Extracting rootdomains from URL string in Google Sheets

Question

Hi I am trying to extract the rootdomain from URL string in Google Sheets. I know how to get the domain and I have the formula to remove www. but now I realize it does not strip subdomain prefixes like 'mysite'.site.com; where mysite is not stripped from the domain name.

Question: How can I retrieve the domain.com rootdomain where the domain string contacts alphanumeric characters, then 1 dot, then alphanumeric characters (and nothing more)

Formula so far in Google Sheets:

=REGEXREPLACE(REGEXREPLACE(D3923;"(http(s)?://)?(www\.)?";"");"/.*";"")

Maybe this can be simplified ...

Test cases

    https://www.domain.com/ => domain.com
    https://domain.com/ => domain.com
    http://www.domain.nl/ => domain.com
    http://domain.de/ => domain.com
    http://www.domain.co.uk/ => domain.co.uk
    http://domain.co.au/ => domain.co.au
    sub.domain.org/ => sub.domain.com
    sub.domain.org => sub.domain.com
    domain.com => domain.com
    http://www.domain.nl?par=1  => domain.com
    https://www.domain.nl/test/?par=1  => domain.com
    http2://sub2.startpagina.nl/test/?par=1  => domain.com

Answer 1

Currently using:

=trim(REGEXEXTRACT(REGEXREPLACE(REGEXREPLACE(A2;"https?://";"");"^(w{3}\.)?";"")&"/";"([^/?]+)"))

Seems to work fine

Updated:7-7-2016

(thanks for all the help!)

Answer 2

I think that a most reliable way is to check over TLD list because of TLDs like co.uk, gov.uk and so on that are impossible to extract via a simple regex.

You can define these functions in Tools -> Script editor

function endsWith(str, searchString) {
    position = str.length - searchString.length;
    var lastIndex = str.lastIndexOf(searchString);
    return lastIndex !== -1 && lastIndex === position;
}

function rawToTlds(raw) {
    var letter = new RegExp(/^\w/);
    return raw.split(/\n/).filter(function (t) { return letter.test(t) })
}

function compressString(s) {
    var zippedBlob = Utilities.gzip(Utilities.newBlob(s))
    return Utilities.base64Encode(zippedBlob.getBytes())
}

function uncompressString(x) {
    var zippedBytes = Utilities.base64Decode(x)
    var zippedBlob = Utilities.newBlob(zippedBytes, 'application/x-gzip')
    var stringBlob = Utilities.ungzip(zippedBlob)
    return stringBlob.getDataAsString()
}

function getTlds() {
    var cacheName = 'TLDs'
    var cache = CacheService.getScriptCache();
    var base64Encoded = cache.get(cacheName);
    if (base64Encoded != null) {
        return uncompressString(base64Encoded).split(',')
    }
    var raw = UrlFetchApp.fetch('https://publicsuffix.org/list/public_suffix_list.dat').getContentText()
    var tlds = rawToTlds(raw)
    cache.put(cacheName, compressString(tlds.join()), 21600)
    return tlds
}

function getDomainName(url, level) {
    var tlds = getTlds()

    var domain = url
        .replace(/^http(s)?:\/\//i, "")
        .replace(/^www\./i, "")
        .replace(/\/.*$/, "")
        .replace(/\?.*/, "");

    if (typeof level === 'undefined') {
        return domain
    }

    var result = domain
    var longest = 0
    for (i in tlds) {
        var tld = '.' + tlds[i]
        if (endsWith(domain, tld) && tld.length > longest) {
            var parts = domain.substring(0, domain.length - tld.length).split('.')
            result = parts.slice(parts.length-level+1, parts.length).join('.') + tld
            longest = tld.length
        }
    }

    return result
}

To get second-level domian of A1 use it like this

=getDomainName(A1, 2)

To get full domain of A1 just do

=getDomainName(A1)

EDIT

Public Suffix List has exceeded 100KB. It doesn't fit in Apps Script cache anymore. So I'm gzipping it now.

Answer 3

try:

=INDEX(IFERROR(REGEXEXTRACT(A1:A, 
 "^(?:https?:\/\/)?(?:ftp:\/\/)?(?:www\.)?([^\/]+)")))