R524809
263 lines
9.4 KiB
JavaScript
263 lines
9.4 KiB
JavaScript
const fs = require('fs');
|
||
const path = require('path');
|
||
const https = require('https');
|
||
const crypto = require('crypto');
|
||
require('dotenv').config({ path: path.join(__dirname, '../.env') });
|
||
|
||
// 腾讯云混元大模型配置
|
||
// 请在 .env 文件中配置 TENCENT_SECRET_ID 和 TENCENT_SECRET_KEY
|
||
const SECRET_ID = process.env.TENCENT_SECRET_ID;
|
||
const SECRET_KEY = process.env.TENCENT_SECRET_KEY;
|
||
|
||
// API配置
|
||
const HOST = 'hunyuan.tencentcloudapi.com';
|
||
const SERVICE = 'hunyuan';
|
||
const REGION = 'ap-guangzhou';
|
||
const ACTION = 'ChatCompletions';
|
||
const VERSION = '2023-09-01';
|
||
const MODEL = 'hunyuan-lite'; // 使用混元lite模型，可根据需要更换
|
||
|
||
// 签名相关函数
|
||
function sha256(message, secret = '', encoding = 'hex') {
|
||
const hmac = secret ? crypto.createHmac('sha256', secret) : crypto.createHash('sha256');
|
||
return hmac.update(message).digest(encoding);
|
||
}
|
||
|
||
function getDate(timestamp) {
|
||
const date = new Date(timestamp * 1000);
|
||
const year = date.getUTCFullYear();
|
||
const month = ('0' + (date.getUTCMonth() + 1)).slice(-2);
|
||
const day = ('0' + date.getUTCDate()).slice(-2);
|
||
return `${year}-${month}-${day}`;
|
||
}
|
||
|
||
function generateSignature(secretId, secretKey, host, payload, timestamp) {
|
||
const date = getDate(timestamp);
|
||
|
||
// 步骤1：拼接规范请求串
|
||
const httpRequestMethod = 'POST';
|
||
const canonicalUri = '/';
|
||
const canonicalQueryString = '';
|
||
const contentType = 'application/json';
|
||
const canonicalHeaders = `content-type:${contentType}\nhost:${host}\nx-tc-action:${ACTION.toLowerCase()}\n`;
|
||
const signedHeaders = 'content-type;host;x-tc-action';
|
||
const hashedRequestPayload = sha256(payload);
|
||
const canonicalRequest = `${httpRequestMethod}\n${canonicalUri}\n${canonicalQueryString}\n${canonicalHeaders}\n${signedHeaders}\n${hashedRequestPayload}`;
|
||
|
||
// 步骤2：拼接待签名字符串
|
||
const algorithm = 'TC3-HMAC-SHA256';
|
||
const credentialScope = `${date}/${SERVICE}/tc3_request`;
|
||
const hashedCanonicalRequest = sha256(canonicalRequest);
|
||
const stringToSign = `${algorithm}\n${timestamp}\n${credentialScope}\n${hashedCanonicalRequest}`;
|
||
|
||
// 步骤3：计算签名
|
||
const secretDate = sha256(date, 'TC3' + secretKey, 'buffer');
|
||
const secretService = sha256(SERVICE, secretDate, 'buffer');
|
||
const secretSigning = sha256('tc3_request', secretService, 'buffer');
|
||
const signature = sha256(stringToSign, secretSigning);
|
||
|
||
// 步骤4：拼接Authorization
|
||
const authorization = `${algorithm} Credential=${secretId}/${credentialScope}, SignedHeaders=${signedHeaders}, Signature=${signature}`;
|
||
|
||
return authorization;
|
||
}
|
||
|
||
// 调用混元API
|
||
async function callHunyuan(prompt) {
|
||
return new Promise((resolve, reject) => {
|
||
const timestamp = Math.floor(Date.now() / 1000);
|
||
|
||
const payload = JSON.stringify({
|
||
Model: MODEL,
|
||
Messages: [
|
||
{
|
||
Role: 'user',
|
||
Content: prompt,
|
||
},
|
||
],
|
||
Stream: false,
|
||
});
|
||
|
||
const authorization = generateSignature(SECRET_ID, SECRET_KEY, HOST, payload, timestamp);
|
||
|
||
const options = {
|
||
hostname: HOST,
|
||
port: 443,
|
||
path: '/',
|
||
method: 'POST',
|
||
headers: {
|
||
'Content-Type': 'application/json',
|
||
Host: HOST,
|
||
'X-TC-Action': ACTION,
|
||
'X-TC-Version': VERSION,
|
||
'X-TC-Timestamp': timestamp.toString(),
|
||
'X-TC-Region': REGION,
|
||
Authorization: authorization,
|
||
},
|
||
};
|
||
|
||
const req = https.request(options, (res) => {
|
||
let data = '';
|
||
res.on('data', (chunk) => (data += chunk));
|
||
res.on('end', () => {
|
||
try {
|
||
const result = JSON.parse(data);
|
||
if (result.Response && result.Response.Choices) {
|
||
resolve(result.Response.Choices[0].Message.Content.trim());
|
||
} else if (result.Response && result.Response.Error) {
|
||
reject(new Error(result.Response.Error.Message));
|
||
} else {
|
||
reject(new Error('Unknown API response: ' + data));
|
||
}
|
||
} catch (e) {
|
||
reject(e);
|
||
}
|
||
});
|
||
});
|
||
|
||
req.on('error', reject);
|
||
req.write(payload);
|
||
req.end();
|
||
});
|
||
}
|
||
|
||
// 批量为多个汉字生成介绍（单次 API 调用处理多个汉字）
|
||
async function generateBatchIntroduce(chars) {
|
||
const charList = chars.join('、');
|
||
const prompt = `请为以下汉字分别生成简短的词组介绍。
|
||
|
||
汉字列表：${charList}
|
||
|
||
要求：
|
||
1. 每个汉字的介绍格式为："字，词组的字"，其中词组是一个包含该汉字的常见词语
|
||
2. 请严格按照 JSON 数组格式输出，不要有其他内容
|
||
|
||
示例输出格式：
|
||
[{"char":"天","introduce":"天，天空的天"},{"char":"地","introduce":"地，大地的地"},{"char":"人","introduce":"人，人民的人"}]
|
||
|
||
请直接输出 JSON 数组，不要包含任何解释或 markdown 代码块标记。`;
|
||
|
||
try {
|
||
const response = await callHunyuan(prompt);
|
||
// 尝试解析 JSON 响应
|
||
let jsonStr = response.trim();
|
||
|
||
// 移除可能的 markdown 代码块标记
|
||
jsonStr = jsonStr.replace(/^```json?\s*/i, '').replace(/\s*```$/i, '');
|
||
|
||
// 尝试提取 JSON 数组
|
||
const jsonMatch = jsonStr.match(/\[[\s\S]*\]/);
|
||
if (jsonMatch) {
|
||
jsonStr = jsonMatch[0];
|
||
}
|
||
|
||
const parsed = JSON.parse(jsonStr);
|
||
|
||
// 验证并修正结果
|
||
const results = [];
|
||
for (const char of chars) {
|
||
const found = parsed.find((item) => item.char === char);
|
||
if (found && found.introduce) {
|
||
// 确保格式正确
|
||
let introduce = found.introduce.replace(/["""]/g, '').trim();
|
||
if (!introduce.startsWith(char + '，')) {
|
||
introduce = `${char}，${introduce}`;
|
||
}
|
||
results.push({ char, introduce });
|
||
} else {
|
||
// 如果没找到，使用默认格式
|
||
results.push({ char, introduce: `${char}，${char}字的${char}` });
|
||
}
|
||
}
|
||
|
||
return results;
|
||
} catch (error) {
|
||
console.error(`批量生成介绍失败:`, error.message);
|
||
// 失败时返回默认格式
|
||
return chars.map((char) => ({
|
||
char,
|
||
introduce: `${char}，${char}字的${char}`,
|
||
}));
|
||
}
|
||
}
|
||
|
||
// 延时函数
|
||
function delay(ms) {
|
||
return new Promise((resolve) => setTimeout(resolve, ms));
|
||
}
|
||
|
||
// 主函数
|
||
async function main() {
|
||
// 检查环境变量
|
||
if (!SECRET_ID || !SECRET_KEY) {
|
||
console.error('请在 .env 文件中配置 TENCENT_SECRET_ID 和 TENCENT_SECRET_KEY');
|
||
console.error('');
|
||
console.error('1. 复制 .env.example 为 .env：');
|
||
console.error(' cp .env.example .env');
|
||
console.error('');
|
||
console.error('2. 编辑 .env 文件，填入你的腾讯云密钥：');
|
||
console.error(' TENCENT_SECRET_ID=your_secret_id');
|
||
console.error(' TENCENT_SECRET_KEY=your_secret_key');
|
||
process.exit(1);
|
||
}
|
||
|
||
// 读取汉字数据
|
||
const charStringPath = path.join(__dirname, '../data/char/char_string.json');
|
||
const charString = JSON.parse(fs.readFileSync(charStringPath, 'utf-8'));
|
||
|
||
// 获取所有汉字
|
||
const chars = charString['3500'].split('');
|
||
console.log(`共有 ${chars.length} 个汉字需要处理`);
|
||
|
||
// 检查是否有已存在的进度文件
|
||
const outputPath = path.join(__dirname, '../data/char/char_introduce.json');
|
||
let results = [];
|
||
let startIndex = 0;
|
||
|
||
if (fs.existsSync(outputPath)) {
|
||
try {
|
||
results = JSON.parse(fs.readFileSync(outputPath, 'utf-8'));
|
||
startIndex = results.length;
|
||
console.log(`发现已有进度，从第 ${startIndex + 1} 个汉字继续`);
|
||
} catch (e) {
|
||
console.log('已有文件解析失败，从头开始');
|
||
}
|
||
}
|
||
|
||
// 批量处理配置
|
||
const BATCH_SIZE = 20; // 每次 API 调用处理 20 个汉字
|
||
const DELAY_BETWEEN_BATCHES = 1000; // 批次间延迟1秒
|
||
|
||
for (let i = startIndex; i < chars.length; i += BATCH_SIZE) {
|
||
const batch = chars.slice(i, Math.min(i + BATCH_SIZE, chars.length));
|
||
console.log(
|
||
`\n处理第 ${i + 1} - ${Math.min(i + BATCH_SIZE, chars.length)} 个汉字（共 ${batch.length} 个）...`
|
||
);
|
||
console.log(`汉字列表: ${batch.join('')}`);
|
||
|
||
// 单次 API 调用处理整批汉字
|
||
const batchResults = await generateBatchIntroduce(batch);
|
||
|
||
// 打印结果
|
||
for (const item of batchResults) {
|
||
console.log(` ${item.char}: ${item.introduce}`);
|
||
}
|
||
|
||
results.push(...batchResults);
|
||
|
||
// 每批处理完后保存进度
|
||
fs.writeFileSync(outputPath, JSON.stringify(results, null, 2), 'utf-8');
|
||
console.log(`已保存进度：${results.length}/${chars.length}`);
|
||
|
||
// 批次间延迟，避免API限流
|
||
if (i + BATCH_SIZE < chars.length) {
|
||
await delay(DELAY_BETWEEN_BATCHES);
|
||
}
|
||
}
|
||
|
||
console.log(`\n完成！共生成 ${results.length} 个汉字介绍`);
|
||
console.log(`结果已保存到: ${outputPath}`);
|
||
}
|
||
|
||
main().catch(console.error);
|