Reading XML with Javascript

Question

I am reading a XML file through SpineTix which is built on Javascript. I've created a custom parser for this.

This is the XML structure

<ArlandaExpress>
<info>
<updated date="2013-06-10" time="10:28:42"/>
<message priority="1">
<![CDATA[
Biljettköp ombord kostar 100:- extra, ticket on board 100 SEK extra
]]>
</message>
<message priority="2">
<![CDATA[
Biljettköp ombord kostar 100:- extra, ticket on board 100 SEK extra
]]>
</message>
<message priority="3">
<![CDATA[
Restid 20 min. 2 för 380 kr tor-sön och röda dagar t.o.m.16/6// Traveltime 20 min. 2 for 380 SEK Thu-Sun and bank holidays until 16/6
]]>
</message>
</info>
  <StockholmC>
    <next minutes="6"/>
    <upcoming datetime="2013-06-10 10:50"/>
    <upcoming datetime="2013-06-10 11:05"/>
    <upcoming datetime="2013-06-10 11:20"/>
  </StockholmC>
</ArlandaExpress>

I need to grab the minutes and datetime attributes in the StockholmC tag. How can I do this? This is my code so far which by the way works if you remove all other tags from the document but StockholmC

function custom_parser( response, records ){
   var rssDocument = parseXML( response );
   if ( rssDocument==null ) return;
   for ( var row=rssDocument.documentElement.firstElementChild; row!=null;   row=row.nextElementSibling ) {
      var r =  new Object();      
      r.next = row.getAttribute('minutes');
      r.date = row.getAttribute('datetime');
      records.push( r );
   }
}

I want to point out that the above code works in a XML file with ONLY the StockholmC tag.

Thank you

Answer 1

You are iterating over every child of the root, starting with <info> (rssDocument.documentElement.firstElementChild). Those elements don't have minutes or datetime attributes.

If you only want to iterate over the children of <StockholmC>, you have to get a reference to it instead of documentElement:

function custom_parser( response, records ){
   var rssDocument = parseXML( response );
   if (!rssDocument) return;
   var stockholmc = rssDocument.getElementsByTagName('StockholmC')[0];
   if (!stockholmc) return;
   for (var row=stockholmc.firstElementChild; row!=null; row = row.nextElementSibling) {
       // ...
   }
}

The DOM interface provides many methods to traverse the DOM tree. Have a look at the introduction to DOM on MDN.

Answer 2

As explained by Felix, you first need to iterate over the children of <StockholmC>. However, as mentioned in your question, as you are using JavaScript inside a SpinetX HMP, you can only rely on a subset of the DOM level 3 (see SVG Micro DOM) and thus, cannot use the getElementsByTagName().

Here is an example of iterating the DOM until you find the correct element.

function custom_parser( response, records ){
  var rssDocument = parseXML( response );
  if ( rssDocument==null ) return;
  var elem = rssDocument.documentElement.firstElementChild;
  while ( elem!=null ) {
    if ( elem.localName=="StockholmC" ){
      elem = elem.firstElementChild;
    } else {
      if ( elem.localName=="next" ) {
        var r =  new Object();     
        r.next = elem.getAttribute('minutes');
        records.push( r );
      } else if ( elem.localName=="upcoming" ) {
        var r =  new Object();     
        r.datetime = elem.getAttribute('datetime');
        records.push( r );
      }
      elem = elem.nextElementSibling;
    }
  }
}

Note that the sample code puts the next and datetime attribute in two different rows of your array.