109 lines
3.0 KiB
TypeScript
109 lines
3.0 KiB
TypeScript
// Developer-only scripts moved out of infrastructure: DOM exporter for local HTML dumps.
|
|
// NOT for production automation; intended as a developer utility to generate compact DOM exports
|
|
// for manual inspection and to aid writing Playwright automations.
|
|
|
|
import { chromium } from "playwright";
|
|
import { promises as fs } from "fs";
|
|
import path from "path";
|
|
import { extractDom, ExportedElement } from "./domExtractor";
|
|
|
|
const INPUT_DIR = path.join(process.cwd(), "html-dumps");
|
|
const OUTPUT_DIR = path.join(process.cwd(), "html-dumps-optimized");
|
|
|
|
async function ensureDir(dir: string): Promise<void> {
|
|
try {
|
|
await fs.mkdir(dir, { recursive: true });
|
|
} catch {
|
|
// ignore
|
|
}
|
|
}
|
|
|
|
async function exportAll(): Promise<void> {
|
|
await ensureDir(OUTPUT_DIR);
|
|
|
|
async function collectHtmlFiles(dir: string): Promise<string[]> {
|
|
const entries = await fs.readdir(dir, { withFileTypes: true });
|
|
const results: string[] = [];
|
|
for (const ent of entries) {
|
|
const p = path.join(dir, ent.name);
|
|
if (ent.isDirectory()) {
|
|
results.push(...(await collectHtmlFiles(p)));
|
|
} else if (ent.isFile() && ent.name.endsWith(".html")) {
|
|
results.push(path.relative(INPUT_DIR, p));
|
|
}
|
|
}
|
|
return results;
|
|
}
|
|
|
|
let htmlFiles: string[] = [];
|
|
try {
|
|
htmlFiles = await collectHtmlFiles(INPUT_DIR);
|
|
} catch (err) {
|
|
console.error(
|
|
"Could not read input directory recursively:",
|
|
INPUT_DIR,
|
|
err
|
|
);
|
|
process.exit(1);
|
|
return;
|
|
}
|
|
|
|
if (htmlFiles.length === 0) {
|
|
console.log("No .html files found in", INPUT_DIR);
|
|
return;
|
|
}
|
|
|
|
const browser = await chromium.launch({ headless: true });
|
|
try {
|
|
for (const file of htmlFiles) {
|
|
const abs = path.join(INPUT_DIR, file);
|
|
const url = "file://" + abs;
|
|
const page = await browser.newPage();
|
|
try {
|
|
await page.addInitScript({
|
|
content: "window.__name = window.__name || (fn => fn);",
|
|
});
|
|
await page.goto(url, {
|
|
waitUntil: "domcontentloaded",
|
|
timeout: 10000,
|
|
});
|
|
const items = (await page.evaluate(
|
|
extractDom as () => ExportedElement[]
|
|
)) as unknown as ExportedElement[];
|
|
const outPath = path.join(
|
|
OUTPUT_DIR,
|
|
file.replace(/\.html$/, ".json")
|
|
);
|
|
await fs.mkdir(path.dirname(outPath), { recursive: true });
|
|
await fs.writeFile(
|
|
outPath,
|
|
JSON.stringify(items, null, 2),
|
|
"utf8"
|
|
);
|
|
console.log(
|
|
"exported " +
|
|
file +
|
|
" -> " +
|
|
path.relative(process.cwd(), outPath) +
|
|
" (elements: " +
|
|
(Array.isArray(items) ? items.length : 0) +
|
|
")"
|
|
);
|
|
} catch (e) {
|
|
console.error("Failed processing", file, e);
|
|
} finally {
|
|
await page.close();
|
|
}
|
|
}
|
|
} finally {
|
|
await browser.close();
|
|
}
|
|
}
|
|
|
|
if (typeof require !== "undefined" && require.main === module) {
|
|
exportAll().catch((err) => {
|
|
console.error(err);
|
|
process.exit(1);
|
|
});
|
|
}
|