-
Notifications
You must be signed in to change notification settings - Fork 83
/
SearchUpdater.php
192 lines (163 loc) · 6.95 KB
/
SearchUpdater.php
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
<?php
namespace SilverStripe\FullTextSearch\Search\Updaters;
use SilverStripe\Core\Config\Configurable;
use SilverStripe\Core\Injector\Injector;
use SilverStripe\Dev\SapphireTest;
use SilverStripe\ORM\Connect\Database;
use SilverStripe\ORM\DataObject;
use SilverStripe\ORM\DB;
use SilverStripe\FullTextSearch\Search\FullTextSearch;
use SilverStripe\FullTextSearch\Search\SearchIntrospection;
use SilverStripe\FullTextSearch\Search\Variants\SearchVariant;
use SilverStripe\FullTextSearch\Search\Processors\SearchUpdateProcessor;
use ReflectionClass;
/**
* This class is responsible for capturing changes to DataObjects and triggering index updates of the resulting dirty
* index items.
*
* Attached automatically by Injector configuration that overloads your flavour of Database class. The
* SearchManipulateCapture_[type] classes overload the manipulate method - basically we need to capture a
* manipulation _after_ all the augmentManipulation code (for instance Version's) is run
*
* Pretty closely tied to the field structure of SearchIndex.
*/
class SearchUpdater
{
use Configurable;
/**
* Whether to register the shutdown function to flush. Can be disabled for example in unit testing.
*
* @config
* @var bool
*/
private static $flush_on_shutdown = true;
/**
* Whether the updater is enabled. Set to false for local development if you don't have a Solr server.
*
* @config
* @var bool
*/
private static $enabled = true;
public static $registered = false;
/** @var SearchUpdateProcessor */
public static $processor = null;
/**
* Called by the ProxyDBExtension database connector with every manipulation made against the database.
*
* Check every index to see what objects need re-inserting into what indexes to keep the index fresh,
* but doesn't actually do it yet.
*
* TODO: This is pretty sensitive to the format of manipulation that DataObject::write produces. Specifically,
* it expects the actual class of the object to be present as a table, regardless of if any fields changed in that table
* (so a class => array( 'fields' => array() ) item), in order to find the actual class for a set of table manipulations
*/
public static function handle_manipulation($manipulation)
{
if (!static::config()->get('enabled')) {
return;
}
// First, extract any state that is in the manipulation itself
foreach ($manipulation as $table => $details) {
if (!isset($manipulation[$table]['class'])) {
$manipulation[$table]['class'] = DataObject::getSchema()->tableClass($table);
}
$manipulation[$table]['state'] = array();
}
SearchVariant::call('extractManipulationState', $manipulation);
// Then combine the manipulation back into object field sets
$writes = array();
foreach ($manipulation as $table => $details) {
if (!isset($details['id'])) {
continue;
}
$id = $details['id'];
$state = $details['state'];
$class = $details['class'];
$command = $details['command'];
$fields = isset($details['fields']) ? $details['fields'] : array();
$base = DataObject::getSchema()->baseDataClass($class);
$key = "$id:$base:" . serialize($state);
$statefulids = array(array('id' => $id, 'state' => $state));
// Is this the first table for this particular object? Then add an item to $writes
if (!isset($writes[$key])) {
$writes[$key] = array(
'base' => $base,
'class' => $class,
'id' => $id,
'statefulids' => $statefulids,
'command' => $command,
'fields' => array()
);
} elseif (is_subclass_of($class, $writes[$key]['class'] ?? '')) {
// Otherwise update the class label if it's more specific than the currently recorded one
$writes[$key]['class'] = $class;
}
// Update the fields
foreach ($fields as $field => $value) {
$writes[$key]['fields']["$class:$field"] = $value;
}
}
// Trim non-delete records without fields
foreach (array_keys($writes ?? []) as $key) {
if ($writes[$key]['command'] !== 'delete' && empty($writes[$key]['fields'])) {
unset($writes[$key]);
}
}
// Then extract any state that is needed for the writes
SearchVariant::call('extractManipulationWriteState', $writes);
// Submit all of these writes to the search processor
static::process_writes($writes);
}
/**
* Send updates to the current search processor for execution
*
* @param array $writes
*/
public static function process_writes($writes)
{
foreach ($writes as $write) {
// For every index
foreach (FullTextSearch::get_indexes() as $index => $instance) {
// If that index as a field from this class
if (SearchIntrospection::is_subclass_of($write['class'], $instance->dependancyList)) {
// Get the dirty IDs
$dirtyids = $instance->getDirtyIDs($write['class'], $write['id'], $write['statefulids'], $write['fields']);
// Then add then then to the global list to deal with later
foreach ($dirtyids as $dirtyclass => $ids) {
if ($ids) {
if (!self::$processor) {
self::$processor = Injector::inst()->create(SearchUpdateProcessor::class);
}
self::$processor->addDirtyIDs($dirtyclass, $ids, $index);
}
}
}
}
}
// If we do have some work to do register the shutdown function to actually do the work
if (self::$processor && !self::$registered && self::config()->get('flush_on_shutdown')) {
register_shutdown_function(array(SearchUpdater::class, "flush_dirty_indexes"));
self::$registered = true;
}
}
/**
* Throw away the recorded dirty IDs without doing anything with them.
*/
public static function clear_dirty_indexes()
{
self::$processor = null;
}
/**
* Do something with the recorded dirty IDs, where that "something" depends on the value of self::$update_method,
* either immediately update the indexes, queue a messsage to update the indexes at some point in the future, or
* just throw the dirty IDs away.
*/
public static function flush_dirty_indexes()
{
if (!self::$processor) {
return;
}
self::$processor->triggerProcessing();
self::$processor = null;
}
}