1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
|
<?php
/**
* @file
* @ingroup Testing
*/
class ParserTestResultNormalizer {
protected $doc, $xpath, $invalid;
public static function normalize( $text, $funcs ) {
$norm = new self( $text );
if ( $norm->invalid ) {
return $text;
}
foreach ( $funcs as $func ) {
$norm->$func();
}
return $norm->serialize();
}
protected function __construct( $text ) {
$this->doc = new DOMDocument( '1.0', 'utf-8' );
// Note: parsing a supposedly XHTML document with an XML parser is not
// guaranteed to give accurate results. For example, it may introduce
// differences in the number of line breaks in <pre> tags.
Wikimedia\suppressWarnings();
if ( !$this->doc->loadXML( '<html><body>' . $text . '</body></html>' ) ) {
$this->invalid = true;
}
Wikimedia\restoreWarnings();
$this->xpath = new DOMXPath( $this->doc );
$this->body = $this->xpath->query( '//body' )->item( 0 );
}
protected function removeTbody() {
foreach ( $this->xpath->query( '//tbody' ) as $tbody ) {
while ( $tbody->firstChild ) {
$child = $tbody->firstChild;
$tbody->removeChild( $child );
$tbody->parentNode->insertBefore( $child, $tbody );
}
$tbody->parentNode->removeChild( $tbody );
}
}
/**
* The point of this function is to produce a normalized DOM in which
* Tidy's output matches the output of html5depurate. Tidy both trims
* and pretty-prints, so this requires fairly aggressive treatment.
*
* In particular, note that Tidy converts <pre>x</pre> to <pre>\nx\n</pre>,
* which theoretically affects display since the second line break is not
* ignored by compliant HTML parsers.
*
* This function also removes empty elements, as does Tidy.
*/
protected function trimWhitespace() {
foreach ( $this->xpath->query( '//text()' ) as $child ) {
if ( strtolower( $child->parentNode->nodeName ) === 'pre' ) {
// Just trim one line break from the start and end
if ( substr_compare( $child->data, "\n", 0 ) === 0 ) {
$child->data = substr( $child->data, 1 );
}
if ( substr_compare( $child->data, "\n", -1 ) === 0 ) {
$child->data = substr( $child->data, 0, -1 );
}
} else {
// Trim all whitespace
$child->data = trim( $child->data );
}
if ( $child->data === '' ) {
$child->parentNode->removeChild( $child );
}
}
}
/**
* Serialize the XML DOM for comparison purposes. This does not generate HTML.
* @return string
*/
protected function serialize() {
return strtr( $this->doc->saveXML( $this->body ),
[ '<body>' => '', '</body>' => '' ] );
}
}
|