Migrate project to TypeScript and Playwright scraping
This commit is contained in:
@@ -0,0 +1,53 @@
|
||||
// @ts-nocheck
|
||||
import { chromium } from "playwright";
|
||||
|
||||
const visibleBrowsers = new Set();
|
||||
|
||||
export async function openInBrowser(url) {
|
||||
const browser = await chromium.launch({ headless: false });
|
||||
const page = await browser.newPage({ viewport: { width: 1440, height: 1000 } });
|
||||
await page.goto(url, { waitUntil: "domcontentloaded", timeout: 30_000 });
|
||||
visibleBrowsers.add(browser);
|
||||
return { browser, page };
|
||||
}
|
||||
|
||||
export async function searchInBrowser({ url, headless = true, waitAfterLoadMs = 3000 }) {
|
||||
const browser = await chromium.launch({ headless });
|
||||
const page = await browser.newPage({ viewport: { width: 1440, height: 1000 } });
|
||||
|
||||
try {
|
||||
const response = await page.goto(url, {
|
||||
waitUntil: "domcontentloaded",
|
||||
timeout: 30_000,
|
||||
});
|
||||
await page.waitForTimeout(Math.max(250, Number(waitAfterLoadMs) || 3000));
|
||||
|
||||
const [currentUrl, title, pageSource] = await Promise.all([
|
||||
page.url(),
|
||||
page.title(),
|
||||
page.content(),
|
||||
]);
|
||||
const cleanText = await page.locator("body").innerText().catch(() => "");
|
||||
const failed = /something has gone wrong|reference id|access denied/i.test(cleanText);
|
||||
|
||||
return {
|
||||
ok: Boolean(response?.ok()) && !failed,
|
||||
status: response?.status() || 0,
|
||||
url: currentUrl,
|
||||
title,
|
||||
text: cleanText.slice(0, 1200),
|
||||
pageSource,
|
||||
};
|
||||
} finally {
|
||||
await browser.close();
|
||||
}
|
||||
}
|
||||
|
||||
export async function scrapePage(url, options = {}) {
|
||||
return searchInBrowser({ url, headless: true, waitAfterLoadMs: 750, ...options });
|
||||
}
|
||||
|
||||
export async function closeVisibleBrowsers() {
|
||||
await Promise.all([...visibleBrowsers].map((browser) => browser.close().catch(() => {})));
|
||||
visibleBrowsers.clear();
|
||||
}
|
||||
@@ -1,5 +1,6 @@
|
||||
import { executeEndpointWrite, queryEndpoint } from "../endpoint/endpoint-client.js";
|
||||
import { openInFirefox, searchInFirefox } from "./firefox-adapter.js";
|
||||
// @ts-nocheck
|
||||
import { executeEndpointWrite, queryEndpoint } from "../endpoint/endpoint-client.ts";
|
||||
import { openInBrowser, scrapePage, searchInBrowser } from "./browser-adapter.ts";
|
||||
import {
|
||||
prepareManufacturerCatalogSqls,
|
||||
mappingSourcesSql,
|
||||
@@ -8,7 +9,7 @@ import {
|
||||
productCombinationsByReferenceSql,
|
||||
productToDoByManufacturerSql,
|
||||
suggestedProductsSql,
|
||||
} from "../sql/catalog-maker.js";
|
||||
} from "../sql/catalog-maker.ts";
|
||||
|
||||
export async function loadManufacturerCatalogState(config, { manufacturerId, position = 1 }) {
|
||||
const refreshResults = [];
|
||||
@@ -181,9 +182,9 @@ export async function findProductSources(
|
||||
const candidate = item.candidates?.[0];
|
||||
if (candidate?.url) {
|
||||
try {
|
||||
openInFirefox(candidate.url);
|
||||
await openInBrowser(candidate.url);
|
||||
} catch {
|
||||
// Picture extraction can continue even if visible Firefox cannot start.
|
||||
// Picture extraction can continue even if the visible browser cannot start.
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -249,13 +250,13 @@ async function findSourceCandidates(source, product, { includePictures = false }
|
||||
query: "",
|
||||
searchUrl: lastSearchUrl,
|
||||
status: "error",
|
||||
error: "EAN is required for Idealo Firefox search.",
|
||||
error: "EAN is required for Idealo Playwright search.",
|
||||
candidates: [],
|
||||
};
|
||||
}
|
||||
|
||||
try {
|
||||
const result = await searchInFirefox({
|
||||
const result = await searchInBrowser({
|
||||
url: attempt.url,
|
||||
headless: source.browserHeadless,
|
||||
waitAfterLoadMs: source.waitAfterLoadMs,
|
||||
@@ -283,14 +284,14 @@ async function findSourceCandidates(source, product, { includePictures = false }
|
||||
query: attempt.query,
|
||||
searchUrl: result.url || attempt.url,
|
||||
status: "error",
|
||||
error: "Idealo returned an error page in Firefox.",
|
||||
error: "Idealo returned an error page in Playwright.",
|
||||
candidates: buildManualSearchCandidates(searchAttempts, product),
|
||||
};
|
||||
} catch (error) {
|
||||
try {
|
||||
openInFirefox(attempt.url);
|
||||
await openInBrowser(attempt.url);
|
||||
} catch {
|
||||
// The manual candidate below remains available even if Firefox cannot start.
|
||||
// The manual candidate below remains available even if the browser cannot start.
|
||||
}
|
||||
|
||||
return {
|
||||
@@ -298,7 +299,7 @@ async function findSourceCandidates(source, product, { includePictures = false }
|
||||
query: attempt.query,
|
||||
searchUrl: attempt.url,
|
||||
status: "manual-search",
|
||||
error: `Firefox automation unavailable: ${error.message || "search failed"}`,
|
||||
error: `Playwright automation unavailable: ${error.message || "search failed"}`,
|
||||
candidates: buildManualSearchCandidates(searchAttempts, product),
|
||||
};
|
||||
}
|
||||
@@ -307,19 +308,12 @@ async function findSourceCandidates(source, product, { includePictures = false }
|
||||
for (const attempt of searchAttempts) {
|
||||
lastSearchUrl = attempt.url;
|
||||
try {
|
||||
const response = await fetch(attempt.url, {
|
||||
method: attempt.method || "GET",
|
||||
redirect: "follow",
|
||||
headers: {
|
||||
"User-Agent": "CatalogMaker/0.1",
|
||||
...(attempt.body ? { "Content-Type": "application/x-www-form-urlencoded" } : {}),
|
||||
},
|
||||
body: attempt.body,
|
||||
signal: AbortSignal.timeout(9000),
|
||||
const response = await scrapePage(attempt.url, {
|
||||
waitAfterLoadMs: source.waitAfterLoadMs,
|
||||
});
|
||||
const html = await response.text();
|
||||
const html = response.pageSource;
|
||||
const candidates = [
|
||||
...extractResponseUrlCandidate(response.url || attempt.url, html, product),
|
||||
...extractResponseUrlCandidate(response.url || attempt.url, html, product),
|
||||
...extractDirectCandidates(html, source, product),
|
||||
].slice(0, 3);
|
||||
|
||||
@@ -604,12 +598,8 @@ async function verifyHudyVariant(url, product) {
|
||||
if (visited.has(page.url)) continue;
|
||||
visited.add(page.url);
|
||||
|
||||
const response = await fetch(page.url, {
|
||||
headers: { "User-Agent": "CatalogMaker/0.1" },
|
||||
redirect: "follow",
|
||||
signal: AbortSignal.timeout(9000),
|
||||
});
|
||||
const html = await response.text();
|
||||
const response = await scrapePage(page.url, { waitAfterLoadMs: 750 });
|
||||
const html = response.pageSource;
|
||||
lastStatus = response.status;
|
||||
const pageEans = extractPageEans(html);
|
||||
const matchedEan = pageEans.find((value) => wantedEans.has(value)) || "";
|
||||
@@ -712,12 +702,8 @@ function extractPictureUrls(html, pageUrl) {
|
||||
|
||||
async function extractPicturesFromPage(url) {
|
||||
try {
|
||||
const response = await fetch(url, {
|
||||
headers: { "User-Agent": "CatalogMaker/0.1" },
|
||||
redirect: "follow",
|
||||
signal: AbortSignal.timeout(9000),
|
||||
});
|
||||
return extractHudyGalleryPictureUrls(await response.text(), response.url || url);
|
||||
const response = await scrapePage(url, { waitAfterLoadMs: 750 });
|
||||
return extractHudyGalleryPictureUrls(response.pageSource, response.url || url);
|
||||
} catch {
|
||||
return [];
|
||||
}
|
||||
@@ -878,22 +864,7 @@ async function checkMappingSourceUrls(items) {
|
||||
async function checkUrl(url) {
|
||||
try {
|
||||
const target = normalizeSourceUrl(url);
|
||||
let response = await fetch(target, {
|
||||
method: "HEAD",
|
||||
redirect: "follow",
|
||||
signal: AbortSignal.timeout(6000),
|
||||
});
|
||||
|
||||
if (!response.ok) {
|
||||
response = await fetch(target, {
|
||||
method: "GET",
|
||||
redirect: "follow",
|
||||
headers: {
|
||||
"User-Agent": "CatalogMaker/0.1",
|
||||
},
|
||||
signal: AbortSignal.timeout(6000),
|
||||
});
|
||||
}
|
||||
const response = await scrapePage(target, { waitAfterLoadMs: 250 });
|
||||
|
||||
return {
|
||||
health: response.ok ? "ok" : "warning",
|
||||
@@ -1,82 +0,0 @@
|
||||
import { spawn } from "node:child_process";
|
||||
import { Builder } from "selenium-webdriver";
|
||||
import firefox from "selenium-webdriver/firefox.js";
|
||||
import { start } from "geckodriver";
|
||||
|
||||
const FIREFOX_PATHS = [
|
||||
"C:\\Program Files\\Mozilla Firefox\\firefox.exe",
|
||||
"C:\\Program Files (x86)\\Mozilla Firefox\\firefox.exe",
|
||||
];
|
||||
|
||||
export function openInFirefox(url) {
|
||||
const firefoxPath = FIREFOX_PATHS.find((candidate) => candidate);
|
||||
if (!firefoxPath) throw new Error("Firefox executable was not found.");
|
||||
|
||||
const process = spawn(firefoxPath, [url], {
|
||||
detached: true,
|
||||
stdio: "ignore",
|
||||
windowsHide: true,
|
||||
});
|
||||
process.unref();
|
||||
}
|
||||
|
||||
export async function searchInFirefox({ url, headless = false, waitAfterLoadMs = 3000 }) {
|
||||
const firefoxPath = FIREFOX_PATHS.find((candidate) => candidate);
|
||||
const port = 4444;
|
||||
const geckodriver = await start({ port, binary: firefoxPath, log: "error" });
|
||||
let driver;
|
||||
|
||||
try {
|
||||
await waitForWebDriver(port);
|
||||
const options = new firefox.Options();
|
||||
if (firefoxPath) options.setBinary(firefoxPath);
|
||||
if (headless) options.addArguments("-headless");
|
||||
|
||||
driver = await new Builder()
|
||||
.forBrowser("firefox")
|
||||
.setFirefoxOptions(options)
|
||||
.usingServer(`http://127.0.0.1:${port}`)
|
||||
.build();
|
||||
|
||||
await driver.get(url);
|
||||
await driver.sleep(Math.max(1000, Number(waitAfterLoadMs) || 3000));
|
||||
|
||||
const [currentUrl, title, pageSource] = await Promise.all([
|
||||
driver.getCurrentUrl(),
|
||||
driver.getTitle(),
|
||||
driver.getPageSource(),
|
||||
]);
|
||||
const cleanText = stripHtml(pageSource).slice(0, 1200);
|
||||
const failed = /something has gone wrong|reference id/i.test(cleanText);
|
||||
|
||||
return { ok: !failed, url: currentUrl, title, text: cleanText, pageSource };
|
||||
} finally {
|
||||
if (driver) await driver.quit().catch(() => {});
|
||||
geckodriver.kill();
|
||||
}
|
||||
}
|
||||
|
||||
async function waitForWebDriver(port) {
|
||||
const deadline = Date.now() + 10000;
|
||||
while (Date.now() < deadline) {
|
||||
try {
|
||||
const response = await fetch(`http://127.0.0.1:${port}/status`);
|
||||
if (response.ok) return;
|
||||
} catch {
|
||||
// Geckodriver needs a short moment to open its local port.
|
||||
}
|
||||
await new Promise((resolve) => setTimeout(resolve, 150));
|
||||
}
|
||||
throw new Error("Firefox WebDriver did not start.");
|
||||
}
|
||||
|
||||
function stripHtml(value) {
|
||||
return String(value ?? "")
|
||||
.replace(/<script[\s\S]*?<\/script>/gi, " ")
|
||||
.replace(/<style[\s\S]*?<\/style>/gi, " ")
|
||||
.replace(/<[^>]+>/g, " ")
|
||||
.replace(/ /gi, " ")
|
||||
.replace(/&/gi, "&")
|
||||
.replace(/\s+/g, " ")
|
||||
.trim();
|
||||
}
|
||||
@@ -1,5 +1,6 @@
|
||||
import { queryEndpoint } from "../endpoint/endpoint-client.js";
|
||||
import { catalogMakerSql } from "../sql/catalog-maker.js";
|
||||
// @ts-nocheck
|
||||
import { queryEndpoint } from "../endpoint/endpoint-client.ts";
|
||||
import { catalogMakerSql } from "../sql/catalog-maker.ts";
|
||||
|
||||
export async function listLanguages(config) {
|
||||
const result = await queryEndpoint(config, catalogMakerSql.languages);
|
||||
@@ -1,5 +1,6 @@
|
||||
import { queryEndpoint } from "../endpoint/endpoint-client.js";
|
||||
import { catalogMakerSql } from "../sql/catalog-maker.js";
|
||||
// @ts-nocheck
|
||||
import { queryEndpoint } from "../endpoint/endpoint-client.ts";
|
||||
import { catalogMakerSql } from "../sql/catalog-maker.ts";
|
||||
|
||||
export async function listManufacturers(config) {
|
||||
const result = await queryEndpoint(config, catalogMakerSql.manufacturers);
|
||||
Reference in New Issue
Block a user