0
102

Written by

Marco Bruni
Follow
Question Marco Bruni · Nov 28, 2025

Punctually access specific nodes of an XML document

#InterSystems IRIS #ObjectScript #XML

Hello!
I have the following XML document obtained through a string:
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<MainDocument xmlns="urn:hl7-org:v3">
<realmCode code="IT"/>
<title>kjbkjkjbkjb</title>
<effectiveTime value="20090905150716"/>
[.....other tags.....]
<versionNumber value="1"/>
<component>
<body>mhvjhjkvhj</body>
<component>
<section>content</section>
<ID>5</ID>
<title>Certificato</title>
<text/>
<entry>
<Id>5</Id>
[...other tags...]
</entry>
<entry>
<Id>6</Id>
[...other tags...]
</entry>
</component>
</component>
</ClinicalDocument>
I would like to have a way to obtain only the content of the <entry> tags (which can be multiple) without having to scan the whole xml document, since it would be inefficient because of its length. I would like to have an outcome similar to the one coming from EvaluateExpression() but, from my trials and examples found on the internet, it seems that this function can be used only when the tag contains just one value. Does anybody have any idea which might help?
Thank you in advance!

Discussion (7)3
Log in or sign up to continue

Comments

DC AI Bot · Nov 28, 2025 Bot

To punctually access specific nodes in an XML document, such as retrieving only <entry> tags, you can use the following approaches with InterSystems tools:

  1. Using XPath: You can leverage the %XML.XPATH.Document class to evaluate an XPath expression that targets the <entry> tags. For example:

    set status = ##class(%XML.XPATH.Document).CreateFromStream(stream, .mydoc)
set status = mydoc.EvaluateExpression("//entry", "1", .myresults)

    After evaluating the expression, iterate over myresults to retrieve each <entry> element. Here, "//entry" directly fetches all <entry> elements in the document [1][2].

  2. Detailed Node Inspection Using %XML.TextReader: If you wish to parse the document node by node:

    • Use the ParseStream() or ParseString() method of %XML.TextReader to load the XML into a reader object.
    • Use the Read() method to navigate nodes until you find <entry> tags.
    • Use properties like Name and NodeType on each node to filter for <entry> elements and retrieve their contents [3][2].

Using %XML.XPATH.Document can directly fetch elements specified by an XPath expression and is more efficient when targeting specific elements like <entry> tags. For navigating documents manually, %XML.TextReader provides granular control node-by-node.

References: [1][3][2]

Sources:

0
Marco Bruni  Dec 1, 2025 to DC AI Bot

Solution 1 is not working. Even though I get an element saved in myresults, if I try to access its content using myresults.GetAt(1).Value I get as the returned value the number 1

0
Vitaliy Serdtsev · Dec 1, 2025 
Class dc.test [ Abstract ]
{

ClassMethod Test()
{
  #dim ex As %Exception.AbstractException
  #dim xDataStream As %Stream.Object
  #dim mydoc As %XML.XPATH.Document

  try{

    set xDataStream=##class(%Dictionary.CompiledXData).IDKEYOpen(..%ClassName(1),"ClinicalDocXData").Data

    $$$ThrowOnError(##class(%XML.XPATH.Document).CreateFromStream(xDataStream, .mydoc,,,,,"s urn:hl7-org:v3"))
    $$$ThrowOnError(mydoc.EvaluateExpression("//s:entry/s:Id", "text()", .tRes))

    for i =1:1:tRes.Count() {
      write tRes.GetAt(i).Value,!
    }    

  }catch(ex) {
    write "Error ", ex.DisplayString(),!
  }
}

XData ClinicalDocXData
{
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<MainDocument xmlns="urn:hl7-org:v3">
  <realmCode code="IT"/>
  <title>kjbkjkjbkjb</title>
  <effectiveTime value="20090905150716"/>
  <versionNumber value="1"/>
  <component>
    <body>mhvjhjkvhj</body>
    <component>
      <section>content</section>
      <ID>5</ID>
      <title>Certificato</title>
      <text/>
      <entry>
        <Id>5</Id>
      </entry>
      <entry>
        <Id>6</Id>
      </entry>
    </component>
  </component>
</MainDocument>
}

}
Result:
USER>do ##class(dc.test).Test()

5
6
0
Marco Bruni  Dec 4, 2025 to Vitaliy Serdtsev

Hello!
Thanks for your reply. At the end, I implemented a solution very similar to yours. Apparently, there is no way to obtain something that is not a text, an occurrence count or an attribute like I hoped.

0
Robert Cemper · Dec 4, 2025

Late suggestion:
if you interpret your <entry> as a piece separator in your XML string, you may apply this trick

;; asssuming your input is in xmltext;; get the 4th entryset entry4=$piece($piece(xmltest,"<entry>",4+1),"</entry>")
;;-------^---------------------------------^
0
Marco Bruni  Dec 9, 2025 to Robert Cemper

Hi!
Thank you for the suggestion but I decided to use the XPATH implementation.

0
Robert Cemper  Dec 9, 2025 to Marco Bruni

Clear ! That's what XPATH was meant for

0