Does the javascript split function have a special meaning in mirth?

Question

I am maintaining a legacy mirth system. We get some incoming HL7 messages with a repeating ZP1.36 segment. As far as I can tell, mirth is dividing these repeating segments into an array of repeating segments via a split function.

var repeat36 = msg['ZP1']['ZP1.36'].toString().split("</ZP1.36>"); //I think returns an array of ZP1.36 segments

But in the raw data, I don't see the string ZP1.36...

I'm used to using split functions in VB/Java/C# that divide strings into tokens around a certain character, like this:

var myTokens = "hello^world".toString().Split("^"); //returns a 2 item array {"hello", "world"}

Mirth's split function doesn't seem to be working this way. It seems to be parsing the messages based on the assumed structure of the HL7 message.

Is this what's going on? Am I missing something? What are the rules for the split function in mirth?

Cross-posted on mirth community http://www.mirthcorp.com/community/forums/showthread.php?p=26203#post26203

Answer 1

I know you already found and accepted a sufficient answer to your question (reposted from the Mirth forums), but I thought it might be helpful to give a bit more information in a separate answer.

At its heart, Mirth leverages a couple of pre-existing technologies in order to work with HL7 data.

1. HAPI (pronounced "Happy"): This is a library of Java code for re-interpretting pipe-delimited V2.x HL7 as XML. The V2.XML format itself is also an official HL7 specification (not to be confused with V3 HL7 which is natively XML). The product name "Mirth" was obviously chosen as a synonym for "happy."
2. E4X: This is a JavaScript standard for working with XML. It failed to gain significant traction as a standard web technology, however, it is leveraged by Mirth due to its availability in the Rhino JavaScript engine.

From the examples that Dans provided you on the Mirth forums:

This snippet is in the standard XML format for HL7 v.2x:

<ZP1.36>
    <ZP1.36.1>Hello</ZP1.36.1>
    <ZP1.36.2>World</ZP1.36.2>
<ZP1.36>

This JavaScript code is utilizing E4X notation to parse and access HL7 data that has been reinterpreted into E4X XML and XMLList data types:

var zp1361 = msg['ZP1']['ZP1.36']['ZP1.36.1'].toString();
var zp1362 = msg['ZP1']['ZP1.36']['ZP1.36.2'].toString();

• msg is the E4X XML object holding the HL7 data
• msg['ZP1'] produces an E4X XMLList object that holds a list of ZP1 segments.

Answer 2

Dans, a Mirth employee on the Mirth message boards explains that the split function works on the xml translation of the incoming message.

So basically, you are working with xml and the ZP1.36 is already split up:

<ZP1.36>
   <ZP1.36.1>Hello</ZP1.36.1>
   <ZP1.36.2>World</ZP1.36.2>
<ZP1.36>

If you want to get the various sub elements you can do something like this:

var zp1361 = msg['ZP1']['ZP1.36']['ZP1.36.1'].toString();
var zp1362 = msg['ZP1']['ZP1.36']['ZP1.36.2'].toString();