tests/phpunit/CRM/Dedupe/MergerTest.php

   1 <?php
   2
   3 /**
   4  * Class CRM_Dedupe_DedupeMergerTest
   5  * @group headless
   6  */
   7 class CRM_Dedupe_MergerTest extends CiviUnitTestCase {
   8
   9   protected $_groupId;
  10   protected $_contactIds = array();
  11
  12   public function createDupeContacts() {
  13     // create a group to hold contacts, so that dupe checks don't consider any other contacts in the DB
  14     $params = array(
  15       'name'       => 'Test Dupe Merger Group',
  16       'title'      => 'Test Dupe Merger Group',
  17       'domain_id'  => 1,
  18       'is_active'  => 1,
  19       'visibility' => 'Public Pages',
  20       'version'    => 3,
  21     );
  22     // TODO: This is not an API test!!
  23     $result = civicrm_api('group', 'create', $params);
  24     $this->_groupId = $result['id'];
  25
  26     // contact data set
  27
  28     // make dupe checks based on based on following contact sets:
  29     // FIRST - LAST - EMAIL
  30     // ---------------------------------
  31     // robin  - hood - robin@example.com
  32     // robin  - hood - robin@example.com
  33     // robin  - hood - hood@example.com
  34     // robin  - dale - robin@example.com
  35     // little - dale - dale@example.com
  36     // little - dale - dale@example.com
  37     // will   - dale - dale@example.com
  38     // will   - dale - will@example.com
  39     // will   - dale - will@example.com
  40     $params = array(
  41       array(
  42         'first_name' => 'robin',
  43         'last_name' => 'hood',
  44         'email' => 'robin@example.com',
  45         'contact_type' => 'Individual',
  46       ),
  47       array(
  48         'first_name' => 'robin',
  49         'last_name' => 'hood',
  50         'email' => 'robin@example.com',
  51         'contact_type' => 'Individual',
  52       ),
  53       array(
  54         'first_name' => 'robin',
  55         'last_name' => 'hood',
  56         'email' => 'hood@example.com',
  57         'contact_type' => 'Individual',
  58       ),
  59       array(
  60         'first_name' => 'robin',
  61         'last_name' => 'dale',
  62         'email' => 'robin@example.com',
  63         'contact_type' => 'Individual',
  64       ),
  65       array(
  66         'first_name' => 'little',
  67         'last_name' => 'dale',
  68         'email' => 'dale@example.com',
  69         'contact_type' => 'Individual',
  70       ),
  71       array(
  72         'first_name' => 'little',
  73         'last_name' => 'dale',
  74         'email' => 'dale@example.com',
  75         'contact_type' => 'Individual',
  76       ),
  77       array(
  78         'first_name' => 'will',
  79         'last_name' => 'dale',
  80         'email' => 'dale@example.com',
  81         'contact_type' => 'Individual',
  82       ),
  83       array(
  84         'first_name' => 'will',
  85         'last_name' => 'dale',
  86         'email' => 'will@example.com',
  87         'contact_type' => 'Individual',
  88       ),
  89       array(
  90         'first_name' => 'will',
  91         'last_name' => 'dale',
  92         'email' => 'will@example.com',
  93         'contact_type' => 'Individual',
  94       ),
  95     );
  96
  97     $count = 1;
  98     foreach ($params as $param) {
  99       $param['version'] = 3;
 100       $contact = civicrm_api('contact', 'create', $param);
 101       $this->_contactIds[$count++] = $contact['id'];
 102
 103       $grpParams = array(
 104         'contact_id' => $contact['id'],
 105         'group_id'   => $this->_groupId,
 106         'version'    => 3,
 107       );
 108       $res = civicrm_api('group_contact', 'create', $grpParams);
 109     }
 110   }
 111
 112   public function deleteDupeContacts() {
 113     // delete all created contacts
 114     foreach ($this->_contactIds as $contactId) {
 115       Contact::delete($contactId);
 116     }
 117
 118     // delete dupe group
 119     $params = array('id' => $this->_groupId, 'version' => 3);
 120     civicrm_api('group', 'delete', $params);
 121   }
 122
 123   /**
 124    * Test the batch merge.
 125    */
 126   public function testBatchMergeSelectedDuplicates() {
 127     $this->createDupeContacts();
 128
 129     // verify that all contacts have been created separately
 130     $this->assertEquals(count($this->_contactIds), 9, 'Check for number of contacts.');
 131
 132     $dao = new CRM_Dedupe_DAO_RuleGroup();
 133     $dao->contact_type = 'Individual';
 134     $dao->name = 'IndividualSupervised';
 135     $dao->is_default = 1;
 136     $dao->find(TRUE);
 137
 138     $foundDupes = CRM_Dedupe_Finder::dupesInGroup($dao->id, $this->_groupId);
 139
 140     // -------------------------------------------------------------------------
 141     // Name and Email (reserved) Matches ( 3 pairs )
 142     // --------------------------------------------------------------------------
 143     // robin  - hood - robin@example.com
 144     // robin  - hood - robin@example.com
 145     // little - dale - dale@example.com
 146     // little - dale - dale@example.com
 147     // will   - dale - will@example.com
 148     // will   - dale - will@example.com
 149     // so 3 pairs for - first + last + mail
 150     $this->assertEquals(count($foundDupes), 3, 'Check Individual-Supervised dupe rule for dupesInGroup().');
 151
 152     // Run dedupe finder as the browser would
 153     $_SERVER['REQUEST_METHOD'] = 'GET'; //avoid invalid key error
 154     $object = new CRM_Contact_Page_DedupeFind();
 155     $object->set('gid', $this->_groupId);
 156     $object->set('rgid', $dao->id);
 157     $object->set('action', CRM_Core_Action::UPDATE);
 158     @$object->run();
 159
 160     // Retrieve pairs from prev next cache table
 161     $select = array('pn.is_selected' => 'is_selected');
 162     $cacheKeyString = "merge Individual_{$dao->id}_{$this->_groupId}";
 163     $pnDupePairs = CRM_Core_BAO_PrevNextCache::retrieve($cacheKeyString, NULL, NULL, 0, 0, $select);
 164
 165     $this->assertEquals(count($foundDupes), count($pnDupePairs), 'Check number of dupe pairs in prev next cache.');
 166
 167     // mark first two pairs as selected
 168     CRM_Core_DAO::singleValueQuery("UPDATE civicrm_prevnext_cache SET is_selected = 1 WHERE id IN ({$pnDupePairs[0]['prevnext_id']}, {$pnDupePairs[1]['prevnext_id']})");
 169
 170     $pnDupePairs = CRM_Core_BAO_PrevNextCache::retrieve($cacheKeyString, NULL, NULL, 0, 0, $select);
 171     $this->assertEquals($pnDupePairs[0]['is_selected'], 1, 'Check if first record in dupe pairs is marked as selected.');
 172     $this->assertEquals($pnDupePairs[0]['is_selected'], 1, 'Check if second record in dupe pairs is marked as selected.');
 173
 174     // batch merge selected dupes
 175     $result = CRM_Dedupe_Merger::batchMerge($dao->id, $this->_groupId, 'safe', TRUE, 5, 1);
 176     $this->assertEquals(count($result['merged']), 2, 'Check number of merged pairs.');
 177
 178     // retrieve pairs from prev next cache table
 179     $pnDupePairs = CRM_Core_BAO_PrevNextCache::retrieve($cacheKeyString, NULL, NULL, 0, 0, $select);
 180     $this->assertEquals(count($pnDupePairs), 1, 'Check number of remaining dupe pairs in prev next cache.');
 181
 182     $this->deleteDupeContacts();
 183   }
 184
 185   /**
 186    * Test the batch merge.
 187    */
 188   public function testBatchMergeAllDuplicates() {
 189     $this->createDupeContacts();
 190
 191     // verify that all contacts have been created separately
 192     $this->assertEquals(count($this->_contactIds), 9, 'Check for number of contacts.');
 193
 194     $dao = new CRM_Dedupe_DAO_RuleGroup();
 195     $dao->contact_type = 'Individual';
 196     $dao->name = 'IndividualSupervised';
 197     $dao->is_default = 1;
 198     $dao->find(TRUE);
 199
 200     $foundDupes = CRM_Dedupe_Finder::dupesInGroup($dao->id, $this->_groupId);
 201
 202     // -------------------------------------------------------------------------
 203     // Name and Email (reserved) Matches ( 3 pairs )
 204     // --------------------------------------------------------------------------
 205     // robin  - hood - robin@example.com
 206     // robin  - hood - robin@example.com
 207     // little - dale - dale@example.com
 208     // little - dale - dale@example.com
 209     // will   - dale - will@example.com
 210     // will   - dale - will@example.com
 211     // so 3 pairs for - first + last + mail
 212     $this->assertEquals(count($foundDupes), 3, 'Check Individual-Supervised dupe rule for dupesInGroup().');
 213
 214     // Run dedupe finder as the browser would
 215     $_SERVER['REQUEST_METHOD'] = 'GET'; //avoid invalid key error
 216     $object = new CRM_Contact_Page_DedupeFind();
 217     $object->set('gid', $this->_groupId);
 218     $object->set('rgid', $dao->id);
 219     $object->set('action', CRM_Core_Action::UPDATE);
 220     @$object->run();
 221
 222     // Retrieve pairs from prev next cache table
 223     $select = array('pn.is_selected' => 'is_selected');
 224     $cacheKeyString = "merge Individual_{$dao->id}_{$this->_groupId}";
 225     $pnDupePairs = CRM_Core_BAO_PrevNextCache::retrieve($cacheKeyString, NULL, NULL, 0, 0, $select);
 226
 227     $this->assertEquals(count($foundDupes), count($pnDupePairs), 'Check number of dupe pairs in prev next cache.');
 228
 229     // batch merge all dupes
 230     $result = CRM_Dedupe_Merger::batchMerge($dao->id, $this->_groupId, 'safe', TRUE, 5, 2);
 231     $this->assertEquals(count($result['merged']), 3, 'Check number of merged pairs.');
 232
 233     // retrieve pairs from prev next cache table
 234     $pnDupePairs = CRM_Core_BAO_PrevNextCache::retrieve($cacheKeyString, NULL, NULL, 0, 0, $select);
 235     $this->assertEquals(count($pnDupePairs), 0, 'Check number of remaining dupe pairs in prev next cache.');
 236
 237     $this->deleteDupeContacts();
 238   }
 239
 240   /**
 241    * The goal of this function is to test that all required tables are returned.
 242    */
 243   public function testGetCidRefs() {
 244     $this->entityCustomGroupWithSingleFieldCreate(__FUNCTION__, 'Contacts');
 245     $this->assertEquals(array_merge($this->getStaticCIDRefs(), $this->getHackedInCIDRef()), CRM_Dedupe_Merger::cidRefs());
 246     $this->assertEquals(array_merge($this->getCalculatedCIDRefs(), $this->getHackedInCIDRef()), CRM_Dedupe_Merger::cidRefs());
 247   }
 248
 249   /**
 250    * Get the list of not-really-cid-refs that are currently hacked in.
 251    *
 252    * This is hacked into getCIDs function.
 253    *
 254    * @return array
 255    */
 256   public function getHackedInCIDRef() {
 257     return array(
 258       'civicrm_entity_tag' => array(
 259         0 => 'entity_id',
 260       ),
 261     );
 262   }
 263
 264   /**
 265    * Get the list of tables that refer to the CID.
 266    *
 267    * This is a statically maintained (in this test list).
 268    *
 269    * There is also a check against an automated list but having both seems to add extra stability to me. They do
 270    * not change often.
 271    */
 272   public function getStaticCIDRefs() {
 273     return array(
 274       'civicrm_acl_cache' => array(
 275         0 => 'contact_id',
 276       ),
 277       'civicrm_acl_contact_cache' => array(
 278         0 => 'user_id',
 279         1 => 'contact_id',
 280       ),
 281       'civicrm_action_log' => array(
 282         0 => 'contact_id',
 283       ),
 284       'civicrm_activity_contact' => array(
 285         0 => 'contact_id',
 286       ),
 287       'civicrm_address' => array(
 288         0 => 'contact_id',
 289       ),
 290       'civicrm_batch' => array(
 291         0 => 'created_id',
 292         1 => 'modified_id',
 293       ),
 294       'civicrm_campaign' => array(
 295         0 => 'created_id',
 296         1 => 'last_modified_id',
 297       ),
 298       'civicrm_case_contact' => array(
 299         0 => 'contact_id',
 300       ),
 301       'civicrm_contact' => array(
 302         0 => 'primary_contact_id',
 303         1 => 'employer_id',
 304       ),
 305       'civicrm_contribution' => array(
 306         0 => 'contact_id',
 307       ),
 308       'civicrm_contribution_page' => array(
 309         0 => 'created_id',
 310       ),
 311       'civicrm_contribution_recur' => array(
 312         0 => 'contact_id',
 313       ),
 314       'civicrm_contribution_soft' => array(
 315         0 => 'contact_id',
 316       ),
 317       'civicrm_custom_group' => array(
 318         0 => 'created_id',
 319       ),
 320       'civicrm_dashboard_contact' => array(
 321         0 => 'contact_id',
 322       ),
 323       'civicrm_dedupe_exception' => array(
 324         0 => 'contact_id1',
 325         1 => 'contact_id2',
 326       ),
 327       'civicrm_domain' => array(
 328         0 => 'contact_id',
 329       ),
 330       'civicrm_email' => array(
 331         0 => 'contact_id',
 332       ),
 333       'civicrm_event' => array(
 334         0 => 'created_id',
 335       ),
 336       'civicrm_event_carts' => array(
 337         0 => 'user_id',
 338       ),
 339       'civicrm_financial_account' => array(
 340         0 => 'contact_id',
 341       ),
 342       'civicrm_financial_item' => array(
 343         0 => 'contact_id',
 344       ),
 345       'civicrm_grant' => array(
 346         0 => 'contact_id',
 347       ),
 348       'civicrm_group' => array(
 349         0 => 'created_id',
 350         1 => 'modified_id',
 351       ),
 352       'civicrm_group_contact' => array(
 353         0 => 'contact_id',
 354       ),
 355       'civicrm_group_contact_cache' => array(
 356         0 => 'contact_id',
 357       ),
 358       'civicrm_group_organization' => array(
 359         0 => 'organization_id',
 360       ),
 361       'civicrm_im' => array(
 362         0 => 'contact_id',
 363       ),
 364       'civicrm_log' => array(
 365         0 => 'modified_id',
 366       ),
 367       'civicrm_mailing' => array(
 368         0 => 'created_id',
 369         1 => 'scheduled_id',
 370         2 => 'approver_id',
 371       ),
 372       'civicrm_mailing_abtest' => array(
 373         0 => 'created_id',
 374       ),
 375       'civicrm_mailing_event_queue' => array(
 376         0 => 'contact_id',
 377       ),
 378       'civicrm_mailing_event_subscribe' => array(
 379         0 => 'contact_id',
 380       ),
 381       'civicrm_mailing_recipients' => array(
 382         0 => 'contact_id',
 383       ),
 384       'civicrm_membership' => array(
 385         0 => 'contact_id',
 386       ),
 387       'civicrm_membership_log' => array(
 388         0 => 'modified_id',
 389       ),
 390       'civicrm_membership_type' => array(
 391         0 => 'member_of_contact_id',
 392       ),
 393       'civicrm_note' => array(
 394         0 => 'contact_id',
 395       ),
 396       'civicrm_openid' => array(
 397         0 => 'contact_id',
 398       ),
 399       'civicrm_participant' => array(
 400         0 => 'contact_id',
 401         1 => 'transferred_to_contact_id', //CRM-16761
 402       ),
 403       'civicrm_payment_token' => array(
 404         0 => 'contact_id',
 405         1 => 'created_id',
 406       ),
 407       'civicrm_pcp' => array(
 408         0 => 'contact_id',
 409       ),
 410       'civicrm_phone' => array(
 411         0 => 'contact_id',
 412       ),
 413       'civicrm_pledge' => array(
 414         0 => 'contact_id',
 415       ),
 416       'civicrm_print_label' => array(
 417         0 => 'created_id',
 418       ),
 419       'civicrm_relationship' => array(
 420         0 => 'contact_id_a',
 421         1 => 'contact_id_b',
 422       ),
 423       'civicrm_report_instance' => array(
 424         0 => 'created_id',
 425         1 => 'owner_id',
 426       ),
 427       'civicrm_setting' => array(
 428         0 => 'contact_id',
 429         1 => 'created_id',
 430       ),
 431       'civicrm_subscription_history' => array(
 432         0 => 'contact_id',
 433       ),
 434       'civicrm_survey' => array(
 435         0 => 'created_id',
 436         1 => 'last_modified_id',
 437       ),
 438       'civicrm_tag' => array(
 439         0 => 'created_id',
 440       ),
 441       'civicrm_uf_group' => array(
 442         0 => 'created_id',
 443       ),
 444       'civicrm_uf_match' => array(
 445         0 => 'contact_id',
 446       ),
 447       'civicrm_value_testgetcidref_1' => array(
 448         0 => 'entity_id',
 449       ),
 450       'civicrm_website' => array(
 451         0 => 'contact_id',
 452       ),
 453     );
 454   }
 455
 456   /**
 457    * Get a list of CIDs that is calculated off the schema.
 458    *
 459    * Note this is an expensive and table locking query. Should be safe in tests though.
 460    */
 461   public function getCalculatedCIDRefs() {
 462     $cidRefs = array();
 463     $sql = "
 464 SELECT
 465     table_name,
 466     column_name
 467 FROM information_schema.key_column_usage
 468 WHERE
 469     referenced_table_schema = database() AND
 470     referenced_table_name = 'civicrm_contact' AND
 471     referenced_column_name = 'id';
 472       ";
 473     $dao = CRM_Core_DAO::executeQuery($sql);
 474     while ($dao->fetch()) {
 475       $cidRefs[$dao->table_name][] = $dao->column_name;
 476     }
 477     // Do specific re-ordering changes to make this the same as the ref validated one.
 478     // The above query orders by FK alphabetically.
 479     // There might be cleverer ways to do this but it shouldn't change much.
 480     $cidRefs['civicrm_contact'][0] = 'primary_contact_id';
 481     $cidRefs['civicrm_contact'][1] = 'employer_id';
 482     $cidRefs['civicrm_acl_contact_cache'][0] = 'user_id';
 483     $cidRefs['civicrm_acl_contact_cache'][1] = 'contact_id';
 484     $cidRefs['civicrm_mailing'][0] = 'created_id';
 485     $cidRefs['civicrm_mailing'][1] = 'scheduled_id';
 486     $cidRefs['civicrm_mailing'][2] = 'approver_id';
 487     return $cidRefs;
 488   }
 489
 490 }