Complex classes like Reader 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 Reader, and based on these observations, apply Extract Interface, too.
1 | <?php |
||
41 | class Reader extends AbstractCsv implements TabularDataReader, JsonSerializable |
||
42 | { |
||
43 | /** |
||
44 | * header offset. |
||
45 | * |
||
46 | * @var int|null |
||
47 | */ |
||
48 | protected $header_offset; |
||
49 | |||
50 | /** |
||
51 | * header record. |
||
52 | * |
||
53 | * @var string[] |
||
54 | */ |
||
55 | protected $header = []; |
||
56 | |||
57 | /** |
||
58 | * records count. |
||
59 | * |
||
60 | * @var int |
||
61 | */ |
||
62 | protected $nb_records = -1; |
||
63 | |||
64 | /** |
||
65 | * {@inheritdoc} |
||
66 | */ |
||
67 | protected $stream_filter_mode = STREAM_FILTER_READ; |
||
68 | |||
69 | /** |
||
70 | * @var bool |
||
71 | */ |
||
72 | protected $is_empty_records_included = false; |
||
73 | |||
74 | /** |
||
75 | * {@inheritdoc} |
||
76 | */ |
||
77 | 3 | public static function createFromPath(string $path, string $open_mode = 'r', $context = null) |
|
78 | { |
||
79 | 3 | return parent::createFromPath($path, $open_mode, $context); |
|
80 | } |
||
81 | |||
82 | /** |
||
83 | * {@inheritdoc} |
||
84 | */ |
||
85 | 39 | protected function resetProperties(): void |
|
86 | { |
||
87 | 39 | parent::resetProperties(); |
|
88 | 39 | $this->nb_records = -1; |
|
89 | 39 | $this->header = []; |
|
90 | 39 | } |
|
91 | |||
92 | /** |
||
93 | * Returns the header offset. |
||
94 | * |
||
95 | * If no CSV header offset is set this method MUST return null |
||
96 | * |
||
97 | */ |
||
98 | 21 | public function getHeaderOffset(): ?int |
|
99 | { |
||
100 | 21 | return $this->header_offset; |
|
101 | } |
||
102 | |||
103 | /** |
||
104 | * {@inheritDoc} |
||
105 | */ |
||
106 | 33 | public function getHeader(): array |
|
107 | { |
||
108 | 33 | if (null === $this->header_offset) { |
|
109 | 18 | return $this->header; |
|
110 | } |
||
111 | |||
112 | 18 | if ([] !== $this->header) { |
|
113 | 9 | return $this->header; |
|
114 | } |
||
115 | |||
116 | 18 | $this->header = $this->setHeader($this->header_offset); |
|
117 | |||
118 | 12 | return $this->header; |
|
119 | } |
||
120 | |||
121 | /** |
||
122 | * Determine the CSV record header. |
||
123 | * |
||
124 | * @throws Exception If the header offset is set and no record is found or is the empty array |
||
125 | * |
||
126 | * @return string[] |
||
127 | */ |
||
128 | 21 | protected function setHeader(int $offset): array |
|
129 | { |
||
130 | 21 | $header = $this->seekRow($offset); |
|
131 | 21 | if (in_array($header, [[], [null]], true)) { |
|
132 | 6 | throw new SyntaxError(sprintf('The header record does not exist or is empty at offset: `%s`', $offset)); |
|
133 | } |
||
134 | |||
135 | 15 | if (0 !== $offset) { |
|
136 | 9 | return $header; |
|
137 | } |
||
138 | |||
139 | 6 | $header = $this->removeBOM($header, mb_strlen($this->getInputBOM()), $this->enclosure); |
|
140 | 6 | if ([''] === $header) { |
|
141 | 3 | throw new SyntaxError(sprintf('The header record does not exist or is empty at offset: `%s`', $offset)); |
|
142 | } |
||
143 | |||
144 | 3 | return $header; |
|
145 | } |
||
146 | |||
147 | /** |
||
148 | * Returns the row at a given offset. |
||
149 | */ |
||
150 | 21 | protected function seekRow(int $offset): array |
|
151 | { |
||
152 | 21 | foreach ($this->getDocument() as $index => $record) { |
|
153 | 21 | if ($offset === $index) { |
|
154 | 15 | return $record; |
|
155 | } |
||
156 | } |
||
157 | |||
158 | 6 | return []; |
|
159 | } |
||
160 | |||
161 | /** |
||
162 | * Returns the document as an Iterator. |
||
163 | */ |
||
164 | 30 | protected function getDocument(): Iterator |
|
165 | { |
||
166 | 30 | if (70400 > PHP_VERSION_ID && '' === $this->escape) { |
|
167 | 4 | $this->document->setCsvControl($this->delimiter, $this->enclosure); |
|
168 | |||
169 | 4 | return EmptyEscapeParser::parse($this->document); |
|
170 | } |
||
171 | |||
172 | 26 | $this->document->setFlags(SplFileObject::READ_CSV | SplFileObject::READ_AHEAD); |
|
173 | 26 | $this->document->setCsvControl($this->delimiter, $this->enclosure, $this->escape); |
|
174 | 26 | $this->document->rewind(); |
|
175 | |||
176 | 26 | return $this->document; |
|
177 | } |
||
178 | |||
179 | /** |
||
180 | * Strip the BOM sequence from a record. |
||
181 | * |
||
182 | * @param string[] $record |
||
183 | * |
||
184 | * @return string[] |
||
185 | */ |
||
186 | 18 | protected function removeBOM(array $record, int $bom_length, string $enclosure): array |
|
187 | { |
||
188 | 18 | if (0 === $bom_length) { |
|
189 | 3 | return $record; |
|
190 | } |
||
191 | |||
192 | 15 | $record[0] = mb_substr($record[0], $bom_length); |
|
193 | 15 | if ($enclosure.$enclosure != substr($record[0].$record[0], strlen($record[0]) - 1, 2)) { |
|
194 | 12 | return $record; |
|
195 | } |
||
196 | |||
197 | 3 | $record[0] = substr($record[0], 1, -1); |
|
198 | |||
199 | 3 | return $record; |
|
200 | } |
||
201 | |||
202 | /** |
||
203 | * {@inheritdoc} |
||
204 | */ |
||
205 | 3 | public function fetchColumn($index = 0): Iterator |
|
206 | { |
||
207 | 3 | return ResultSet::createFromTabularDataReader($this)->fetchColumn($index); |
|
208 | } |
||
209 | |||
210 | /** |
||
211 | * {@inheritdoc} |
||
212 | */ |
||
213 | 9 | public function fetchOne(int $nth_record = 0): array |
|
214 | { |
||
215 | 9 | return ResultSet::createFromTabularDataReader($this)->fetchOne($nth_record); |
|
216 | } |
||
217 | |||
218 | /** |
||
219 | * {@inheritdoc} |
||
220 | */ |
||
221 | 3 | public function fetchPairs($offset_index = 0, $value_index = 1): Iterator |
|
222 | { |
||
223 | 3 | return ResultSet::createFromTabularDataReader($this)->fetchPairs($offset_index, $value_index); |
|
224 | } |
||
225 | |||
226 | /** |
||
227 | * {@inheritdoc} |
||
228 | */ |
||
229 | 9 | public function count(): int |
|
230 | { |
||
231 | 9 | if (-1 === $this->nb_records) { |
|
232 | 9 | $this->nb_records = iterator_count($this->getRecords()); |
|
233 | } |
||
234 | |||
235 | 9 | return $this->nb_records; |
|
236 | } |
||
237 | |||
238 | /** |
||
239 | * {@inheritdoc} |
||
240 | */ |
||
241 | 6 | public function getIterator(): Iterator |
|
245 | |||
246 | /** |
||
247 | * {@inheritdoc} |
||
248 | */ |
||
249 | 3 | public function jsonSerialize(): array |
|
250 | { |
||
251 | 3 | return iterator_to_array($this->getRecords(), false); |
|
252 | } |
||
253 | |||
254 | /** |
||
255 | * {@inheritDoc} |
||
256 | */ |
||
257 | 42 | public function getRecords(array $header = []): Iterator |
|
291 | |||
292 | /** |
||
293 | * Returns the header to be used for iteration. |
||
294 | * |
||
295 | * @param string[] $header |
||
296 | * |
||
297 | * @throws Exception If the header contains non unique column name |
||
298 | * |
||
299 | * @return string[] |
||
300 | */ |
||
301 | 36 | protected function computeHeader(array $header) |
|
313 | |||
314 | /** |
||
315 | * Combine the CSV header to each record if present. |
||
316 | * |
||
317 | * @param string[] $header |
||
318 | */ |
||
319 | 42 | protected function combineHeader(Iterator $iterator, array $header): Iterator |
|
339 | |||
340 | /** |
||
341 | * Strip the BOM sequence from the returned records if necessary. |
||
342 | */ |
||
343 | 36 | protected function stripBOM(Iterator $iterator, string $bom): Iterator |
|
344 | { |
||
345 | 36 | if ('' === $bom) { |
|
346 | 24 | return $iterator; |
|
369 | |||
370 | /** |
||
371 | * Selects the record to be used as the CSV header. |
||
372 | * |
||
373 | * Because the header is represented as an array, to be valid |
||
374 | * a header MUST contain only unique string value. |
||
375 | * |
||
376 | * @param int|null $offset the header record offset |
||
377 | * |
||
378 | * @throws Exception if the offset is a negative integer |
||
379 | * |
||
380 | * @return static |
||
381 | */ |
||
382 | 36 | public function setHeaderOffset(?int $offset): self |
|
397 | |||
398 | /** |
||
399 | * Enable skipping empty records. |
||
400 | */ |
||
401 | 12 | public function skipEmptyRecords(): self |
|
410 | |||
411 | /** |
||
412 | * Disable skipping empty records. |
||
413 | */ |
||
414 | 18 | public function includeEmptyRecords(): self |
|
423 | |||
424 | /** |
||
425 | * Tells whether empty records are skipped by the instance. |
||
426 | */ |
||
427 | 12 | public function isEmptyRecordsIncluded(): bool |
|
431 | } |
||
432 |