"Meta tag scraper" JSON API using @cloudflare workers. https://meta.motyar.workers.dev Twitter thread here https://twitter.com/motyar/status/1439208342711197696

meta.motyar.workers.dev.js

async function handleRequest(request) {

    var inputUrl = request.url.replace("https://meta.motyar.workers.dev/", "")
    // return error if the url is not passed
    if(!inputUrl){
        return new Response(JSON.stringify({"error":"Input URL missing! Pass it like https://meta.motyar.workers.dev/example.com"}), {
            headers: {
                "content-type": "application/json;charset=UTF-8"
            }
        })
    }

    // add http if missing
    if (!inputUrl.startsWith("http")) {
        inputUrl = "http://" + inputUrl;
    }

    const response = await fetch(inputUrl)


    var meta = {}

    await new HTMLRewriter()
        .on('meta', {
            element(e) {
                let name = e.getAttribute("name")
                let property = e.getAttribute("property")
                let content = e.getAttribute("content")

                let key = (name || property);

                if (key) {
                    meta[key] = content;
                }
            }
        }).transform(response.clone());

    // if title is missing in meta tags, get the title tag using regex
    if (!meta["title"]) {
        const html = await response.text()
        var titleM = html.match(/<title[^>]*>([^<]+)<\/title>/);
        if (titleM) {
            meta["title"] = titleM[1];
        }
    }

    let title = meta['title'] || meta['og:title'];
    let description = meta['description'] || meta['og:description'];

    return new Response(JSON.stringify(meta), {
        headers: {
            "content-type": "application/json;charset=UTF-8"
        }
    })
}

addEventListener("fetch", event => {
    return event.respondWith(handleRequest(event.request))
})