Complex classes like Lexer 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 Lexer, and based on these observations, apply Extract Interface, too.
1 | <?php |
||
19 | class Lexer |
||
20 | { |
||
21 | /** |
||
22 | * Whitespace tokens. |
||
23 | */ |
||
24 | private const SPACE_TOKENS = [ |
||
25 | Token::T_SPACE, |
||
26 | Token::T_COMMENT, |
||
27 | Token::T_NEWLINE, |
||
28 | ]; |
||
29 | |||
30 | /** |
||
31 | * Tag map. |
||
32 | */ |
||
33 | private const TAG_MAP = [ |
||
34 | ':' => Token::T_COLON, |
||
35 | ';' => Token::T_SEMICOLON, |
||
36 | '$' => Token::T_DOLLAR, |
||
37 | '%%' => Token::T_MARK, |
||
38 | '%{' => Token::T_BEGIN_INC, |
||
39 | '%}' => Token::T_END_INC, |
||
40 | '%token' => Token::T_TOKEN, |
||
41 | '%term' => Token::T_TOKEN, |
||
42 | '%left' => Token::T_LEFT, |
||
43 | '%right' => Token::T_RIGHT, |
||
44 | '%nonassoc' => Token::T_NON_ASSOC, |
||
45 | '%prec' => Token::T_PRECTOK, |
||
46 | '%type' => Token::T_TYPE, |
||
47 | '%union' => Token::T_UNION, |
||
48 | '%start' => Token::T_START, |
||
49 | '%expect' => Token::T_EXPECT, |
||
50 | '%pure_parser' => Token::T_PURE_PARSER, |
||
51 | ]; |
||
52 | |||
53 | /** |
||
54 | * @var string |
||
55 | */ |
||
56 | protected $buffer; |
||
57 | |||
58 | /** |
||
59 | * @var string |
||
60 | */ |
||
61 | protected $fileName; |
||
62 | |||
63 | /** |
||
64 | * @var int |
||
65 | */ |
||
66 | protected $line; |
||
67 | |||
68 | /** |
||
69 | * @var int |
||
70 | */ |
||
71 | protected $offset; |
||
72 | |||
73 | /** |
||
74 | * @var Token |
||
75 | */ |
||
76 | protected $currentToken; |
||
77 | |||
78 | /** |
||
79 | * @var Token |
||
80 | */ |
||
81 | protected $backToken; |
||
82 | |||
83 | /** |
||
84 | * @var string |
||
85 | */ |
||
86 | protected $backChar; |
||
87 | |||
88 | /** |
||
89 | * @var bool |
||
90 | */ |
||
91 | protected $prevIsDollar; |
||
92 | |||
93 | /** |
||
94 | * @var string |
||
95 | */ |
||
96 | protected $char; |
||
97 | |||
98 | /** |
||
99 | * @var string |
||
100 | */ |
||
101 | protected $value; |
||
102 | |||
103 | /** |
||
104 | * @param string $code |
||
105 | * @param string $filename |
||
106 | */ |
||
107 | public function startLexing(string $code, string $filename = '') |
||
108 | { |
||
109 | $this->buffer = $code; |
||
110 | $this->fileName = $filename; |
||
111 | |||
112 | $this->reset(); |
||
113 | } |
||
114 | |||
115 | /** |
||
116 | * @return void |
||
117 | */ |
||
118 | protected function reset() |
||
119 | { |
||
120 | $this->line = 1; |
||
121 | $this->offset = 0; |
||
122 | $this->backChar = null; |
||
123 | $this->backToken = null; |
||
124 | $this->prevIsDollar = false; |
||
125 | } |
||
126 | |||
127 | /** |
||
128 | * @throws LexingException |
||
129 | * @throws ParseException |
||
130 | * |
||
131 | * @return Token |
||
132 | */ |
||
133 | public function getToken(): Token |
||
134 | { |
||
135 | $this->currentToken = $this->getRawToken(); |
||
136 | |||
137 | while (\in_array($this->currentToken->getType(), self::SPACE_TOKENS)) { |
||
138 | $this->currentToken = $this->getRawToken(); |
||
139 | } |
||
140 | |||
141 | return $this->currentToken; |
||
142 | } |
||
143 | |||
144 | /** |
||
145 | * @throws LexingException |
||
146 | */ |
||
147 | public function ungetToken() |
||
148 | { |
||
149 | if ($this->backToken !== null) { |
||
150 | throw new LexingException('Too many ungetToken calls'); |
||
151 | } |
||
152 | |||
153 | $this->backToken = $this->currentToken; |
||
154 | } |
||
155 | |||
156 | /** |
||
157 | * @throws LexingException |
||
158 | * @throws ParseException |
||
159 | * |
||
160 | * @return Token |
||
161 | */ |
||
162 | public function peek(): Token |
||
163 | { |
||
164 | $result = $this->getToken(); |
||
165 | $this->ungetToken(); |
||
166 | |||
167 | return $result; |
||
168 | } |
||
169 | |||
170 | /** |
||
171 | * @throws LexingException |
||
172 | * @throws ParseException |
||
173 | * |
||
174 | * @return Token |
||
175 | */ |
||
176 | public function getRawToken() |
||
177 | { |
||
178 | if ($this->backToken !== null) { |
||
179 | $this->currentToken = $this->backToken; |
||
180 | $this->backToken = null; |
||
181 | |||
182 | return $this->currentToken; |
||
183 | } |
||
184 | |||
185 | $this->char = $this->getChar(); |
||
186 | $this->value = ''; |
||
187 | |||
188 | switch (true) { |
||
189 | case $this->isWhitespace(): |
||
190 | return $this->token(Token::T_SPACE, $this->value); |
||
191 | case $this->isNewline(): |
||
192 | return $this->token(Token::T_NEWLINE, $this->value); |
||
193 | case $this->isComment(): |
||
194 | return $this->token(Token::T_COMMENT, $this->value); |
||
195 | case $this->isEof(): |
||
196 | return $this->token(Token::T_EOF, $this->value); |
||
197 | } |
||
198 | |||
199 | $tag = $this->detectToken(); |
||
200 | |||
201 | switch (true) { |
||
202 | case isset(self::TAG_MAP[$this->value]): |
||
203 | return $this->token(self::TAG_MAP[$this->value], $this->value); |
||
204 | default: |
||
205 | return $this->token($tag, $this->value); |
||
206 | } |
||
207 | } |
||
208 | |||
209 | /** |
||
210 | * @throws LexingException |
||
211 | * |
||
212 | * @return bool |
||
213 | */ |
||
214 | protected function isWhitespace(): bool |
||
215 | { |
||
216 | if (Utils::isWhite($this->char)) { |
||
217 | while (Utils::isWhite($this->char)) { |
||
218 | $this->value .= $this->char; |
||
219 | $this->char = $this->getChar(); |
||
220 | } |
||
221 | $this->ungetChar($this->char); |
||
222 | |||
223 | return true; |
||
224 | } |
||
225 | |||
226 | return false; |
||
227 | } |
||
228 | |||
229 | /** |
||
230 | * @return bool |
||
231 | */ |
||
232 | protected function isNewline(): bool |
||
233 | { |
||
234 | if ($this->char === "\n") { |
||
235 | $this->value = $this->char; |
||
236 | |||
237 | return true; |
||
238 | } |
||
239 | |||
240 | return false; |
||
241 | } |
||
242 | |||
243 | /** |
||
244 | * @throws LexingException |
||
245 | * @throws ParseException |
||
246 | * |
||
247 | * @return bool |
||
248 | */ |
||
249 | protected function isComment(): bool |
||
292 | |||
293 | /** |
||
294 | * @return bool |
||
295 | */ |
||
296 | protected function isEof(): bool |
||
306 | |||
307 | /** |
||
308 | * @throws LexingException |
||
309 | * @throws ParseException |
||
310 | * |
||
311 | * @return int |
||
312 | */ |
||
313 | protected function detectToken() |
||
393 | |||
394 | /** |
||
395 | * @return string |
||
396 | */ |
||
397 | protected function getChar(): string |
||
418 | |||
419 | /** |
||
420 | * @param string $char |
||
421 | * |
||
422 | * @throws LexingException |
||
423 | */ |
||
424 | protected function ungetChar(string $char) |
||
436 | |||
437 | /** |
||
438 | * @param int $type |
||
439 | * @param string $value |
||
440 | * |
||
441 | * @return Token |
||
442 | */ |
||
443 | protected function token(int $type, string $value): Token |
||
447 | } |
||
448 |
When comparing the result of a bit operation, we suggest to add explicit parenthesis and not to rely on PHP’s built-in operator precedence to ensure the code behaves as intended and to make it more readable.
Let’s take a look at these examples: