|
1
|
|
|
<?php |
|
2
|
|
|
namespace Masterminds\HTML5\Serializer; |
|
3
|
|
|
|
|
4
|
|
|
/** |
|
5
|
|
|
* Traverser for walking a DOM tree. |
|
6
|
|
|
* |
|
7
|
|
|
* This is a concrete traverser designed to convert a DOM tree into an |
|
8
|
|
|
* HTML5 document. It is not intended to be a generic DOMTreeWalker |
|
9
|
|
|
* implementation. |
|
10
|
|
|
* |
|
11
|
|
|
* @see http://www.w3.org/TR/2012/CR-html5-20121217/syntax.html#serializing-html-fragments |
|
12
|
|
|
*/ |
|
13
|
|
|
class Traverser |
|
14
|
|
|
{ |
|
15
|
|
|
|
|
16
|
|
|
/** |
|
17
|
|
|
* Namespaces that should be treated as "local" to HTML5. |
|
18
|
|
|
*/ |
|
19
|
|
|
static $local_ns = array( |
|
20
|
|
|
'http://www.w3.org/1999/xhtml' => 'html', |
|
21
|
|
|
'http://www.w3.org/1998/Math/MathML' => 'math', |
|
22
|
|
|
'http://www.w3.org/2000/svg' => 'svg' |
|
23
|
|
|
); |
|
24
|
|
|
|
|
25
|
|
|
protected $dom; |
|
26
|
|
|
|
|
27
|
|
|
protected $options; |
|
28
|
|
|
|
|
29
|
|
|
protected $encode = false; |
|
30
|
|
|
|
|
31
|
|
|
protected $rules; |
|
32
|
|
|
|
|
33
|
|
|
protected $out; |
|
34
|
|
|
|
|
35
|
|
|
/** |
|
36
|
|
|
* Create a traverser. |
|
37
|
|
|
* |
|
38
|
|
|
* @param DOMNode|DOMNodeList $dom |
|
39
|
|
|
* The document or node to traverse. |
|
40
|
|
|
* @param resource $out |
|
41
|
|
|
* A stream that allows writing. The traverser will output into this |
|
42
|
|
|
* stream. |
|
43
|
|
|
* @param array $options |
|
44
|
|
|
* An array or options for the traverser as key/value pairs. These include: |
|
45
|
|
|
* - encode_entities: A bool to specify if full encding should happen for all named |
|
46
|
|
|
* charachter references. Defaults to false which escapes &'<>". |
|
47
|
|
|
* - output_rules: The path to the class handling the output rules. |
|
48
|
|
|
*/ |
|
49
|
64 |
|
public function __construct($dom, $out, RulesInterface $rules, $options = array()) |
|
50
|
|
|
{ |
|
51
|
64 |
|
$this->dom = $dom; |
|
52
|
64 |
|
$this->out = $out; |
|
53
|
64 |
|
$this->rules = $rules; |
|
54
|
64 |
|
$this->options = $options; |
|
55
|
|
|
|
|
56
|
64 |
|
$this->rules->setTraverser($this); |
|
57
|
64 |
|
} |
|
58
|
|
|
|
|
59
|
|
|
/** |
|
60
|
|
|
* Tell the traverser to walk the DOM. |
|
61
|
|
|
* |
|
62
|
|
|
* @return resource $out |
|
63
|
|
|
* Returns the output stream. |
|
64
|
|
|
*/ |
|
65
|
22 |
|
public function walk() |
|
66
|
|
|
{ |
|
67
|
22 |
|
if ($this->dom instanceof \DOMDocument) { |
|
68
|
16 |
|
$this->rules->document($this->dom); |
|
69
|
22 |
|
} elseif ($this->dom instanceof \DOMDocumentFragment) { |
|
70
|
|
|
// Document fragments are a special case. Only the children need to |
|
71
|
|
|
// be serialized. |
|
72
|
13 |
|
if ($this->dom->hasChildNodes()) { |
|
73
|
13 |
|
$this->children($this->dom->childNodes); |
|
74
|
13 |
|
} |
|
75
|
13 |
|
} // If NodeList, loop |
|
76
|
|
|
elseif ($this->dom instanceof \DOMNodeList) { |
|
77
|
|
|
// If this is a NodeList of DOMDocuments this will not work. |
|
78
|
|
|
$this->children($this->dom); |
|
79
|
|
|
} // Else assume this is a DOMNode-like datastructure. |
|
80
|
|
|
else { |
|
81
|
|
|
$this->node($this->dom); |
|
82
|
|
|
} |
|
83
|
|
|
|
|
84
|
22 |
|
return $this->out; |
|
85
|
|
|
} |
|
86
|
|
|
|
|
87
|
|
|
/** |
|
88
|
|
|
* Process a node in the DOM. |
|
89
|
|
|
* |
|
90
|
|
|
* @param mixed $node |
|
91
|
|
|
* A node implementing \DOMNode. |
|
92
|
|
|
*/ |
|
93
|
27 |
|
public function node($node) |
|
94
|
|
|
{ |
|
95
|
|
|
// A listing of types is at http://php.net/manual/en/dom.constants.php |
|
96
|
27 |
|
switch ($node->nodeType) { |
|
97
|
27 |
|
case XML_ELEMENT_NODE: |
|
98
|
24 |
|
$this->rules->element($node); |
|
99
|
24 |
|
break; |
|
100
|
27 |
|
case XML_TEXT_NODE: |
|
101
|
23 |
|
$this->rules->text($node); |
|
102
|
23 |
|
break; |
|
103
|
19 |
|
case XML_CDATA_SECTION_NODE: |
|
104
|
1 |
|
$this->rules->cdata($node); |
|
105
|
1 |
|
break; |
|
106
|
19 |
|
case XML_PI_NODE: |
|
107
|
2 |
|
$this->rules->processorInstruction($node); |
|
108
|
2 |
|
break; |
|
109
|
17 |
|
case XML_COMMENT_NODE: |
|
110
|
2 |
|
$this->rules->comment($node); |
|
111
|
2 |
|
break; |
|
112
|
|
|
// Currently we don't support embedding DTDs. |
|
113
|
17 |
|
default: |
|
114
|
|
|
//print '<!-- Skipped -->'; |
|
115
|
17 |
|
break; |
|
116
|
27 |
|
} |
|
117
|
27 |
|
} |
|
118
|
|
|
|
|
119
|
|
|
/** |
|
120
|
|
|
* Walk through all the nodes on a node list. |
|
121
|
|
|
* |
|
122
|
|
|
* @param \DOMNodeList $nl |
|
123
|
|
|
* A list of child elements to walk through. |
|
124
|
|
|
*/ |
|
125
|
27 |
|
public function children($nl) |
|
126
|
|
|
{ |
|
127
|
27 |
|
foreach ($nl as $node) { |
|
128
|
27 |
|
$this->node($node); |
|
129
|
27 |
|
} |
|
130
|
27 |
|
} |
|
131
|
|
|
|
|
132
|
|
|
/** |
|
133
|
|
|
* Is an element local? |
|
134
|
|
|
* |
|
135
|
|
|
* @param mixed $ele |
|
136
|
|
|
* An element that implement \DOMNode. |
|
137
|
|
|
* |
|
138
|
|
|
* @return bool True if local and false otherwise. |
|
139
|
|
|
*/ |
|
140
|
28 |
|
public function isLocalElement($ele) |
|
141
|
|
|
{ |
|
142
|
28 |
|
$uri = $ele->namespaceURI; |
|
143
|
28 |
|
if (empty($uri)) { |
|
144
|
2 |
|
return false; |
|
145
|
|
|
} |
|
146
|
|
|
|
|
147
|
28 |
|
return isset(static::$local_ns[$uri]); |
|
148
|
|
|
} |
|
149
|
|
|
} |
|
150
|
|
|
|