0

I am trying to write a browser-based Javascript client for a ReST application which responds with XML (so it seems JSONP is out of the questions).

I am trying to retrieve the data using dojo.io.script.get but the parameter that is passed to the callback function is an object from which it seems I cannot retrieve the XML data of the response.

dojo.io.script.get({url:"http://enterpriseapp.enterprisedomain/path/to/rest/collection",
    load:function (data) {
        // 'data' does not contain the actual response (which is XML)
    }
});

What is the correct way to retrieve this data?

mydoghasworms
  • 18,233
  • 11
  • 61
  • 95

1 Answers1

0

The dojo.io.script.get method will inject a <script> from the specified web address. The data content from this script will be passed to your load function; hence, the content must validate as Javascript. You can't load XML into a script tag.

If you want to load XML, you'll need to use dojo.xhrGet; however, this will not allow requests to 3rd party urls. The advantage of using dojo.io.script.get is that you can use different origin address' than the page loading them.

dojo.xhrGet({
    handleAs: "xml",
    load: function(dom){
        // do something with the DOM XML object
    },
    error: function(error){
    }
});

See: dojo.xhrGet Documentation

If you are trying to load the XML from another website it's a bit of a dead-end. You can use the Access-Control-Allow-Origin header if you have access to the sending server.

Another solution that I have used is to write a proxy script (in PHP or other server language) to mirror the XML on the correct domain. You'll need to be careful if you do this to include good checks so that your server code is not abused by someone for proxying.

See the following Stackoverflow conversation for more about Access-Control-Allow-Origin:
jQuery XML REST Access-Control-Allow-Origin

Community
  • 1
  • 1
Stephen Simpson
  • 1,381
  • 1
  • 11
  • 23
  • Thanks, I ended up going the route you described: writing a proxy web application that does the querying on my behalf. Not great, but it seems like the only way to overcome this. – mydoghasworms Nov 14 '12 at 12:25