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', $grpParams);
 109     }
 110   }
 111
 112   /**
 113    * Delete all created contacts.
 114    */
 115   public function deleteDupeContacts() {
 116     foreach ($this->_contactIds as $contactId) {
 117       $this->contactDelete($contactId);
 118     }
 119
 120     // delete dupe group
 121     $params = array('id' => $this->_groupId, 'version' => 3);
 122     civicrm_api('group', 'delete', $params);
 123   }
 124
 125   /**
 126    * Test the batch merge.
 127    */
 128   public function testBatchMergeSelectedDuplicates() {
 129     $this->createDupeContacts();
 130
 131     // verify that all contacts have been created separately
 132     $this->assertEquals(count($this->_contactIds), 9, 'Check for number of contacts.');
 133
 134     $dao = new CRM_Dedupe_DAO_RuleGroup();
 135     $dao->contact_type = 'Individual';
 136     $dao->name = 'IndividualSupervised';
 137     $dao->is_default = 1;
 138     $dao->find(TRUE);
 139
 140     $foundDupes = CRM_Dedupe_Finder::dupesInGroup($dao->id, $this->_groupId);
 141
 142     // -------------------------------------------------------------------------
 143     // Name and Email (reserved) Matches ( 3 pairs )
 144     // --------------------------------------------------------------------------
 145     // robin  - hood - robin@example.com
 146     // robin  - hood - robin@example.com
 147     // little - dale - dale@example.com
 148     // little - dale - dale@example.com
 149     // will   - dale - will@example.com
 150     // will   - dale - will@example.com
 151     // so 3 pairs for - first + last + mail
 152     $this->assertEquals(count($foundDupes), 3, 'Check Individual-Supervised dupe rule for dupesInGroup().');
 153
 154     // Run dedupe finder as the browser would
 155     $_SERVER['REQUEST_METHOD'] = 'GET'; //avoid invalid key error
 156     $object = new CRM_Contact_Page_DedupeFind();
 157     $object->set('gid', $this->_groupId);
 158     $object->set('rgid', $dao->id);
 159     $object->set('action', CRM_Core_Action::UPDATE);
 160     $object->setEmbedded(TRUE);
 161     @$object->run();
 162
 163     // Retrieve pairs from prev next cache table
 164     $select = array('pn.is_selected' => 'is_selected');
 165     $cacheKeyString = "merge Individual_{$dao->id}_{$this->_groupId}";
 166     $pnDupePairs = CRM_Core_BAO_PrevNextCache::retrieve($cacheKeyString, NULL, NULL, 0, 0, $select);
 167
 168     $this->assertEquals(count($foundDupes), count($pnDupePairs), 'Check number of dupe pairs in prev next cache.');
 169
 170     // mark first two pairs as selected
 171     CRM_Core_DAO::singleValueQuery("UPDATE civicrm_prevnext_cache SET is_selected = 1 WHERE id IN ({$pnDupePairs[0]['prevnext_id']}, {$pnDupePairs[1]['prevnext_id']})");
 172
 173     $pnDupePairs = CRM_Core_BAO_PrevNextCache::retrieve($cacheKeyString, NULL, NULL, 0, 0, $select);
 174     $this->assertEquals($pnDupePairs[0]['is_selected'], 1, 'Check if first record in dupe pairs is marked as selected.');
 175     $this->assertEquals($pnDupePairs[0]['is_selected'], 1, 'Check if second record in dupe pairs is marked as selected.');
 176
 177     // batch merge selected dupes
 178     $result = CRM_Dedupe_Merger::batchMerge($dao->id, $this->_groupId, 'safe', TRUE, 5, 1);
 179     $this->assertEquals(count($result['merged']), 2, 'Check number of merged pairs.');
 180
 181     // retrieve pairs from prev next cache table
 182     $pnDupePairs = CRM_Core_BAO_PrevNextCache::retrieve($cacheKeyString, NULL, NULL, 0, 0, $select);
 183     $this->assertEquals(count($pnDupePairs), 1, 'Check number of remaining dupe pairs in prev next cache.');
 184
 185     $this->deleteDupeContacts();
 186   }
 187
 188   /**
 189    * Test the batch merge.
 190    */
 191   public function testBatchMergeAllDuplicates() {
 192     $this->createDupeContacts();
 193
 194     // verify that all contacts have been created separately
 195     $this->assertEquals(count($this->_contactIds), 9, 'Check for number of contacts.');
 196
 197     $dao = new CRM_Dedupe_DAO_RuleGroup();
 198     $dao->contact_type = 'Individual';
 199     $dao->name = 'IndividualSupervised';
 200     $dao->is_default = 1;
 201     $dao->find(TRUE);
 202
 203     $foundDupes = CRM_Dedupe_Finder::dupesInGroup($dao->id, $this->_groupId);
 204
 205     // -------------------------------------------------------------------------
 206     // Name and Email (reserved) Matches ( 3 pairs )
 207     // --------------------------------------------------------------------------
 208     // robin  - hood - robin@example.com
 209     // robin  - hood - robin@example.com
 210     // little - dale - dale@example.com
 211     // little - dale - dale@example.com
 212     // will   - dale - will@example.com
 213     // will   - dale - will@example.com
 214     // so 3 pairs for - first + last + mail
 215     $this->assertEquals(count($foundDupes), 3, 'Check Individual-Supervised dupe rule for dupesInGroup().');
 216
 217     // Run dedupe finder as the browser would
 218     $_SERVER['REQUEST_METHOD'] = 'GET'; //avoid invalid key error
 219     $object = new CRM_Contact_Page_DedupeFind();
 220     $object->set('gid', $this->_groupId);
 221     $object->set('rgid', $dao->id);
 222     $object->set('action', CRM_Core_Action::UPDATE);
 223     $object->setEmbedded(TRUE);
 224     @$object->run();
 225
 226     // Retrieve pairs from prev next cache table
 227     $select = array('pn.is_selected' => 'is_selected');
 228     $cacheKeyString = "merge Individual_{$dao->id}_{$this->_groupId}";
 229     $pnDupePairs = CRM_Core_BAO_PrevNextCache::retrieve($cacheKeyString, NULL, NULL, 0, 0, $select);
 230
 231     $this->assertEquals(count($foundDupes), count($pnDupePairs), 'Check number of dupe pairs in prev next cache.');
 232
 233     // batch merge all dupes
 234     $result = CRM_Dedupe_Merger::batchMerge($dao->id, $this->_groupId, 'safe', TRUE, 5, 2);
 235     $this->assertEquals(count($result['merged']), 3, 'Check number of merged pairs.');
 236
 237     // retrieve pairs from prev next cache table
 238     $pnDupePairs = CRM_Core_BAO_PrevNextCache::retrieve($cacheKeyString, NULL, NULL, 0, 0, $select);
 239     $this->assertEquals(count($pnDupePairs), 0, 'Check number of remaining dupe pairs in prev next cache.');
 240
 241     $this->deleteDupeContacts();
 242   }
 243
 244   /**
 245    * The goal of this function is to test that all required tables are returned.
 246    */
 247   public function testGetCidRefs() {
 248     $this->entityCustomGroupWithSingleFieldCreate(__FUNCTION__, 'Contacts');
 249     $this->assertEquals(array_merge($this->getStaticCIDRefs(), $this->getHackedInCIDRef()), CRM_Dedupe_Merger::cidRefs());
 250     $this->assertEquals(array_merge($this->getCalculatedCIDRefs(), $this->getHackedInCIDRef()), CRM_Dedupe_Merger::cidRefs());
 251   }
 252
 253   /**
 254    * Get the list of not-really-cid-refs that are currently hacked in.
 255    *
 256    * This is hacked into getCIDs function.
 257    *
 258    * @return array
 259    */
 260   public function getHackedInCIDRef() {
 261     return array(
 262       'civicrm_entity_tag' => array(
 263         0 => 'entity_id',
 264       ),
 265     );
 266   }
 267
 268   /**
 269    * Get the list of tables that refer to the CID.
 270    *
 271    * This is a statically maintained (in this test list).
 272    *
 273    * There is also a check against an automated list but having both seems to add extra stability to me. They do
 274    * not change often.
 275    */
 276   public function getStaticCIDRefs() {
 277     return array(
 278       'civicrm_acl_cache' => array(
 279         0 => 'contact_id',
 280       ),
 281       'civicrm_acl_contact_cache' => array(
 282         0 => 'user_id',
 283         1 => 'contact_id',
 284       ),
 285       'civicrm_action_log' => array(
 286         0 => 'contact_id',
 287       ),
 288       'civicrm_activity_contact' => array(
 289         0 => 'contact_id',
 290       ),
 291       'civicrm_address' => array(
 292         0 => 'contact_id',
 293       ),
 294       'civicrm_batch' => array(
 295         0 => 'created_id',
 296         1 => 'modified_id',
 297       ),
 298       'civicrm_campaign' => array(
 299         0 => 'created_id',
 300         1 => 'last_modified_id',
 301       ),
 302       'civicrm_case_contact' => array(
 303         0 => 'contact_id',
 304       ),
 305       'civicrm_contact' => array(
 306         0 => 'primary_contact_id',
 307         1 => 'employer_id',
 308       ),
 309       'civicrm_contribution' => array(
 310         0 => 'contact_id',
 311       ),
 312       'civicrm_contribution_page' => array(
 313         0 => 'created_id',
 314       ),
 315       'civicrm_contribution_recur' => array(
 316         0 => 'contact_id',
 317       ),
 318       'civicrm_contribution_soft' => array(
 319         0 => 'contact_id',
 320       ),
 321       'civicrm_custom_group' => array(
 322         0 => 'created_id',
 323       ),
 324       'civicrm_dashboard_contact' => array(
 325         0 => 'contact_id',
 326       ),
 327       'civicrm_dedupe_exception' => array(
 328         0 => 'contact_id1',
 329         1 => 'contact_id2',
 330       ),
 331       'civicrm_domain' => array(
 332         0 => 'contact_id',
 333       ),
 334       'civicrm_email' => array(
 335         0 => 'contact_id',
 336       ),
 337       'civicrm_event' => array(
 338         0 => 'created_id',
 339       ),
 340       'civicrm_event_carts' => array(
 341         0 => 'user_id',
 342       ),
 343       'civicrm_financial_account' => array(
 344         0 => 'contact_id',
 345       ),
 346       'civicrm_financial_item' => array(
 347         0 => 'contact_id',
 348       ),
 349       'civicrm_grant' => array(
 350         0 => 'contact_id',
 351       ),
 352       'civicrm_group' => array(
 353         0 => 'created_id',
 354         1 => 'modified_id',
 355       ),
 356       'civicrm_group_contact' => array(
 357         0 => 'contact_id',
 358       ),
 359       'civicrm_group_contact_cache' => array(
 360         0 => 'contact_id',
 361       ),
 362       'civicrm_group_organization' => array(
 363         0 => 'organization_id',
 364       ),
 365       'civicrm_im' => array(
 366         0 => 'contact_id',
 367       ),
 368       'civicrm_log' => array(
 369         0 => 'modified_id',
 370       ),
 371       'civicrm_mailing' => array(
 372         0 => 'created_id',
 373         1 => 'scheduled_id',
 374         2 => 'approver_id',
 375       ),
 376       'civicrm_mailing_abtest' => array(
 377         0 => 'created_id',
 378       ),
 379       'civicrm_mailing_event_queue' => array(
 380         0 => 'contact_id',
 381       ),
 382       'civicrm_mailing_event_subscribe' => array(
 383         0 => 'contact_id',
 384       ),
 385       'civicrm_mailing_recipients' => array(
 386         0 => 'contact_id',
 387       ),
 388       'civicrm_membership' => array(
 389         0 => 'contact_id',
 390       ),
 391       'civicrm_membership_log' => array(
 392         0 => 'modified_id',
 393       ),
 394       'civicrm_membership_type' => array(
 395         0 => 'member_of_contact_id',
 396       ),
 397       'civicrm_note' => array(
 398         0 => 'contact_id',
 399       ),
 400       'civicrm_openid' => array(
 401         0 => 'contact_id',
 402       ),
 403       'civicrm_participant' => array(
 404         0 => 'contact_id',
 405         1 => 'transferred_to_contact_id', //CRM-16761
 406       ),
 407       'civicrm_payment_token' => array(
 408         0 => 'contact_id',
 409         1 => 'created_id',
 410       ),
 411       'civicrm_pcp' => array(
 412         0 => 'contact_id',
 413       ),
 414       'civicrm_phone' => array(
 415         0 => 'contact_id',
 416       ),
 417       'civicrm_pledge' => array(
 418         0 => 'contact_id',
 419       ),
 420       'civicrm_print_label' => array(
 421         0 => 'created_id',
 422       ),
 423       'civicrm_relationship' => array(
 424         0 => 'contact_id_a',
 425         1 => 'contact_id_b',
 426       ),
 427       'civicrm_report_instance' => array(
 428         0 => 'created_id',
 429         1 => 'owner_id',
 430       ),
 431       'civicrm_setting' => array(
 432         0 => 'contact_id',
 433         1 => 'created_id',
 434       ),
 435       'civicrm_subscription_history' => array(
 436         0 => 'contact_id',
 437       ),
 438       'civicrm_survey' => array(
 439         0 => 'created_id',
 440         1 => 'last_modified_id',
 441       ),
 442       'civicrm_tag' => array(
 443         0 => 'created_id',
 444       ),
 445       'civicrm_uf_group' => array(
 446         0 => 'created_id',
 447       ),
 448       'civicrm_uf_match' => array(
 449         0 => 'contact_id',
 450       ),
 451       'civicrm_value_testgetcidref_1' => array(
 452         0 => 'entity_id',
 453       ),
 454       'civicrm_website' => array(
 455         0 => 'contact_id',
 456       ),
 457     );
 458   }
 459
 460   /**
 461    * Get a list of CIDs that is calculated off the schema.
 462    *
 463    * Note this is an expensive and table locking query. Should be safe in tests though.
 464    */
 465   public function getCalculatedCIDRefs() {
 466     $cidRefs = array();
 467     $sql = "
 468 SELECT
 469     table_name,
 470     column_name
 471 FROM information_schema.key_column_usage
 472 WHERE
 473     referenced_table_schema = database() AND
 474     referenced_table_name = 'civicrm_contact' AND
 475     referenced_column_name = 'id';
 476       ";
 477     $dao = CRM_Core_DAO::executeQuery($sql);
 478     while ($dao->fetch()) {
 479       $cidRefs[$dao->table_name][] = $dao->column_name;
 480     }
 481     // Do specific re-ordering changes to make this the same as the ref validated one.
 482     // The above query orders by FK alphabetically.
 483     // There might be cleverer ways to do this but it shouldn't change much.
 484     $cidRefs['civicrm_contact'][0] = 'primary_contact_id';
 485     $cidRefs['civicrm_contact'][1] = 'employer_id';
 486     $cidRefs['civicrm_acl_contact_cache'][0] = 'user_id';
 487     $cidRefs['civicrm_acl_contact_cache'][1] = 'contact_id';
 488     $cidRefs['civicrm_mailing'][0] = 'created_id';
 489     $cidRefs['civicrm_mailing'][1] = 'scheduled_id';
 490     $cidRefs['civicrm_mailing'][2] = 'approver_id';
 491     return $cidRefs;
 492   }
 493
 494 }