Flourish PHP Unframework


The fXML class provides functionality to read and create XML. An fXML object represents an XML string for reading/traversing, but does not allow modification. The static methods encode() and sendHeader() are useful when creating XML.


The fXML constructor will accept four different types of sources for the XML to represent:

  1. An fFile object
  2. A file path
  3. A URL
  4. An XML string
// From an fFile
$file = new fFile('/path/to/file');
$xml  = new fXML($file);

// From a file path
$xml  = new fXML('/path/to/file');

// From a URL
$xml  = new fXML('http://example.com/rss');

// From an XML string
$xml  = new fXML('<?xml ');

XML files with invalid data will cause an fValidationException to be thrown.

When fetching XML from a URL, an optional second parameter, $timeout, can be specified. If not specified, the default_socket_timeout ini setting is used.

// Fetch the XML with a 5 second timeout
$xml = new fXML('http://examples.com/rss', 5);

Fixing Invalid XML

Two of the most common errors when creating XML are encoding characters as HTML entities for XML and delivering ISO-8859-1 (or Windows-1252) encoded XML as UTF-8. An optional boolean parameter, $fix_entities_encoding can fix both of these problems.

// Used with an HTTP timeout
$xml = new fXML('http://example.com/rss', 5, TRUE);

// Used without a timeout
$xml = new fXML('./foo.xml', TRUE);

Invalid Named Entities

In XML, there are only five pre-defined named entities: &amp;, &gt;, &lt;, &quot; and &apos;. All other named entities from HTML will cause a parse error if included in XML without further encoding. For instance, &rsquo; is invalid, but &amp;rsquo; or is valid.

$fix_entities_encoding will take

<book><title>Isn&rsquo;t Valid XML</title></book>

and convert it to

<book><title>Isnt Valid XML</title></book>

before passing the XML to the parser.

Incorrectly Encoded Content

If no encoding is specified for an XML document, the encoded is assumed to be UTF-8. Many developers not familiar with XML and issues related to encoding will omit the encoding attribute of the <?xml> tag and will insert ISO-8859-1 or Windows-1252 (also called Latin or Latin 1) content.

fXML will throw an exception when such an XML document is parsed, since the parser being used will find invalid UTF-8 characters and give up encoding. The $fix_entities_encoding parameter will detect non-UTF-8 characters for documents defined as UTF-8 (whether explicitly or by omission of the encoding attribute) and convert the content.

Element Information

Information about the root XML element that was passed into the constructor can be accessed by the following methods:

Method Description
getName() Returns the name of the element without any preceding namespace prefix
getNamespace() Returns the full namespace URI for the element, if any
getPrefix() Returns the namespace prefix for the element, if any
getText() Returns the text content of the element


The raw XML being modeled by the fXML object can be retrieved by calling the method toXML().

echo $xml->toXML();

Attribute Values

Attributes of the XML element can be accessed using array notation.

// Normal attribute
echo $xml['my_attribute'];

// Attribute with namespace prefix
echo $xml['ns:attribute'];

Child Element Text Content

The text content of children of the XML element can be accessed by requesting object memebers.

// The content of the child element "firstName"
echo $xml->firstName;

// Child "title" in the "ns" prefix
echo $xml->{'ns:title'};


For anything beyond attribute and child element text content of the root XML element, the xpath() method will need to be used. This method returns an array of all matching parts of the XML file.

If you aren't familiar with XPath, the Wikipedia page about XPath 1.0 is a good place to start. XPath allow traversal of the XML file using element names combined with predicates and functions.

xpath() accepts two parameter, the XPath $path, and optionally a boolean flag, $first_only, to return only the first match.

// Select every "item" element in the "channel"
foreach ($xml->xpath("channel/item") as $item) {
    echo $item->title;

The items matched by xpath() may be child elements, text content or attributes. Here are the data types for various types of matches:

It is also possible to pull back just the first matched element by passing the second parameter as TRUE.

// Pull the first attribute only, thus returning a string
echo $xml->xpath('@attribute', TRUE);

Custom Prefixes

When using XPath, array-style attribute access or child element text content access, it can be useful to set a custom namespace prefix to allow for compatibility with various XML sources. The method addCustomPrefix() accepts two parameters, the $ns_prefix to register and its corresponding $namespace. One registered, this prefix will be available throughout fXML.

$xml = new fXML('http://example.com/rss');
$xml->addCustomPrefix('pf', 'http://namspace');

echo $xml['pf:attribute'];
echo $xml->{'pf:child'};
foreach ($xml->xpath('pf:*') as $element) {
    echo $element->getText();

Encoding Data

When creating XML documents, such as RSS feeds, it is necessary to create properly encoded XML, otherwise strict XML parsers will not be able to parse the document. The encode() method will ensure that all content is properly encoded for including in a UTF-8 XML file.

Here is an example of usage:

$xml  = '<?xml version="1.0" encoding="UTF-8"?>' . "\n";
$xml .= "<articles>";

foreach ($articles as $article) {
    $xml .= "<article><title>";
    $xml .= fXML::encode($article->getTitle());
    $xml .= "</title><description>";
    $xml .= fXML::encode($article->getDescription());
    $xml .= "</description></article>";

$xml .= "</articles>";

Sending the Content Type Header

When sending XML files over HTTP, the method sendHeader() should be called ensure that the Content-Type header is set to the correct value of text/xml; charset=utf-8. The utf-8 character set encoding is specified since all of Flourish is built to work with UTF-8 and all XML parsers must support that character set.