add reversible prospect deduplication

apps/api/tests/test_phase6.py

@@ -0,0 +1,30 @@
+import unittest
+from app.domain import normalize_phone, normalize_location, normalize_business, match_businesses
+
+class Phase6DomainTests(unittest.TestCase):
+    def test_south_african_phone_formats_share_canonical_value(self):
+        self.assertEqual(normalize_phone("082 555 1234"), "+27825551234")
+        self.assertEqual(normalize_phone("0027 82 555 1234"), "+27825551234")
+        self.assertEqual(normalize_phone("+27 (82) 555-1234"), "+27825551234")
+
+    def test_unknown_international_phone_is_not_rewritten(self):
+        self.assertEqual(normalize_phone("+44 (20) 1234 5678"), "+442012345678")
+        self.assertEqual(normalize_phone("555-1234"), "5551234")
+
+    def test_location_normalization_and_business_fields(self):
+        self.assertEqual(normalize_location({"province": " Gauteng ", "city": " Johannesburg ", "suburb": " Sandton "}), {"province": "gauteng", "city": "johannesburg", "suburb": "sandton"})
+        b = normalize_business({"name":" Acme ", "province":" Gauteng ", "city":" Johannesburg ", "suburb":" Sandton "})
+        self.assertEqual((b["province"], b["city"], b["suburb"]), ("gauteng", "johannesburg", "sandton"))
+
+    def test_matching_has_deterministic_confidence_and_reasons(self):
+        a = {"name":"Acme Consulting", "email":"hello@acme.co.za", "city":"Johannesburg"}
+        b = {"name":"Acme Consultng", "email":"hello@acme.co.za", "city":"Johannesburg"}
+        first = match_businesses(a, [dict(b, id=2), {"id":3,"name":"Unrelated Shop"}], threshold=0.5)
+        second = match_businesses(a, [dict(b, id=2), {"id":3,"name":"Unrelated Shop"}], threshold=0.5)
+        self.assertEqual(first, second)
+        self.assertEqual(first[0]["id"], 2)
+        self.assertGreaterEqual(first[0]["confidence"], 0.5)
+        self.assertTrue(first[0]["reasons"])
+        self.assertEqual([x["id"] for x in match_businesses(a, [{"id":3,"name":"Unrelated Shop"}], threshold=0.8)], [])
+
+if __name__ == "__main__": unittest.main()

apps/api/tests/test_phase6_api.py

@@ -0,0 +1,35 @@
+import json, os, sqlite3, threading, unittest
+from http.client import HTTPConnection
+from tempfile import TemporaryDirectory
+from app.main import create_server
+
+class Phase6ApiTests(unittest.TestCase):
+    def setUp(self):
+        self.tmp=TemporaryDirectory(); os.environ['BOOTSTRAP_ADMIN_EMAIL']='owner@phase6.test'; os.environ['BOOTSTRAP_ADMIN_PASSWORD']='password'
+        self.server=create_server('127.0.0.1',0,self.tmp.name+'/db.sqlite'); self.thread=threading.Thread(target=self.server.serve_forever,daemon=True); self.thread.start(); self.conn=HTTPConnection('127.0.0.1',self.server.server_port,timeout=3); self.cookie=None
+        self.request('POST','/api/v1/auth/login',{'email':'owner@phase6.test','password':'password'})
+    def tearDown(self): self.server.shutdown(); self.server.server_close(); self.thread.join(2); self.tmp.cleanup()
+    def request(self,method,path,payload=None):
+        body=json.dumps(payload).encode() if payload is not None else None; headers={'Content-Type':'application/json'} if body else {}
+        if self.cookie: headers['Cookie']=self.cookie
+        self.conn.request(method,path,body,headers); r=self.conn.getresponse(); c=r.getheader('Set-Cookie');
+        if c: self.cookie=c.split(';',1)[0]
+        return r.status,json.loads(r.read() or b'{}')
+    def test_merge_preserves_children_and_reverse_restores_ownership(self):
+        _, source=self.request('POST','/api/v1/businesses',{'name':'Acme Consulting','phone':'082 555 1234'})
+        _, target=self.request('POST','/api/v1/businesses',{'name':'Acme Consulting HQ','city':'Johannesburg'})
+        self.assertEqual(self.request('POST',f"/api/v1/businesses/{source['id']}/notes",{'body':'evidence'})[0],201)
+        self.assertEqual(self.request('POST',f"/api/v1/businesses/{source['id']}/evidence",{'kind':'source','claim':'claim'})[0],201)
+        status, merged=self.request('POST',f"/api/v1/businesses/{source['id']}/merge",{'target_business_id':target['id']})
+        self.assertEqual(status,200); self.assertEqual(merged['status'],'merged')
+        status, detail=self.request('GET',f"/api/v1/businesses/{target['id']}"); self.assertEqual(status,200); self.assertEqual(len(detail['notes']),1); self.assertEqual(len(detail['evidence']),1)
+        status, source_detail=self.request('GET',f"/api/v1/businesses/{source['id']}"); self.assertEqual(status,200); self.assertEqual(source_detail['merge_status'],'merged')
+        status, reversed_=self.request('POST',f"/api/v1/merge-history/{merged['merge_history_id']}/reverse",{}); self.assertEqual(status,200); self.assertEqual(reversed_['status'],'reversed')
+        _, restored=self.request('GET',f"/api/v1/businesses/{source['id']}"); self.assertEqual(len(restored['notes']),1); self.assertEqual(len(restored['evidence']),1)
+    def test_matches_endpoint_is_thresholded_and_history_is_tenant_scoped(self):
+        _, a=self.request('POST','/api/v1/businesses',{'name':'Bright Co','email':'hello@bright.test'})
+        _, b=self.request('POST','/api/v1/businesses',{'name':'Bright Company','website':'https://bright.test'})
+        status, matches=self.request('GET',f"/api/v1/businesses/{a['id']}/matches?threshold=0.4"); self.assertEqual(status,200); self.assertEqual(matches['items'][0]['id'],b['id']); self.assertIn('similar_name',matches['items'][0]['reasons'])
+        self.assertEqual(self.request('GET','/api/v1/merge-history')[0],200)
+
+if __name__=='__main__': unittest.main()