Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
91.07% |
51 / 56 |
|
50.00% |
3 / 6 |
CRAP | |
0.00% |
0 / 1 |
| DatabaseQueryBuilder | |
91.07% |
51 / 56 |
|
50.00% |
3 / 6 |
16.18 | |
0.00% |
0 / 1 |
| __construct | |
100.00% |
2 / 2 |
|
100.00% |
1 / 1 |
1 | |||
| buildQuery | |
100.00% |
3 / 3 |
|
100.00% |
1 / 1 |
2 | |||
| buildRangeQuery | |
93.55% |
29 / 31 |
|
0.00% |
0 / 1 |
4.00 | |||
| buildDiscreteQuery | |
90.00% |
9 / 10 |
|
0.00% |
0 / 1 |
3.01 | |||
| makeOresClassificationTableAlias | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| validateSelected | |
77.78% |
7 / 9 |
|
0.00% |
0 / 1 |
5.27 | |||
| 1 | <?php |
| 2 | /** |
| 3 | * This program is free software: you can redistribute it and/or modify |
| 4 | * it under the terms of the GNU General Public License as published by |
| 5 | * the Free Software Foundation, either version 3 of the License, or |
| 6 | * (at your option) any later version. |
| 7 | * |
| 8 | * This program is distributed in the hope that it will be useful, |
| 9 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 10 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 11 | * GNU General Public License for more details. |
| 12 | * |
| 13 | * You should have received a copy of the GNU General Public License |
| 14 | * along with this program. If not, see <http://www.gnu.org/licenses/>. |
| 15 | */ |
| 16 | |
| 17 | namespace ORES\Storage; |
| 18 | |
| 19 | use ORES\Range; |
| 20 | use Wikimedia\Rdbms\IExpression; |
| 21 | use Wikimedia\Rdbms\IReadableDatabase; |
| 22 | use Wikimedia\Rdbms\OrExpressionGroup; |
| 23 | |
| 24 | class DatabaseQueryBuilder { |
| 25 | |
| 26 | /** |
| 27 | * @var ThresholdLookup |
| 28 | */ |
| 29 | private $thresholdLookup; |
| 30 | |
| 31 | /** |
| 32 | * @var IReadableDatabase |
| 33 | */ |
| 34 | private $db; |
| 35 | |
| 36 | public function __construct( ThresholdLookup $thresholdLookup, IReadableDatabase $db ) { |
| 37 | $this->thresholdLookup = $thresholdLookup; |
| 38 | $this->db = $db; |
| 39 | } |
| 40 | |
| 41 | /** |
| 42 | * Build a WHERE clause for selecting only the ores_classification rows |
| 43 | * that match the specified classes for a model. |
| 44 | * |
| 45 | * @param string $modelName Model to query |
| 46 | * @param string|string[] $selected Array (or comma-separated string) of class names to select |
| 47 | * @param bool $isDiscrete |
| 48 | * True for a model with distinct rows for each class |
| 49 | * False for a model with thresholds within the score of a single row |
| 50 | * @return IExpression|OrExpressionGroup|false SQL Condition that can be used in WHERE directly, or false when there |
| 51 | * is nothing to filter on |
| 52 | */ |
| 53 | public function buildQuery( $modelName, $selected, $isDiscrete = false ) { |
| 54 | return $isDiscrete ? |
| 55 | $this->buildDiscreteQuery( $modelName, $selected ) : |
| 56 | $this->buildRangeQuery( $modelName, $selected ); |
| 57 | } |
| 58 | |
| 59 | /** |
| 60 | * Build a WHERE clause for selecting only the ores_classification rows |
| 61 | * that match the specified classes for a model with thresholds. |
| 62 | * |
| 63 | * NOTE: This is used by PageTriage to filter on 'articlequality' |
| 64 | * |
| 65 | * @param string $modelName Model to filter |
| 66 | * @param string|string[] $selected Array (or comma-separated string) of class names to select |
| 67 | * @return OrExpressionGroup|false SQL Condition that can be used in WHERE directly, or false when there |
| 68 | * is nothing to filter on |
| 69 | */ |
| 70 | private function buildRangeQuery( $modelName, $selected ) { |
| 71 | $thresholds = $this->thresholdLookup->getThresholds( $modelName ); |
| 72 | $tableAlias = $this->makeOresClassificationTableAlias( $modelName ); |
| 73 | |
| 74 | $selected = $this->validateSelected( $selected, array_keys( $thresholds ) ); |
| 75 | if ( !$selected ) { |
| 76 | return false; |
| 77 | } |
| 78 | |
| 79 | $ranges = []; |
| 80 | foreach ( $selected as $className ) { |
| 81 | $range = new Range( |
| 82 | $thresholds[$className]['min'], |
| 83 | $thresholds[$className]['max'] |
| 84 | ); |
| 85 | |
| 86 | $result = array_filter( |
| 87 | $ranges, |
| 88 | static function ( Range $r ) use ( $range ) { |
| 89 | return $r->overlaps( $range ); |
| 90 | } |
| 91 | ); |
| 92 | $overlap = reset( $result ); |
| 93 | if ( $overlap ) { |
| 94 | /** @var Range $overlap */ |
| 95 | $overlap->combineWith( $range ); |
| 96 | } else { |
| 97 | $ranges[] = $range; |
| 98 | } |
| 99 | } |
| 100 | |
| 101 | $betweenConditions = array_map( |
| 102 | function ( Range $range ) use ( $tableAlias ) { |
| 103 | $min = $range->getMin(); |
| 104 | $max = $range->getMax(); |
| 105 | return $this->db->expr( "$tableAlias.oresc_probability", '>=', $min ) |
| 106 | ->and( "$tableAlias.oresc_probability", '<=', $max ); |
| 107 | }, |
| 108 | $ranges |
| 109 | ); |
| 110 | |
| 111 | return new OrExpressionGroup( ...$betweenConditions ); |
| 112 | } |
| 113 | |
| 114 | /** |
| 115 | * Build a WHERE clause for selecting only the ores_classification rows |
| 116 | * that match the specified levels for a model with discrete classification entries. |
| 117 | * |
| 118 | * NOTE: This is used by PageTriage to filter on 'draftquality' |
| 119 | * |
| 120 | * @param string $modelName Model to filter |
| 121 | * @param string|string[] $selected Array (or comma-separated string) of class names to select |
| 122 | * @return IExpression|false SQL Condition that can be used in WHERE directly, or false when there |
| 123 | * is nothing to filter on |
| 124 | */ |
| 125 | private function buildDiscreteQuery( $modelName, $selected ) { |
| 126 | global $wgOresModelClasses; |
| 127 | $modelClasses = $wgOresModelClasses[ $modelName ]; |
| 128 | $tableAlias = $this->makeOresClassificationTableAlias( $modelName ); |
| 129 | |
| 130 | $selected = $this->validateSelected( $selected, array_keys( $modelClasses ) ); |
| 131 | if ( !$selected ) { |
| 132 | return false; |
| 133 | } |
| 134 | |
| 135 | $classIds = []; |
| 136 | foreach ( $selected as $className ) { |
| 137 | $classIds[] = $modelClasses[ $className ]; |
| 138 | } |
| 139 | |
| 140 | return $this->db->expr( "$tableAlias.oresc_is_predicted", '=', 1 ) |
| 141 | ->and( "$tableAlias.oresc_class", '=', $classIds ); |
| 142 | } |
| 143 | |
| 144 | private function makeOresClassificationTableAlias( $modelName ) { |
| 145 | return "ores_{$modelName}_cls"; |
| 146 | } |
| 147 | |
| 148 | /** |
| 149 | * @param string|string[] $selected Selected class names, can be a comma-separated |
| 150 | * @param string[] $possible All available class names for the model |
| 151 | * @return string[]|false Valid and unique selected class names or false if |
| 152 | * no filters should be created |
| 153 | */ |
| 154 | private function validateSelected( $selected, $possible ) { |
| 155 | $selected = is_array( $selected ) ? $selected : |
| 156 | explode( ',', $selected ); |
| 157 | $selectedValid = array_intersect( $selected, $possible ); |
| 158 | $selectedValidUnique = array_unique( $selectedValid ); |
| 159 | |
| 160 | if ( count( $selectedValidUnique ) === 0 ) { |
| 161 | // none selected |
| 162 | return false; |
| 163 | } |
| 164 | |
| 165 | // all filters selected, and more than one possible filter exists. |
| 166 | // For filters like e.g. revertrisklanguageagnostic, where only one threshold exists, |
| 167 | // allow for selecting a single filter. |
| 168 | if ( count( $selectedValidUnique ) === count( $possible ) && count( $possible ) > 1 ) { |
| 169 | return false; |
| 170 | } |
| 171 | |
| 172 | return $selectedValidUnique; |
| 173 | } |
| 174 | |
| 175 | } |