为什么在Scala中解析XML时为什么会得到额外的“节点”？

Question

Here is the XML that's returned by my web server

<root>
    <item attr="test1">
    <item attr="test2">
</root>

Here is the Scala Code

var url = "http://mywebserver/my.xml"
val xml = XML.load(url)
logger.info(s"The XML is: $xml")
var items = xml.child
logger.info(s"items: $items")
items.foreach(item => {
  logger.info(item.toString())
  val attr = item.attribute("attr")
  logger.info(s"Attribute: $attr")
})

The XML logged is the copy of the XML returned by the web server However the foreach loop prints

INFO myclassName - 
INFO myclassName - Attribute: None
INFO myclassName - <item attr="test1">
INFO myclassName - Attribute: test1
INFO myclassName - 
INFO myclassName - Attribute: None
INFO myclassName - <item attr="test2">
INFO myclassName - Attribute: test2

Where are the extra blank nodes coming from? Am I missing any configuration?

Answer 1

<foo><bar></bar></foo> and <foo> <bar> </bar> </foo> are different XML. In the XML spec, the characters between tags--including whitespace--form a "text element" child of the enclosing tag, and Scala's XML parsing is simply honoring the spec by reading your whitespace as text elements.