Commit | Line | Data |
---|---|---|
6a488035 TO |
1 | <?php |
2 | /* | |
3 | +--------------------------------------------------------------------+ | |
39de6fd5 | 4 | | CiviCRM version 4.6 | |
6a488035 | 5 | +--------------------------------------------------------------------+ |
06b69b18 | 6 | | Copyright CiviCRM LLC (c) 2004-2014 | |
6a488035 TO |
7 | +--------------------------------------------------------------------+ |
8 | | This file is a part of CiviCRM. | | |
9 | | | | |
10 | | CiviCRM is free software; you can copy, modify, and distribute it | | |
11 | | under the terms of the GNU Affero General Public License | | |
12 | | Version 3, 19 November 2007 and the CiviCRM Licensing Exception. | | |
13 | | | | |
14 | | CiviCRM is distributed in the hope that it will be useful, but | | |
15 | | WITHOUT ANY WARRANTY; without even the implied warranty of | | |
16 | | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. | | |
17 | | See the GNU Affero General Public License for more details. | | |
18 | | | | |
19 | | You should have received a copy of the GNU Affero General Public | | |
20 | | License and the CiviCRM Licensing Exception along | | |
21 | | with this program; if not, contact CiviCRM LLC | | |
22 | | at info[AT]civicrm[DOT]org. If you have questions about the | | |
23 | | GNU Affero General Public License or the licensing of CiviCRM, | | |
24 | | see the CiviCRM license FAQ at http://civicrm.org/licensing | | |
25 | +--------------------------------------------------------------------+ | |
26 | */ | |
27 | ||
28 | /** | |
29 | * | |
30 | * @package CRM | |
06b69b18 | 31 | * @copyright CiviCRM LLC (c) 2004-2014 |
6a488035 TO |
32 | * $Id$ |
33 | * | |
34 | */ | |
35 | ||
36 | /** | |
37 | * The CiviCRM duplicate discovery engine is based on an | |
38 | * algorithm designed by David Strauss <david@fourkitchens.com>. | |
39 | */ | |
40 | class CRM_Dedupe_Finder { | |
41 | ||
42 | /** | |
43 | * Return a contact_id-keyed array of arrays of possible dupes | |
44 | * (of the key contact_id) - limited to dupes of $cids if provided. | |
45 | * | |
98997235 TO |
46 | * @param int $rgid |
47 | * Rule group id. | |
48 | * @param array $cids | |
49 | * Contact ids to limit the search to. | |
6a488035 TO |
50 | * |
51 | * @return array array of (cid1, cid2, weight) dupe triples | |
52 | */ | |
00be9182 | 53 | public static function dupes($rgid, $cids = array()) { |
6a488035 TO |
54 | $rgBao = new CRM_Dedupe_BAO_RuleGroup(); |
55 | $rgBao->id = $rgid; | |
56 | $rgBao->contactIds = $cids; | |
57 | if (!$rgBao->find(TRUE)) { | |
16254ae1 | 58 | CRM_Core_Error::fatal("Dedupe rule not found for selected contacts"); |
6a488035 TO |
59 | } |
60 | ||
61 | $rgBao->fillTable(); | |
62 | $dao = new CRM_Core_DAO(); | |
63 | $dao->query($rgBao->thresholdQuery()); | |
64 | $dupes = array(); | |
65 | while ($dao->fetch()) { | |
66 | $dupes[] = array($dao->id1, $dao->id2, $dao->weight); | |
67 | } | |
68 | $dao->query($rgBao->tableDropQuery()); | |
69 | ||
70 | return $dupes; | |
71 | } | |
72 | ||
73 | /** | |
74 | * Return an array of possible dupes, based on the provided array of | |
75 | * params, using the default rule group for the given contact type and | |
76 | * usage. | |
77 | * | |
78 | * check_permission is a boolean flag to indicate if permission should be considered. | |
79 | * default is to always check permissioning but public pages for example might not want | |
80 | * permission to be checked for anonymous users. Refer CRM-6211. We might be beaking | |
81 | * Multi-Site dedupe for public pages. | |
82 | * | |
98997235 TO |
83 | * @param array $params |
84 | * Array of params of the form $params[$table][$field] == $value. | |
85 | * @param string $ctype | |
86 | * Contact type to match against. | |
87 | * @param string $used | |
88 | * Dedupe rule group usage ('Unsupervised' or 'Supervised' or 'General'). | |
89 | * @param array $except | |
90 | * Array of contacts that shouldn't be considered dupes. | |
91 | * @param int $ruleGroupID | |
92 | * The id of the dedupe rule we should be using. | |
6a488035 TO |
93 | * |
94 | * @return array matching contact ids | |
95 | */ | |
96 | static function dupesByParams($params, | |
97 | $ctype, | |
d58a19a1 TO |
98 | $used = 'Unsupervised', |
99 | $except = array(), | |
6a488035 TO |
100 | $ruleGroupID = NULL |
101 | ) { | |
102 | // If $params is empty there is zero reason to proceed. | |
103 | if (!$params) { | |
104 | return array(); | |
105 | } | |
106 | ||
107 | $foundByID = FALSE; | |
108 | if ($ruleGroupID) { | |
109 | $rgBao = new CRM_Dedupe_BAO_RuleGroup(); | |
110 | $rgBao->id = $ruleGroupID; | |
111 | $rgBao->contact_type = $ctype; | |
112 | if ($rgBao->find(TRUE)) { | |
113 | $foundByID = TRUE; | |
114 | } | |
115 | } | |
116 | ||
117 | if (!$foundByID) { | |
118 | $rgBao = new CRM_Dedupe_BAO_RuleGroup(); | |
119 | $rgBao->contact_type = $ctype; | |
120 | $rgBao->used = $used; | |
121 | if (!$rgBao->find(TRUE)) { | |
122 | CRM_Core_Error::fatal("$used rule for $ctype does not exist"); | |
123 | } | |
124 | } | |
125 | $params['check_permission'] = CRM_Utils_Array::value('check_permission', $params, TRUE); | |
126 | ||
127 | $rgBao->params = $params; | |
128 | $rgBao->fillTable(); | |
129 | $dao = new CRM_Core_DAO(); | |
130 | $dao->query($rgBao->thresholdQuery($params['check_permission'])); | |
131 | $dupes = array(); | |
132 | while ($dao->fetch()) { | |
133 | if (isset($dao->id) && $dao->id) { | |
134 | $dupes[] = $dao->id; | |
135 | } | |
136 | } | |
137 | $dao->query($rgBao->tableDropQuery()); | |
138 | return array_diff($dupes, $except); | |
139 | } | |
140 | ||
141 | /** | |
142 | * Return a contact_id-keyed array of arrays of possible dupes in the given group. | |
143 | * | |
98997235 TO |
144 | * @param int $rgid |
145 | * Rule group id. | |
146 | * @param int $gid | |
147 | * Contact group id (currently, works only with non-smart groups). | |
6a488035 TO |
148 | * |
149 | * @return array array of (cid1, cid2, weight) dupe triples | |
150 | */ | |
00be9182 | 151 | public static function dupesInGroup($rgid, $gid) { |
6a488035 TO |
152 | $cids = array_keys(CRM_Contact_BAO_Group::getMember($gid)); |
153 | if ( !empty($cids) ) { | |
d58a19a1 TO |
154 | return self::dupes($rgid, $cids); |
155 | } | |
6a488035 TO |
156 | return array(); |
157 | } | |
158 | ||
159 | /** | |
160 | * Return dupes of a given contact, using the default rule group (of a provided usage). | |
161 | * | |
98997235 TO |
162 | * @param int $cid |
163 | * Contact id of the given contact. | |
164 | * @param string $used | |
165 | * Dedupe rule group usage ('Unsupervised' or 'Supervised' or 'General'). | |
166 | * @param string $ctype | |
167 | * Contact type of the given contact. | |
6a488035 TO |
168 | * |
169 | * @return array array of dupe contact_ids | |
170 | */ | |
00be9182 | 171 | public static function dupesOfContact($cid, $used = 'Unsupervised', $ctype = NULL) { |
6a488035 TO |
172 | // if not provided, fetch the contact type from the database |
173 | if (!$ctype) { | |
174 | $dao = new CRM_Contact_DAO_Contact(); | |
175 | $dao->id = $cid; | |
176 | if (!$dao->find(TRUE)) { | |
177 | CRM_Core_Error::fatal("contact id of $cid does not exist"); | |
178 | } | |
179 | $ctype = $dao->contact_type; | |
180 | } | |
181 | $rgBao = new CRM_Dedupe_BAO_RuleGroup(); | |
182 | $rgBao->used = $used; | |
183 | $rgBao->contact_type = $ctype; | |
184 | if (!$rgBao->find(TRUE)) { | |
185 | CRM_Core_Error::fatal("$used rule for $ctype does not exist"); | |
186 | } | |
187 | $dupes = self::dupes($rgBao->id, array($cid)); | |
188 | ||
189 | // get the dupes for this cid | |
190 | $result = array(); | |
191 | foreach ($dupes as $dupe) { | |
192 | if ($dupe[0] == $cid) { | |
193 | $result[] = $dupe[1]; | |
194 | } | |
195 | elseif ($dupe[1] == $cid) { | |
196 | $result[] = $dupe[0]; | |
197 | } | |
198 | } | |
199 | return $result; | |
200 | } | |
201 | ||
202 | /** | |
203 | * A hackish function needed to massage CRM_Contact_Form_$ctype::formRule() | |
204 | * object into a valid $params array for dedupe | |
205 | * | |
98997235 TO |
206 | * @param array $fields |
207 | * Contact structure from formRule(). | |
208 | * @param string $ctype | |
209 | * Contact type of the given contact. | |
6a488035 TO |
210 | * |
211 | * @return array valid $params array for dedupe | |
212 | */ | |
00be9182 | 213 | public static function formatParams($fields, $ctype) { |
6a488035 TO |
214 | $flat = array(); |
215 | CRM_Utils_Array::flatten($fields, $flat); | |
216 | ||
309a09df | 217 | // FIXME: This may no longer be necessary - check inputs |
6a488035 TO |
218 | $replace_these = array( |
219 | 'individual_prefix' => 'prefix_id', | |
220 | 'individual_suffix' => 'suffix_id', | |
221 | 'gender' => 'gender_id', | |
222 | ); | |
309a09df | 223 | foreach (array('individual_suffix', 'individual_prefix', 'gender') as $name) { |
a7488080 | 224 | if (!empty($fields[$name])) { |
6a488035 TO |
225 | $flat[$replace_these[$name]] = $flat[$name]; |
226 | unset($flat[$name]); | |
227 | } | |
228 | } | |
229 | ||
230 | // handle {birth,deceased}_date | |
231 | foreach (array( | |
232 | 'birth_date', 'deceased_date') as $date) { | |
a7488080 | 233 | if (!empty($fields[$date])) { |
6a488035 TO |
234 | $flat[$date] = $fields[$date]; |
235 | if (is_array($flat[$date])) { | |
236 | $flat[$date] = CRM_Utils_Date::format($flat[$date]); | |
237 | } | |
238 | $flat[$date] = CRM_Utils_Date::processDate($flat[$date]); | |
239 | } | |
240 | } | |
241 | ||
a7488080 | 242 | if (!empty($flat['contact_source'])) { |
6a488035 TO |
243 | $flat['source'] = $flat['contact_source']; |
244 | unset($flat['contact_source']); | |
245 | } | |
246 | ||
247 | // handle preferred_communication_method | |
df5ad245 | 248 | if (!empty($fields['preferred_communication_method'])) { |
6a488035 TO |
249 | $methods = array_intersect($fields['preferred_communication_method'], array('1')); |
250 | $methods = array_keys($methods); | |
251 | sort($methods); | |
252 | if ($methods) { | |
253 | $flat['preferred_communication_method'] = CRM_Core_DAO::VALUE_SEPARATOR . implode(CRM_Core_DAO::VALUE_SEPARATOR, $methods) . CRM_Core_DAO::VALUE_SEPARATOR; | |
254 | } | |
255 | } | |
256 | ||
257 | // handle custom data | |
258 | $tree = CRM_Core_BAO_CustomGroup::getTree($ctype, CRM_Core_DAO::$_nullObject, NULL, -1); | |
259 | CRM_Core_BAO_CustomGroup::postProcess($tree, $fields, TRUE); | |
260 | foreach ($tree as $key => $cg) { | |
261 | if (!is_int($key)) { | |
262 | continue; | |
263 | } | |
264 | foreach ($cg['fields'] as $cf) { | |
265 | $flat[$cf['column_name']] = CRM_Utils_Array::value('data', $cf['customValue']); | |
266 | } | |
267 | } | |
268 | ||
269 | // if the key is dotted, keep just the last part of it | |
270 | foreach ($flat as $key => $value) { | |
271 | if (substr_count($key, '.')) { | |
272 | $last = explode('.', $key); | |
273 | $last = array_pop($last); | |
274 | // make sure the first occurence is kept, not the last | |
275 | if (!isset($flat[$last])) { | |
276 | $flat[$last] = $value; | |
277 | } | |
278 | unset($flat[$key]); | |
279 | } | |
280 | } | |
281 | ||
282 | // drop the -digit (and -Primary, for CRM-3902) postfixes (so event registration's $flat['email-5'] becomes $flat['email']) | |
283 | // FIXME: CRM-5026 should be fixed here; the below clobbers all address info; we should split off address fields and match | |
284 | // the -digit to civicrm_address.location_type_id and -Primary to civicrm_address.is_primary | |
285 | foreach ($flat as $key => $value) { | |
286 | $matches = array(); | |
287 | if (preg_match('/(.*)-(\d+|Primary)$/', $key, $matches)) { | |
288 | $flat[$matches[1]] = $value; | |
289 | unset($flat[$key]); | |
290 | } | |
291 | } | |
292 | ||
293 | $params = array(); | |
294 | $supportedFields = CRM_Dedupe_BAO_RuleGroup::supportedFields($ctype); | |
295 | if (is_array($supportedFields)) { | |
296 | foreach ($supportedFields as $table => $fields) { | |
297 | if ($table == 'civicrm_address') { | |
298 | // for matching on civicrm_address fields, we also need the location_type_id | |
299 | $fields['location_type_id'] = ''; | |
300 | // FIXME: we also need to do some hacking for id and name fields, see CRM-3902’s comments | |
301 | $fixes = array( | |
d58a19a1 TO |
302 | 'address_name' => 'name', |
303 | 'country' => 'country_id', | |
304 | 'state_province' => 'state_province_id', | |
305 | 'county' => 'county_id', | |
6a488035 TO |
306 | ); |
307 | foreach ($fixes as $orig => $target) { | |
a7488080 | 308 | if (!empty($flat[$orig])) { |
6a488035 TO |
309 | $params[$table][$target] = $flat[$orig]; |
310 | } | |
311 | } | |
312 | } | |
313 | foreach ($fields as $field => $title) { | |
a7488080 | 314 | if (!empty($flat[$field])) { |
6a488035 TO |
315 | $params[$table][$field] = $flat[$field]; |
316 | } | |
317 | } | |
318 | } | |
319 | } | |
320 | return $params; | |
321 | } | |
322 | } |