Complex classes like DocParser 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 DocParser, and based on these observations, apply Extract Interface, too.
1 | <?php |
||
23 | final class DocParser implements DocParserInterface |
||
24 | { |
||
25 | /** |
||
26 | * @var EnvironmentInterface |
||
27 | */ |
||
28 | private $environment; |
||
29 | |||
30 | /** |
||
31 | * @var InlineParserEngine |
||
32 | */ |
||
33 | private $inlineParserEngine; |
||
34 | |||
35 | /** |
||
36 | * @var int|float |
||
37 | */ |
||
38 | private $maxNestingLevel; |
||
39 | |||
40 | /** |
||
41 | * @param EnvironmentInterface $environment |
||
42 | */ |
||
43 | 2070 | public function __construct(EnvironmentInterface $environment) |
|
49 | |||
50 | /** |
||
51 | * @param string $input |
||
52 | * |
||
53 | * @return string[] |
||
54 | */ |
||
55 | 2061 | private function preProcessInput(string $input): array |
|
56 | { |
||
57 | 2061 | $lines = \preg_split('/\r\n|\n|\r/', $input); |
|
58 | |||
59 | // Remove any newline which appears at the very end of the string. |
||
60 | // We've already split the document by newlines, so we can simply drop |
||
61 | // any empty element which appears on the end. |
||
62 | 2061 | if (\end($lines) === '') { |
|
63 | 2013 | \array_pop($lines); |
|
64 | } |
||
65 | |||
66 | 2061 | return $lines; |
|
67 | } |
||
68 | |||
69 | /** |
||
70 | * @param string $input |
||
71 | * |
||
72 | * @return Document |
||
73 | */ |
||
74 | 2061 | public function parse(string $input): Document |
|
75 | { |
||
76 | 2061 | $context = new Context(new Document(), $this->environment); |
|
77 | |||
78 | 2061 | $lines = $this->preProcessInput($input); |
|
79 | 2061 | foreach ($lines as $line) { |
|
80 | 2061 | $context->setNextLine($line); |
|
81 | 2061 | $this->incorporateLine($context); |
|
82 | } |
||
83 | |||
84 | 2061 | $lineCount = \count($lines); |
|
85 | 2061 | while ($tip = $context->getTip()) { |
|
86 | 2061 | $tip->finalize($context, $lineCount); |
|
87 | } |
||
88 | |||
89 | 2061 | $this->processInlines($context); |
|
90 | |||
91 | 2061 | $this->processDocument($context); |
|
92 | |||
93 | 2061 | return $context->getDocument(); |
|
94 | } |
||
95 | |||
96 | 2061 | private function incorporateLine(ContextInterface $context) |
|
97 | { |
||
98 | 2061 | $context->getBlockCloser()->resetTip(); |
|
99 | 2061 | $context->setBlocksParsed(false); |
|
100 | |||
101 | 2061 | $cursor = new Cursor($context->getLine()); |
|
102 | |||
103 | 2061 | $this->resetContainer($context, $cursor); |
|
104 | 2061 | $context->getBlockCloser()->setLastMatchedContainer($context->getContainer()); |
|
105 | |||
106 | 2061 | $this->parseBlocks($context, $cursor); |
|
107 | |||
108 | // What remains at the offset is a text line. Add the text to the appropriate container. |
||
109 | // First check for a lazy paragraph continuation: |
||
110 | 2061 | if ($this->handleLazyParagraphContinuation($context, $cursor)) { |
|
111 | 36 | return; |
|
112 | } |
||
113 | |||
114 | // not a lazy continuation |
||
115 | // finalize any blocks not matched |
||
116 | 2061 | $context->getBlockCloser()->closeUnmatchedBlocks(); |
|
117 | |||
118 | // Determine whether the last line is blank, updating parents as needed |
||
119 | 2061 | $this->setAndPropagateLastLineBlank($context, $cursor); |
|
120 | |||
121 | // Handle any remaining cursor contents |
||
122 | 2061 | if ($context->getContainer() instanceof StringContainerInterface) { |
|
123 | 753 | $context->getContainer()->handleRemainingContents($context, $cursor); |
|
124 | 1812 | } elseif (!$cursor->isBlank()) { |
|
125 | // Create paragraph container for line |
||
126 | 1731 | $p = new Paragraph(); |
|
127 | 1731 | $context->addBlock($p); |
|
128 | 1731 | $cursor->advanceToNextNonSpaceOrTab(); |
|
129 | 1731 | $p->addLine($cursor->getRemainder()); |
|
130 | } |
||
131 | 2061 | } |
|
132 | |||
133 | 2061 | private function processDocument(ContextInterface $context) |
|
134 | { |
||
135 | 2061 | foreach ($this->environment->getDocumentProcessors() as $documentProcessor) { |
|
136 | $documentProcessor->processDocument($context->getDocument()); |
||
137 | } |
||
138 | 2061 | } |
|
139 | |||
140 | 2061 | private function processInlines(ContextInterface $context) |
|
141 | { |
||
142 | 2061 | $walker = $context->getDocument()->walker(); |
|
143 | |||
144 | 2061 | while ($event = $walker->next()) { |
|
145 | 2061 | if (!$event->isEntering()) { |
|
146 | 2061 | continue; |
|
147 | } |
||
148 | |||
149 | 2061 | $node = $event->getNode(); |
|
150 | 2061 | if ($node instanceof AbstractStringContainerBlock) { |
|
151 | 2019 | $this->inlineParserEngine->parse($node, $context->getDocument()->getReferenceMap()); |
|
152 | } |
||
153 | } |
||
154 | 2061 | } |
|
155 | |||
156 | /** |
||
157 | * Sets the container to the last open child (or its parent) |
||
158 | * |
||
159 | * @param ContextInterface $context |
||
160 | * @param Cursor $cursor |
||
161 | */ |
||
162 | 2061 | private function resetContainer(ContextInterface $context, Cursor $cursor) |
|
163 | { |
||
164 | 2061 | $container = $context->getDocument(); |
|
165 | |||
166 | 2061 | while ($lastChild = $container->lastChild()) { |
|
167 | 1119 | if (!($lastChild instanceof AbstractBlock)) { |
|
168 | break; |
||
169 | } |
||
170 | |||
171 | 1119 | if (!$lastChild->isOpen()) { |
|
172 | 453 | break; |
|
173 | } |
||
174 | |||
175 | 1110 | $container = $lastChild; |
|
176 | 1110 | if (!$container->matchesNextLine($cursor)) { |
|
177 | 708 | $container = $container->parent(); // back up to the last matching block |
|
178 | 708 | break; |
|
179 | } |
||
180 | } |
||
181 | |||
182 | 2061 | $context->setContainer($container); |
|
183 | 2061 | } |
|
184 | |||
185 | /** |
||
186 | * Parse blocks |
||
187 | * |
||
188 | * @param ContextInterface $context |
||
189 | * @param Cursor $cursor |
||
190 | */ |
||
191 | 2061 | private function parseBlocks(ContextInterface $context, Cursor $cursor) |
|
192 | { |
||
193 | 2061 | while (!$context->getContainer()->isCode() && !$context->getBlocksParsed()) { |
|
194 | 2061 | $parsed = false; |
|
195 | 2061 | foreach ($this->environment->getBlockParsers() as $parser) { |
|
196 | 2061 | if ($parser->parse($context, $cursor)) { |
|
197 | 819 | $parsed = true; |
|
198 | 1233 | break; |
|
199 | } |
||
200 | } |
||
201 | |||
202 | 2061 | if (!$parsed || $context->getContainer() instanceof StringContainerInterface || (($tip = $context->getTip()) && $tip->getDepth() >= $this->maxNestingLevel)) { |
|
203 | 2034 | $context->setBlocksParsed(true); |
|
204 | 2034 | break; |
|
205 | } |
||
206 | } |
||
207 | 2061 | } |
|
208 | |||
209 | /** |
||
210 | * @param ContextInterface $context |
||
211 | * @param Cursor $cursor |
||
212 | * |
||
213 | * @return bool |
||
214 | */ |
||
215 | 2061 | private function handleLazyParagraphContinuation(ContextInterface $context, Cursor $cursor): bool |
|
216 | { |
||
217 | 2061 | $tip = $context->getTip(); |
|
218 | |||
219 | 2061 | if ($tip instanceof Paragraph && |
|
220 | 2061 | !$context->getBlockCloser()->areAllClosed() && |
|
221 | 2061 | !$cursor->isBlank() && |
|
222 | 2061 | \count($tip->getStrings()) > 0) { |
|
223 | |||
224 | // lazy paragraph continuation |
||
225 | 36 | $tip->addLine($cursor->getRemainder()); |
|
226 | |||
227 | 36 | return true; |
|
228 | } |
||
229 | |||
230 | 2061 | return false; |
|
231 | } |
||
232 | |||
233 | /** |
||
234 | * @param ContextInterface $context |
||
235 | * @param Cursor $cursor |
||
236 | */ |
||
237 | 2061 | private function setAndPropagateLastLineBlank(ContextInterface $context, Cursor $cursor) |
|
255 | } |
||
256 |