Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
100.00% |
106 / 106 |
|
100.00% |
5 / 5 |
CRAP | |
100.00% |
1 / 1 |
| GenerateEquivset | |
100.00% |
106 / 106 |
|
100.00% |
5 / 5 |
29 | |
100.00% |
1 / 1 |
| __construct | |
100.00% |
3 / 3 |
|
100.00% |
1 / 1 |
3 | |||
| configure | |
100.00% |
4 / 4 |
|
100.00% |
1 / 1 |
1 | |||
| execute | |
100.00% |
87 / 87 |
|
100.00% |
1 / 1 |
20 | |||
| compareCodePoints | |
100.00% |
5 / 5 |
|
100.00% |
1 / 1 |
3 | |||
| generatePHP | |
100.00% |
7 / 7 |
|
100.00% |
1 / 1 |
2 | |||
| 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 | |
| 19 | namespace Wikimedia\Equivset\Command; |
| 20 | |
| 21 | use RuntimeException; |
| 22 | use Symfony\Component\Console\Command\Command; |
| 23 | use Symfony\Component\Console\Input\InputInterface; |
| 24 | use Symfony\Component\Console\Output\OutputInterface; |
| 25 | |
| 26 | /** |
| 27 | * Generate Equivset Command. |
| 28 | */ |
| 29 | class GenerateEquivset extends Command { |
| 30 | |
| 31 | /** |
| 32 | * @var string |
| 33 | */ |
| 34 | protected string $dataDir; |
| 35 | |
| 36 | /** |
| 37 | * @var string |
| 38 | */ |
| 39 | protected string $distDir; |
| 40 | |
| 41 | /** |
| 42 | * Generate Equivset |
| 43 | * |
| 44 | * @param string $dataDir Data Directory |
| 45 | * @param string $distDir Distribution Directory |
| 46 | */ |
| 47 | public function __construct( string $dataDir = '', string $distDir = '' ) { |
| 48 | parent::__construct(); |
| 49 | |
| 50 | $this->dataDir = $dataDir ?: __DIR__ . '/../../data'; |
| 51 | $this->distDir = $distDir ?: __DIR__ . '/../../dist'; |
| 52 | } |
| 53 | |
| 54 | /** {@inheritdoc} */ |
| 55 | protected function configure() { |
| 56 | $this->setName( 'generate-equivset' ); |
| 57 | $this->setDescription( |
| 58 | 'Generate the JSON, PHP, and plain text versions of the equivset in `./dist`' |
| 59 | ); |
| 60 | } |
| 61 | |
| 62 | /** {@inheritdoc} */ |
| 63 | public function execute( InputInterface $input, OutputInterface $output ) { |
| 64 | // phpcs:ignore Generic.PHP.NoSilencedErrors |
| 65 | $fp = @fopen( $this->dataDir . '/equivset.in', 'rb' ); |
| 66 | if ( $fp === false ) { |
| 67 | throw new RuntimeException( "Unable to open equivset.in" ); |
| 68 | } |
| 69 | |
| 70 | $lineNum = 0; |
| 71 | $setsByChar = []; |
| 72 | $sets = []; |
| 73 | $exitStatus = 0; |
| 74 | $lastChar = null; |
| 75 | |
| 76 | // phpcs:ignore Generic.CodeAnalysis.AssignmentInCondition.FoundInWhileCondition |
| 77 | while ( ( $line = fgets( $fp ) ) !== false ) { |
| 78 | $lineNum++; |
| 79 | $line = trim( $line ); |
| 80 | |
| 81 | # Filter comments |
| 82 | if ( !$line || $line[0] == '#' ) { |
| 83 | continue; |
| 84 | } |
| 85 | |
| 86 | # Process line |
| 87 | if ( !preg_match( |
| 88 | '/^(?P<hexleft>[0-9A-F]+) +(?P<charleft>.) +=> +' . |
| 89 | '(?:(?P<hexright>[0-9A-F]+) +(?P<charright>.)|(?P<invisible>invisible)|(?P<space>space))$/u', |
| 90 | $line, $m, PREG_UNMATCHED_AS_NULL |
| 91 | ) |
| 92 | ) { |
| 93 | $output->writeln( "<error>Error: invalid entry at line $lineNum: $line</error>" ); |
| 94 | $exitStatus = 1; |
| 95 | continue; |
| 96 | } |
| 97 | $error = false; |
| 98 | |
| 99 | if ( mb_chr( hexdec( $m['hexleft'] ) ) !== $m['charleft'] ) { |
| 100 | $actual = strtoupper( dechex( mb_ord( $m['charleft'] ) ) ); |
| 101 | $output->writeln( "<error>Error: left number ({$m['hexleft']}) does not match left " . |
| 102 | "character ($actual) at line $lineNum: $line</error>" ); |
| 103 | $error = true; |
| 104 | } |
| 105 | if ( isset( $m['invisible'] ) ) { |
| 106 | $m['charright'] = ''; |
| 107 | } elseif ( isset( $m['space'] ) ) { |
| 108 | $m['charright'] = ' '; |
| 109 | } elseif ( mb_chr( hexdec( $m['hexright'] ) ) !== $m['charright'] ) { |
| 110 | $actual = strtoupper( dechex( mb_ord( $m['charright'] ) ) ); |
| 111 | $output->writeln( "<error>Error: right number ({$m['hexright']}) does not match right " . |
| 112 | "character ($actual) at line $lineNum: $line</error>" ); |
| 113 | $error = true; |
| 114 | } |
| 115 | if ( $m['charleft'] === $m['charright'] ) { |
| 116 | $output->writeln( "<error>Error: {$m['hexright']} maps to itself</error>" ); |
| 117 | $error = true; |
| 118 | } |
| 119 | if ( isset( $setsByChar[$m['charleft']] ) ) { |
| 120 | $output->writeln( "<error>Error: Duplicate character ({$m['charleft']}) " . |
| 121 | "at line $lineNum: $line</error>" ); |
| 122 | $error = true; |
| 123 | } |
| 124 | if ( $lastChar !== null && $m['charleft'] < $lastChar ) { |
| 125 | $output->writeln( "<error>Error: Characters not in order based on hex-value ({$m['charleft']}) " . |
| 126 | "at line $lineNum: $line</error>" ); |
| 127 | $error = true; |
| 128 | } else { |
| 129 | $lastChar = $m['charleft']; |
| 130 | } |
| 131 | if ( $error ) { |
| 132 | $exitStatus = 1; |
| 133 | continue; |
| 134 | } |
| 135 | |
| 136 | # Find the set for the right character, add a new one if necessary |
| 137 | $setName = $setsByChar[$m['charright']] ?? $m['charright']; |
| 138 | |
| 139 | // Circle detected, one edge in every circle is redundant and can just be ignored |
| 140 | if ( $setName === $m['charleft'] ) { |
| 141 | continue; |
| 142 | } |
| 143 | |
| 144 | $sets[$setName] ??= [ $setName ]; |
| 145 | |
| 146 | // When a mapping between two chars exists before one of them gets the final set, a merge is needed |
| 147 | if ( isset( $sets[$m['charleft']] ) ) { |
| 148 | foreach ( $sets[$m['charleft']] as $char ) { |
| 149 | $setsByChar[$char] = $setName; |
| 150 | $sets[$setName][] = $char; |
| 151 | } |
| 152 | unset( $sets[$m['charleft']] ); |
| 153 | } else { |
| 154 | $setsByChar[$m['charleft']] = $setName; |
| 155 | $sets[$setName][] = $m['charleft']; |
| 156 | } |
| 157 | } |
| 158 | |
| 159 | $header = [ |
| 160 | 'This file is generated by `bin/console generate-equivset`', |
| 161 | 'It contains a map of characters, encoded in UTF-8, such that running', |
| 162 | 'strtr() on a string with this map will cause confusable characters to', |
| 163 | 'be reduced to a canonical representation.', |
| 164 | ]; |
| 165 | |
| 166 | // JSON |
| 167 | $data = json_encode( |
| 168 | [ '_readme' => implode( ' ', $header ) ] + $setsByChar, |
| 169 | JSON_UNESCAPED_SLASHES | JSON_PRETTY_PRINT | JSON_UNESCAPED_UNICODE |
| 170 | ); |
| 171 | $data = preg_replace( '/^ +/m', "\t", $data ); |
| 172 | file_put_contents( $this->distDir . '/equivset.json', $data ); |
| 173 | |
| 174 | // PHP file |
| 175 | file_put_contents( $this->distDir . '/equivset.php', self::generatePHP( $setsByChar, $header ) ); |
| 176 | |
| 177 | // Text File. |
| 178 | uksort( $sets, [ self::class, 'compareCodePoints' ] ); |
| 179 | touch( $this->distDir . '/equivset.txt' ); |
| 180 | $textFile = fopen( $this->distDir . '/equivset.txt', 'w' ); |
| 181 | foreach ( $sets as $members ) { |
| 182 | $setName = array_shift( $members ); |
| 183 | usort( $members, [ self::class, 'compareCodePoints' ] ); |
| 184 | fwrite( $textFile, $setName . ' ' . implode( ' ', $members ) . "\n" ); |
| 185 | } |
| 186 | fclose( $textFile ); |
| 187 | |
| 188 | if ( $exitStatus > 0 ) { |
| 189 | $output->writeln( '<error>Finished with errors</error>' ); |
| 190 | } else { |
| 191 | $output->writeln( '<info>Finished</info>' ); |
| 192 | } |
| 193 | |
| 194 | return $exitStatus; |
| 195 | } |
| 196 | |
| 197 | private static function compareCodePoints( string $a, string $b ): int { |
| 198 | if ( $a === '' ) { |
| 199 | return -1; |
| 200 | } elseif ( $b === '' ) { |
| 201 | return 1; |
| 202 | } |
| 203 | return mb_ord( $a ) - mb_ord( $b ); |
| 204 | } |
| 205 | |
| 206 | /** |
| 207 | * @param string[] $data |
| 208 | * @param string[] $header |
| 209 | * @return string |
| 210 | */ |
| 211 | private static function generatePHP( array $data, array $header ): string { |
| 212 | $s = "<?php\n" |
| 213 | . "// " . implode( "\n// ", $header ) . "\n" |
| 214 | . "return [\n"; |
| 215 | foreach ( $data as $key => $value ) { |
| 216 | $s .= "\t" . var_export( (string)$key, true ) . ' => ' . var_export( $value, true ) . ",\n"; |
| 217 | } |
| 218 | $s .= "];\n"; |
| 219 | return $s; |
| 220 | } |
| 221 | } |