Recently, I was working on a web service which required us to parse the data from the XML feed and to store it in the database. Normally this is a simple task which can be achieved by using PHP’s simple_xml library to parse the data. However, if the document has not been encoded properly, simple_xml will generate an XML.

Whenever an XML document is encoded, the encoding used should be provided in the document.
If the document was encoded with unicode, example, UTF-8, the following would be the first line of the:
<?xml version=”1.0″ encoding=”UTF-8″?>

In my case, the xml document was label as UTF-8, however it was an ascii document which contained non-acii characters. This created a major problem with the parser. The quick solution is to strip the non-ascii characters from the document.
This can be achieved with the following php code:

$jobs = file_get_contents(‘/home/mydir/doc.xml’);
$jobs = preg_replace(‘/[^(\x20-\x7F)]*/’,”, $jobs);

At first glance, one would probably think that using CSS or JavaScript to control the zooming would be the solution. However, this is not the case. Apple has implemented a special META tag to handle this.

<meta content=”width=320; initial-scale=1.0; maximum-scale=1.0; user-scalable=0;”/>

Other mobile browsers are implementing this as well, including the upcoming Mobile Firefox(Fennec) for the Android OS. This should help to make it easier to design sites with consistent behaviors across multiple mobile browsers.