Files
lead-scraper/lib/utils/domains.ts
Timo Uttenweiler facf8c9f69 Initial commit: LeadFlow lead generation platform
Full-stack Next.js 16 app with three scraping pipelines:
- AirScale CSV → Anymailfinder Bulk Decision Maker search
- LinkedIn Sales Navigator → Vayne → Anymailfinder email enrichment
- Apify Google SERP → domain extraction → Anymailfinder bulk enrichment

Includes Docker multi-stage build + docker-compose for Coolify deployment.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-17 11:21:11 +01:00

35 lines
1.1 KiB
TypeScript

export function cleanDomain(raw: string): string {
if (!raw) return "";
let domain = raw.trim().toLowerCase();
// Remove protocol
domain = domain.replace(/^https?:\/\//i, "");
// Remove www.
domain = domain.replace(/^www\./i, "");
// Remove paths, query strings, fragments
domain = domain.split("/")[0];
domain = domain.split("?")[0];
domain = domain.split("#")[0];
// Remove trailing dots
domain = domain.replace(/\.$/, "");
return domain;
}
export function extractDomainFromUrl(url: string): string {
try {
const parsed = new URL(url.startsWith("http") ? url : `https://${url}`);
return parsed.hostname.replace(/^www\./i, "").toLowerCase();
} catch {
return cleanDomain(url);
}
}
const SOCIAL_DIRS = [
"linkedin.com", "facebook.com", "instagram.com", "twitter.com",
"x.com", "yelp.com", "google.com", "maps.google.com", "wikipedia.org",
"xing.com", "youtube.com", "tiktok.com", "pinterest.com",
];
export function isSocialOrDirectory(domain: string): boolean {
return SOCIAL_DIRS.some(d => domain === d || domain.endsWith(`.${d}`));
}