|
1
|
|
|
<?php |
|
2
|
|
|
namespace vipnytt; |
|
3
|
|
|
|
|
4
|
|
|
use DOMDocument; |
|
5
|
|
|
use SimpleXMLElement; |
|
6
|
|
|
use vipnytt\OPMLParser\Exceptions; |
|
7
|
|
|
|
|
8
|
|
|
class OPMLParser |
|
9
|
|
|
{ |
|
10
|
|
|
/** |
|
11
|
|
|
* Encoding |
|
12
|
|
|
*/ |
|
13
|
|
|
const ENCODING = 'UTF-8'; |
|
14
|
|
|
|
|
15
|
|
|
/** |
|
16
|
|
|
* Optional <head> elements |
|
17
|
|
|
*/ |
|
18
|
|
|
const OPTIONAL_HEAD_ELEMENTS = [ |
|
19
|
|
|
'title', |
|
20
|
|
|
'dateCreated', |
|
21
|
|
|
'dateModified', |
|
22
|
|
|
'ownerName', |
|
23
|
|
|
'ownerEmail', |
|
24
|
|
|
'ownerId', |
|
25
|
|
|
'docs', |
|
26
|
|
|
'expansionState', |
|
27
|
|
|
'vertScrollState', |
|
28
|
|
|
'windowTop', |
|
29
|
|
|
'windowLeft', |
|
30
|
|
|
'windowBottom', |
|
31
|
|
|
'windowRight' |
|
32
|
|
|
]; |
|
33
|
|
|
|
|
34
|
|
|
/** |
|
35
|
|
|
* XML content |
|
36
|
|
|
* @var string |
|
37
|
|
|
*/ |
|
38
|
|
|
protected $xml; |
|
39
|
|
|
|
|
40
|
|
|
/** |
|
41
|
|
|
* Array containing the parsed XML |
|
42
|
|
|
* @var array |
|
43
|
|
|
*/ |
|
44
|
|
|
protected $result = []; |
|
45
|
|
|
|
|
46
|
|
|
/** |
|
47
|
|
|
* Constructor |
|
48
|
|
|
* |
|
49
|
|
|
* @param string $xml is the string we want to parse |
|
50
|
|
|
* @throws Exceptions\ParseException |
|
51
|
|
|
*/ |
|
52
|
|
|
public function __construct($xml) |
|
53
|
|
|
{ |
|
54
|
|
|
$this->xml = $xml; |
|
55
|
|
|
$dom = new DOMDocument(); |
|
56
|
|
|
$dom->recover = true; |
|
57
|
|
|
$dom->strictErrorChecking = false; |
|
58
|
|
|
$dom->loadXML($this->xml, LIBXML_NOCDATA); |
|
59
|
|
|
$dom->encoding = self::ENCODING; |
|
60
|
|
|
|
|
61
|
|
|
$opml = simplexml_import_dom($dom); |
|
62
|
|
|
|
|
63
|
|
|
if ($opml === false) { |
|
64
|
|
|
throw new Exceptions\ParseException('Provided XML document is not valid'); |
|
65
|
|
|
} |
|
66
|
|
|
|
|
67
|
|
|
$this->result = [ |
|
68
|
|
|
'version' => (string)$opml['version'], |
|
69
|
|
|
'head' => [], |
|
70
|
|
|
'body' => [] |
|
71
|
|
|
]; |
|
72
|
|
|
|
|
73
|
|
|
if (!isset($opml->head)) { |
|
74
|
|
|
throw new Exceptions\ParseException('Provided XML is not an valid OPML document'); |
|
75
|
|
|
} |
|
76
|
|
|
// First, we get all "head" elements. Head is required but its sub-elements are optional. |
|
77
|
|
|
foreach ($opml->head->children() as $key => $value) { |
|
78
|
|
|
if (in_array($key, self::OPTIONAL_HEAD_ELEMENTS, true)) { |
|
79
|
|
|
$this->result['head'][$key] = (string)$value; |
|
80
|
|
|
} |
|
81
|
|
|
} |
|
82
|
|
|
if (!isset($opml->body)) { |
|
83
|
|
|
return; |
|
84
|
|
|
} |
|
85
|
|
|
// Then, we get body outlines. Body must contain at least one outline element. |
|
86
|
|
|
foreach ($opml->body->children() as $key => $value) { |
|
87
|
|
|
if ($key === 'outline') { |
|
88
|
|
|
$this->result['body'][] = $this->parse_outline($value); |
|
89
|
|
|
} |
|
90
|
|
|
} |
|
91
|
|
|
} |
|
92
|
|
|
|
|
93
|
|
|
/** |
|
94
|
|
|
* Parse an XML object as an outline object and return corresponding array |
|
95
|
|
|
* |
|
96
|
|
|
* @param SimpleXMLElement $outline_xml the XML object we want to parse |
|
97
|
|
|
* @return array corresponding to an outline and following format described above |
|
98
|
|
|
*/ |
|
99
|
|
|
protected function parse_outline(SimpleXMLElement $outline_xml) |
|
100
|
|
|
{ |
|
101
|
|
|
$outline = []; |
|
102
|
|
|
foreach ($outline_xml->attributes() as $key => $value) { |
|
103
|
|
|
$outline[$key] = (string)$value; |
|
104
|
|
|
} |
|
105
|
|
|
// Bug fix for OPMLs witch contains `title` but not the required `text` |
|
106
|
|
|
if (empty($outline['text']) && isset($outline['title'])) { |
|
107
|
|
|
$outline['text'] = $outline['title']; |
|
108
|
|
|
} |
|
109
|
|
|
foreach ($outline_xml->children() as $key => $value) { |
|
110
|
|
|
// An outline may contain any number of outline children |
|
111
|
|
|
if ($key === 'outline') { |
|
112
|
|
|
$outline['@outlines'][] = $this->parse_outline($value); |
|
113
|
|
|
} |
|
114
|
|
|
} |
|
115
|
|
|
return $outline; |
|
116
|
|
|
} |
|
117
|
|
|
|
|
118
|
|
|
/** |
|
119
|
|
|
* Return the parsed XML as an Array |
|
120
|
|
|
* |
|
121
|
|
|
* @return array |
|
122
|
|
|
*/ |
|
123
|
|
|
public function getResult() |
|
124
|
|
|
{ |
|
125
|
|
|
return $this->result; |
|
126
|
|
|
} |
|
127
|
|
|
|
|
128
|
|
|
/** |
|
129
|
|
|
* Validate the parsed XML array |
|
130
|
|
|
* Note: The parser support parsing of OPMLs with missing content |
|
131
|
|
|
* |
|
132
|
|
|
* @return string|false Validated string on success, false on failure |
|
133
|
|
|
*/ |
|
134
|
|
|
public function validate() |
|
135
|
|
|
{ |
|
136
|
|
|
try { |
|
137
|
|
|
$render = new OPMLParser\Render($this->result); |
|
138
|
|
|
} catch (Exceptions\RenderException $e) { |
|
139
|
|
|
return false; |
|
140
|
|
|
} |
|
141
|
|
|
return $render->asString(); |
|
142
|
|
|
} |
|
143
|
|
|
} |
|
144
|
|
|
|
VIPnytt /
OPMLParser
