|
1
|
|
|
<?php |
|
2
|
|
|
/** |
|
3
|
|
|
* This file is part of dispositif/wikibot application |
|
4
|
|
|
* 2019 : Philippe M. <[email protected]> |
|
5
|
|
|
* For the full copyright and MIT license information, please view the LICENSE file. |
|
6
|
|
|
*/ |
|
7
|
|
|
|
|
8
|
|
|
declare(strict_types=1); |
|
9
|
|
|
|
|
10
|
|
|
namespace App\Domain\Utils; |
|
11
|
|
|
|
|
12
|
|
|
class WikiTextUtil extends TextUtil |
|
13
|
|
|
{ |
|
14
|
|
|
/** |
|
15
|
|
|
* remove wiki encoding : italic, bold, links [ ] and [[fu|bar]] => bar |
|
16
|
|
|
* replace non-breaking spaces |
|
17
|
|
|
* replace {{lang|en|fubar}} => fubar. |
|
18
|
|
|
* |
|
19
|
|
|
* @param $text |
|
20
|
|
|
* @param bool $stripcomment |
|
21
|
|
|
* |
|
22
|
|
|
* @return string |
|
23
|
|
|
*/ |
|
24
|
|
|
public static function unWikify(string $text, ?bool $stripcomment = true): string |
|
25
|
|
|
{ |
|
26
|
|
|
// todo remove HTML tags ? |
|
27
|
|
|
if (true === $stripcomment) { |
|
28
|
|
|
$text = self::removeHTMLcomments($text); |
|
29
|
|
|
} |
|
30
|
|
|
|
|
31
|
|
|
$text = str_replace( |
|
32
|
|
|
['[', ']', "'''", "''", ' '], |
|
33
|
|
|
['', '', '', '', ' '], |
|
34
|
|
|
preg_replace( |
|
35
|
|
|
[ |
|
36
|
|
|
"#\[\[[^|\]]*\|([^]]*)]]#", |
|
37
|
|
|
'#{{ ?(?:lang|langue) ?\|[^|]+\| ?(?:texte=)?([^{}=]+)(?:\|dir=rtl)?}}#i', |
|
38
|
|
|
"#&[\w\d]{2,7};#", |
|
39
|
|
|
], |
|
40
|
|
|
['$1', '$1', ''], |
|
41
|
|
|
$text |
|
42
|
|
|
) |
|
43
|
|
|
); |
|
44
|
|
|
$text = str_replace(['<small>', '</small>'], '', $text); // ?? |
|
45
|
|
|
|
|
46
|
|
|
return $text; |
|
47
|
|
|
} |
|
48
|
|
|
|
|
49
|
|
|
public static function isWikify(string $text): bool |
|
50
|
|
|
{ |
|
51
|
|
|
if (self::unWikify($text) !== $text) { |
|
52
|
|
|
return true; |
|
53
|
|
|
} |
|
54
|
|
|
|
|
55
|
|
|
return false; |
|
56
|
|
|
} |
|
57
|
|
|
|
|
58
|
|
|
/** |
|
59
|
|
|
* Get page titles from wiki encoded links. |
|
60
|
|
|
* (but not others projects links like [[wikt:bla]]. |
|
61
|
|
|
* |
|
62
|
|
|
* @param string $text |
|
63
|
|
|
* |
|
64
|
|
|
* @return array|null |
|
65
|
|
|
*/ |
|
66
|
|
|
public static function getWikilinkPages(string $text): ?array |
|
67
|
|
|
{ |
|
68
|
|
|
if (preg_match_all('#\[\[([^:|\]]+)(?:\|[^|\]]*)?]]#', $text, $matches) > 0) { |
|
69
|
|
|
return $matches[1]; |
|
70
|
|
|
} |
|
71
|
|
|
|
|
72
|
|
|
return null; |
|
73
|
|
|
} |
|
74
|
|
|
|
|
75
|
|
|
/** |
|
76
|
|
|
* Strip external links (http://) from wiki text. |
|
77
|
|
|
* "[http://google.fr Google]" => "Google" |
|
78
|
|
|
* "bla [http://google.fr]" => "bla" |
|
79
|
|
|
* |
|
80
|
|
|
* @param string $text |
|
81
|
|
|
* |
|
82
|
|
|
* @return string |
|
83
|
|
|
*/ |
|
84
|
|
|
public static function stripExternalLink(string $text): string |
|
85
|
|
|
{ |
|
86
|
|
|
$text = preg_replace('#\[(https?://[^][<>\s"]+) *((?<= )[^\n\]]*|)\]#i', '${2}', $text); |
|
87
|
|
|
|
|
88
|
|
|
return trim($text); |
|
89
|
|
|
} |
|
90
|
|
|
|
|
91
|
|
|
/** |
|
92
|
|
|
* @param string $text |
|
93
|
|
|
* |
|
94
|
|
|
* @return bool |
|
95
|
|
|
*/ |
|
96
|
|
|
public static function isCommented(string $text): bool |
|
97
|
|
|
{ |
|
98
|
|
|
//ou preg_match('#<\!--(?!-->).*-->#s', '', $text); // plus lourd mais précis |
|
99
|
|
|
return (preg_match('#<!--[^>]*-->#', $text) > 0) ? true : false; |
|
100
|
|
|
} |
|
101
|
|
|
|
|
102
|
|
|
/** |
|
103
|
|
|
* Remove '<!--', '-->', and everything between. |
|
104
|
|
|
* To avoid leaving blank lines, when a comment is both preceded |
|
105
|
|
|
* and followed by a newline (ignoring spaces), trim leading and |
|
106
|
|
|
* trailing spaces and one of the newlines. |
|
107
|
|
|
* (c) WikiMedia /includes/parser/Sanitizer.php. |
|
108
|
|
|
* |
|
109
|
|
|
* @param string $text |
|
110
|
|
|
* |
|
111
|
|
|
* @return string |
|
112
|
|
|
*/ |
|
113
|
|
|
public static function removeHTMLcomments(string $text) |
|
114
|
|
|
{ |
|
115
|
|
|
while (false !== ($start = mb_strpos($text, '<!--'))) { |
|
116
|
|
|
$end = mb_strpos($text, '-->', $start + 4); |
|
117
|
|
|
if (false === $end) { |
|
118
|
|
|
// Unterminated comment; bail out |
|
119
|
|
|
break; |
|
120
|
|
|
} |
|
121
|
|
|
$end += 3; |
|
122
|
|
|
// Trim space and newline if the comment is both |
|
123
|
|
|
// preceded and followed by a newline |
|
124
|
|
|
$spaceStart = max($start - 1, 0); |
|
125
|
|
|
$spaceLen = $end - $spaceStart; |
|
126
|
|
|
while (' ' === substr($text, $spaceStart, 1) && $spaceStart > 0) { |
|
127
|
|
|
--$spaceStart; |
|
128
|
|
|
++$spaceLen; |
|
129
|
|
|
} |
|
130
|
|
|
while (' ' === substr($text, $spaceStart + $spaceLen, 1)) { |
|
131
|
|
|
++$spaceLen; |
|
132
|
|
|
} |
|
133
|
|
|
if ("\n" === substr($text, $spaceStart, 1) |
|
134
|
|
|
&& "\n" === substr($text, $spaceStart + $spaceLen, 1) |
|
135
|
|
|
) { |
|
136
|
|
|
// Remove the comment, leading and trailing |
|
137
|
|
|
// spaces, and leave only one newline. |
|
138
|
|
|
$text = substr_replace($text, "\n", $spaceStart, $spaceLen + 1); |
|
139
|
|
|
} else { |
|
140
|
|
|
// Remove just the comment. |
|
141
|
|
|
$text = substr_replace($text, '', $start, $end - $start); |
|
142
|
|
|
} |
|
143
|
|
|
} |
|
144
|
|
|
|
|
145
|
|
|
return $text; |
|
146
|
|
|
} |
|
147
|
|
|
} |
|
148
|
|
|
|
Dispositif /
Wikibot
