Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
0.00% |
0 / 111 |
|
0.00% |
0 / 7 |
CRAP | |
0.00% |
0 / 1 |
| JsonToMathML | |
0.00% |
0 / 108 |
|
0.00% |
0 / 7 |
930 | |
0.00% |
0 / 1 |
| __construct | |
0.00% |
0 / 12 |
|
0.00% |
0 / 1 |
2 | |||
| execute | |
0.00% |
0 / 4 |
|
0.00% |
0 / 1 |
2 | |||
| readJsonAndGenerateMML | |
0.00% |
0 / 35 |
|
0.00% |
0 / 1 |
156 | |||
| formatInput | |
0.00% |
0 / 43 |
|
0.00% |
0 / 1 |
156 | |||
| getJSON | |
0.00% |
0 / 3 |
|
0.00% |
0 / 1 |
2 | |||
| writeToFile | |
0.00% |
0 / 2 |
|
0.00% |
0 / 1 |
2 | |||
| fetchMathML | |
0.00% |
0 / 9 |
|
0.00% |
0 / 1 |
6 | |||
| 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 2 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 along |
| 14 | * with this program; if not, write to the Free Software Foundation, Inc., |
| 15 | * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. |
| 16 | * http://www.gnu.org/copyleft/gpl.html |
| 17 | * |
| 18 | * @ingroup Maintenance |
| 19 | */ |
| 20 | |
| 21 | use MediaWiki\Extension\Math\MathRenderer; |
| 22 | use MediaWiki\MediaWikiServices; |
| 23 | |
| 24 | require_once __DIR__ . '/../../../maintenance/Maintenance.php'; |
| 25 | |
| 26 | /** |
| 27 | * From a specified json file with (La)TeX formula as input, |
| 28 | * create a json file with the Tex and corresponding MathML. |
| 29 | * This is mostly used for generating Test-Content for the MathML features of WikiTexVC. |
| 30 | * |
| 31 | * The script fetches: |
| 32 | * - Mathoid MathML (mode: 'mathml') |
| 33 | * - LaTeXML MathML (mode: 'latexml') |
| 34 | * |
| 35 | * @author Johannes Stegmüller |
| 36 | */ |
| 37 | class JsonToMathML extends Maintenance { |
| 38 | /** @var string */ |
| 39 | private $inputPath = ""; |
| 40 | |
| 41 | /** @var string */ |
| 42 | private $outputPath = ""; |
| 43 | /** @var int */ |
| 44 | private $inputFormat = 0; |
| 45 | |
| 46 | public function __construct() { |
| 47 | parent::__construct(); |
| 48 | $this->addDescription( 'From a JSON file containing (La)TeX math inputs, create a JSON file with LaTeX and ' . |
| 49 | 'corresponding MathML.' ); |
| 50 | $this->addArg( 'input-path', "Path (with filename) of the json file read by this script.", |
| 51 | true ); |
| 52 | $this->addArg( 'output-path', "Path (with filename) of the output json file created by this script.", |
| 53 | true ); |
| 54 | $this->addOption( 'inputformat', 'Custom parsing how to format the input-json ( see formatInput function)', |
| 55 | false, true, 'i' ); |
| 56 | $this->addOption( 'chem-fallback', 'If the json read does not define input-type (tex or chem), check ' . |
| 57 | 'expressions as Tex and then as chem', false, true, 'c' ); |
| 58 | $this->requireExtension( 'Math' ); |
| 59 | } |
| 60 | |
| 61 | public function execute() { |
| 62 | $this->inputFormat = $this->getOption( "inputformat", 0 ); |
| 63 | $this->inputPath = $this->getArg( 0 ); |
| 64 | $this->outputPath = $this->getArg( 1 ); |
| 65 | $this->readJsonAndGenerateMML(); |
| 66 | } |
| 67 | |
| 68 | public function readJsonAndGenerateMML() { |
| 69 | $inputTex = $this->getJSON( $this->inputPath ); |
| 70 | if ( $inputTex == null ) { |
| 71 | throw new InvalidArgumentException( "Provide a json file as input which has content" ); |
| 72 | } |
| 73 | $inputTexF = $this->formatInput( $inputTex ); |
| 74 | $allEntries = []; |
| 75 | |
| 76 | foreach ( $inputTexF as $entry ) { |
| 77 | try { |
| 78 | $mmlMathoid = $this->fetchMathML( $entry['tex'], $entry['type'], 'mathml' ); |
| 79 | if ( ( $this->getOption( "chem-fallback", 0 ) && !$mmlMathoid ) || $mmlMathoid == "" ) { |
| 80 | $mmlMathoid = $this->fetchMathML( $entry['tex'], "chem", 'mathml' ); |
| 81 | if ( $mmlMathoid && $mmlMathoid != "" ) { |
| 82 | $entry['type'] = "chem"; |
| 83 | } |
| 84 | } |
| 85 | $mmlLaTeXML = $this->fetchMathML( $entry['tex'], $entry['type'], 'latexml' ); |
| 86 | $entryToAdd = [ |
| 87 | "tex" => $entry['tex'], |
| 88 | "type" => $entry['type'], |
| 89 | "mmlMathoid" => $mmlMathoid, |
| 90 | "mmlLaTeXML" => $mmlLaTeXML |
| 91 | ]; |
| 92 | |
| 93 | if ( array_key_exists( "texNew", $entry ) ) { |
| 94 | $entryToAdd["texNew"] = $entry["texNew"]; |
| 95 | } |
| 96 | if ( array_key_exists( "typeC", $entry ) ) { |
| 97 | $entryToAdd["typeC"] = $entry["typeC"]; |
| 98 | } |
| 99 | if ( array_key_exists( "description", $entry ) ) { |
| 100 | $entryToAdd["description"] = $entry["description"]; |
| 101 | } |
| 102 | |
| 103 | $allEntries[] = $entryToAdd; |
| 104 | } catch ( Exception $e ) { |
| 105 | $allEntries[] = [ |
| 106 | "tex" => $entry['tex'], |
| 107 | "type" => $entry['type'], |
| 108 | "mmlMathoid" => "skipped (Exception)", |
| 109 | "mmlLaTeXML" => "skipped (Exception)", |
| 110 | "skipped" => true, |
| 111 | ]; |
| 112 | $this->output( "Exception occurred during rendering:" . $entry['tex'] . " render:" . $e . "\n" ); |
| 113 | } |
| 114 | } |
| 115 | |
| 116 | $this->writeToFile( $this->outputPath, $allEntries ); |
| 117 | } |
| 118 | |
| 119 | /** |
| 120 | * Creates a uniform array of data from files for more convenient parsing. |
| 121 | * @param array $fileData input read from json, format can differ |
| 122 | * @return array uniform array |
| 123 | */ |
| 124 | private function formatInput( $fileData ) { |
| 125 | $inputF = []; |
| 126 | switch ( $this->inputFormat ) { |
| 127 | case 0: |
| 128 | // Example file ParserTest135.json |
| 129 | foreach ( $fileData as $entry ) { |
| 130 | $inputF[] = [ |
| 131 | "tex" => $entry['input'], |
| 132 | "type" => "tex", |
| 133 | ]; |
| 134 | } |
| 135 | break; |
| 136 | case 1: |
| 137 | // Example file TexUtilMMLLookup.json |
| 138 | foreach ( $fileData as $tex => $mml ) { |
| 139 | $inputF[] = [ |
| 140 | "tex" => $tex, |
| 141 | "type" => "tex", |
| 142 | ]; |
| 143 | } |
| 144 | break; |
| 145 | case 2: |
| 146 | // Example file ExportedTexUtilKeys.json |
| 147 | foreach ( $fileData as $tex => $type ) { |
| 148 | $inputF[] = [ |
| 149 | "tex" => $tex, |
| 150 | "type" => $type, |
| 151 | ]; |
| 152 | } |
| 153 | break; |
| 154 | case 3: |
| 155 | // Example file Mhchemv4tex.json |
| 156 | foreach ( $fileData as $group => $cases ) { |
| 157 | foreach ( $cases as $case ) { |
| 158 | $inputF[] = [ |
| 159 | "description" => $group, |
| 160 | "tex" => $case["tex"], |
| 161 | "texNew" => $case["texNew"], |
| 162 | "type" => "chem", |
| 163 | "typeC" => $case["type"] |
| 164 | ]; |
| 165 | } |
| 166 | } |
| 167 | break; |
| 168 | case 4: |
| 169 | // Example file ExamplesNewCommandsMhchem.json |
| 170 | foreach ( $fileData as $entry ) { |
| 171 | $inputF[] = [ |
| 172 | "description" => $entry["description"], |
| 173 | "tex" => $entry['tex'], |
| 174 | "type" => $entry['type'], |
| 175 | ]; |
| 176 | } |
| 177 | break; |
| 178 | } |
| 179 | return $inputF; |
| 180 | } |
| 181 | |
| 182 | /** |
| 183 | * Reads the json file to an object |
| 184 | * @param string $filePath filepath to the json-file |
| 185 | * @return array |
| 186 | */ |
| 187 | private function getJSON( string $filePath ) { |
| 188 | $file = file_get_contents( $filePath ); |
| 189 | $json = json_decode( $file, true ); |
| 190 | return $json; |
| 191 | } |
| 192 | |
| 193 | public function writeToFile( string $fullPath, array $allEntries ): void { |
| 194 | $jsonData = json_encode( $allEntries, JSON_PRETTY_PRINT ); |
| 195 | file_put_contents( $fullPath, $jsonData ); |
| 196 | } |
| 197 | |
| 198 | /** |
| 199 | * Creates a renderer and fetches the generated MathML |
| 200 | * @param string $tex input tex |
| 201 | * @param string $type input type ('tex' or 'chem') |
| 202 | * @param string $renderingMode mode for rendering (latexml, mathml) |
| 203 | * @return string MathML as string |
| 204 | */ |
| 205 | public function fetchMathML( string $tex, string $type, string $renderingMode ): string { |
| 206 | $params = [ "type" => $type ]; |
| 207 | if ( $type == "chem" ) { |
| 208 | $params["chem"] = true; |
| 209 | $tex = "\\ce{ " . $tex . " }"; |
| 210 | } |
| 211 | /** @var MathRenderer $renderer */ |
| 212 | $renderer = MediaWikiServices::getInstance()->get( 'Math.RendererFactory' ) |
| 213 | ->getRenderer( $tex, $params, $renderingMode ); |
| 214 | $renderer->render(); |
| 215 | $mml = $renderer->getMathml(); |
| 216 | return $mml; |
| 217 | } |
| 218 | } |
| 219 | |
| 220 | $maintClass = JsonToMathML::class; |
| 221 | require_once RUN_MAINTENANCE_IF_MAIN; |