1
|
|
|
<?php |
2
|
|
|
/** |
3
|
|
|
* This file is part of the League.csv library |
4
|
|
|
* |
5
|
|
|
* @license http://opensource.org/licenses/MIT |
6
|
|
|
* @link https://github.com/thephpleague/csv/ |
7
|
|
|
* @version 9.0.0 |
8
|
|
|
* @package League.csv |
9
|
|
|
* |
10
|
|
|
* For the full copyright and license information, please view the LICENSE |
11
|
|
|
* file that was distributed with this source code. |
12
|
|
|
*/ |
13
|
|
|
declare(strict_types=1); |
14
|
|
|
|
15
|
|
|
namespace League\Csv; |
16
|
|
|
|
17
|
|
|
use BadMethodCallException; |
18
|
|
|
use CallbackFilterIterator; |
19
|
|
|
use Countable; |
20
|
|
|
use Iterator; |
21
|
|
|
use IteratorAggregate; |
22
|
|
|
use JsonSerializable; |
23
|
|
|
use League\Csv\Exception\RuntimeException; |
24
|
|
|
use SplFileObject; |
25
|
|
|
|
26
|
|
|
/** |
27
|
|
|
* A class to manage records selection from a CSV document |
28
|
|
|
* |
29
|
|
|
* @package League.csv |
30
|
|
|
* @since 3.0.0 |
31
|
|
|
* |
32
|
|
|
* @method array fetchOne(int $nth_record = 0) Returns a single record from the CSV |
33
|
|
|
* @method Generator fetchColumn(string|int $column_index) Returns the next value from a single CSV record field |
34
|
|
|
* @method Generator fetchPairs(string|int $offset_index = 0, string|int $value_index = 1) Fetches the next key-value pairs from the CSV document |
35
|
|
|
*/ |
36
|
|
|
class Reader extends AbstractCsv implements Countable, IteratorAggregate, JsonSerializable |
37
|
|
|
{ |
38
|
|
|
/** |
39
|
|
|
* CSV Document header offset |
40
|
|
|
* |
41
|
|
|
* @var int|null |
42
|
|
|
*/ |
43
|
|
|
protected $header_offset; |
44
|
|
|
|
45
|
|
|
/** |
46
|
|
|
* CSV Document Header record |
47
|
|
|
* |
48
|
|
|
* @var string[] |
49
|
|
|
*/ |
50
|
|
|
protected $header = []; |
51
|
|
|
|
52
|
|
|
/** |
53
|
|
|
* Tell whether the CSV records offset must be kept on Json serialization |
54
|
|
|
* |
55
|
|
|
* @var bool |
56
|
|
|
*/ |
57
|
|
|
protected $json_preserve_offset = false; |
58
|
|
|
|
59
|
|
|
/** |
60
|
|
|
* Records count |
61
|
|
|
* |
62
|
|
|
* @var int |
63
|
|
|
*/ |
64
|
|
|
protected $nb_records = -1; |
65
|
|
|
|
66
|
|
|
/** |
67
|
|
|
* @inheritdoc |
68
|
|
|
*/ |
69
|
|
|
protected $stream_filter_mode = STREAM_FILTER_READ; |
70
|
|
|
|
71
|
|
|
/** |
72
|
|
|
* Returns the record offset used as header |
73
|
|
|
* |
74
|
|
|
* If no CSV record is used this method MUST return null |
75
|
|
|
* |
76
|
|
|
* @return int|null |
77
|
|
|
*/ |
78
|
10 |
|
public function getHeaderOffset() |
79
|
|
|
{ |
80
|
10 |
|
return $this->header_offset; |
81
|
|
|
} |
82
|
|
|
|
83
|
|
|
/** |
84
|
|
|
* Returns the CSV record header |
85
|
|
|
* |
86
|
|
|
* The returned header is represented as an array of string values |
87
|
|
|
* |
88
|
|
|
* @return string[] |
89
|
|
|
*/ |
90
|
10 |
|
public function getHeader(): array |
91
|
|
|
{ |
92
|
10 |
|
if (null === $this->header_offset) { |
93
|
8 |
|
return $this->header; |
94
|
|
|
} |
95
|
|
|
|
96
|
4 |
|
if (!empty($this->header)) { |
97
|
2 |
|
return $this->header; |
98
|
|
|
} |
99
|
|
|
|
100
|
4 |
|
$this->header = $this->setHeader($this->header_offset); |
101
|
|
|
|
102
|
4 |
|
return $this->header; |
103
|
|
|
} |
104
|
|
|
|
105
|
|
|
/** |
106
|
|
|
* Determine the CSV record header |
107
|
|
|
* |
108
|
|
|
* @param int $offset |
109
|
|
|
* |
110
|
|
|
* @throws RuntimeException If the header offset is an integer |
111
|
|
|
* and the corresponding record is missing |
112
|
|
|
* or is an empty array |
113
|
|
|
* |
114
|
|
|
* @return string[] |
115
|
|
|
*/ |
116
|
6 |
|
protected function setHeader(int $offset): array |
117
|
|
|
{ |
118
|
6 |
|
$this->document->setFlags(SplFileObject::READ_CSV | SplFileObject::READ_AHEAD | SplFileObject::SKIP_EMPTY); |
119
|
6 |
|
$this->document->setCsvControl($this->delimiter, $this->enclosure, $this->escape); |
120
|
6 |
|
$this->document->seek($offset); |
121
|
6 |
|
if (empty($header = $this->document->current())) { |
122
|
2 |
|
throw new RuntimeException(sprintf('The header record does not exist or is empty at offset: `%s`', $offset)); |
123
|
|
|
} |
124
|
|
|
|
125
|
4 |
|
if (0 === $offset) { |
126
|
2 |
|
return $this->removeBOM($header, mb_strlen($this->getInputBOM()), $this->enclosure); |
127
|
|
|
} |
128
|
|
|
|
129
|
2 |
|
return $header; |
130
|
|
|
} |
131
|
|
|
|
132
|
|
|
/** |
133
|
|
|
* Strip the BOM sequence from a record |
134
|
|
|
* |
135
|
|
|
* @param string[] $record |
136
|
|
|
* @param int $bom_length |
137
|
|
|
* @param string $enclosure |
138
|
|
|
* |
139
|
|
|
* @return string[] |
140
|
|
|
*/ |
141
|
8 |
|
protected function removeBOM(array $record, int $bom_length, string $enclosure): array |
142
|
|
|
{ |
143
|
8 |
|
if (0 == $bom_length) { |
144
|
2 |
|
return $record; |
145
|
|
|
} |
146
|
|
|
|
147
|
6 |
|
$record[0] = mb_substr($record[0], $bom_length); |
148
|
6 |
|
if ($enclosure == mb_substr($record[0], 0, 1) && $enclosure == mb_substr($record[0], -1, 1)) { |
149
|
2 |
|
$record[0] = mb_substr($record[0], 1, -1); |
150
|
|
|
} |
151
|
|
|
|
152
|
6 |
|
return $record; |
153
|
|
|
} |
154
|
|
|
|
155
|
|
|
/** |
156
|
|
|
* @inheritdoc |
157
|
|
|
*/ |
158
|
6 |
|
public function __call($method, array $arguments) |
159
|
|
|
{ |
160
|
6 |
|
$whitelisted = ['fetchColumn' => 1, 'fetchPairs' => 1, 'fetchOne' => 1]; |
161
|
6 |
|
if (isset($whitelisted[$method])) { |
162
|
2 |
|
return (new ResultSet($this->getRecords(), $this->getHeader()))->$method(...$arguments); |
163
|
|
|
} |
164
|
|
|
|
165
|
4 |
|
throw new BadMethodCallException(sprintf('%s::%s() method does not exist', __CLASS__, $method)); |
166
|
|
|
} |
167
|
|
|
|
168
|
|
|
/** |
169
|
|
|
* @inheritdoc |
170
|
|
|
*/ |
171
|
2 |
|
public function count(): int |
172
|
|
|
{ |
173
|
2 |
|
if (-1 === $this->nb_records) { |
174
|
2 |
|
$this->nb_records = iterator_count($this->getRecords()); |
175
|
|
|
} |
176
|
|
|
|
177
|
2 |
|
return $this->nb_records; |
178
|
|
|
} |
179
|
|
|
|
180
|
|
|
/** |
181
|
|
|
* @inheritdoc |
182
|
|
|
*/ |
183
|
2 |
|
public function getIterator(): Iterator |
184
|
|
|
{ |
185
|
2 |
|
return $this->getRecords(); |
186
|
|
|
} |
187
|
|
|
|
188
|
|
|
/** |
189
|
|
|
* @inheritdoc |
190
|
|
|
*/ |
191
|
2 |
|
public function jsonSerialize(): array |
192
|
|
|
{ |
193
|
2 |
|
return iterator_to_array($this->getRecords(), $this->json_preserve_offset); |
194
|
|
|
} |
195
|
|
|
|
196
|
|
|
/** |
197
|
|
|
* Returns the CSV records in an iterator object. |
198
|
|
|
* |
199
|
|
|
* Each CSV record is represented as a simple array of string or null values. |
200
|
|
|
* |
201
|
|
|
* If the CSV document has a header record then each record is combined |
202
|
|
|
* to each header record and the header record is removed from the iterator. |
203
|
|
|
* |
204
|
|
|
* If the CSV document is inconsistent. Missing record fields are |
205
|
|
|
* filled with null values while extra record fields are strip from |
206
|
|
|
* the returned object. |
207
|
|
|
* |
208
|
|
|
* @param string[] $header an optional header to use instead of the CSV document header |
209
|
|
|
* |
210
|
|
|
* @return Iterator |
211
|
|
|
*/ |
212
|
12 |
|
public function getRecords(array $header = []): Iterator |
213
|
|
|
{ |
214
|
12 |
|
$header = $this->computeHeader($header); |
215
|
|
|
$normalized = function ($record): bool { |
216
|
10 |
|
return is_array($record) && $record != [null]; |
217
|
10 |
|
}; |
218
|
10 |
|
$bom = $this->getInputBOM(); |
219
|
10 |
|
$this->document->setFlags(SplFileObject::READ_CSV | SplFileObject::READ_AHEAD | SplFileObject::SKIP_EMPTY); |
220
|
10 |
|
$this->document->setCsvControl($this->delimiter, $this->enclosure, $this->escape); |
221
|
|
|
|
222
|
10 |
|
$records = $this->stripBOM(new CallbackFilterIterator($this->document, $normalized), $bom); |
223
|
10 |
|
if (null !== $this->header_offset) { |
224
|
|
|
$records = new CallbackFilterIterator($records, function (array $record, int $offset): bool { |
225
|
4 |
|
return $offset !== $this->header_offset; |
226
|
4 |
|
}); |
227
|
|
|
} |
228
|
|
|
|
229
|
10 |
|
return $this->combineHeader($records, $header); |
230
|
|
|
} |
231
|
|
|
|
232
|
|
|
/** |
233
|
|
|
* Returns the header to be used for iteration |
234
|
|
|
* |
235
|
|
|
* @param string[] $header |
236
|
|
|
* |
237
|
|
|
* @throws RuntimeException If the header contains non unique column name |
238
|
|
|
* |
239
|
|
|
* @return string[] |
240
|
|
|
*/ |
241
|
16 |
|
protected function computeHeader(array $header) |
242
|
|
|
{ |
243
|
16 |
|
if (empty($header)) { |
244
|
14 |
|
$header = $this->getHeader(); |
245
|
|
|
} |
246
|
|
|
|
247
|
16 |
|
if ($header === array_unique(array_filter($header, 'is_string'))) { |
248
|
14 |
|
return $header; |
249
|
|
|
} |
250
|
|
|
|
251
|
2 |
|
throw new RuntimeException('The header record must be empty or a flat array with unique string values'); |
252
|
|
|
} |
253
|
|
|
|
254
|
|
|
/** |
255
|
|
|
* Add the CSV header if present and valid |
256
|
|
|
* |
257
|
|
|
* @param Iterator $iterator |
258
|
|
|
* @param string[] $header |
259
|
|
|
* |
260
|
|
|
* @return Iterator |
261
|
|
|
*/ |
262
|
20 |
|
protected function combineHeader(Iterator $iterator, array $header): Iterator |
263
|
|
|
{ |
264
|
20 |
|
if (empty($header)) { |
265
|
14 |
|
return $iterator; |
266
|
|
|
} |
267
|
|
|
|
268
|
8 |
|
$field_count = count($header); |
269
|
|
|
$mapper = function (array $record) use ($header, $field_count): array { |
270
|
8 |
|
if (count($record) != $field_count) { |
271
|
4 |
|
$record = array_slice(array_pad($record, $field_count, null), 0, $field_count); |
272
|
|
|
} |
273
|
|
|
|
274
|
8 |
|
return array_combine($header, $record); |
275
|
8 |
|
}; |
276
|
|
|
|
277
|
8 |
|
return new MapIterator($iterator, $mapper); |
278
|
|
|
} |
279
|
|
|
|
280
|
|
|
/** |
281
|
|
|
* Strip the BOM sequence if present |
282
|
|
|
* |
283
|
|
|
* @param Iterator $iterator |
284
|
|
|
* @param string $bom |
285
|
|
|
* |
286
|
|
|
* @return Iterator |
287
|
|
|
*/ |
288
|
16 |
|
protected function stripBOM(Iterator $iterator, string $bom): Iterator |
289
|
|
|
{ |
290
|
16 |
|
if ('' === $bom) { |
291
|
10 |
|
return $iterator; |
292
|
|
|
} |
293
|
|
|
|
294
|
6 |
|
$bom_length = mb_strlen($bom); |
295
|
6 |
|
$mapper = function (array $record, int $index) use ($bom_length): array { |
296
|
6 |
|
if (0 != $index) { |
297
|
2 |
|
return $record; |
298
|
|
|
} |
299
|
|
|
|
300
|
6 |
|
return $this->removeBOM($record, $bom_length, $this->enclosure); |
301
|
6 |
|
}; |
302
|
|
|
|
303
|
6 |
|
return new MapIterator($iterator, $mapper); |
304
|
|
|
} |
305
|
|
|
|
306
|
|
|
/** |
307
|
|
|
* Selects the record to be used as the CSV header |
308
|
|
|
* |
309
|
|
|
* Because of the header is represented as an array, to be valid |
310
|
|
|
* a header MUST contain only unique string value. |
311
|
|
|
* |
312
|
|
|
* @param int|null $offset the header record offset |
313
|
|
|
* |
314
|
|
|
* @return static |
315
|
|
|
*/ |
316
|
10 |
|
public function setHeaderOffset($offset): self |
317
|
|
|
{ |
318
|
10 |
|
$this->filterNullableInteger($offset, 0, __METHOD__.'() expects the header offset index to be a positive integer or 0'); |
319
|
10 |
|
if ($offset !== $this->header_offset) { |
320
|
2 |
|
$this->header_offset = $offset; |
321
|
2 |
|
$this->resetProperties(); |
322
|
|
|
} |
323
|
|
|
|
324
|
10 |
|
return $this; |
325
|
|
|
} |
326
|
|
|
|
327
|
|
|
/** |
328
|
|
|
* Whether we should preserve the CSV document record offset on json serialization. |
329
|
|
|
* |
330
|
|
|
* If set to true CSV document record offset will be preserve when calling jsonSerialize |
331
|
|
|
* |
332
|
|
|
* @param bool $status |
333
|
|
|
* |
334
|
|
|
* @return self |
335
|
|
|
*/ |
336
|
2 |
|
public function jsonPreserveOffset(bool $status): self |
337
|
|
|
{ |
338
|
2 |
|
$this->json_preserve_offset = $status; |
339
|
|
|
|
340
|
2 |
|
return $this; |
341
|
|
|
} |
342
|
|
|
|
343
|
|
|
/** |
344
|
|
|
* @inheritdoc |
345
|
|
|
*/ |
346
|
8 |
|
protected function resetProperties() |
347
|
|
|
{ |
348
|
8 |
|
$this->nb_records = -1; |
349
|
8 |
|
$this->header = []; |
350
|
8 |
|
} |
351
|
|
|
} |
352
|
|
|
|