|
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 OutOfRangeException; |
|
17
|
|
|
use php_user_filter; |
|
18
|
|
|
use function array_combine; |
|
19
|
|
|
use function array_map; |
|
20
|
|
|
use function array_walk; |
|
21
|
|
|
use function in_array; |
|
22
|
|
|
use function is_numeric; |
|
23
|
|
|
use function mb_convert_encoding; |
|
24
|
|
|
use function mb_list_encodings; |
|
25
|
|
|
use function preg_match; |
|
26
|
|
|
use function sprintf; |
|
27
|
|
|
use function stream_bucket_append; |
|
28
|
|
|
use function stream_bucket_make_writeable; |
|
29
|
|
|
use function stream_filter_register; |
|
30
|
|
|
use function stream_get_filters; |
|
31
|
|
|
use function strpos; |
|
32
|
|
|
use function strtolower; |
|
33
|
|
|
use function substr; |
|
34
|
|
|
|
|
35
|
|
|
/** |
|
36
|
|
|
* Converts resource stream or tabular data content charset. |
|
37
|
|
|
*/ |
|
38
|
|
|
class CharsetConverter extends php_user_filter |
|
39
|
|
|
{ |
|
40
|
|
|
const FILTERNAME = 'convert.league.csv'; |
|
41
|
|
|
|
|
42
|
|
|
/** |
|
43
|
|
|
* the filter name used to instantiate the class with. |
|
44
|
|
|
* |
|
45
|
|
|
* @var string |
|
46
|
|
|
*/ |
|
47
|
|
|
public $filtername; |
|
48
|
|
|
|
|
49
|
|
|
/** |
|
50
|
|
|
* @var mixed value passed to passed to stream_filter_append or stream_filter_prepend functions. |
|
51
|
|
|
*/ |
|
52
|
|
|
public $params; |
|
53
|
|
|
|
|
54
|
|
|
/** |
|
55
|
|
|
* The records input encoding charset. |
|
56
|
|
|
* |
|
57
|
|
|
* @var string |
|
58
|
|
|
*/ |
|
59
|
|
|
protected $input_encoding = 'UTF-8'; |
|
60
|
|
|
|
|
61
|
|
|
/** |
|
62
|
|
|
* The records output encoding charset. |
|
63
|
|
|
* |
|
64
|
|
|
* @var string |
|
65
|
|
|
*/ |
|
66
|
|
|
protected $output_encoding = 'UTF-8'; |
|
67
|
|
|
|
|
68
|
|
|
/** |
|
69
|
|
|
* Static method to add the stream filter to a {@link AbstractCsv} object. |
|
70
|
|
|
*/ |
|
71
|
3 |
|
public static function addTo(AbstractCsv $csv, string $input_encoding, string $output_encoding): AbstractCsv |
|
72
|
|
|
{ |
|
73
|
3 |
|
self::register(); |
|
74
|
|
|
|
|
75
|
3 |
|
return $csv->addStreamFilter(self::getFiltername($input_encoding, $output_encoding)); |
|
76
|
|
|
} |
|
77
|
|
|
|
|
78
|
|
|
/** |
|
79
|
|
|
* Static method to register the class as a stream filter. |
|
80
|
|
|
*/ |
|
81
|
3 |
|
public static function register(): void |
|
82
|
|
|
{ |
|
83
|
3 |
|
$filtername = self::FILTERNAME.'.*'; |
|
84
|
3 |
|
if (!in_array($filtername, stream_get_filters(), true)) { |
|
85
|
3 |
|
stream_filter_register($filtername, self::class); |
|
86
|
|
|
} |
|
87
|
3 |
|
} |
|
88
|
|
|
|
|
89
|
|
|
/** |
|
90
|
|
|
* Static method to return the stream filter filtername. |
|
91
|
|
|
*/ |
|
92
|
3 |
|
public static function getFiltername(string $input_encoding, string $output_encoding): string |
|
93
|
|
|
{ |
|
94
|
3 |
|
return sprintf( |
|
95
|
3 |
|
'%s.%s/%s', |
|
96
|
3 |
|
self::FILTERNAME, |
|
97
|
3 |
|
self::filterEncoding($input_encoding), |
|
98
|
3 |
|
self::filterEncoding($output_encoding) |
|
99
|
|
|
); |
|
100
|
|
|
} |
|
101
|
|
|
|
|
102
|
|
|
/** |
|
103
|
|
|
* Filter encoding charset. |
|
104
|
|
|
* |
|
105
|
|
|
* @throws OutOfRangeException if the charset is malformed or unsupported |
|
106
|
|
|
*/ |
|
107
|
6 |
|
protected static function filterEncoding(string $encoding): string |
|
108
|
|
|
{ |
|
109
|
6 |
|
static $encoding_list; |
|
110
|
6 |
|
if (null === $encoding_list) { |
|
111
|
3 |
|
$list = mb_list_encodings(); |
|
112
|
3 |
|
$encoding_list = array_combine(array_map('strtolower', $list), $list); |
|
113
|
|
|
} |
|
114
|
|
|
|
|
115
|
6 |
|
$key = strtolower($encoding); |
|
116
|
6 |
|
if (isset($encoding_list[$key])) { |
|
117
|
3 |
|
return $encoding_list[$key]; |
|
118
|
|
|
} |
|
119
|
|
|
|
|
120
|
3 |
|
throw new OutOfRangeException(sprintf('The submitted charset %s is not supported by the mbstring extension', $encoding)); |
|
121
|
|
|
} |
|
122
|
|
|
|
|
123
|
|
|
/** |
|
124
|
|
|
* {@inheritdoc} |
|
125
|
|
|
*/ |
|
126
|
12 |
|
public function onCreate(): bool |
|
127
|
|
|
{ |
|
128
|
12 |
|
$prefix = self::FILTERNAME.'.'; |
|
129
|
12 |
|
if (0 !== strpos($this->filtername, $prefix)) { |
|
130
|
3 |
|
return false; |
|
131
|
|
|
} |
|
132
|
|
|
|
|
133
|
9 |
|
$encodings = substr($this->filtername, strlen($prefix)); |
|
134
|
9 |
|
if (1 !== preg_match(',^(?<input>[-\w]+)\/(?<output>[-\w]+)$,', $encodings, $matches)) { |
|
135
|
3 |
|
return false; |
|
136
|
|
|
} |
|
137
|
|
|
|
|
138
|
|
|
try { |
|
139
|
6 |
|
$this->input_encoding = self::filterEncoding($matches['input']); |
|
140
|
3 |
|
$this->output_encoding = self::filterEncoding($matches['output']); |
|
141
|
3 |
|
} catch (OutOfRangeException $e) { |
|
142
|
3 |
|
return false; |
|
143
|
|
|
} |
|
144
|
|
|
|
|
145
|
3 |
|
return true; |
|
146
|
|
|
} |
|
147
|
|
|
|
|
148
|
|
|
/** |
|
149
|
|
|
* @param resource $in |
|
150
|
|
|
* @param resource $out |
|
151
|
|
|
* @param int $consumed |
|
152
|
|
|
* @param bool $closing |
|
153
|
|
|
*/ |
|
154
|
3 |
|
public function filter($in, $out, &$consumed, $closing): int |
|
155
|
|
|
{ |
|
156
|
3 |
|
while ($res = stream_bucket_make_writeable($in)) { |
|
157
|
3 |
|
$res->data = @mb_convert_encoding($res->data, $this->output_encoding, $this->input_encoding); |
|
158
|
3 |
|
$consumed += $res->datalen; |
|
159
|
3 |
|
stream_bucket_append($out, $res); |
|
160
|
|
|
} |
|
161
|
|
|
|
|
162
|
3 |
|
return PSFS_PASS_ON; |
|
163
|
|
|
} |
|
164
|
|
|
|
|
165
|
|
|
/** |
|
166
|
|
|
* Convert Csv records collection into UTF-8. |
|
167
|
|
|
*/ |
|
168
|
21 |
|
public function convert(iterable $records): iterable |
|
169
|
|
|
{ |
|
170
|
21 |
|
if ($this->output_encoding === $this->input_encoding) { |
|
171
|
3 |
|
return $records; |
|
172
|
|
|
} |
|
173
|
|
|
|
|
174
|
18 |
|
if (is_array($records)) { |
|
175
|
15 |
|
return array_map($this, $records); |
|
176
|
|
|
} |
|
177
|
|
|
|
|
178
|
3 |
|
return new MapIterator($records, $this); |
|
179
|
|
|
} |
|
180
|
|
|
|
|
181
|
|
|
/** |
|
182
|
|
|
* Enable using the class as a formatter for the {@link Writer}. |
|
183
|
|
|
*/ |
|
184
|
3 |
|
public function __invoke(array $record): array |
|
185
|
|
|
{ |
|
186
|
3 |
|
array_walk($record, [$this, 'encodeField']); |
|
187
|
|
|
|
|
188
|
3 |
|
return $record; |
|
189
|
|
|
} |
|
190
|
|
|
|
|
191
|
|
|
/** |
|
192
|
|
|
* Walker method to convert the offset and the value of a CSV record field. |
|
193
|
|
|
* |
|
194
|
|
|
* @param mixed $value can be a scalar type or null |
|
195
|
|
|
* @param mixed $offset can be a string or an int |
|
196
|
|
|
*/ |
|
197
|
15 |
|
protected function encodeField(&$value, &$offset): void |
|
198
|
|
|
{ |
|
199
|
15 |
|
if (null !== $value && !is_numeric($value)) { |
|
200
|
3 |
|
$value = mb_convert_encoding((string) $value, $this->output_encoding, $this->input_encoding); |
|
201
|
|
|
} |
|
202
|
|
|
|
|
203
|
15 |
|
if (!is_numeric($offset)) { |
|
204
|
3 |
|
$offset = mb_convert_encoding((string) $offset, $this->output_encoding, $this->input_encoding); |
|
205
|
|
|
} |
|
206
|
15 |
|
} |
|
207
|
|
|
|
|
208
|
|
|
/** |
|
209
|
|
|
* Sets the records input encoding charset. |
|
210
|
|
|
*/ |
|
211
|
12 |
|
public function inputEncoding(string $encoding): self |
|
212
|
|
|
{ |
|
213
|
12 |
|
$encoding = self::filterEncoding($encoding); |
|
214
|
9 |
|
if ($encoding === $this->input_encoding) { |
|
215
|
6 |
|
return $this; |
|
216
|
|
|
} |
|
217
|
|
|
|
|
218
|
6 |
|
$clone = clone $this; |
|
219
|
6 |
|
$clone->input_encoding = $encoding; |
|
220
|
|
|
|
|
221
|
6 |
|
return $clone; |
|
222
|
|
|
} |
|
223
|
|
|
|
|
224
|
|
|
/** |
|
225
|
|
|
* Sets the records output encoding charset. |
|
226
|
|
|
*/ |
|
227
|
3 |
|
public function outputEncoding(string $encoding): self |
|
228
|
|
|
{ |
|
229
|
3 |
|
$encoding = self::filterEncoding($encoding); |
|
230
|
3 |
|
if ($encoding === $this->output_encoding) { |
|
231
|
3 |
|
return $this; |
|
232
|
|
|
} |
|
233
|
|
|
|
|
234
|
3 |
|
$clone = clone $this; |
|
235
|
3 |
|
$clone->output_encoding = $encoding; |
|
236
|
|
|
|
|
237
|
3 |
|
return $clone; |
|
238
|
|
|
} |
|
239
|
|
|
} |
|
240
|
|
|
|