• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 419
  • Last Modified:

xpath test on "

hi,

I used xpath expression:
//Object1/@name[contains(.,'"')]
to test if the Object1 name attribute contains "
It does not return the first Object1.

However, if I use expression:
//Object1/@name[contains(.,'"')]
It gives correct object.

Why?

<test>
<Object1 name="&quot;AA&quot; BB">11</Object1>
<Object2 name="&quot;AA&quot;">22</Object2>
<Object1 name = "CC">33</Object1>
</test>
0
danclemson
Asked:
danclemson
  • 2
1 Solution
 
dfiala13Commented:
It's your parser.  it's not unescaping the value in the literal.  The .NET parser (which I tested with) performed as you expected:

//Object1/@name[contains(.,'&quot;')] worked and found the match
while

//Object1/@name[contains(.,'"')]
threw a parser error
0
 
rdcproCommented:
No, it's not a matter of unescaping.  I'll bet he's using DOM XPath methods.

If he's using the XPath in a SelectSingleNode() or SelectNodes() expression, then his behavior is as expected.  

If he's using XSLT, then the XPath expression

//Object1/@name[contains(.,'"')]

is not well-formed, hence your parser throwing the exception.  All conforming parsers will throw this exception, including the one danclemson is using.  But he's probably using the XPath as SelectSingleNode, where the XPath does not get parsed!

Here's why...

The XML parser reads this:

<Object1 name="&quot;AA&quot; BB">11</Object1>

and then it expands the entity in the name attribute, so that the value, in memory, is:

"AA" BB

Once the attribute has been parsed, then entity isn't there any more.  Instead, the *actual* character represented by the entity is there. This isn't a problem, because the attribute has already been parsed, and all knowledge of the delimiters originally used in the attribute is lost.  

Now you need an XPath expression to select this attribute.  If you use XSLT, then the XPath expression you use must be well-formed, and your XPath expression will contain a &quot; instead of a doublequote character, because the doublequote character is not well formed.  So in my test, I used:

<xsl:value-of select="@*[contains(., '&quot;')]"/>

but when you're using selectSingleNode(), as long as you don't violate your language's rules for nesting of quotes and apostrophes, you DON'T use the entity reference.  The XPath in the selectSingleNode is never parsed by the XML parser!

selectSingleNode(\"@*[contains(., '"')]\")
or possibly
selectSingleNode("@*[contains(., '"')]")

Regards,
Mike Sharp
0
 
danclemsonAuthor Commented:
Hi,

Thanks for the reply.
I think both of you are right.
I was using XMLSPY xpath evaluation.
Don't know what's the underline implementation of this xpath evalutation function. But seems it's not escaping the entity reference.
0
 
rdcproCommented:
In my version of XML Spy (4.4), it finds the node with

//Object1/@name[contains(.,'&quot;')]

and not

//Object1/@name[contains(.,'"')]

But XML Spy has some peculiarities.  I believe they're treating it as an XSLT XPath, because that's what you'd be doing with Spy.  I'm surprised you don't get it with the first expression and you do with the second...what version are you using?

I created an XPath evaluator some years ago that worked through the web.  That's when I discovered that the selectSingleNode method does not parse the XPath as "XML", and entities are not expanded.  

Regards,
Mike Sharp
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

Join & Write a Comment

Featured Post

Cloud Class® Course: Microsoft Windows 7 Basic

This introductory course to Windows 7 environment will teach you about working with the Windows operating system. You will learn about basic functions including start menu; the desktop; managing files, folders, and libraries.

  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now