Confusion: Unexpectedly XPath seach starts at wrong element

[GoogleCodeExporter]

Further elaborating at http://code.google.com/p/omnixml/issues/detail?id=9

~~~~~~ Node ~~~~~
  <hasp id="1299527092" type="HASP-HL">
    <feature id="0" />
    <feature id="16" />
    <vendor id="54321"/>
    <license_manager >
      <ip>127.0.0.1</ip>
      <osname>Windows 7 Professional</osname>
    </license_manager>
  </hasp>

I try to extract values using expressions like
"hasp/license_manager/ip"
"hasp/vendor/@id"
"hasp/@id"

Correct results returned by

1. http://www.freeformatter.com/xpath-tester.html
2. http://xpath.online-toolz.com/tools/xpath-editor.php
3. http://www.softpedia.com/get/Programming/File-Editors/XMLPad.shtml

No matches returned by OmniXML XPathSelect

However the broken expressions
"license_manager/ip"
"vendor/@id"
"@id"

Extract correct results in OmniXML and no results in reference tools

From the hindsight, the reference tools (that are normally used to debug and
settle the expressions), so they kind add invisible virtual "document-root)
node on top of it. While OmniXML works inside the node with the document.
But that is very confusing and unexpected.

I'd propose to add
XPathSelect(NodeList, expression) overload
and
XPathSelect(Node, expression, TreatAsDocument: boolean = false)
or XPathSelectAsDoc(Node, expression)

To make transplanting expressions from editors/debuggers into sources easier

Original issue reported on code.google.com by `[email protected]` on 10 Dec 2014 at 2:51