// author: simon-pierre boucher <contact@spboucher.ai>
import { err, ok, type Result } from "./result.js";
import { tendrilError } from "./errors.js";

const DEFAULT_PORTS: Readonly<Record<string, string>> = {
  "http:": "80",
  "https:": "443",
};

const TRACKING_PARAMS = new Set(["fbclid", "gclid", "gclsrc", "ref", "mc_cid", "mc_eid", "_ga", "_gl"]);

function isTrackingParam(key: string): boolean {
  const lower = key.toLowerCase();
  return lower.startsWith("utm_") || TRACKING_PARAMS.has(lower);
}

/**
 * Canonicalize a URL for dedup and cache keys (§10 rule 1, §10.1). The result is
 * intended for comparison only — keep the original string for output. Returns
 * ERR_INVALID_URL for anything that is not an absolute http(s) URL.
 */
export function normalizeUrl(raw: string, base?: string): Result<string> {
  let u: URL;
  try {
    u = base !== undefined ? new URL(raw, base) : new URL(raw);
  } catch {
    return err(tendrilError("ERR_INVALID_URL", { details: { url: raw } }));
  }

  if (u.protocol !== "http:" && u.protocol !== "https:") {
    return err(tendrilError("ERR_INVALID_URL", { details: { url: raw, scheme: u.protocol } }));
  }

  u.hostname = u.hostname.toLowerCase();

  if (u.port !== "" && DEFAULT_PORTS[u.protocol] === u.port) {
    u.port = "";
  }

  const kept: Array<[string, string]> = [];
  for (const [key, value] of u.searchParams.entries()) {
    if (!isTrackingParam(key)) kept.push([key, value]);
  }
  kept.sort((a, b) => (a[0] === b[0] ? (a[1] < b[1] ? -1 : a[1] > b[1] ? 1 : 0) : a[0] < b[0] ? -1 : 1));
  u.search = "";
  for (const [key, value] of kept) u.searchParams.append(key, value);

  const hashbang = u.hash.startsWith("#!");
  if (!hashbang) u.hash = "";

  if (u.pathname.length > 1 && u.pathname.endsWith("/")) {
    u.pathname = u.pathname.replace(/\/+$/, "");
    if (u.pathname === "") u.pathname = "/";
  }

  return ok(u.toString());
}

export interface ParsedUrl {
  readonly url: URL;
  readonly host: string;
}

export function parseHost(raw: string): Result<ParsedUrl> {
  try {
    const url = new URL(raw);
    return ok({ url, host: url.hostname.toLowerCase() });
  } catch {
    return err(tendrilError("ERR_INVALID_URL", { details: { url: raw } }));
  }
}