From ea677b7aba9149cdbf6a433f7e925151402a16cc Mon Sep 17 00:00:00 2001
From: lorentz <lorentz@wulfconsulting.com>
Date: Wed, 15 Jul 2026 19:19:44 -0400
Subject: [PATCH] test(18-01): add campaign-grouping tier-key helpers + test
 scaffold

- normalizeSubject strips repeated Re:/Fwd:/Fw: prefixes case-insensitively, lowercases, trims (D-03)
- extractUrlDomain returns hostname or null (never throws) for malformed URLs (Pitfall 4)
- test file mocks ./postgres-client before import, mirroring phishing-eml-service.test.ts's discipline
---
 .../campaign-grouping-service.test.ts         | 50 ++++++++++++++++++
 lib/services/campaign-grouping-service.ts     | 52 +++++++++++++++++++
 2 files changed, 102 insertions(+)
 create mode 100644 lib/services/campaign-grouping-service.test.ts
 create mode 100644 lib/services/campaign-grouping-service.ts

diff --git a/lib/services/campaign-grouping-service.test.ts b/lib/services/campaign-grouping-service.test.ts
new file mode 100644
index 0000000..bc92fbc
--- /dev/null
+++ b/lib/services/campaign-grouping-service.test.ts
@@ -0,0 +1,50 @@
+import { describe, it, expect, vi } from 'vitest';
+
+// Mock postgresClient BEFORE importing the module under test.
+const queryMock = vi.fn();
+const transactionMock = vi.fn();
+vi.mock('./postgres-client', () => ({
+  postgresClient: {
+    query: (...args: unknown[]) => queryMock(...args),
+    transaction: (...args: unknown[]) => transactionMock(...args),
+  },
+}));
+
+// eslint-disable-next-line import/first -- imported after vi.mock hoisting
+import { normalizeSubject, extractUrlDomain } from './campaign-grouping-service';
+
+describe('normalizeSubject', () => {
+  it('strips a single Re: prefix, lowercases, trims', () => {
+    expect(normalizeSubject('Re: Your Invoice ')).toBe('your invoice');
+  });
+
+  it('strips repeated Re:/Fwd:/Fw: prefixes case-insensitively', () => {
+    expect(normalizeSubject('FW: Re: fwd: Urgent Payment')).toBe('urgent payment');
+  });
+
+  it('handles null', () => {
+    expect(normalizeSubject(null)).toBe('');
+  });
+
+  it('handles empty string', () => {
+    expect(normalizeSubject('')).toBe('');
+  });
+
+  it('lowercases and trims a subject with no prefix', () => {
+    expect(normalizeSubject('  Urgent Payment  ')).toBe('urgent payment');
+  });
+});
+
+describe('extractUrlDomain', () => {
+  it('extracts hostname from a full URL', () => {
+    expect(extractUrlDomain('https://evil.example.com/path?x=1')).toBe('evil.example.com');
+  });
+
+  it('returns null for a malformed URL instead of throwing', () => {
+    expect(extractUrlDomain('not-a-url')).toBeNull();
+  });
+
+  it('extracts hostname regardless of scheme', () => {
+    expect(extractUrlDomain('http://another.example.net')).toBe('another.example.net');
+  });
+});
diff --git a/lib/services/campaign-grouping-service.ts b/lib/services/campaign-grouping-service.ts
new file mode 100644
index 0000000..399d769
--- /dev/null
+++ b/lib/services/campaign-grouping-service.ts
@@ -0,0 +1,52 @@
+/**
+ * Campaign Grouping Service
+ *
+ * Shared grouping core called by all 3 trigger sites: the webhook path
+ * (`webhook-service.ts`'s `triggerPhishingDetection()`), the cron sweep
+ * (`phishing-sweep-service.ts`), and the on-demand
+ * `POST /api/phishing/tickets/{id}/analyze` route. One deterministic,
+ * testable function computes a tiered campaign key (CAMP-01) and
+ * finds-or-creates a `campaigns` row (CAMP-02) — no duplicated matching
+ * logic between callers, mirroring `phishing-detector.ts`'s shared-core
+ * architecture.
+ *
+ * D-07 limitation (load-bearing, stated explicitly): `parseAndStoreMessage`
+ * (the only writer of `messages`/`indicators` rows — Phase 16) is not wired
+ * into the automatic webhook/cron path this phase. That means the automatic
+ * path only ever has `reports`/`contacts` data available, so Tier 1
+ * (Message-ID) and Tier 2 (attachment-hash/URL-domain) can only ever match
+ * for a report that has already been through an explicit `/analyze` call at
+ * least once. Until then, automatic grouping effectively only reaches
+ * Tier 3 (sender + normalized subject + client + 24h window).
+ */
+
+import { postgresClient } from './postgres-client';
+
+// =============================================================================
+// Pure logic — tier-key helpers (mirrors matchesPhishingPatterns / computePhishingContentHash)
+// =============================================================================
+
+/**
+ * D-03: strip leading Re:/Fwd:/Fw: (repeated, case-insensitive), lowercase, trim.
+ */
+export function normalizeSubject(subject: string | null): string {
+  let s = (subject ?? '').trim();
+  const prefixRe = /^(re|fwd|fw):\s*/i;
+  while (prefixRe.test(s)) {
+    s = s.replace(prefixRe, '').trim();
+  }
+  return s.toLowerCase();
+}
+
+/**
+ * Pitfall 4: indicators.value for indicator_type='url' is a bare URL string,
+ * not a domain — extract at read time, guarded (malformed/relative URLs are
+ * possible in real-world phishing emails).
+ */
+export function extractUrlDomain(url: string): string | null {
+  try {
+    return new URL(url).hostname || null;
+  } catch {
+    return null;
+  }
+}