1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
|
import assert from "assert";
import { Readability } from "@mozilla/readability";
import { Mutex } from "async-mutex";
import { Job, Worker } from "bullmq";
import DOMPurify from "dompurify";
import { eq } from "drizzle-orm";
import { isShuttingDown, shutdownPromise } from "exit";
import { JSDOM } from "jsdom";
import metascraper from "metascraper";
import metascraperDescription from "metascraper-description";
import metascraperImage from "metascraper-image";
import metascraperLogo from "metascraper-logo-favicon";
import metascraperReadability from "metascraper-readability";
import metascraperTitle from "metascraper-title";
import metascraperTwitter from "metascraper-twitter";
import metascraperUrl from "metascraper-url";
import { Browser } from "puppeteer";
import puppeteer from "puppeteer-extra";
import AdblockerPlugin from "puppeteer-extra-plugin-adblocker";
import StealthPlugin from "puppeteer-extra-plugin-stealth";
import { db } from "@hoarder/db";
import { bookmarkLinks } from "@hoarder/db/schema";
import serverConfig from "@hoarder/shared/config";
import logger from "@hoarder/shared/logger";
import {
LinkCrawlerQueue,
OpenAIQueue,
queueConnectionDetails,
SearchIndexingQueue,
ZCrawlLinkRequest,
zCrawlLinkRequestSchema,
} from "@hoarder/shared/queues";
const metascraperParser = metascraper([
metascraperReadability(),
metascraperTitle(),
metascraperDescription(),
metascraperTwitter(),
metascraperImage(),
metascraperLogo(),
metascraperUrl(),
]);
let browser: Browser | undefined;
// Guards the interactions with the browser instance.
// This is needed given that most of the browser APIs are async.
const browserMutex = new Mutex();
async function launchBrowser() {
browser = undefined;
await browserMutex.runExclusive(async () => {
browser = await puppeteer.launch({
headless: serverConfig.crawler.headlessBrowser,
executablePath: serverConfig.crawler.browserExecutablePath,
userDataDir: serverConfig.crawler.browserUserDataDir,
});
browser.on("disconnected", async () => {
if (isShuttingDown) {
logger.info(
"The puppeteer browser got disconnected. But we're shutting down so won't restart it.",
);
return;
}
logger.info(
"The puppeteer browser got disconnected. Will attempt to launch it again.",
);
await launchBrowser();
});
});
}
export class CrawlerWorker {
static async build() {
puppeteer.use(StealthPlugin());
puppeteer.use(
AdblockerPlugin({
blockTrackersAndAnnoyances: true,
}),
);
await launchBrowser();
logger.info("Starting crawler worker ...");
const worker = new Worker<ZCrawlLinkRequest, void>(
LinkCrawlerQueue.name,
runCrawler,
{
connection: queueConnectionDetails,
autorun: false,
},
);
worker.on("completed", (job) => {
const jobId = job?.id || "unknown";
logger.info(`[Crawler][${jobId}] Completed successfully`);
});
worker.on("failed", (job, error) => {
const jobId = job?.id || "unknown";
logger.error(`[Crawler][${jobId}] Crawling job failed: ${error}`);
});
return worker;
}
}
async function getBookmarkUrl(bookmarkId: string) {
const bookmark = await db.query.bookmarkLinks.findFirst({
where: eq(bookmarkLinks.id, bookmarkId),
});
if (!bookmark) {
throw new Error("The bookmark either doesn't exist or not a link");
}
return bookmark.url;
}
/**
* This provides some "basic" protection from malicious URLs. However, all of those
* can be easily circumvented by pointing dns of origin to localhost, or with
* redirects.
*/
function validateUrl(url: string) {
const urlParsed = new URL(url);
if (urlParsed.protocol != "http:" && urlParsed.protocol != "https:") {
throw new Error(`Unsupported URL protocol: ${urlParsed.protocol}`);
}
if (["localhost", "127.0.0.1", "0.0.0.0"].includes(urlParsed.hostname)) {
throw new Error(`Link hostname rejected: ${urlParsed.hostname}`);
}
}
async function crawlPage(url: string) {
assert(browser);
const context = await browser.createBrowserContext();
try {
const page = await context.newPage();
await page.goto(url, {
timeout: 10000, // 10 seconds
});
// Wait until there's at most two connections for 2 seconds
// Attempt to wait only for 5 seconds
await Promise.race([
page.waitForNetworkIdle({
idleTime: 1000, // 1 sec
concurrency: 2,
}),
new Promise((f) => setTimeout(f, 5000)),
]);
const htmlContent = await page.content();
return htmlContent;
} finally {
await context.close();
}
}
async function runCrawler(job: Job<ZCrawlLinkRequest, void>) {
const jobId = job.id || "unknown";
const request = zCrawlLinkRequestSchema.safeParse(job.data);
if (!request.success) {
logger.error(
`[Crawler][${jobId}] Got malformed job request: ${request.error.toString()}`,
);
return;
}
const { bookmarkId } = request.data;
const url = await getBookmarkUrl(bookmarkId);
logger.info(
`[Crawler][${jobId}] Will crawl "${url}" for link with id "${bookmarkId}"`,
);
validateUrl(url);
const htmlContent = await crawlPage(url);
const meta = await metascraperParser({
url,
html: htmlContent,
});
const window = new JSDOM("").window;
const purify = DOMPurify(window);
const purifiedHTML = purify.sanitize(htmlContent);
const purifiedDOM = new JSDOM(purifiedHTML, { url });
const readableContent = new Readability(purifiedDOM.window.document).parse();
// TODO(important): Restrict the size of content to store
await db
.update(bookmarkLinks)
.set({
title: meta.title,
description: meta.description,
imageUrl: meta.image,
favicon: meta.logo,
content: readableContent?.textContent,
htmlContent: readableContent?.content,
crawledAt: new Date(),
})
.where(eq(bookmarkLinks.id, bookmarkId));
// Enqueue openai job
OpenAIQueue.add("openai", {
bookmarkId,
});
// Update the search index
SearchIndexingQueue.add("search_indexing", {
bookmarkId,
type: "index",
});
}
|
