Complex classes like BrillTagger often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes. You can also have a look at the cohesion graph to spot any un-connected, or weakly-connected components.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use BrillTagger, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 13 | class BrillTagger |
||
| 14 | { |
||
| 15 | private $dictionary = LEXICON; |
||
| 16 | |||
| 17 | /** |
||
| 18 | * @param $text |
||
| 19 | * @return array |
||
| 20 | */ |
||
| 21 | 21 | public function tag($text) { |
|
| 22 | |||
| 23 | 21 | preg_match_all("/[\w\d\.'%@]+/", $text, $matches); |
|
| 24 | |||
| 25 | 21 | $tags = []; |
|
| 26 | 21 | $i = 0; |
|
| 27 | |||
| 28 | 21 | foreach ($matches[0] as $token) { |
|
| 29 | # default to a common noun |
||
| 30 | 21 | $tags[$i] = ['token' => $token, 'tag' => 'NN']; |
|
| 31 | |||
| 32 | # remove trailing full stops |
||
| 33 | 21 | if (substr(trim($token), -1) == '.') { |
|
| 34 | 12 | $token = preg_replace('/\.+$/', '', $token); |
|
| 35 | } |
||
| 36 | |||
| 37 | # get from dictionary if set |
||
| 38 | 21 | if ($this->tokenExists($token)) { |
|
| 39 | 21 | $tags[$i]['tag'] = $this->dictionary[strtolower($token)][0]; |
|
| 40 | } |
||
| 41 | |||
| 42 | 21 | $tags[$i]['tag'] = $this->transformNumerics($tags[$i]['tag'], $token); |
|
| 43 | |||
| 44 | # Anything that ends 'ly' is an adverb |
||
| 45 | 21 | if ($this->isAdverb($token)) { |
|
| 46 | 1 | $tags[$i]['tag'] = 'RB'; |
|
| 47 | } |
||
| 48 | |||
| 49 | 21 | if ($this->isNoun($tags[$i]['tag'])) { |
|
| 50 | 18 | $tags[$i]['tag'] = $this->transformNoun($tags[$i]['tag'], $token); |
|
| 51 | } |
||
| 52 | |||
| 53 | 21 | if ($i > 0) { |
|
| 54 | 12 | $tags[$i]['tag'] = $this->transformNounToVerb($tags, $i, $token); |
|
| 55 | 12 | $tags[$i]['tag'] = $this->transformVerbToNoun($tags, $i); |
|
| 56 | } |
||
| 57 | |||
| 58 | 21 | $i++; |
|
| 59 | } |
||
| 60 | |||
| 61 | 21 | return $tags; |
|
| 62 | } |
||
| 63 | |||
| 64 | /** |
||
| 65 | * @param string $token |
||
| 66 | * @return bool |
||
| 67 | */ |
||
| 68 | 22 | public function tokenExists($token) { |
|
| 69 | 22 | return isset($this->dictionary[strtolower($token)]); |
|
| 70 | } |
||
| 71 | |||
| 72 | /** |
||
| 73 | * @param string $tag |
||
| 74 | * @return bool |
||
| 75 | */ |
||
| 76 | 21 | public function isNoun($tag) { |
|
| 77 | 21 | return substr(trim($tag), 0, 1) == 'N'; |
|
| 78 | } |
||
| 79 | |||
| 80 | /** |
||
| 81 | * @param string $tag |
||
| 82 | * @return bool |
||
| 83 | */ |
||
| 84 | 13 | public function isSingularNoun($tag) { |
|
| 85 | 13 | return $tag == 'NN'; |
|
| 86 | } |
||
| 87 | |||
| 88 | /** |
||
| 89 | * @param string $tag |
||
| 90 | * @param string $token |
||
| 91 | * @return bool |
||
| 92 | */ |
||
| 93 | 15 | public function isPluralNoun($tag, $token) { |
|
| 96 | |||
| 97 | /** |
||
| 98 | * @param string $tag |
||
| 99 | * @return bool |
||
| 100 | */ |
||
| 101 | 2 | public function isVerb($tag) { |
|
| 102 | 2 | return substr(trim($tag), 0, 2) == 'VB'; |
|
| 103 | } |
||
| 104 | |||
| 105 | /** |
||
| 106 | * @param string $tag |
||
| 107 | * @return bool |
||
| 108 | */ |
||
| 109 | 1 | public function isPronoun($tag) { |
|
| 110 | 1 | return substr(trim($tag), 0, 1) == 'P'; |
|
| 111 | } |
||
| 112 | |||
| 113 | /** |
||
| 114 | * @param string $token |
||
| 115 | * @return bool |
||
| 116 | */ |
||
| 117 | 4 | public function isPastTenseVerb($token) { |
|
| 118 | 4 | return in_array('VBN', $this->dictionary[strtolower($token)]); |
|
| 119 | } |
||
| 120 | |||
| 121 | /** |
||
| 122 | * @param string $token |
||
| 123 | * @return bool |
||
| 124 | */ |
||
| 125 | 4 | public function isPresentTenseVerb($token) { |
|
| 126 | 4 | return in_array('VBZ', $this->dictionary[strtolower($token)]); |
|
| 127 | } |
||
| 128 | |||
| 129 | /** it him me us you 'em thee we'uns |
||
| 130 | * @param string $tag |
||
| 131 | * @return bool |
||
| 132 | */ |
||
| 133 | 1 | public function isAccusativePronoun($tag) { |
|
| 134 | 1 | return $tag === 'PPO'; |
|
| 135 | } |
||
| 136 | |||
| 137 | /** it he she thee |
||
| 138 | * @param string $tag |
||
| 139 | * @return bool |
||
| 140 | */ |
||
| 141 | 1 | public function isThirdPersonPronoun($tag) { |
|
| 142 | 1 | return $tag === 'PPS'; |
|
| 143 | } |
||
| 144 | |||
| 145 | /** they we I you ye thou you'uns |
||
| 146 | * @param string $tag |
||
| 147 | * @return bool |
||
| 148 | */ |
||
| 149 | 1 | public function isSingularPersonalPronoun($tag) { |
|
| 150 | 1 | return $tag === 'PPSS'; |
|
| 151 | } |
||
| 152 | |||
| 153 | /** itself himself myself yourself herself oneself ownself |
||
| 154 | * @param string $tag |
||
| 155 | * @return bool |
||
| 156 | */ |
||
| 157 | 1 | public function isSingularReflexivePronoun($tag) { |
|
| 160 | |||
| 161 | /** themselves ourselves yourselves |
||
| 162 | * @param string $tag |
||
| 163 | * @return bool |
||
| 164 | */ |
||
| 165 | 1 | public function isPluralReflexivePronoun($tag) { |
|
| 168 | |||
| 169 | /** ours mine his her/hers their/theirs our its my your/yours out thy thine |
||
| 170 | * @param string $tag |
||
| 171 | * @return bool |
||
| 172 | */ |
||
| 173 | 1 | public function isPossessivePronoun($tag) { |
|
| 174 | 1 | return in_array($tag, ['PP$$', 'PP$']); |
|
| 175 | } |
||
| 176 | |||
| 177 | /** |
||
| 178 | * @param string $token |
||
| 179 | * @return bool |
||
| 180 | */ |
||
| 181 | 19 | public function isAdjective($token) { |
|
| 184 | |||
| 185 | /** |
||
| 186 | * @param string $token |
||
| 187 | * @return bool |
||
| 188 | */ |
||
| 189 | 19 | public function isGerund($token) { |
|
| 192 | |||
| 193 | /** |
||
| 194 | * @param string $token |
||
| 195 | * @return bool |
||
| 196 | */ |
||
| 197 | 19 | public function isPastParticiple($token) { |
|
| 200 | |||
| 201 | /** |
||
| 202 | * @param string $token |
||
| 203 | * @return bool |
||
| 204 | */ |
||
| 205 | 22 | public function isAdverb($token) { |
|
| 208 | |||
| 209 | /** Common noun to adj. if it ends with 'al', |
||
| 210 | * to gerund if 'ing', to past tense if 'ed' |
||
| 211 | * |
||
| 212 | * @param string $tag |
||
| 213 | * @param string $token |
||
| 214 | * @return string |
||
| 215 | */ |
||
| 216 | 18 | public function transformNoun($tag, $token) { |
|
| 237 | |||
| 238 | /** |
||
| 239 | * @param array $tags |
||
| 240 | * @param int $i |
||
| 241 | * @param string $token |
||
| 242 | * @return mixed |
||
| 243 | */ |
||
| 244 | 12 | public function transformNounToVerb($tags, $i, $token) { |
|
| 264 | |||
| 265 | /** |
||
| 266 | * @param array $tags |
||
| 267 | * @param int $i |
||
| 268 | * @return mixed |
||
| 269 | */ |
||
| 270 | 12 | public function transformVerbToNoun($tags, $i) { |
|
| 278 | |||
| 279 | /** |
||
| 280 | * @param string $tag |
||
| 281 | * @param string $token |
||
| 282 | * @return string |
||
| 283 | */ |
||
| 284 | 21 | public function transformNumerics($tag, $token) { |
|
| 302 | } |
||
| 303 |