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
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
|
<?php
/***************************************************************************
* functions_xml.inc.php *
* ------------------- *
* application : PhpWebGallery 1.3 <http://phpwebgallery.net> *
* author : Pierrick LE GALL <pierrick@z0rglub.com> *
* *
* $Id$
* *
***************************************************************************
***************************************************************************
* *
* This program is free software; you can redistribute it and/or modify *
* it under the terms of the GNU General Public License as published by *
* the Free Software Foundation; *
* *
***************************************************************************/
//------------------------------------------------------------------- functions
// getContent returns the content of a tag
//
// example : getContent( "<name>Joe</name>" ) returns "Joe"
//
// It also works with strings containing themself sub-tags :
// <perso><name>Jean</name><firstname>Billie</fisrtname></perso> ->
// <name>Jean</name><firstname>Billie</firstname>
function getContent( $element )
{
// deleting start of the tag
$content = preg_replace( '/^<[^>]+>/', '', $element );
// deleting end of the tag
$content = preg_replace( '/<\/[^>]+>$/', '', $content );
// replacing multiple instance of space character
$content = preg_replace( '/\s+/', ' ', $content );
return $content;
}
// The function get Attribute returns the value corresponding to the
// attribute $attribute for the tag $element.
function getAttribute( $element, $attribute )
{
$regex = '/^<\w+[^>]*'.$attribute.'\s*=\s*"('.VAL_REG.')"/i';
if ( preg_match( $regex, $element, $out ) ) return $out[1];
else return '';
}
function deprecated_getAttribute( $element, $attribute )
{
// Retrieving string with tag name and all attributes
$regex = '/^<\w+( '.ATT_REG.'="'.VAL_REG.'")*/';
preg_match( $regex, $element, $out );
// Splitting string for retrieving separately attributes
// and corresponding values
$regex = '/('.ATT_REG.')="('.VAL_REG.')"/';
preg_match_all( $regex, $out[0], $out );
// Searching and returning the value of the requested attribute
for ( $i = 0; $i < sizeof( $out[0] ); $i++ )
{
if ( $out[1][$i] == $attribute )
{
return $out[2][$i];
}
}
return '';
}
// The function getChild returns the first child
// exemple : getChild( "<table><tr>XXX</tr><tr>YYY</tr></table>", "tr" )
// returns "<tr>XXX</tr>"
function getChild( $document, $node )
{
$regex = '/<'.$node.'(\s+'.ATT_REG.'="'.VAL_REG.'")*';
$regex.= '(\s*\/>|>.*<\/'.$node.'>)/U';
preg_match( $regex, $document, $out );
return $out[0];
}
// getChildren returns a list of the children identified by the $node
// example :
// getChild( "<table><tr>XXX</tr><tr>YYY</tr></table>", "tr" )
// returns an array with :
// $array[0] equals "<tr>XXX</tr>"
// $array[1] equals "<tr>YYY</tr>"
function getChildren( $document, $node )
{
$regex = '/<'.$node.'(\s+'.ATT_REG.'="'.VAL_REG.'")*';
$regex.= '(\s*\/>|>.*<\/'.$node.'>)/U';
preg_match_all( $regex, $document, $out );
return $out[0];
}
// get_CodeXML places the content of a text file in a PHP variable and
// return it. If the file can't be opened, returns false.
function getXmlCode( $filename )
{
$file = fopen( $filename, 'r' );
if ( !$file )
{
return false;
}
while ( !feof( $file ) )
{
$xml_content .= fgets( $file, 1024 );
}
fclose( $file );
$xml_content = str_replace( "\n", '', $xml_content );
$xml_content = str_replace( "\t", '', $xml_content );
return $xml_content;
}
?>
|