1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
/** |
4
|
|
|
* League.Csv (https://csv.thephpleague.com) |
5
|
|
|
* |
6
|
|
|
* (c) Ignace Nyamagana Butera <[email protected]> |
7
|
|
|
* |
8
|
|
|
* For the full copyright and license information, please view the LICENSE |
9
|
|
|
* file that was distributed with this source code. |
10
|
|
|
*/ |
11
|
|
|
|
12
|
|
|
declare(strict_types=1); |
13
|
|
|
|
14
|
|
|
namespace League\Csv; |
15
|
|
|
|
16
|
|
|
use BadMethodCallException; |
17
|
|
|
use CallbackFilterIterator; |
18
|
|
|
use Countable; |
19
|
|
|
use Generator; |
20
|
|
|
use Iterator; |
21
|
|
|
use IteratorAggregate; |
22
|
|
|
use JsonSerializable; |
23
|
|
|
use League\Csv\Polyfill\EmptyEscapeParser; |
24
|
|
|
use SplFileObject; |
25
|
|
|
use TypeError; |
26
|
|
|
use function array_combine; |
27
|
|
|
use function array_filter; |
28
|
|
|
use function array_pad; |
29
|
|
|
use function array_slice; |
30
|
|
|
use function array_unique; |
31
|
|
|
use function gettype; |
32
|
|
|
use function is_array; |
33
|
|
|
use function iterator_count; |
34
|
|
|
use function iterator_to_array; |
35
|
|
|
use function mb_strlen; |
36
|
|
|
use function mb_substr; |
37
|
|
|
use function sprintf; |
38
|
|
|
use function strlen; |
39
|
|
|
use function substr; |
40
|
|
|
use const PHP_VERSION_ID; |
41
|
|
|
use const STREAM_FILTER_READ; |
42
|
|
|
|
43
|
|
|
/** |
44
|
|
|
* A class to parse and read records from a CSV document. |
45
|
|
|
* |
46
|
|
|
* @method array fetchOne(int $nth_record = 0) Returns a single record from the CSV |
47
|
|
|
* @method Generator fetchColumn(string|int $column_index) Returns the next value from a single CSV record field |
48
|
|
|
* @method Generator fetchPairs(string|int $offset_index = 0, string|int $value_index = 1) Fetches the next key-value pairs from the CSV document |
49
|
|
|
*/ |
50
|
|
|
class Reader extends AbstractCsv implements Countable, IteratorAggregate, JsonSerializable |
51
|
|
|
{ |
52
|
|
|
/** |
53
|
|
|
* header offset. |
54
|
|
|
* |
55
|
|
|
* @var int|null |
56
|
|
|
*/ |
57
|
|
|
protected $header_offset; |
58
|
|
|
|
59
|
|
|
/** |
60
|
|
|
* header record. |
61
|
|
|
* |
62
|
|
|
* @var string[] |
63
|
|
|
*/ |
64
|
|
|
protected $header = []; |
65
|
|
|
|
66
|
|
|
/** |
67
|
|
|
* records count. |
68
|
|
|
* |
69
|
|
|
* @var int |
70
|
|
|
*/ |
71
|
|
|
protected $nb_records = -1; |
72
|
|
|
|
73
|
|
|
/** |
74
|
|
|
* {@inheritdoc} |
75
|
|
|
*/ |
76
|
|
|
protected $stream_filter_mode = STREAM_FILTER_READ; |
77
|
|
|
|
78
|
|
|
/** |
79
|
|
|
* {@inheritdoc} |
80
|
|
|
*/ |
81
|
3 |
|
public static function createFromPath(string $path, string $open_mode = 'r', $context = null) |
82
|
|
|
{ |
83
|
3 |
|
return parent::createFromPath($path, $open_mode, $context); |
84
|
|
|
} |
85
|
|
|
|
86
|
|
|
/** |
87
|
|
|
* {@inheritdoc} |
88
|
|
|
*/ |
89
|
27 |
|
protected function resetProperties() |
90
|
|
|
{ |
91
|
27 |
|
parent::resetProperties(); |
92
|
27 |
|
$this->nb_records = -1; |
93
|
27 |
|
$this->header = []; |
94
|
27 |
|
} |
95
|
|
|
|
96
|
|
|
/** |
97
|
|
|
* Returns the header offset. |
98
|
|
|
* |
99
|
|
|
* If no CSV header offset is set this method MUST return null |
100
|
|
|
* |
101
|
|
|
* @return int|null |
102
|
|
|
*/ |
103
|
15 |
|
public function getHeaderOffset() |
104
|
|
|
{ |
105
|
15 |
|
return $this->header_offset; |
106
|
|
|
} |
107
|
|
|
|
108
|
|
|
/** |
109
|
|
|
* Returns the CSV record used as header. |
110
|
|
|
* |
111
|
|
|
* The returned header is represented as an array of string values |
112
|
|
|
* |
113
|
|
|
* @return string[] |
114
|
|
|
*/ |
115
|
21 |
|
public function getHeader(): array |
116
|
|
|
{ |
117
|
21 |
|
if (null === $this->header_offset) { |
118
|
15 |
|
return $this->header; |
119
|
|
|
} |
120
|
|
|
|
121
|
9 |
|
if ([] !== $this->header) { |
122
|
3 |
|
return $this->header; |
123
|
|
|
} |
124
|
|
|
|
125
|
9 |
|
$this->header = $this->setHeader($this->header_offset); |
126
|
|
|
|
127
|
6 |
|
return $this->header; |
128
|
|
|
} |
129
|
|
|
|
130
|
|
|
/** |
131
|
|
|
* Determine the CSV record header. |
132
|
|
|
* |
133
|
|
|
* @throws Exception If the header offset is set and no record is found or is the empty array |
134
|
|
|
* |
135
|
|
|
* @return string[] |
136
|
|
|
*/ |
137
|
12 |
|
protected function setHeader(int $offset): array |
138
|
|
|
{ |
139
|
12 |
|
$header = $this->seekRow($offset); |
140
|
12 |
|
if (false === $header || [] === $header) { |
141
|
6 |
|
throw new Exception(sprintf('The header record does not exist or is empty at offset: `%s`', $offset)); |
142
|
|
|
} |
143
|
|
|
|
144
|
6 |
|
if (0 === $offset) { |
145
|
3 |
|
return $this->removeBOM($header, mb_strlen($this->getInputBOM()), $this->enclosure); |
146
|
|
|
} |
147
|
|
|
|
148
|
3 |
|
return $header; |
149
|
|
|
} |
150
|
|
|
|
151
|
|
|
/** |
152
|
|
|
* Returns the row at a given offset. |
153
|
|
|
* |
154
|
|
|
* @return array|false |
155
|
|
|
*/ |
156
|
12 |
|
protected function seekRow(int $offset) |
157
|
|
|
{ |
158
|
12 |
|
foreach ($this->getDocument() as $index => $record) { |
159
|
12 |
|
if ($offset === $index) { |
160
|
9 |
|
return $record; |
161
|
|
|
} |
162
|
|
|
} |
163
|
|
|
|
164
|
9 |
|
return false; |
165
|
|
|
} |
166
|
|
|
|
167
|
|
|
/** |
168
|
|
|
* Returns the document as an Iterator. |
169
|
|
|
*/ |
170
|
18 |
|
protected function getDocument(): Iterator |
171
|
|
|
{ |
172
|
18 |
|
if (70400 > PHP_VERSION_ID && '' === $this->escape) { |
173
|
6 |
|
$this->document->setCsvControl($this->delimiter, $this->enclosure); |
174
|
|
|
|
175
|
6 |
|
return EmptyEscapeParser::parse($this->document); |
176
|
|
|
} |
177
|
|
|
|
178
|
12 |
|
$this->document->setFlags(SplFileObject::READ_CSV | SplFileObject::READ_AHEAD | SplFileObject::SKIP_EMPTY); |
179
|
12 |
|
$this->document->setCsvControl($this->delimiter, $this->enclosure, $this->escape); |
180
|
12 |
|
$this->document->rewind(); |
181
|
|
|
|
182
|
12 |
|
return $this->document; |
183
|
|
|
} |
184
|
|
|
|
185
|
|
|
/** |
186
|
|
|
* Strip the BOM sequence from a record. |
187
|
|
|
* |
188
|
|
|
* @param string[] $record |
189
|
|
|
* |
190
|
|
|
* @return string[] |
191
|
|
|
*/ |
192
|
12 |
|
protected function removeBOM(array $record, int $bom_length, string $enclosure): array |
193
|
|
|
{ |
194
|
12 |
|
if (0 === $bom_length) { |
195
|
3 |
|
return $record; |
196
|
|
|
} |
197
|
|
|
|
198
|
9 |
|
$record[0] = mb_substr($record[0], $bom_length); |
199
|
9 |
|
if ($enclosure.$enclosure != substr($record[0].$record[0], strlen($record[0]) - 1, 2)) { |
200
|
6 |
|
return $record; |
201
|
|
|
} |
202
|
|
|
|
203
|
3 |
|
$record[0] = substr($record[0], 1, -1); |
204
|
|
|
|
205
|
3 |
|
return $record; |
206
|
|
|
} |
207
|
|
|
|
208
|
|
|
/** |
209
|
|
|
* {@inheritdoc} |
210
|
|
|
*/ |
211
|
9 |
|
public function __call($method, array $arguments) |
212
|
|
|
{ |
213
|
9 |
|
static $whitelisted = ['fetchColumn' => 1, 'fetchOne' => 1, 'fetchPairs' => 1]; |
214
|
9 |
|
if (isset($whitelisted[$method])) { |
215
|
3 |
|
return (new ResultSet($this->getRecords(), $this->getHeader()))->$method(...$arguments); |
216
|
|
|
} |
217
|
|
|
|
218
|
6 |
|
throw new BadMethodCallException(sprintf('%s::%s() method does not exist', static::class, $method)); |
219
|
|
|
} |
220
|
|
|
|
221
|
|
|
/** |
222
|
|
|
* {@inheritdoc} |
223
|
|
|
*/ |
224
|
3 |
|
public function count(): int |
225
|
|
|
{ |
226
|
3 |
|
if (-1 === $this->nb_records) { |
227
|
3 |
|
$this->nb_records = iterator_count($this->getRecords()); |
228
|
|
|
} |
229
|
|
|
|
230
|
3 |
|
return $this->nb_records; |
231
|
|
|
} |
232
|
|
|
|
233
|
|
|
/** |
234
|
|
|
* {@inheritdoc} |
235
|
|
|
*/ |
236
|
6 |
|
public function getIterator(): Iterator |
237
|
|
|
{ |
238
|
6 |
|
return $this->getRecords(); |
239
|
|
|
} |
240
|
|
|
|
241
|
|
|
/** |
242
|
|
|
* {@inheritdoc} |
243
|
|
|
*/ |
244
|
3 |
|
public function jsonSerialize(): array |
245
|
|
|
{ |
246
|
3 |
|
return iterator_to_array($this->getRecords(), false); |
247
|
|
|
} |
248
|
|
|
|
249
|
|
|
/** |
250
|
|
|
* Returns the CSV records as an iterator object. |
251
|
|
|
* |
252
|
|
|
* Each CSV record is represented as a simple array containing strings or null values. |
253
|
|
|
* |
254
|
|
|
* If the CSV document has a header record then each record is combined |
255
|
|
|
* to the header record and the header record is removed from the iterator. |
256
|
|
|
* |
257
|
|
|
* If the CSV document is inconsistent. Missing record fields are |
258
|
|
|
* filled with null values while extra record fields are strip from |
259
|
|
|
* the returned object. |
260
|
|
|
* |
261
|
|
|
* @param string[] $header an optional header to use instead of the CSV document header |
262
|
|
|
*/ |
263
|
21 |
|
public function getRecords(array $header = []): Iterator |
264
|
|
|
{ |
265
|
21 |
|
$header = $this->computeHeader($header); |
266
|
18 |
|
$normalized = static function ($record): bool { |
267
|
18 |
|
return is_array($record) && $record != [null]; |
268
|
18 |
|
}; |
269
|
|
|
|
270
|
18 |
|
$bom = ''; |
271
|
18 |
|
if ($this->is_input_bom_skipped) { |
272
|
15 |
|
$bom = $this->getInputBOM(); |
273
|
|
|
} |
274
|
|
|
|
275
|
18 |
|
$document = $this->getDocument(); |
276
|
18 |
|
$records = $this->stripBOM(new CallbackFilterIterator($document, $normalized), $bom); |
277
|
18 |
|
if (null !== $this->header_offset) { |
278
|
6 |
|
$records = new CallbackFilterIterator($records, function (array $record, int $offset): bool { |
279
|
6 |
|
return $offset !== $this->header_offset; |
280
|
6 |
|
}); |
281
|
|
|
} |
282
|
|
|
|
283
|
18 |
|
return $this->combineHeader($records, $header); |
284
|
|
|
} |
285
|
|
|
|
286
|
|
|
/** |
287
|
|
|
* Returns the header to be used for iteration. |
288
|
|
|
* |
289
|
|
|
* @param string[] $header |
290
|
|
|
* |
291
|
|
|
* @throws Exception If the header contains non unique column name |
292
|
|
|
* |
293
|
|
|
* @return string[] |
294
|
|
|
*/ |
295
|
27 |
|
protected function computeHeader(array $header) |
296
|
|
|
{ |
297
|
27 |
|
if ([] === $header) { |
298
|
24 |
|
$header = $this->getHeader(); |
299
|
|
|
} |
300
|
|
|
|
301
|
27 |
|
if ($header === array_unique(array_filter($header, 'is_string'))) { |
302
|
24 |
|
return $header; |
303
|
|
|
} |
304
|
|
|
|
305
|
3 |
|
throw new Exception('The header record must be empty or a flat array with unique string values'); |
306
|
|
|
} |
307
|
|
|
|
308
|
|
|
/** |
309
|
|
|
* Combine the CSV header to each record if present. |
310
|
|
|
* |
311
|
|
|
* @param string[] $header |
312
|
|
|
*/ |
313
|
33 |
|
protected function combineHeader(Iterator $iterator, array $header): Iterator |
314
|
|
|
{ |
315
|
33 |
|
if ([] === $header) { |
316
|
24 |
|
return $iterator; |
317
|
|
|
} |
318
|
|
|
|
319
|
12 |
|
$field_count = count($header); |
320
|
12 |
|
$mapper = static function (array $record) use ($header, $field_count): array { |
321
|
12 |
|
if (count($record) != $field_count) { |
322
|
6 |
|
$record = array_slice(array_pad($record, $field_count, null), 0, $field_count); |
323
|
|
|
} |
324
|
|
|
|
325
|
12 |
|
return array_combine($header, $record); |
326
|
12 |
|
}; |
327
|
|
|
|
328
|
12 |
|
return new MapIterator($iterator, $mapper); |
329
|
|
|
} |
330
|
|
|
|
331
|
|
|
/** |
332
|
|
|
* Strip the BOM sequence from the returned records if necessary. |
333
|
|
|
*/ |
334
|
27 |
|
protected function stripBOM(Iterator $iterator, string $bom): Iterator |
335
|
|
|
{ |
336
|
27 |
|
if ('' === $bom) { |
337
|
18 |
|
return $iterator; |
338
|
|
|
} |
339
|
|
|
|
340
|
9 |
|
$bom_length = mb_strlen($bom); |
341
|
9 |
|
$mapper = function (array $record, int $index) use ($bom_length): array { |
342
|
9 |
|
if (0 !== $index) { |
343
|
3 |
|
return $record; |
344
|
|
|
} |
345
|
|
|
|
346
|
9 |
|
return $this->removeBOM($record, $bom_length, $this->enclosure); |
347
|
9 |
|
}; |
348
|
|
|
|
349
|
9 |
|
return new MapIterator($iterator, $mapper); |
350
|
|
|
} |
351
|
|
|
|
352
|
|
|
/** |
353
|
|
|
* Selects the record to be used as the CSV header. |
354
|
|
|
* |
355
|
|
|
* Because the header is represented as an array, to be valid |
356
|
|
|
* a header MUST contain only unique string value. |
357
|
|
|
* |
358
|
|
|
* @param int|null $offset the header record offset |
359
|
|
|
* |
360
|
|
|
* @throws Exception if the offset is a negative integer |
361
|
|
|
* |
362
|
|
|
* @return static |
363
|
|
|
*/ |
364
|
24 |
|
public function setHeaderOffset($offset): self |
365
|
|
|
{ |
366
|
24 |
|
if ($offset === $this->header_offset) { |
367
|
12 |
|
return $this; |
368
|
|
|
} |
369
|
|
|
|
370
|
12 |
|
if (!is_nullable_int($offset)) { |
371
|
3 |
|
throw new TypeError(sprintf(__METHOD__.'() expects 1 Argument to be null or an integer %s given', gettype($offset))); |
372
|
|
|
} |
373
|
|
|
|
374
|
9 |
|
if (null !== $offset && 0 > $offset) { |
375
|
3 |
|
throw new Exception(__METHOD__.'() expects 1 Argument to be greater or equal to 0'); |
376
|
|
|
} |
377
|
|
|
|
378
|
6 |
|
$this->header_offset = $offset; |
379
|
6 |
|
$this->resetProperties(); |
380
|
|
|
|
381
|
6 |
|
return $this; |
382
|
|
|
} |
383
|
|
|
} |
384
|
|
|
|