Function for converting an xml report from Yahoo Web Analytics into a json format

report_to_json.php

<?php

# Function for converting an xml report into a json format
function report_to_json($path)
{
    # Check if the file exists
    if(!file_exists($path)) return false;

    # Set content type to json and charset to utf-8
    # header('Content-type: application/json; charset=utf-8');

    # Variable to hold the output data array
    $outputData = array();

    # Load the xml file as a DOMDocument
    $dom = DOMDocument::load($path);
    $domxpath = new DOMXpath($dom);

    # Ready some xpath queries
    $headersQuery = '/GetReportResponse/TopHeadRows/Row/Item/Name';
    $leftItemsQuery = '/GetReportResponse/LeftHeadItems/Item';
    $dataRowsQuery = '/GetReportResponse/DataRows/Row';

    # Fetch the DOM lists for above queries
    $headersDom = $domxpath->query($headersQuery);
    $leftItemsDom = $domxpath->query($leftItemsQuery);
    $dataRowsDom = $domxpath->query($dataRowsQuery);
    
    # Prepare the headers
    $headers = array("id", "Category", "children");
    $extraHeaders = count($headers);

    # Loop through the headers and fetch the value
    for($i = 0; $i < $headersDom->length; $i++)
        $headers[] = trim($headersDom->item($i)->nodeValue);
    
    #####################
    # Fetch item values #
    #####################

    # Ultimate row container
    $rows = array();

    # Always contains the previous occurrence of a row per indentation level.
    # Used for calculating parent IDs.
    $last = array();

    # Loop once per data row
    for($i = 0; $i < $dataRowsDom->length; $i++)
    {
        # Prepare an array for this row
        unset($row);
        $row = array();

        # Fetch the datarow and get the cells
        $dataRow = $dataRowsDom->item($i);
        $dataCells = $domxpath->query('./Cell', $dataRow);

        # Fetch the left head item
        $leftItem = $leftItemsDom->item($i);

        # The indent for this row
        $indent = $domxpath->query('./Indent', $leftItem)
                           ->item(0)->nodeValue;
        $indent = intval(trim($indent));

        # Find the parent of this row
        if($indent > 0) $parent = &$last[$indent-1];

        # Last
        $last[$indent] = &$row;

        # Loop through all headers and set the values on this row accordingly
        foreach($headers as $h => $header)
        {
            # Header 0 is the id
            if($h === 0)
            {
                $row[$header] = $i+1;
            }
            # Header 1 is the category, value taken from leftItem
            else if($h == 1)
            {
                $cat = $domxpath->query('./Name', $leftItem)
                                ->item(0)->nodeValue;
                $cat = trim($cat);
                $row[$header] = $cat;
            }
            # Header 2 is the children, just set it to an array
            else if($h == 2)
            {
                $row[$header] = array();
            }
            # The rest of the headers have their values in the datarow
            else if($h > 2)
            {
                $dataCell = $dataCells->item($h-$extraHeaders);
                $cellValue = $domxpath->query('./FormattedValue', $dataCell)
                                      ->item(0)->nodeValue;
                $cellValue = trim($cellValue);
                $row[$header] = $cellValue;
            }
        }

        # Add this row to the collected data rows or as the child of it's parent
        if($indent < 1) $rows[] = &$row;
        else $parent["children"][] = &$row;
    }

    // var_dump($rows);
    return prettify(json_encode(($rows)));
}

# Function to prettify the output json
function prettify($json)
{
    $result      = '';
    $pos         = 0;
    $strLen      = strlen($json);
    $indentStr   = '    ';
    $newLine     = "\n";
    $prevChar    = '';
    $outOfQuotes = true;

    for ($i=0; $i<=$strLen; $i++) {

        // Grab the next character in the string.
        $char = substr($json, $i, 1);

        // Are we inside a quoted string?
        if ($char == '"' && $prevChar != '\\') {
            $outOfQuotes = !$outOfQuotes;
        
        // If this character is the end of an element, 
        // output a new line and indent the next line.
        } else if(($char == '}' || $char == ']') && $outOfQuotes) {
            $result .= $newLine;
            $pos --;
            for ($j=0; $j<$pos; $j++) {
                $result .= $indentStr;
            }
        }
        
        // Add the character to the result string.
        $result .= $char;

        // If the last character was the beginning of an element, 
        // output a new line and indent the next line.
        if (($char == ',' || $char == '{' || $char == '[') && $outOfQuotes) {
            $result .= $newLine;
            if ($char == '{' || $char == '[') {
                $pos ++;
            }
            
            for ($j = 0; $j < $pos; $j++) {
                $result .= $indentStr;
            }
        }
        
        $prevChar = $char;
    }

    return $result;
}

# Debug
if(!defined('ASTRUPS'))
{
    header('Content-type: application/json');
    echo report_to_json('data/example-report.xml');
}