There is a common "trick" often proposed to convert a SimpleXML object to an array, by running it through json_encode() and then json_decode(). I'd like to explain why this is a bad idea.
Most simply, because the whole point of SimpleXML is to be easier to use and more powerful than a plain array. For instance, you can write <?php $foo->bar->baz['bing'] ?> and it means the same thing as <?php $foo->bar[0]->baz[0]['bing'] ?>, regardless of how many bar or baz elements there are in the XML; and if you write <?php (string)$foo->bar[0]->baz[0] ?> you get all the string content of that node - including CDATA sections - regardless of whether it also has child elements or attributes. You also have access to namespace information, the ability to make simple edits to the XML, and even the ability to "import" into a DOM object, for much more powerful manipulation. All of this is lost by turning the object into an array rather than reading understanding the examples on this page.
Additionally, because it is not designed for this purpose, the conversion to JSON and back will actually lose information in some situations. For instance, any elements or attributes in a namespace will simply be discarded, and any text content will be discarded if an element also has children or attributes. Sometimes, this won't matter, but if you get in the habit of converting everything to arrays, it's going to sting you eventually.
Of course, you could write a smarter conversion, which didn't have these limitations, but at that point, you are getting no value out of SimpleXML at all, and should just use the lower level XML Parser functions, or the XMLReader class, to create your structure. You still won't have the extra convenience functionality of SimpleXML, but that's your loss.
Utilisation de base SimpleXML
Plusieurs exemples de ce chapitre requièrent une chaîne XML. Plutôt que de la répéter à chaque exemple, nous allons la placer dans un fichier que nous inclurons dans chacun d'entre eux. Le contenu de ce fichier est illustré par l'exemple qui suit. Autrement, vous pouvez créer un document XML et le lire avec simplexml_load_file().
Exemple #1 Fichier avec une chaîne XML qui sera inclus partout
<?php
$xmlstr = <<<XML
<?xml version='1.0' standalone='yes'?>
<movies>
<movie>
<title>PHP: Behind the Parser</title>
<characters>
<character>
<name>Ms. Coder</name>
<actor>Onlivia Actora</actor>
</character>
<character>
<name>Mr. Coder</name>
<actor>El ActÓr</actor>
</character>
</characters>
<plot>
So, this language. It's like, a programming language. Or is it a
scripting language? All is revealed in this thrilling horror spoof
of a documentary.
</plot>
<great-lines>
<line>PHP solves all my web problems</line>
</great-lines>
<rating type="thumbs">7</rating>
<rating type="stars">5</rating>
</movie>
</movies>
XML;
?>La simplicité de SimpleXML apparaît plus clairement lorsqu'on essaye d'extraire une chaîne ou un nombre d'un document XML basique.
Exemple #2 Lecture de <plot>
<?php
include 'example.php';
$movies = new SimpleXMLElement($xmlstr);
echo $movies->movie[0]->plot;
?>L'exemple ci-dessus va afficher :
So, this language. It's like, a programming language. Or is it a scripting language? All is revealed in this thrilling horror spoof of a documentary.
L'accès aux éléments d'un document XML qui contient des caractères non permis par rapport à la convention de nommage de PHP (e.g. les mots clés) est possible en encapsulant le nom de l'élément entre crochets et apostrophes.
Exemple #3 Récupération de <line>
<?php
include 'example.php';
$movies = new SimpleXMLElement($xmlstr);
echo $movies->movie->{'great-lines'}->line;
?>L'exemple ci-dessus va afficher :
PHP solves all my web problems
Exemple #4 Accéder à un élément non-unique avec SimpleXML
Lorsque plusieurs instances d'un élément existent en tant que fils d'un élément père unique, les techniques normales d'itération peuvent être appliquées.
<?php
include 'example.php';
$movies = new SimpleXMLElement($xmlstr);
/* Pour chaque <character>, nous affichons un <name>. */
foreach ($movies->movie->characters->character as $character) {
echo $character->name, ' played by ', $character->actor, PHP_EOL;
}
?>L'exemple ci-dessus va afficher :
Ms. Coder played by Onlivia Actora Mr. Coder played by El ActÓr
Note:
Les propriétés (
$movies->moviedans notre précédent exemple) ne sont pas des tableaux. Ceux sont des objets itérables et accessibles.
Exemple #5 Utilisation des attributs
Jusque là, nous n'avons couvert que la lecture des noms d'éléments et leurs valeurs. SimpleXML peut aussi atteindre leurs attributs. L'accès aux attributs d'un élément se fait de la même façon que l'accès aux éléments d'un tableau.
<?php
include 'example.php';
$movies = new SimpleXMLElement($xmlstr);
/* Accès au node <rating> du premier <movie>.
* Affichage des attributs de <rating> également. */
foreach ($movies->movie[0]->rating as $rating) {
switch((string) $rating['type']) { // Récupération des attributs comme indices d'éléments
case 'thumbs':
echo $rating, ' thumbs up';
break;
case 'stars':
echo $rating, ' stars';
break;
}
}
?>L'exemple ci-dessus va afficher :
7 thumbs up5 stars
Exemple #6 Comparaison des éléments et des attributs avec du texte
Pour comparer un élément ou un attribut avec une chaîne de caractères
ou pour le passer à une fonction qui nécessite une chaîne de caractères,
vous devez le transtyper en une chaîne en utilisant
(string).
Sinon, PHP traitera l'élément comme un objet.
<?php
include 'example.php';
$movies = new SimpleXMLElement($xmlstr);
if ((string) $movies->movie->title == 'PHP: Behind the Parser') {
print 'My favorite movie.';
}
echo htmlentities((string) $movies->movie->title);
?>L'exemple ci-dessus va afficher :
My favorite movie.PHP: Behind the Parser
Exemple #7 Comparaison de 2 éléments
Deux objets SimpleXMLElement sont considérés comme différents même s'ils pointent vers le même élément.
<?php
include 'example.php';
$movies1 = new SimpleXMLElement($xmlstr);
$movies2 = new SimpleXMLElement($xmlstr);
var_dump($movies1 == $movies2);
?>L'exemple ci-dessus va afficher :
bool(false)
Exemple #8 Utilisation de XPath
SimpleXML inclut le support embarqué de XPath.
Pour trouver tous les éléments <character> :
<?php
include 'example.php';
$movies = new SimpleXMLElement($xmlstr);
foreach ($movies->xpath('//character') as $character) {
echo $character->name, ' played by ', $character->actor, PHP_EOL;
}
?>
'//' sert de joker. Pour spécifier un chemin absolu,
enlevez un slash.
L'exemple ci-dessus va afficher :
Ms. Coder played by Onlivia Actora Mr. Coder played by El ActÓr
Exemple #9 Attribuer des valeurs
Les données dans SimpleXML n'ont pas à être constantes. L'objet permet la manipulation de tous ces éléments.
<?php
include 'example.php';
$movies = new SimpleXMLElement($xmlstr);
$movies->movie[0]->characters->character[0]->name = 'Miss Coder';
echo $movies->asXML();
?>L'exemple ci-dessus va afficher :
<?xml version="1.0" standalone="yes"?>
<movies>
<movie>
<title>PHP: Behind the Parser</title>
<characters>
<character>
<name>Miss Coder</name>
<actor>Onlivia Actora</actor>
</character>
<character>
<name>Mr. Coder</name>
<actor>El ActÓr</actor>
</character>
</characters>
<plot>
So, this language. It's like, a programming language. Or is it a
scripting language? All is revealed in this thrilling horror spoof
of a documentary.
</plot>
<great-lines>
<line>PHP solves all my web problems</line>
</great-lines>
<rating type="thumbs">7</rating>
<rating type="stars">5</rating>
</movie>
</movies>
Exemple #10 Ajout d'éléments et d'attributs
SimpleXML est capable d'ajouter facilement des enfants et des attributs.
<?php
include 'example.php';
$movies = new SimpleXMLElement($xmlstr);
$character = $movies->movie[0]->characters->addChild('character');
$character->addChild('name', 'Mr. Parser');
$character->addChild('actor', 'John Doe');
$rating = $movies->movie[0]->addChild('rating', 'PG');
$rating->addAttribute('type', 'mpaa');
echo $movies->asXML();
?>L'exemple ci-dessus va afficher :
<?xml version="1.0" standalone="yes"?>
<movies>
<movie>
<title>PHP: Behind the Parser</title>
<characters>
<character>
<name>Ms. Coder</name>
<actor>Onlivia Actora</actor>
</character>
<character>
<name>Mr. Coder</name>
<actor>El ActÓr</actor>
</character>
<character><name>Mr. Parser</name><actor>John Doe</actor></character></characters>
<plot>
So, this language. It's like, a programming language. Or is it a
scripting language? All is revealed in this thrilling horror spoof
of a documentary.
</plot>
<great-lines>
<line>PHP solves all my web problems</line>
</great-lines>
<rating type="thumbs">7</rating>
<rating type="stars">5</rating>
<rating type="mpaa">PG</rating></movie>
</movies>
Exemple #11 Interopérabilité DOM
PHP possède un mécanisme pour convertir les nœuds XML entre les formats SimpleXML et DOM. Cet exemple montre comment changer un élément DOM en SimpleXML.
<?php
$dom = new DOMDocument;
$dom->loadXML('<books><book><title>blah</title></book></books>');
if (!$dom) {
echo 'Erreur lors de l\'analyse du document';
exit;
}
$books = simplexml_import_dom($dom);
echo $books->book[0]->title;
?>L'exemple ci-dessus va afficher :
blah
add a note
User Contributed Notes 10 notes
rowan dot collins at gmail dot com ¶
9 years ago
jishcem at gmail dot com ¶
11 years ago
For me it was easier to use arrays than objects,
So, I used this code,
$xml = simplexml_load_file('xml_file.xml');
$json_string = json_encode($xml);
$result_array = json_decode($json_string, TRUE);
Hope it would help someone
Anonymous ¶
7 years ago
If your xml string contains booleans encoded with "0" and "1", you will run into problems when you cast the element directly to bool:
$xmlstr = <<<XML
<?xml version='1.0' standalone='yes'?>
<values>
<truevalue>1</truevalue>
<falsevalue>0</falsevalue>
</values>
XML;
$values = new SimpleXMLElement($xmlstr);
$truevalue = (bool)$values->truevalue; // true
$falsevalue = (bool)$values->falsevalue; // also true!!!
Instead you need to cast to string or int first:
$truevalue = (bool)(int)$values->truevalue; // true
$falsevalue = (bool)(int)$values->falsevalue; // false
gkokmdam at zonnet dot nl ¶
13 years ago
A quick tip on xpath queries and default namespaces. It looks like the XML-system behind SimpleXML has the same workings as I believe the XML-system .NET uses: when one needs to address something in the default namespace, one will have to declare the namespace using registerXPathNamespace and then use its prefix to address the otherwise in the default namespace living element.
<?php
$string = <<<XML
<?xml version='1.0'?>
<document xmlns="http://www.w3.org/2005/Atom">
<title>Forty What?</title>
<from>Joe</from>
<to>Jane</to>
<body>
I know that's the answer -- but what's the question?
</body>
</document>
XML;
$xml = simplexml_load_string($string);
$xml->registerXPathNamespace("def", "http://www.w3.org/2005/Atom");
$nodes = $xml->xpath("//def:document/def:title");
?>
ie dot raymond at gmail dot com ¶
14 years ago
If you need to output valid xml in your response, don't forget to set your header content type to xml in addition to echoing out the result of asXML():
<?php
$xml=simplexml_load_file('...');
...
...xml stuff
...
//output xml in your response:
header('Content-Type: text/xml');
echo $xml->asXML();
?>
kdos ¶
13 years ago
Using stuff like: is_object($xml->module->admin) to check if there actually is a node called "admin", doesn't seem to work as expected, since simplexml always returns an object- in that case an empty one - even if a particular node does not exist.
For me good old empty() function seems to work just fine in such cases.
Cheers
Max K. ¶
14 years ago
From the README file:
SimpleXML is meant to be an easy way to access XML data.
SimpleXML objects follow four basic rules:
1) properties denote element iterators
2) numeric indices denote elements
3) non numeric indices denote attributes
4) string conversion allows to access TEXT data
When iterating properties then the extension always iterates over
all nodes with that element name. Thus method children() must be
called to iterate over subnodes. But also doing the following:
foreach ($obj->node_name as $elem) {
// do something with $elem
}
always results in iteration of 'node_name' elements. So no further
check is needed to distinguish the number of nodes of that type.
When an elements TEXT data is being accessed through a property
then the result does not include the TEXT data of subelements.
Known issues
============
Due to engine problems it is currently not possible to access
a subelement by index 0: $object->property[0].
php at keith tyler dot com ¶
14 years ago
[Editor's Note: The SimpleXMLIterator class, however, does implement these methods.]
While SimpleXMLElement claims to be iterable, it does not seem to implement the standard Iterator interface functions like ::next and ::reset properly. Therefore while foreach() works, functions like next(), current(), or each() don't seem to work as you would expect -- the pointer never seems to move or keeps getting reset.
Josef ¶
3 years ago
How to find out if a Node exists:
<?xml version='1.0' standalone='yes'?>
<book>
<author>Josef</author>
<isbn></isbn>
</book>
empty($xml->isbn) will be true
isset($xml->isbn) will be true
empty($xml->title) will be true
isset($xml->title) will be false
radams at circlepix com ¶
15 years ago
To test whether an element exists:
<?php
$xml = <<<EOT
<?xml version='1.0' standalone='yes'?>
<root>
<test1></test1>
<test2 />
<test4> </test4>
</root>
EOT;
$xmlDoc = new SimpleXMLElement($xml);
echo "Test1: \n";
var_dump($xmlDoc->test1);
echo "\n(" . (bool)$xmlDoc->test1 . ")";
echo "\n\n";
echo "Test2: \n";
var_dump($xmlDoc->test2);
echo "\n(" . (bool)$xmlDoc->test2 . ")";
echo "\n\n";
echo "Test3: \n";
var_dump($xmlDoc->test3);
echo "\n(" . (bool)$xmlDoc->test3 . ")";
echo "\n\n";
echo "Test4: \n";
var_dump($xmlDoc->test4);
echo "\n(" . (bool)$xmlDoc->test4 . ")";
echo "\n\n";
?>
The var_dumps for test1, test2, and test3 are identical, but the (bool) test gives a '1' for test1 and test2, and a '' for test3.
