|
1
|
|
|
<?php declare(strict_types = 1); |
|
2
|
|
|
namespace ApacheSolrForTypo3\Solr\Domain\Index\Queue\UpdateHandler; |
|
3
|
|
|
|
|
4
|
|
|
/*************************************************************** |
|
5
|
|
|
* Copyright notice |
|
6
|
|
|
* |
|
7
|
|
|
* (c) 2021 Markus Friedrich <[email protected]> |
|
8
|
|
|
* All rights reserved |
|
9
|
|
|
* |
|
10
|
|
|
* This script is part of the TYPO3 project. The TYPO3 project is |
|
11
|
|
|
* free software; you can redistribute it and/or modify |
|
12
|
|
|
* it under the terms of the GNU General Public License as published by |
|
13
|
|
|
* the Free Software Foundation; either version 3 of the License, or |
|
14
|
|
|
* (at your option) any later version. |
|
15
|
|
|
* |
|
16
|
|
|
* The GNU General Public License can be found at |
|
17
|
|
|
* http://www.gnu.org/copyleft/gpl.html. |
|
18
|
|
|
* |
|
19
|
|
|
* This script is distributed in the hope that it will be useful, |
|
20
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
21
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
22
|
|
|
* GNU General Public License for more details. |
|
23
|
|
|
* |
|
24
|
|
|
* This copyright notice MUST APPEAR in all copies of the script! |
|
25
|
|
|
***************************************************************/ |
|
26
|
|
|
|
|
27
|
|
|
use TYPO3\CMS\Backend\Utility\BackendUtility; |
|
28
|
|
|
use ApacheSolrForTypo3\Solr\Domain\Index\Queue\GarbageRemover\StrategyFactory; |
|
29
|
|
|
|
|
30
|
|
|
/** |
|
31
|
|
|
* Garbage handler |
|
32
|
|
|
* |
|
33
|
|
|
* Handles updates on potential relevant records and |
|
34
|
|
|
* collects the garbage, e.g. a deletion might require |
|
35
|
|
|
* index and index queue updates. |
|
36
|
|
|
*/ |
|
37
|
|
|
class GarbageHandler extends AbstractUpdateHandler |
|
38
|
|
|
{ |
|
39
|
|
|
/** |
|
40
|
|
|
* Configuration used to check if recursive updates are required |
|
41
|
|
|
* |
|
42
|
|
|
* Holds the configuration when a recursive page queuing should be triggered, while processing record |
|
43
|
|
|
* updates |
|
44
|
|
|
* |
|
45
|
|
|
* Note: The SQL transaction is already committed, so the current state covers only "non"-changed fields. |
|
46
|
|
|
* |
|
47
|
|
|
* @var array |
|
48
|
|
|
*/ |
|
49
|
|
|
protected $updateSubPagesRecursiveTriggerConfiguration = [ |
|
50
|
|
|
// the current page has the field "extendToSubpages" enabled and the field "hidden" was set to 1 |
|
51
|
|
|
// covers following scenarios: |
|
52
|
|
|
// 'currentState' => ['hidden' => '0', 'extendToSubpages' => '0|1'], 'changeSet' => ['hidden' => '1', (optional)'extendToSubpages' => '1'] |
|
53
|
|
|
'extendToSubpageEnabledAndHiddenFlagWasAdded' => [ |
|
54
|
|
|
'currentState' => ['extendToSubpages' => '1'], |
|
55
|
|
|
'changeSet' => ['hidden' => '1'] |
|
56
|
|
|
], |
|
57
|
|
|
// the current page has the field "hidden" enabled and the field "extendToSubpages" was set to 1 |
|
58
|
|
|
// covers following scenarios: |
|
59
|
|
|
// 'currentState' => ['hidden' => '0|1', 'extendToSubpages' => '0'], 'changeSet' => [(optional)'hidden' => '1', 'extendToSubpages' => '1'] |
|
60
|
|
|
'hiddenIsEnabledAndExtendToSubPagesWasAdded' => [ |
|
61
|
|
|
'currentState' => ['hidden' => '1'], |
|
62
|
|
|
'changeSet' => ['extendToSubpages' => '1'] |
|
63
|
|
|
], |
|
64
|
|
|
// the field "no_search_sub_entries" of current page was set to 1 |
|
65
|
|
|
'no_search_sub_entriesFlagWasAdded' => [ |
|
66
|
|
|
'changeSet' => ['no_search_sub_entries' => '1'] |
|
67
|
|
|
], |
|
68
|
|
|
]; |
|
69
|
|
|
|
|
70
|
|
|
/** |
|
71
|
|
|
* Tracks down index documents belonging to a particular record or page and |
|
72
|
|
|
* removes them from the index and the Index Queue. |
|
73
|
|
|
* |
|
74
|
|
|
* @param string $table The record's table name. |
|
75
|
|
|
* @param int $uid The record's uid. |
|
76
|
|
|
* @throws \UnexpectedValueException if a hook object does not implement interface |
|
77
|
|
|
* \ApacheSolrForTypo3\Solr\GarbageCollectorPostProcessor |
|
78
|
|
|
*/ |
|
79
|
|
|
public function collectGarbage($table, $uid): void |
|
80
|
|
|
{ |
|
81
|
|
|
$garbageRemoverStrategy = StrategyFactory::getByTable($table); |
|
82
|
|
|
$garbageRemoverStrategy->removeGarbageOf($table, $uid); |
|
83
|
|
|
} |
|
84
|
|
|
|
|
85
|
|
|
/** |
|
86
|
|
|
* Handles moved pages |
|
87
|
|
|
* |
|
88
|
|
|
* @param int $uid |
|
89
|
|
|
*/ |
|
90
|
|
|
public function handlePageMovement(int $uid): void |
|
91
|
|
|
{ |
|
92
|
|
|
// TODO the below comment is not valid anymore, pid has been removed from doc ID |
|
93
|
|
|
// ...still needed? |
|
94
|
|
|
|
|
95
|
|
|
// must be removed from index since the pid changes and |
|
96
|
|
|
// is part of the Solr document ID |
|
97
|
|
|
$this->collectGarbage('pages', $uid); |
|
98
|
|
|
|
|
99
|
|
|
// now re-index with new properties |
|
100
|
|
|
$this->indexQueue->updateItem('pages', $uid); |
|
101
|
|
|
} |
|
102
|
|
|
|
|
103
|
|
|
/** |
|
104
|
|
|
* Performs a record garbage check |
|
105
|
|
|
* |
|
106
|
|
|
* @param int $uid |
|
107
|
|
|
* @param string $table |
|
108
|
|
|
* @param array $updatedFields |
|
109
|
|
|
* @param bool $frontendGroupsRemoved |
|
110
|
|
|
*/ |
|
111
|
|
|
public function performRecordGarbageCheck( |
|
112
|
|
|
int $uid, |
|
113
|
|
|
string $table, |
|
114
|
|
|
array $updatedFields, |
|
115
|
|
|
bool $frontendGroupsRemoved |
|
116
|
|
|
): void { |
|
117
|
|
|
$record = $this->getRecordWithFieldRelevantForGarbageCollection($table, $uid); |
|
118
|
|
|
|
|
119
|
|
|
// If no record could be found skip further processing |
|
120
|
|
|
if (empty($record)) { |
|
121
|
|
|
return; |
|
122
|
|
|
} |
|
123
|
|
|
|
|
124
|
|
|
if ($table === 'pages') { |
|
125
|
|
|
$this->deleteSubEntriesWhenRecursiveTriggerIsRecognized($table, $uid, $updatedFields); |
|
126
|
|
|
} |
|
127
|
|
|
|
|
128
|
|
|
$record = $this->tcaService->normalizeFrontendGroupField($table, $record); |
|
129
|
|
|
$isGarbage = $this->getIsGarbageRecord($table, $record, $frontendGroupsRemoved); |
|
130
|
|
|
if (!$isGarbage) { |
|
131
|
|
|
return; |
|
132
|
|
|
} |
|
133
|
|
|
|
|
134
|
|
|
$this->collectGarbage($table, $uid); |
|
135
|
|
|
} |
|
136
|
|
|
|
|
137
|
|
|
/** |
|
138
|
|
|
* @param string $table |
|
139
|
|
|
* @param int $uid |
|
140
|
|
|
* @param array $updatedFields |
|
141
|
|
|
*/ |
|
142
|
|
|
protected function deleteSubEntriesWhenRecursiveTriggerIsRecognized( |
|
143
|
|
|
string $table, |
|
144
|
|
|
int $uid, |
|
145
|
|
|
array $updatedFields |
|
146
|
|
|
): void { |
|
147
|
|
|
if (!$this->isRecursivePageUpdateRequired($uid, $updatedFields)) { |
|
148
|
|
|
return; |
|
149
|
|
|
} |
|
150
|
|
|
|
|
151
|
|
|
// get affected subpages when "extendToSubpages" flag was set |
|
152
|
|
|
$pagesToDelete = $this->getSubPageIds($uid); |
|
153
|
|
|
// we need to at least remove this page |
|
154
|
|
|
foreach ($pagesToDelete as $pageToDelete) { |
|
155
|
|
|
$this->collectGarbage($table, $pageToDelete); |
|
156
|
|
|
} |
|
157
|
|
|
} |
|
158
|
|
|
|
|
159
|
|
|
/** |
|
160
|
|
|
* Determines if a record is garbage and can be deleted. |
|
161
|
|
|
* |
|
162
|
|
|
* @param string $table |
|
163
|
|
|
* @param array $record |
|
164
|
|
|
* @param bool $frontendGroupsRemoved |
|
165
|
|
|
* @return bool |
|
166
|
|
|
*/ |
|
167
|
|
|
protected function getIsGarbageRecord(string $table, array $record, bool $frontendGroupsRemoved): bool |
|
168
|
|
|
{ |
|
169
|
|
|
return $frontendGroupsRemoved |
|
170
|
|
|
|| $this->tcaService->isHidden($table, $record) |
|
171
|
|
|
|| $this->isInvisibleByStartOrEndtime($table, $record) |
|
172
|
|
|
|| ($table === 'pages' && $this->isPageExcludedFromSearch($record)) |
|
173
|
|
|
|| ($table === 'pages' && !$this->isIndexablePageType($record)); |
|
174
|
|
|
} |
|
175
|
|
|
|
|
176
|
|
|
/** |
|
177
|
|
|
* Checks whether a page has a page type that can be indexed. |
|
178
|
|
|
* Currently standard pages and mount pages can be indexed. |
|
179
|
|
|
* |
|
180
|
|
|
* @param array $record A page record |
|
181
|
|
|
* @return bool TRUE if the page can be indexed according to its page type, FALSE otherwise |
|
182
|
|
|
*/ |
|
183
|
|
|
protected function isIndexablePageType(array $record): bool |
|
184
|
|
|
{ |
|
185
|
|
|
return $this->frontendEnvironment->isAllowedPageType($record); |
|
186
|
|
|
} |
|
187
|
|
|
|
|
188
|
|
|
/** |
|
189
|
|
|
* Checks whether the page has been excluded from searching. |
|
190
|
|
|
* |
|
191
|
|
|
* @param array $record An array with record fields that may affect visibility. |
|
192
|
|
|
* @return bool True if the page has been excluded from searching, FALSE otherwise |
|
193
|
|
|
*/ |
|
194
|
|
|
protected function isPageExcludedFromSearch(array $record): bool |
|
195
|
|
|
{ |
|
196
|
|
|
return (bool)$record['no_search']; |
|
197
|
|
|
} |
|
198
|
|
|
|
|
199
|
|
|
|
|
200
|
|
|
/** |
|
201
|
|
|
* Check if a record is getting invisible due to changes in start or endtime. In addition it is checked that the related |
|
202
|
|
|
* queue item was marked as indexed. |
|
203
|
|
|
* |
|
204
|
|
|
* @param string $table |
|
205
|
|
|
* @param array $record |
|
206
|
|
|
* @return bool |
|
207
|
|
|
*/ |
|
208
|
|
|
protected function isInvisibleByStartOrEndtime(string $table, array $record): bool |
|
209
|
|
|
{ |
|
210
|
|
|
return ( |
|
211
|
|
|
($this->tcaService->isStartTimeInFuture($table, $record) |
|
212
|
|
|
|| $this->tcaService->isEndTimeInPast($table, $record)) |
|
213
|
|
|
&& $this->isRelatedQueueRecordMarkedAsIndexed($table, $record) |
|
214
|
|
|
); |
|
215
|
|
|
} |
|
216
|
|
|
|
|
217
|
|
|
/** |
|
218
|
|
|
* Checks if the related index queue item is indexed. |
|
219
|
|
|
* |
|
220
|
|
|
* * For tt_content the page from the pid is checked |
|
221
|
|
|
* * For all other records the table it's self is checked |
|
222
|
|
|
* |
|
223
|
|
|
* @param string $table The table name. |
|
224
|
|
|
* @param array $record An array with record fields that may affect visibility. |
|
225
|
|
|
* @return bool True if the record is marked as being indexed |
|
226
|
|
|
*/ |
|
227
|
|
|
protected function isRelatedQueueRecordMarkedAsIndexed(string $table, array $record): bool |
|
228
|
|
|
{ |
|
229
|
|
|
if ($table === 'tt_content') { |
|
230
|
|
|
$table = 'pages'; |
|
231
|
|
|
$uid = $record['pid']; |
|
232
|
|
|
} else { |
|
233
|
|
|
$uid = $record['uid']; |
|
234
|
|
|
} |
|
235
|
|
|
|
|
236
|
|
|
return $this->indexQueue->containsIndexedItem($table, $uid); |
|
237
|
|
|
} |
|
238
|
|
|
|
|
239
|
|
|
/** |
|
240
|
|
|
* Returns a record with all visibility affecting fields. |
|
241
|
|
|
* |
|
242
|
|
|
* @param string $table |
|
243
|
|
|
* @param int $uid |
|
244
|
|
|
* @return array |
|
245
|
|
|
*/ |
|
246
|
|
|
public function getRecordWithFieldRelevantForGarbageCollection(string $table, int $uid): array |
|
247
|
|
|
{ |
|
248
|
|
|
$garbageCollectionRelevantFields = $this->tcaService->getVisibilityAffectingFieldsByTable($table); |
|
249
|
|
|
return (array)BackendUtility::getRecord($table, $uid, $garbageCollectionRelevantFields, '', false); |
|
250
|
|
|
} |
|
251
|
|
|
} |
|
252
|
|
|
|