|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
declare(strict_types=1); |
|
4
|
|
|
|
|
5
|
|
|
namespace PhpSpellcheck\Utils; |
|
6
|
|
|
|
|
7
|
|
|
use PhpSpellcheck\Misspelling; |
|
8
|
|
|
|
|
9
|
|
|
class IspellParser |
|
10
|
|
|
{ |
|
11
|
|
|
/** |
|
12
|
|
|
* @param string $output assumed to be generated from an input on which {@see IspellParser::adaptInputForTerseModeProcessing()} has been applied |
|
13
|
|
|
* @param array<mixed> $context |
|
14
|
|
|
* |
|
15
|
|
|
* @return iterable<Misspelling> |
|
16
|
|
|
*/ |
|
17
|
|
|
public static function parseMisspellingsFromOutput(string $output, array $context = []): iterable |
|
18
|
|
|
{ |
|
19
|
|
|
$lines = explode(PHP_EOL, $output); |
|
20
|
|
|
$lineNumber = 1; |
|
21
|
|
|
foreach ($lines as $line) { |
|
22
|
|
|
$line = trim($line); |
|
23
|
|
|
if ('' === $line) { |
|
24
|
|
|
++$lineNumber; |
|
25
|
|
|
// Go to the next line |
|
26
|
|
|
continue; |
|
27
|
|
|
} |
|
28
|
|
|
|
|
29
|
|
|
switch ($line[0]) { |
|
30
|
|
|
case '#': |
|
31
|
|
|
[, $word, $offset] = explode(' ', $line); |
|
32
|
|
|
yield new Misspelling( |
|
33
|
|
|
$word, |
|
34
|
|
|
/** |
|
35
|
|
|
* a `^` character is added to each line while sending text as input so it needs to |
|
36
|
|
|
* account for that. {@see IspellParser::adaptInputForTerseModeProcessing}. |
|
37
|
|
|
*/ |
|
38
|
|
|
trim($offset) - 1, |
|
39
|
|
|
$lineNumber, |
|
40
|
|
|
[], |
|
41
|
|
|
$context |
|
42
|
|
|
); |
|
43
|
|
|
|
|
44
|
|
|
break; |
|
45
|
|
|
case '&': |
|
46
|
|
|
$parts = explode(':', $line); |
|
47
|
|
|
[, $word, , $offset] = explode(' ', $parts[0]); |
|
48
|
|
|
yield new Misspelling( |
|
49
|
|
|
$word, |
|
50
|
|
|
/** |
|
51
|
|
|
* a `^` character is added to each line while sending text as input so it needs to |
|
52
|
|
|
* account for that. {@see IspellParser::adaptInputForTerseModeProcessing}. |
|
53
|
|
|
*/ |
|
54
|
|
|
trim($offset) - 1, |
|
55
|
|
|
$lineNumber, |
|
56
|
|
|
explode(', ', trim($parts[1])), |
|
57
|
|
|
$context |
|
58
|
|
|
); |
|
59
|
|
|
|
|
60
|
|
|
break; |
|
61
|
|
|
} |
|
62
|
|
|
} |
|
63
|
|
|
} |
|
64
|
|
|
|
|
65
|
|
|
/** |
|
66
|
|
|
* Preprocess the source text so that aspell/ispell pipe mode instruction is ignored. |
|
67
|
|
|
* |
|
68
|
|
|
* In pipe mode some special characters at the beginning of the line are instructions for aspell/ispell |
|
69
|
|
|
* {@link http://aspell.net/man-html/Through-A-Pipe.html#Through-A-Pipe}. |
|
70
|
|
|
* |
|
71
|
|
|
* Spellchecker must not interpret them, so it escapes them using the ^ symbol. |
|
72
|
|
|
*/ |
|
73
|
|
|
public static function adaptInputForTerseModeProcessing(string $input): string |
|
74
|
|
|
{ |
|
75
|
|
|
return preg_replace('/^/m', '^', $input); |
|
76
|
|
|
} |
|
77
|
|
|
} |
|
78
|
|
|
|
tigitz /
php-spellchecker
