| 1 |
efrain |
1 |
<?php
|
|
|
2 |
|
|
|
3 |
/**
|
|
|
4 |
* HTML parser implementation. It only implements links.
|
|
|
5 |
*
|
|
|
6 |
* @author Josep Arús
|
|
|
7 |
*
|
|
|
8 |
* @license http://www.gnu.org/copyleft/gpl.html GNU Public License
|
|
|
9 |
* @package mod_wiki
|
|
|
10 |
*/
|
|
|
11 |
include_once("nwiki.php");
|
|
|
12 |
|
|
|
13 |
class html_parser extends nwiki_parser {
|
|
|
14 |
protected $blockrules = array();
|
|
|
15 |
|
|
|
16 |
protected $section_editing = true;
|
|
|
17 |
|
|
|
18 |
/** @var int Minimum level of the headers on the page (usually tinymce uses <h1> and atto <h3>) */
|
|
|
19 |
protected $minheaderlevel = null;
|
|
|
20 |
|
|
|
21 |
public function __construct() {
|
|
|
22 |
parent::__construct();
|
|
|
23 |
// The order is important, headers should be parsed before links.
|
|
|
24 |
$this->tagrules = array(
|
|
|
25 |
// Headers are considered tags here.
|
|
|
26 |
'header' => array(
|
|
|
27 |
'expression' => "/<\s*h([1-6])\s*>(.+?)<\/h[1-6]>/is"
|
|
|
28 |
),
|
|
|
29 |
'link' => $this->tagrules['link'],
|
|
|
30 |
'url' => $this->tagrules['url']
|
|
|
31 |
);
|
|
|
32 |
}
|
|
|
33 |
|
|
|
34 |
/**
|
|
|
35 |
* Find minimum header level used on the page (<h1>, <h3>, ...)
|
|
|
36 |
*
|
|
|
37 |
* @param string $text
|
|
|
38 |
* @return int
|
|
|
39 |
*/
|
|
|
40 |
protected function find_min_header_level($text) {
|
|
|
41 |
preg_match_all($this->tagrules['header']['expression'], $text, $matches);
|
|
|
42 |
return !empty($matches[1]) ? min($matches[1]) : 1;
|
|
|
43 |
}
|
|
|
44 |
|
|
|
45 |
protected function before_parsing() {
|
|
|
46 |
parent::before_parsing();
|
|
|
47 |
|
|
|
48 |
$this->minheaderlevel = $this->find_min_header_level($this->string);
|
|
|
49 |
|
|
|
50 |
// Protect all explicit links from further wiki parsing. The link text may contain another URL which would get
|
|
|
51 |
// converted into another link via {@see nwiki_parser::$tagrules} 'url' element.
|
|
|
52 |
if (preg_match_all('/<a\s[^>]+?>(.*?)<\/a>/is', $this->string, $matches)) {
|
|
|
53 |
foreach (array_unique($matches[0]) as $match) {
|
|
|
54 |
$this->string = str_replace($match, $this->protect($match), $this->string);
|
|
|
55 |
}
|
|
|
56 |
}
|
|
|
57 |
|
|
|
58 |
$this->rules($this->string);
|
|
|
59 |
}
|
|
|
60 |
|
|
|
61 |
/**
|
|
|
62 |
* Header tag rule
|
|
|
63 |
* @param array $match Header regex match
|
|
|
64 |
* @return string
|
|
|
65 |
*/
|
|
|
66 |
protected function header_tag_rule($match) {
|
|
|
67 |
return $this->generate_header($match[2], (int)$match[1] - $this->minheaderlevel + 1);
|
|
|
68 |
}
|
|
|
69 |
|
|
|
70 |
/**
|
|
|
71 |
* Section editing: Special for HTML Parser (It parses <h1></h1>)
|
|
|
72 |
*/
|
|
|
73 |
|
|
|
74 |
public function get_section($header, $text, $clean = false) {
|
|
|
75 |
if ($clean) {
|
|
|
76 |
$text = preg_replace('/\r\n/', "\n", $text);
|
|
|
77 |
$text = preg_replace('/\r/', "\n", $text);
|
|
|
78 |
$text .= "\n\n";
|
|
|
79 |
}
|
|
|
80 |
|
|
|
81 |
$minheaderlevel = $this->find_min_header_level($text);
|
|
|
82 |
|
|
|
83 |
$h1 = array("<\s*h{$minheaderlevel}\s*>", "<\/h{$minheaderlevel}>");
|
|
|
84 |
|
|
|
85 |
$regex = "/(.*?)({$h1[0]}\s*".preg_quote($header, '/')."\s*{$h1[1]}.*?)((?:{$h1[0]}.*)|$)/is";
|
|
|
86 |
preg_match($regex, $text, $match);
|
|
|
87 |
|
|
|
88 |
if (!empty($match)) {
|
|
|
89 |
return array($match[1], $match[2], $match[3]);
|
|
|
90 |
} else {
|
|
|
91 |
return false;
|
|
|
92 |
}
|
|
|
93 |
}
|
|
|
94 |
|
|
|
95 |
protected function get_repeated_sections(&$text, $repeated = array()) {
|
|
|
96 |
$this->repeated_sections = $repeated;
|
|
|
97 |
return preg_replace_callback($this->tagrules['header'], array($this, 'get_repeated_sections_callback'), $text);
|
|
|
98 |
}
|
|
|
99 |
|
|
|
100 |
protected function get_repeated_sections_callback($match) {
|
|
|
101 |
$text = trim($match[2]);
|
|
|
102 |
|
|
|
103 |
if (in_array($text, $this->repeated_sections)) {
|
|
|
104 |
$this->returnvalues['repeated_sections'][] = $text;
|
|
|
105 |
return parser_utils::h('p', $text);
|
|
|
106 |
} else {
|
|
|
107 |
$this->repeated_sections[] = $text;
|
|
|
108 |
}
|
|
|
109 |
|
|
|
110 |
return $match[0];
|
|
|
111 |
}
|
|
|
112 |
}
|