall[];
-    delta?: string;
-    usage?: UsageInfo;
-    reasoning?: {
-        summary?: Array<{
-            type: 'summary_text';
-            text: string;
-        }>;
-        content?: any;
-        encrypted_content?: string;
-    };
-}
-export declare function resetOpenAIClient(): void;
-/**
- * 使用 Responses API 创建流式响应（带自动工具调用）
- */
-export declare function createStreamingResponse(options: ResponseOptions, abortSignal?: AbortSignal, onRetry?: (error: Error, attempt: number, nextDelay: number) => void): AsyncGenerator<ResponseStreamChunk, void, unknown>;