Hi,
Friday, July 11, 2003, 4:30:41 AM, you wrote:
MG> Well I've made a signifigant improvement to this function, and it's now the
MG> parsing operation is only something like 26 lines. It works flawlessly (so
MG> far) for XML that does not have repeat element names on the same level.
MG> Which seems to be bad form anyway, using similar element names and
MG> distinguishing them with attributes.
MG> I'd still like the code to work with XML that did have similarly named
MG> elements in the same node level.. can you see anything in my DOMXML calls
MG> that might indicate the problem?
I rearranged the way you did your parsing and came up with this:
function parseBranch($branch, &$doc, &$parent)
{
static $elements = array();
foreach ($branch as $key => $val) {
switch ($key) {
// parent attributes
case ('ATTRIBUTES'):
foreach ($val as $attrName => $attrVal) {
switch ($attrName){
case 'VALUE':
$nodeValue =
$doc->create_text_node($attrVal);
$parentValue =
$parent->append_child($nodeValue);
break;
default:
$parent->set_attribute($attrName, $attrVal);
break;
}
}
break;
// parent value
case ('VALUE'):
$nodeValue = $doc->create_text_node($val);
$parentValue = $parent->append_child($nodeValue);
break;
default:
// add child element
if(!isset($elements[$key])){
$elements[$key] = $doc->create_element($key);
}
// parse children of this node
foreach ($val as $children) {
//create node
$child =
$parent->append_child($elements[$key]->clone_node());
//set its elements
parseBranch($children, $doc, $child);
}
break;
}
}
}
which gives
<?xml version="1.0"?>
<ROOTELEMENT ID="1">
<RECORD ID="1">Value 1</RECORD>
<RECORD ID="2">Value 2</RECORD>
</ROOTELEMENT>
--
regards,
Tom
--
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php
