161 lines
4.8 KiB
JavaScript
161 lines
4.8 KiB
JavaScript
/**
|
||
* 行业趋势监测脚本 — 每日 10:00 运行
|
||
* 从 HackerNews API 拉取热帖,筛选 AI 相关主题
|
||
*/
|
||
import "dotenv/config";
|
||
import fs from "fs";
|
||
import path from "path";
|
||
import { fileURLToPath } from "url";
|
||
|
||
const __dirname = path.dirname(fileURLToPath(import.meta.url));
|
||
const DATA_DIR = path.join(__dirname, "..", "data", "trend-alerts");
|
||
|
||
const HN_TOP_STORIES = "https://hacker-news.firebaseio.com/v0/topstories.json";
|
||
const HN_ITEM = "https://hacker-news.firebaseio.com/v0/item";
|
||
// 主数据源:Algolia HN API(全球可达、单请求返回热帖含分数/评论,避免 firebaseio 从国内服务器超时)
|
||
const HN_ALGOLIA = "https://hn.algolia.com/api/v1/search?tags=front_page&hitsPerPage=100";
|
||
|
||
// AI 相关关键词
|
||
const AI_KEYWORDS = [
|
||
"ai", "llm", "gpt", "machine learning", "deep learning",
|
||
"openai", "anthropic", "claude", "gemini", "copilot",
|
||
"agent", "prompt", "rag", "vector", "embedding",
|
||
"transformer", "LLM", "GPT", "AI", "chatgpt", "ChatGPT",
|
||
"diffusion", "stable diffusion", "langchain", "fine-tun",
|
||
"neural", "inference", "open source ai", "llama", "mistral",
|
||
];
|
||
|
||
function todayDateStr() {
|
||
return new Date().toISOString().slice(0, 10);
|
||
}
|
||
|
||
function log(message) {
|
||
const ts = new Date().toISOString();
|
||
console.log(`[${ts}] ${message}`);
|
||
}
|
||
|
||
async function fetchJson(url) {
|
||
let lastErr;
|
||
for (let attempt = 0; attempt < 3; attempt++) {
|
||
try {
|
||
const res = await fetch(url, { signal: AbortSignal.timeout(12000) });
|
||
if (!res.ok) throw new Error(`HTTP ${res.status}`);
|
||
return await res.json();
|
||
} catch (e) {
|
||
lastErr = e;
|
||
if (attempt < 2) await new Promise((r) => setTimeout(r, 500 * (attempt + 1)));
|
||
}
|
||
}
|
||
throw lastErr;
|
||
}
|
||
|
||
function matchesAIKeywords(title) {
|
||
if (!title) return false;
|
||
const lowerTitle = title.toLowerCase();
|
||
for (const keyword of AI_KEYWORDS) {
|
||
if (lowerTitle.includes(keyword.toLowerCase())) {
|
||
return true;
|
||
}
|
||
}
|
||
return false;
|
||
}
|
||
|
||
async function main() {
|
||
log("行业趋势监测开始...");
|
||
|
||
// 确保数据目录存在
|
||
if (!fs.existsSync(DATA_DIR)) {
|
||
fs.mkdirSync(DATA_DIR, { recursive: true });
|
||
}
|
||
|
||
// 主源:Algolia(单请求、全球可达);失败回退 firebaseio 两段式拉取
|
||
let items = [];
|
||
let source = "algolia";
|
||
try {
|
||
items = await fetchAlgoliaTop();
|
||
log(`[algolia] 获取到 ${items.length} 条热帖`);
|
||
} catch (e) {
|
||
source = "firebase";
|
||
log(`[algolia] 失败(${e.message}),回退 firebaseio...`);
|
||
items = await fetchFirebaseTop();
|
||
log(`[firebase] 获取到 ${items.length} 条热帖`);
|
||
}
|
||
|
||
// 筛选 AI 相关(字段已统一:title/url/score/commentCount/by/time)
|
||
const aiItems = items
|
||
.filter((item) => matchesAIKeywords(item.title))
|
||
.sort((a, b) => (b.score || 0) - (a.score || 0))
|
||
.slice(0, 5);
|
||
|
||
log(`筛选出 ${aiItems.length} 条 AI 相关热帖`);
|
||
|
||
const dateStr = todayDateStr();
|
||
const alertFile = path.join(DATA_DIR, `${dateStr}.json`);
|
||
|
||
const alert = {
|
||
date: dateStr,
|
||
generatedAt: new Date().toISOString(),
|
||
source,
|
||
items: aiItems,
|
||
summary: `从 ${items.length} 条热帖中筛选出 ${aiItems.length} 条 AI 相关话题`,
|
||
};
|
||
|
||
fs.writeFileSync(alertFile, JSON.stringify(alert, null, 2), "utf-8");
|
||
log(`趋势报告已保存: ${alertFile}`);
|
||
|
||
for (const item of aiItems) {
|
||
log(` 🏷 ${item.title} (${item.score} pts, ${item.commentCount} comments)`);
|
||
}
|
||
|
||
log("行业趋势监测完成");
|
||
}
|
||
|
||
async function fetchAlgoliaTop() {
|
||
const data = await fetchJson(HN_ALGOLIA);
|
||
return (data.hits || [])
|
||
.filter((h) => h.title)
|
||
.map((h) => ({
|
||
id: h.objectID,
|
||
title: h.title,
|
||
url: h.url || `https://news.ycombinator.com/item?id=${h.objectID}`,
|
||
score: h.points || 0,
|
||
commentCount: h.num_comments || 0,
|
||
by: h.author || "",
|
||
time: h.created_at || "",
|
||
}));
|
||
}
|
||
|
||
async function fetchFirebaseTop() {
|
||
const storyIds = await fetchJson(HN_TOP_STORIES);
|
||
log(`[firebase] 获取到 ${storyIds.length} 条热帖 ID`);
|
||
const top100Ids = storyIds.slice(0, 100);
|
||
const items = [];
|
||
for (let i = 0; i < top100Ids.length; i += 10) {
|
||
const batch = top100Ids.slice(i, i + 10);
|
||
const results = await Promise.all(
|
||
batch.map((id) => fetchJson(`${HN_ITEM}/${id}.json`).catch(() => null))
|
||
);
|
||
for (const item of results) {
|
||
if (!item) continue;
|
||
items.push({
|
||
id: item.id,
|
||
title: item.title,
|
||
url: item.url || `https://news.ycombinator.com/item?id=${item.id}`,
|
||
score: item.score || 0,
|
||
commentCount: item.descendants || 0,
|
||
by: item.by || "",
|
||
time: item.time ? new Date(item.time * 1000).toISOString() : "",
|
||
});
|
||
}
|
||
if (i + 10 < top100Ids.length) {
|
||
await new Promise((r) => setTimeout(r, 200));
|
||
}
|
||
}
|
||
return items;
|
||
}
|
||
|
||
main().catch((err) => {
|
||
console.error("行业趋势监测失败:", err);
|
||
process.exit(1);
|
||
});
|