FHIR Sandbox

[Kineta Chien]

What would I expect to see with the sample urls in Cerner's FHIR documentation? I'm expecting to see json, but get this message instead: 

The Accept or Content-Type is invalid. See the documentation on the fhir.cerner.com website for acceptable format values for the resource you are attempting to access.

For example:

https://fhir.sandboxcernerpowerchart.com/may2015/open/d075cf8b-3261-481d-97e5-ba6c48d3b41f/MedicationPrescription?patient=2744010&status=active

or:

https://fhir.sandboxcernerpowerchart.com/may2015/open/d075cf8b-3261-481d-97e5-ba6c48d3b41f/Observation?subject:Patient=1490008

Thanks in advance. This is an early look at this material for me, so I may very well be missing something obvious.

[Kineta Chien]

Does Cerner provide any working examples?

[Dennis Patterson]

Kineta, make sure that you're populating the Accept header to request json when you're calling those urls.  (See http://fhir.cerner.com/may2015/#schema).  If you try to access them from a browser, it'll be requesting html, which the server doesn't support.  You'll want to use something like Postman or RESTClient if you're using a browser.  Otherwise, you can use curl from the command line or one of the FHIR libraries if you're writing code.

[Kineta Chien]

I was able to get a result (sort of) with a jquery ajax call. Now I need to figure out the queries to use. 

example:

$.ajax({

    headers: { 

        Accept : "application/json+fhir",

        "Content-Type": "application/json+fhir"

    },

url: "https://fhir.sandboxcernerpowerchart.com/may2015/open/d075cf8b-3261-481d-97e5-ba6c48d3b41f/Observation?subject:Patient=1490008",

dataType: "json",

    success : function(response) {

        alert(response);

}

});

[Kineta Chien]

I have a question about the $ curl thing you mention. It's not something I'm familiar with. I've seen it referred to in your documentation and at the SMART documentation but can't find any reference to it's use. 

I'm retrieving the data with a jquery ajax call, which works fine - but I'd like to understand what you're referring to when you say " use curl from the command line"

Thanks in advance.

On Monday, October 19, 2015 at 6:17:52 AM UTC-7, Dennis Patterson wrote:

[Kevin Shekleton]

curl is a great command-line tool for making web requests. It comes pre-installed on *nix systems (for instance, Linux and Mac computers) and is very useful for playing around.

You can download curl (for all systems, including Windows) here: http://curl.haxx.se/

[Jens Petter Abrahamsen]

It's the +fhir in the content-type that throws the browser off.

application/json+fhir

Web-services which return plain application/json content type are usually displayed correctly in most browsers (even pretty if you install a JSON extension).

It's not a problem in real use, just annoying when pointing the browser at something to check it out quickly.