feat: add ccTLD-specific date formats, noisy-token parsing, and .gg/.je/.br/.ua normalization, closes #20 and #25

- Add `toISOFromTokens` to extract ISO dates from noisy values (e.g. ".ua" `0-UANIC 20111004161638`, `OK-UNTIL 20261004161638`)
- Parse compact `YYYYMMDDHHMMSS` (.ua) and `YYYYMMDD` (.br, including ticket suffix like `20260319 #31066859`) date formats
- Parse ordinal/prose dates used by .gg/.je (`28th December 2018 at 05:54:43.861`)
- Teach `parseKeyValueLines` to collect indented continuation lines for header-style blocks (.gg/.je) without mistaking URLs or times as key separators
- Extract .gg/.je expiry from `Relevant dates:` sentences and split inline registrar URL (`epag (http://www.epag.de)`)
- Map .ua `status: OK-UNTIL <timestamp>` to `expirationDate` when no explicit expiry field is present
- Treat RDAP `pending release` / `release process: waiting` statuses as `isRegistered: false` (previously always `true`)
- Mark WHOIS `.br` `release process: waiting` as available

src/whois/normalize.test.ts

@@ -207,3 +207,76 @@ test("isAvailableByWhois correctly identifies availability patterns", () => {
   const rec = normalizeWhois("example.com", "com", text, "whois.example.com");
   expect(rec.isRegistered).toBe(false);
 });
+
+test("WHOIS .gov.ua extracts dates from noisy values and OK-UNTIL", () => {
+  const text = `
+domain: agrex.gov.ua
+status: OK-UNTIL 20261004161638
+created: 0-UANIC 20111004161638
+modified: UARR149-UANIC 20251004050127
+`;
+  const rec = normalizeWhois("agrex.gov.ua", "gov.ua", text, "whois.gov.ua");
+  expect(rec.creationDate).toBe("2011-10-04T16:16:38Z");
+  expect(rec.updatedDate).toBe("2025-10-04T05:01:27Z");
+  expect(rec.expirationDate).toBe("2026-10-04T16:16:38Z");
+});
+
+test("WHOIS .br 'release process: waiting' is not registered", () => {
+  const text = `
+domain:      iba.com.br
+release process:  waiting
+`;
+  expect(isAvailableByWhois(text)).toBe(true);
+  expect(normalizeWhois("iba.com.br", "com.br", text, "whois.registro.br").isRegistered).toBe(
+    false,
+  );
+});
+
+test("WHOIS .br published domain is registered with compact dates", () => {
+  const text = `
+domain:      iba.com.br
+nserver:     ns11.cloudns.net
+created:     20260319 #31066859
+changed:     20260414
+expires:     20270319
+status:      published
+`;
+  const rec = normalizeWhois("iba.com.br", "com.br", text, "whois.registro.br");
+  expect(rec.isRegistered).toBe(true);
+  expect(rec.creationDate).toBe("2026-03-19T00:00:00Z");
+  expect(rec.updatedDate).toBe("2026-04-14T00:00:00Z");
+  expect(rec.expirationDate).toBe("2027-03-19T00:00:00Z");
+});
+
+test("WHOIS .gg header-style blocks with colons in values", () => {
+  const text = `
+Domain:
+     t3.gg
+
+Domain Status:
+     Active
+
+Registrant:
+     T3 Tools Inc
+
+Registrar:
+     epag (http://www.epag.de)
+
+Relevant dates:
+     Registered on 28th December 2018 at 05:54:43.861
+     Registry fee due on 28th December each year
+
+Registration status:
+     Registered until cancelled
+
+Name servers:
+     ns1.vercel-dns.com
+     ns2.vercel-dns.com
+`;
+  const rec = normalizeWhois("t3.gg", "gg", text, "whois.gg");
+  expect(rec.isRegistered).toBe(true);
+  expect(rec.creationDate).toBe("2018-12-28T05:54:43Z");
+  expect(rec.expirationDate).toBeUndefined();
+  expect(rec.registrar).toEqual({ name: "epag", url: "http://www.epag.de" });
+  expect(rec.nameservers?.map((n) => n.host)).toEqual(["ns1.vercel-dns.com", "ns2.vercel-dns.com"]);
+});

src/whois/normalize.ts

@@ -1,4 +1,4 @@
-import { toISO } from "../lib/dates";
+import { toISOFromTokens } from "../lib/dates";
 import { isPrivacyName } from "../lib/privacy";
 import { parseKeyValueLines, uniq } from "../lib/text";
 import type { Contact, DomainRecord, Nameserver, RegistrarInfo } from "../types";
@@ -31,6 +31,7 @@ const WHOIS_AVAILABLE_PATTERNS: RegExp[] = [
   /\bno se encuentra registrado\b/i, // Spanish: "not found registered"
   /\bnicht gefunden\b/i, // German: "not found"
   /\bpending release\b/i, // often signals not registered/being deleted
+  /\brelease process:\s*waiting\b/i, // .br: expired, awaiting release
 ];
 
 /**
@@ -55,22 +56,26 @@ export function normalizeWhois(
   const map = parseKeyValueLines(whoisText);
 
   // Date extraction across common synonyms
-  const creationDate = anyValue(map, [
-    "creation date",
-    "created on",
-    "created",
-    "registered on",
-    "registered",
-    "registration date",
-    "domain registration date",
-    "domain create date",
-    "domain name commencement date",
-    "registration time", // .cn
-    "domain record activated", // .edu
-    "domain registered",
-    "registered date", // .co.jp
-    "assigned", // .il
-  ]);
+  // .gg/.je list dates as sentences under "Relevant dates:", e.g. "Registered on 28th December 2018 at 05:54:43.861"
+  const relevantDate = (label: RegExp) =>
+    map["relevant dates"]?.find((l) => label.test(l))?.replace(label, "");
+  const creationDate =
+    anyValue(map, [
+      "creation date",
+      "created on",
+      "created",
+      "registered on",
+      "registered",
+      "registration date",
+      "domain registration date",
+      "domain create date",
+      "domain name commencement date",
+      "registration time", // .cn
+      "domain record activated", // .edu
+      "domain registered",
+      "registered date", // .co.jp
+      "assigned", // .il
+    ]) ?? relevantDate(/^registered on\s+/i);
   const updatedDate = anyValue(map, [
     "updated date",
     "updated",
@@ -128,6 +133,11 @@ export function normalizeWhois(
     ]);
     const abuseEmail = anyValue(map, ["registrar abuse contact email", "abuse contact email"]);
     const abusePhone = anyValue(map, ["registrar abuse contact phone", "abuse contact phone"]);
+    // .gg/.je: "epag (http://www.epag.de)"
+    const inlineUrl = name?.match(/^(.*?)\s*\((https?:\/\/[^)\s]+)\)$/);
+    if (inlineUrl?.[1] && !url) {
+      return { name: inlineUrl[1], url: inlineUrl[2] };
+    }
     if (!name && !ianaId && !url && !abuseEmail && !abusePhone) return undefined;
     return {
       name: name || undefined,
@@ -156,6 +166,9 @@ export function normalizeWhois(
         .filter((s): s is { status: string; raw: string } => s !== null)
     : undefined;
 
+  // Some registries (.ua) publish expiry only as a status, e.g. "OK-UNTIL 20261004161638"
+  const okUntil = statuses?.find((s) => /^ok-until$/i.test(s.status))?.raw;
+
   // Nameservers: also appear as "nserver" on some ccTLDs (.de, .ru) and as "name server"
   const nsLines: string[] = [
     ...(map["name server"] || []),
@@ -228,9 +241,9 @@ export function normalizeWhois(
     registrar,
     reseller: anyValue(map, ["reseller"]) || undefined,
     statuses,
-    creationDate: toISO(creationDate || undefined),
-    updatedDate: toISO(updatedDate || undefined),
-    expirationDate: toISO(expirationDate || undefined),
+    creationDate: toISOFromTokens(creationDate),
+    updatedDate: toISOFromTokens(updatedDate),
+    expirationDate: toISOFromTokens(expirationDate) ?? toISOFromTokens(okUntil),
     deletionDate: undefined,
     transferLock,
     dnssec,