fix(monitor): remove non-existent worker columns from job_run_logs query
The job_run_logs table tracks scheduled job orchestration, not individual worker jobs. Worker info (worker_id, worker_hostname) belongs on dispensary_crawl_jobs, not job_run_logs. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
115
backend/new-scrapers/scrape-deeply-rooted-playwright.ts
Normal file
115
backend/new-scrapers/scrape-deeply-rooted-playwright.ts
Normal file
@@ -0,0 +1,115 @@
|
||||
import { chromium, Frame } from 'playwright';
|
||||
|
||||
type Product = {
|
||||
name: string;
|
||||
brand?: string;
|
||||
price?: number;
|
||||
size?: string;
|
||||
category?: string;
|
||||
url?: string;
|
||||
};
|
||||
|
||||
async function getDutchieFrame(page: any): Promise<Frame> {
|
||||
const iframeHandle = await page.waitForSelector(
|
||||
'iframe[src*="dutchie"], iframe[srcdoc*="dutchie"], iframe[id^="iframe-"]',
|
||||
{ timeout: 45000 }
|
||||
);
|
||||
|
||||
const frame = await iframeHandle.contentFrame();
|
||||
if (!frame) {
|
||||
throw new Error('Unable to access embedded Dutchie iframe.');
|
||||
}
|
||||
|
||||
await frame.waitForLoadState('domcontentloaded', { timeout: 30000 });
|
||||
return frame;
|
||||
}
|
||||
|
||||
async function loadAllProducts(frame: Frame): Promise<void> {
|
||||
const maxScrolls = 20;
|
||||
for (let i = 0; i < maxScrolls; i++) {
|
||||
const beforeCount = await frame.$$eval('[data-testid*="product"], [data-testid*="card"]', els => els.length);
|
||||
await frame.mouse.wheel(0, 1200);
|
||||
await frame.waitForTimeout(800);
|
||||
const afterCount = await frame.$$eval('[data-testid*="product"], [data-testid*="card"]', els => els.length);
|
||||
if (afterCount <= beforeCount) break;
|
||||
}
|
||||
await frame.evaluate(() => window.scrollTo({ top: 0 }));
|
||||
}
|
||||
|
||||
async function extractProducts(frame: Frame): Promise<Product[]> {
|
||||
return frame.evaluate(() => {
|
||||
const cards = Array.from(
|
||||
document.querySelectorAll('[data-testid="product-list-item"], [data-testid="card-link"], [data-testid*="product-card"]')
|
||||
);
|
||||
|
||||
return cards.map((card: Element) => {
|
||||
const name =
|
||||
(card.querySelector('[data-testid="product-card-name"]') as HTMLElement)?.innerText?.trim() ||
|
||||
(card.querySelector('[data-testid="product-name"]') as HTMLElement)?.innerText?.trim() ||
|
||||
(card.querySelector('h3, h4') as HTMLElement)?.innerText?.trim() ||
|
||||
(card.textContent || '').split('\n').map(t => t.trim()).find(t => t.length > 3) ||
|
||||
'';
|
||||
|
||||
const brand =
|
||||
(card.querySelector('[data-testid="product-card-brand"]') as HTMLElement)?.innerText?.trim() ||
|
||||
(card.querySelector('[data-testid="product-brand"]') as HTMLElement)?.innerText?.trim() ||
|
||||
undefined;
|
||||
|
||||
const priceText =
|
||||
(card.querySelector('[data-testid="product-card-price"]') as HTMLElement)?.innerText ||
|
||||
(card.textContent || '');
|
||||
const priceMatch = priceText.match(/\$?(\d+(?:\.\d{2})?)/);
|
||||
const price = priceMatch ? parseFloat(priceMatch[1]) : undefined;
|
||||
|
||||
const size =
|
||||
(card.querySelector('[data-testid*="size"]') as HTMLElement)?.innerText?.trim() ||
|
||||
(card.querySelector('[data-testid*="weight"]') as HTMLElement)?.innerText?.trim() ||
|
||||
undefined;
|
||||
|
||||
const category =
|
||||
(card.querySelector('[data-testid*="category"]') as HTMLElement)?.innerText?.trim() ||
|
||||
undefined;
|
||||
|
||||
const link = card.querySelector('a[href*="/product/"]') as HTMLAnchorElement | null;
|
||||
const url = link?.href;
|
||||
|
||||
return { name, brand, price, size, category, url };
|
||||
}).filter(p => p.name);
|
||||
});
|
||||
}
|
||||
|
||||
async function main() {
|
||||
const targetUrl = 'https://azdeeplyrooted.com/menu';
|
||||
|
||||
const browser = await chromium.launch({
|
||||
headless: true,
|
||||
args: ['--no-sandbox', '--disable-setuid-sandbox', '--disable-blink-features=AutomationControlled'],
|
||||
});
|
||||
|
||||
const page = await browser.newPage({
|
||||
viewport: { width: 1300, height: 900 },
|
||||
userAgent:
|
||||
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
|
||||
});
|
||||
|
||||
try {
|
||||
console.log(`Navigating to ${targetUrl}...`);
|
||||
await page.goto(targetUrl, { waitUntil: 'domcontentloaded', timeout: 60000 });
|
||||
|
||||
const frame = await getDutchieFrame(page);
|
||||
await frame.waitForSelector('[data-testid*="product"], [data-testid*="card"]', { timeout: 60000 }).catch(() => undefined);
|
||||
|
||||
await loadAllProducts(frame);
|
||||
const products = await extractProducts(frame);
|
||||
|
||||
console.log(`Found ${products.length} products`);
|
||||
console.log(JSON.stringify(products.slice(0, 20), null, 2));
|
||||
} catch (err) {
|
||||
console.error('Scrape failed:', err);
|
||||
process.exitCode = 1;
|
||||
} finally {
|
||||
await browser.close();
|
||||
}
|
||||
}
|
||||
|
||||
main();
|
||||
Reference in New Issue
Block a user