Complex classes like Tokenizer 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 Tokenizer, and based on these observations, apply Extract Interface, too.
1 | <?php |
||
10 | class Tokenizer |
||
11 | { |
||
12 | protected $source; |
||
13 | protected $pos = 0; |
||
14 | protected $line = 1; |
||
15 | protected $lineStart = 0; |
||
16 | |||
17 | /** @var Token */ |
||
18 | protected $lookAhead; |
||
19 | |||
20 | 44 | public function setSource($source) |
|
21 | { |
||
22 | 44 | $this->source = $source; |
|
23 | 44 | $this->lookAhead = $this->next(); |
|
24 | 44 | } |
|
25 | |||
26 | 44 | protected function next() |
|
27 | { |
||
28 | 44 | $this->skipWhitespace(); |
|
29 | |||
30 | 44 | $line = $this->line; |
|
31 | 44 | $lineStart = $this->lineStart; |
|
32 | |||
33 | /** @var Token $token */ |
||
34 | 44 | $token = $this->scan(); |
|
35 | |||
36 | 44 | $token->setLine($line); |
|
37 | 44 | $token->setColumn($this->pos - $lineStart); |
|
38 | |||
39 | 44 | return $token; |
|
40 | } |
||
41 | |||
42 | 44 | protected function skipWhitespace() |
|
43 | { |
||
44 | 44 | while ($this->pos < strlen($this->source)) { |
|
45 | 44 | $ch = $this->source[$this->pos]; |
|
46 | 44 | if ($ch === ' ' || $ch === "\t") { |
|
47 | 42 | $this->pos++; |
|
48 | 44 | } elseif($ch === '#') { |
|
49 | 1 | $this->pos++; |
|
50 | while ( |
||
51 | 1 | $this->pos < strlen($this->source) && |
|
52 | 1 | ($code = ord($this->source[$this->pos])) && |
|
53 | 1 | $code !== 10 && $code !== 13 && $code !== 0x2028 && $code !== 0x2029 |
|
54 | 1 | ) { |
|
55 | 1 | $this->pos++; |
|
56 | 1 | } |
|
57 | 44 | } elseif ($ch === "\r") { |
|
58 | $this->pos++; |
||
59 | if ($this->source[$this->pos] === "\n") { |
||
60 | $this->pos++; |
||
61 | } |
||
62 | $this->line++; |
||
63 | $this->lineStart = $this->pos; |
||
64 | 44 | } elseif ($ch === "\n") { |
|
65 | 18 | $this->pos++; |
|
66 | 18 | $this->line++; |
|
67 | 18 | $this->lineStart = $this->pos; |
|
68 | 18 | } else { |
|
69 | 44 | break; |
|
70 | } |
||
71 | 42 | } |
|
72 | 44 | } |
|
73 | |||
74 | /** |
||
75 | * @return Token |
||
76 | */ |
||
77 | 44 | protected function scan() |
|
78 | { |
||
79 | 44 | if ($this->pos >= strlen($this->source)) { |
|
80 | 37 | return new Token(Token::TYPE_END); |
|
81 | } |
||
82 | |||
83 | 44 | $ch = $this->source[$this->pos]; |
|
84 | switch ($ch) { |
||
85 | 44 | case Token::TYPE_LPAREN: |
|
86 | 19 | ++$this->pos; |
|
87 | |||
88 | 19 | return new Token(Token::TYPE_LPAREN); |
|
89 | 44 | case Token::TYPE_RPAREN: |
|
90 | 15 | ++$this->pos; |
|
91 | |||
92 | 15 | return new Token(Token::TYPE_RPAREN); |
|
93 | 44 | case Token::TYPE_LBRACE: |
|
94 | 43 | ++$this->pos; |
|
95 | |||
96 | 43 | return new Token(Token::TYPE_LBRACE); |
|
97 | 44 | case Token::TYPE_RBRACE: |
|
98 | 37 | ++$this->pos; |
|
99 | |||
100 | 37 | return new Token(Token::TYPE_RBRACE); |
|
101 | 43 | case Token::TYPE_LT: |
|
102 | 1 | ++$this->pos; |
|
103 | |||
104 | 1 | return new Token(Token::TYPE_LT); |
|
105 | 43 | case Token::TYPE_GT: |
|
106 | 1 | ++$this->pos; |
|
107 | |||
108 | 1 | return new Token(Token::TYPE_GT); |
|
109 | 43 | case Token::TYPE_AMP: |
|
110 | ++$this->pos; |
||
111 | |||
112 | return new Token(Token::TYPE_AMP); |
||
113 | 43 | case Token::TYPE_COMMA: |
|
114 | 23 | ++$this->pos; |
|
115 | |||
116 | 23 | return new Token(Token::TYPE_COMMA); |
|
117 | 43 | case Token::TYPE_LSQUARE_BRACE: |
|
118 | 4 | ++$this->pos; |
|
119 | |||
120 | 4 | return new Token(Token::TYPE_LSQUARE_BRACE); |
|
121 | 43 | case Token::TYPE_RSQUARE_BRACE: |
|
122 | 3 | ++$this->pos; |
|
123 | |||
124 | 3 | return new Token(Token::TYPE_RSQUARE_BRACE); |
|
125 | 43 | case Token::TYPE_COLON: |
|
126 | 24 | ++$this->pos; |
|
127 | |||
128 | 24 | return new Token(Token::TYPE_COLON); |
|
129 | |||
130 | 43 | case Token::TYPE_POINT: |
|
131 | 8 | if ($this->checkFragment()) { |
|
132 | 8 | return new Token(Token::TYPE_FRAGMENT_REFERENCE); |
|
133 | } |
||
134 | |||
135 | break; |
||
136 | |||
137 | 43 | case Token::TYPE_VARIABLE: |
|
138 | 4 | ++$this->pos; |
|
139 | |||
140 | 4 | return new Token(Token::TYPE_VARIABLE); |
|
141 | } |
||
142 | |||
143 | 43 | if ($ch === '_' || 'a' <= $ch && $ch <= 'z' || 'A' <= $ch && $ch <= 'Z') { |
|
144 | 43 | return $this->scanWord(); |
|
145 | } |
||
146 | |||
147 | 11 | if ($ch === '-' || '0' <= $ch && $ch <= '9') { |
|
148 | 7 | return $this->scanNumber(); |
|
149 | } |
||
150 | |||
151 | 8 | if ($ch === '"') { |
|
152 | 8 | return $this->scanString(); |
|
153 | } |
||
154 | |||
155 | throw $this->createIllegal(); |
||
156 | } |
||
157 | |||
158 | 8 | protected function checkFragment() |
|
159 | { |
||
160 | 8 | $this->pos++; |
|
161 | 8 | $ch = $this->source[$this->pos]; |
|
162 | |||
163 | 8 | $this->pos++; |
|
164 | 8 | $nextCh = $this->source[$this->pos]; |
|
165 | |||
166 | 8 | $isset = $ch == Token::TYPE_POINT && $nextCh == Token::TYPE_POINT; |
|
167 | |||
168 | 8 | if ($isset) { |
|
169 | 8 | $this->pos++; |
|
170 | |||
171 | 8 | return true; |
|
172 | } |
||
173 | |||
174 | return false; |
||
175 | } |
||
176 | |||
177 | 43 | protected function scanWord() |
|
178 | { |
||
179 | 43 | $start = $this->pos; |
|
180 | 43 | $this->pos++; |
|
181 | |||
182 | 43 | while ($this->pos < strlen($this->source)) { |
|
183 | 43 | $ch = $this->source[$this->pos]; |
|
184 | |||
185 | 43 | if ($ch === '_' || $ch === '$' || 'a' <= $ch && $ch <= ('z') || 'A' <= $ch && $ch <= 'Z' || '0' <= $ch && $ch <= '9') { |
|
186 | 43 | $this->pos++; |
|
187 | 43 | } else { |
|
188 | 42 | break; |
|
189 | } |
||
190 | 43 | } |
|
191 | |||
192 | 43 | $value = substr($this->source, $start, $this->pos - $start); |
|
193 | |||
194 | 43 | return new Token($this->getKeyword($value), $value); |
|
195 | } |
||
196 | |||
197 | 43 | protected function getKeyword($name) |
|
198 | { |
||
199 | switch ($name) { |
||
200 | 43 | case 'null': |
|
201 | 2 | return Token::TYPE_NULL; |
|
202 | |||
203 | 43 | case 'true': |
|
204 | 2 | return Token::TYPE_TRUE; |
|
205 | |||
206 | 43 | case 'false': |
|
207 | 1 | return Token::TYPE_FALSE; |
|
208 | |||
209 | 43 | case 'query': |
|
210 | 11 | return Token::TYPE_QUERY; |
|
211 | |||
212 | 42 | case 'fragment': |
|
213 | 3 | return Token::TYPE_FRAGMENT; |
|
214 | |||
215 | 42 | case 'mutation': |
|
216 | 5 | return Token::TYPE_MUTATION; |
|
217 | |||
218 | 42 | case 'on': |
|
219 | 6 | return Token::TYPE_ON; |
|
220 | } |
||
221 | |||
222 | 42 | return Token::TYPE_IDENTIFIER; |
|
223 | } |
||
224 | |||
225 | 7 | protected function scanNumber() |
|
247 | |||
248 | 7 | protected function skipInteger() |
|
249 | { |
||
250 | 7 | $start = $this->pos; |
|
251 | |||
252 | 7 | while ($this->pos < strlen($this->source)) { |
|
253 | 7 | $ch = $this->source[$this->pos]; |
|
254 | 7 | if ('0' <= $ch && $ch <= '9') { |
|
255 | 7 | $this->pos++; |
|
256 | 7 | } else { |
|
257 | 7 | break; |
|
258 | } |
||
259 | 7 | } |
|
260 | |||
261 | 7 | if ($this->pos - $start === 0) { |
|
262 | throw $this->createIllegal(); |
||
263 | } |
||
264 | 7 | } |
|
265 | |||
266 | protected function createIllegal() |
||
267 | { |
||
268 | return $this->pos < strlen($this->source) |
||
269 | ? $this->createError("Unexpected {$this->source[$this->pos]}") |
||
270 | : $this->createError('Unexpected end of input'); |
||
271 | } |
||
272 | |||
273 | 2 | protected function createError($message) |
|
277 | |||
278 | 2 | protected function getColumn() |
|
282 | |||
283 | 8 | protected function scanString() |
|
284 | { |
||
285 | 8 | $this->pos++; |
|
286 | |||
287 | 8 | $value = ''; |
|
288 | 8 | while ($this->pos < strlen($this->source)) { |
|
289 | 8 | $ch = $this->source[$this->pos]; |
|
290 | 8 | if ($ch === '"') { |
|
291 | 8 | $this->pos++; |
|
292 | |||
293 | 8 | return new Token(Token::TYPE_STRING, $value); |
|
294 | } |
||
295 | |||
296 | 8 | if ($ch === "\r" || $ch === "\n") { |
|
297 | break; |
||
298 | } |
||
299 | |||
300 | 8 | $value .= $ch; |
|
301 | 8 | $this->pos++; |
|
302 | 8 | } |
|
303 | |||
304 | throw $this->createIllegal(); |
||
305 | } |
||
306 | |||
307 | 44 | protected function end() |
|
311 | |||
312 | 44 | protected function peek() |
|
316 | |||
317 | 43 | protected function lex() |
|
318 | { |
||
319 | 43 | $prev = $this->lookAhead; |
|
320 | 43 | $this->lookAhead = $this->next(); |
|
321 | |||
322 | 43 | return $prev; |
|
323 | } |
||
324 | |||
325 | 6 | protected function createUnexpected(Token $token) |
|
326 | { |
||
327 | 6 | switch ($token->getType()) { |
|
340 | } |