R524809
254 lines
7.8 KiB
JavaScript
254 lines
7.8 KiB
JavaScript
const fs = require("fs");
|
||
const path = require("path");
|
||
const https = require("https");
|
||
const crypto = require("crypto");
|
||
|
||
// 腾讯云混元大模型配置
|
||
// 请设置环境变量 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=${signature}`;
|
||
|
||
return authorization;
|
||
}
|
||
|
||
// 调用混元API
|
||
async function callHunyuan(prompt) {
|
||
return new Promise((resolve, reject) => {
|
||
const timestamp = Math.floor(Date.now() / 1000);
|
||
|
||
const payload = JSON.stringify({
|
||
Model: MODEL,
|
||
Messages: [
|
||
{
|
||
Role: "user",
|
||
Content: prompt,
|
||
},
|
||
],
|
||
Stream: false,
|
||
});
|
||
|
||
const authorization = generateSignature(
|
||
SECRET_ID,
|
||
SECRET_KEY,
|
||
HOST,
|
||
payload,
|
||
timestamp
|
||
);
|
||
|
||
const options = {
|
||
hostname: HOST,
|
||
port: 443,
|
||
path: "/",
|
||
method: "POST",
|
||
headers: {
|
||
"Content-Type": "application/json",
|
||
Host: HOST,
|
||
"X-TC-Action": ACTION,
|
||
"X-TC-Version": VERSION,
|
||
"X-TC-Timestamp": timestamp.toString(),
|
||
"X-TC-Region": REGION,
|
||
Authorization: authorization,
|
||
},
|
||
};
|
||
|
||
const req = https.request(options, (res) => {
|
||
let data = "";
|
||
res.on("data", (chunk) => (data += chunk));
|
||
res.on("end", () => {
|
||
try {
|
||
const result = JSON.parse(data);
|
||
if (result.Response && result.Response.Choices) {
|
||
resolve(result.Response.Choices[0].Message.Content.trim());
|
||
} else if (result.Response && result.Response.Error) {
|
||
reject(new Error(result.Response.Error.Message));
|
||
} else {
|
||
reject(new Error("Unknown API response: " + data));
|
||
}
|
||
} catch (e) {
|
||
reject(e);
|
||
}
|
||
});
|
||
});
|
||
|
||
req.on("error", reject);
|
||
req.write(payload);
|
||
req.end();
|
||
});
|
||
}
|
||
|
||
// 为单个汉字生成介绍
|
||
async function generateIntroduce(char) {
|
||
const prompt = `请为汉字"${char}"生成一个简短的词组介绍，格式为"${char}，XXX的${char}"，其中XXX是一个包含该汉字的常见词语。
|
||
例如：
|
||
- 汉字"天"的介绍是"天，天空的天"
|
||
- 汉字"地"的介绍是"地，大地的地"
|
||
- 汉字"人"的介绍是"人，人民的人"
|
||
|
||
请只输出介绍内容，不要包含其他解释。比如对于"天"，只输出"天，天空的天"。`;
|
||
|
||
try {
|
||
const response = await callHunyuan(prompt);
|
||
// 清理响应，只保留核心内容
|
||
let introduce = response.replace(/["""]/g, "").trim();
|
||
// 如果响应包含多余内容，尝试提取核心部分
|
||
const match = introduce.match(/.，.+的./);
|
||
if (match) {
|
||
introduce = match[0];
|
||
}
|
||
// 确保格式正确：如果没有逗号，添加逗号
|
||
if (!introduce.startsWith(char + "，")) {
|
||
// 尝试提取词组部分
|
||
const wordMatch = introduce.match(/(.+的.)/);
|
||
if (wordMatch) {
|
||
introduce = `${char}，${wordMatch[1]}`;
|
||
} else {
|
||
introduce = `${char}，${char}字的${char}`;
|
||
}
|
||
}
|
||
return introduce;
|
||
} catch (error) {
|
||
console.error(`生成 "${char}" 介绍失败:`, error.message);
|
||
return `${char}，${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("请设置环境变量 TENCENT_SECRET_ID 和 TENCENT_SECRET_KEY");
|
||
console.error("例如：");
|
||
console.error(
|
||
" export TENCENT_SECRET_ID=your_secret_id"
|
||
);
|
||
console.error(
|
||
" export 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
|
||
const BATCH_SIZE = 10; // 每批处理10个
|
||
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(
|
||
`处理第 ${i + 1} - ${Math.min(i + BATCH_SIZE, chars.length)} 个汉字...`
|
||
);
|
||
|
||
// 并行处理当前批次
|
||
const batchResults = await Promise.all(
|
||
batch.map(async (char, idx) => {
|
||
// 每个请求之间稍微错开
|
||
await delay(idx * 100);
|
||
const introduce = await generateIntroduce(char);
|
||
console.log(` ${char}: ${introduce}`);
|
||
return { char, introduce };
|
||
})
|
||
);
|
||
|
||
results.push(...batchResults);
|
||
|
||
// 每批处理完后保存进度
|
||
fs.writeFileSync(outputPath, JSON.stringify(results, null, 2), "utf-8");
|
||
console.log(`已保存进度：${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);
|