You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@xalan.apache.org by Gary L Peskin <ga...@firstech.com> on 2001/07/04 19:53:51 UTC
Re: Addition of a new function
Shruti --
I think you'd be better off by implementing an XSLT extension function
rather than changing the core of XalanJ.
Have a look at http://xml.apache.org/xalan-j/extensions.html and at the
XSLT Recommendation itself.
You'd also find some good examples in org.apache.xalan.lib.Extensions.
It seems like you might be able to use the tokenize extension
(http://xml.apache.org/xalan-j/extensionslib.html#tokenize), already
supplied by XalanJ, to accomplish what you want:
//book[. = xalan:tokenize("hello ho")]
or something like that.
Gary
> shruti wrote:
>
> Hello,
>
> i have to add a new function to the xalan's xpath list of functions. i
> am adding a function for searching a list of words .
> It would be an extension to the xalan's contains function ...
>
> //book[contains(./,"hello")]
> //book[contains-words(./,"hello hi","any")] - this would search for
> words hello and hi and returns all the nodes which have either hello
> or hi in their text.
>
> For this i had to add a class (FunctionContainsWords.java) in the
> org/apache/xpath/functions package and make an appropriate entry for
> this function in the function table and keywords class
>
> Could their be any alternative approach possible by which i could add
> this function by not changing the existing code . But by just deriving
> my own classes from the existing ones..
>
> Please suggest...
>
> thanx
> Shruti.
>
>
Re: Addition of a new function
Posted by Gary L Peskin <ga...@firstech.com>.
Shruti --
Your XPath expression selects all employee nodes that meet the
criteria. The tokenize function has no effect on the output, it's only
used to select employee nodes. And the expression you've chosen will
basically select any non-empty employee node.
It sounds like what you want to do is create some output. You'll
probably need to use the full Xalan and not just XPath API for that.
Gary
shruti wrote:
>
> thanks Gary ....
> Now i am able to use the tokenize function but i am not getting the correct
> output ...
>
> my input xml looks like -
>
> <employees xmlns:xalan="http://xml.apache.org/xalan">
> <employee>Herbert,Johnny,09/01/1998,95000</employee>
> <employee>Hill,Graham,08/20/2000,89000</employee>
> </employees>
>
> and the xpath query is //employee[xalan:tokenize(./,",")]... i am evaluating
> is query using
>
> NodeIterator n1= XPathAPI.selectNodeIterator(doc, xpath) where xpath
> contains the query string and doc is the document object.
> but the resultant nodeset has the same output as the input ... the text is
> not seperated on the basis of the token ....
> can u suggest why ?
>
> ----- Original Message -----
> From: Gary L Peskin <ga...@firstech.com>
> To: <xa...@xml.apache.org>
> Sent: Wednesday, July 11, 2001 12:49 AM
> Subject: Re: Addition of a new function
>
> > Yes, you can do this. Declare the xalan namespace on the root node of
> > your document, like this:
> >
> > xmlns:xalan="http://xml.apache.org/xalan"
> >
> > Gary
> >
> > shruti wrote:
> > >
> > > Can i use this tokenize function in an XPath query ......
> > > if yes, where do i register the xalan namespace in the evaluation
> > > context....i am evaluating the xpath expression using
> > >
> > > XPathAPI.eval(doc,xpath);
> > > If i give the following xpath expression
> file://book[.=xalan:tokenize(hello ho)],
> > > it gives an error -
> > >
> > > Prefix must resolve to a namespace : xalan
> > >
> > > ----- Original Message -----
> > > From: Gary L Peskin <ga...@firstech.com>
> > > To: <xa...@xml.apache.org>
> > > Sent: Wednesday, July 04, 2001 11:23 PM
> > > Subject: Re: Addition of a new function
> > >
> > > > Shruti --
> > > >
> > > > I think you'd be better off by implementing an XSLT extension function
> > > > rather than changing the core of XalanJ.
> > > >
> > > > Have a look at http://xml.apache.org/xalan-j/extensions.html and at
> the
> > > > XSLT Recommendation itself.
> > > >
> > > > You'd also find some good examples in org.apache.xalan.lib.Extensions.
> > > > It seems like you might be able to use the tokenize extension
> > > > (http://xml.apache.org/xalan-j/extensionslib.html#tokenize), already
> > > > supplied by XalanJ, to accomplish what you want:
> > > >
> > > > file://book[. = xalan:tokenize("hello ho")]
> > > >
> > > > or something like that.
> > > >
> > > > Gary
> > > >
> > > > > shruti wrote:
> > > > >
> > > > > Hello,
> > > > >
> > > > > i have to add a new function to the xalan's xpath list of functions.
> i
> > > > > am adding a function for searching a list of words .
> > > > > It would be an extension to the xalan's contains function ...
> > > > >
> > > > > file://book[contains(./,"hello")]
> > > > > file://book[contains-words(./,"hello hi","any")] - this would search
> for
> > > > > words hello and hi and returns all the nodes which have either hello
> > > > > or hi in their text.
> > > > >
> > > > > For this i had to add a class (FunctionContainsWords.java) in the
> > > > > org/apache/xpath/functions package and make an appropriate entry for
> > > > > this function in the function table and keywords class
> > > > >
> > > > > Could their be any alternative approach possible by which i could
> add
> > > > > this function by not changing the existing code . But by just
> deriving
> > > > > my own classes from the existing ones..
> > > > >
> > > > > Please suggest...
> > > > >
> > > > > thanx
> > > > > Shruti.
> > > > >
> > > > >
Re: Addition of a new function
Posted by shruti <sh...@noida.hcltech.com>.
thanks Gary ....
Now i am able to use the tokenize function but i am not getting the correct
output ...
my input xml looks like -
<employees xmlns:xalan="http://xml.apache.org/xalan">
<employee>Herbert,Johnny,09/01/1998,95000</employee>
<employee>Hill,Graham,08/20/2000,89000</employee>
</employees>
and the xpath query is //employee[xalan:tokenize(./,",")]... i am evaluating
is query using
NodeIterator n1= XPathAPI.selectNodeIterator(doc, xpath) where xpath
contains the query string and doc is the document object.
but the resultant nodeset has the same output as the input ... the text is
not seperated on the basis of the token ....
can u suggest why ?
----- Original Message -----
From: Gary L Peskin <ga...@firstech.com>
To: <xa...@xml.apache.org>
Sent: Wednesday, July 11, 2001 12:49 AM
Subject: Re: Addition of a new function
> Yes, you can do this. Declare the xalan namespace on the root node of
> your document, like this:
>
> xmlns:xalan="http://xml.apache.org/xalan"
>
> Gary
>
> shruti wrote:
> >
> > Can i use this tokenize function in an XPath query ......
> > if yes, where do i register the xalan namespace in the evaluation
> > context....i am evaluating the xpath expression using
> >
> > XPathAPI.eval(doc,xpath);
> > If i give the following xpath expression
file://book[.=xalan:tokenize(hello ho)],
> > it gives an error -
> >
> > Prefix must resolve to a namespace : xalan
> >
> > ----- Original Message -----
> > From: Gary L Peskin <ga...@firstech.com>
> > To: <xa...@xml.apache.org>
> > Sent: Wednesday, July 04, 2001 11:23 PM
> > Subject: Re: Addition of a new function
> >
> > > Shruti --
> > >
> > > I think you'd be better off by implementing an XSLT extension function
> > > rather than changing the core of XalanJ.
> > >
> > > Have a look at http://xml.apache.org/xalan-j/extensions.html and at
the
> > > XSLT Recommendation itself.
> > >
> > > You'd also find some good examples in org.apache.xalan.lib.Extensions.
> > > It seems like you might be able to use the tokenize extension
> > > (http://xml.apache.org/xalan-j/extensionslib.html#tokenize), already
> > > supplied by XalanJ, to accomplish what you want:
> > >
> > > file://book[. = xalan:tokenize("hello ho")]
> > >
> > > or something like that.
> > >
> > > Gary
> > >
> > > > shruti wrote:
> > > >
> > > > Hello,
> > > >
> > > > i have to add a new function to the xalan's xpath list of functions.
i
> > > > am adding a function for searching a list of words .
> > > > It would be an extension to the xalan's contains function ...
> > > >
> > > > file://book[contains(./,"hello")]
> > > > file://book[contains-words(./,"hello hi","any")] - this would search
for
> > > > words hello and hi and returns all the nodes which have either hello
> > > > or hi in their text.
> > > >
> > > > For this i had to add a class (FunctionContainsWords.java) in the
> > > > org/apache/xpath/functions package and make an appropriate entry for
> > > > this function in the function table and keywords class
> > > >
> > > > Could their be any alternative approach possible by which i could
add
> > > > this function by not changing the existing code . But by just
deriving
> > > > my own classes from the existing ones..
> > > >
> > > > Please suggest...
> > > >
> > > > thanx
> > > > Shruti.
> > > >
> > > >
Re: Addition of a new function
Posted by Gary L Peskin <ga...@firstech.com>.
Yes, you can do this. Declare the xalan namespace on the root node of
your document, like this:
xmlns:xalan="http://xml.apache.org/xalan"
Gary
shruti wrote:
>
> Can i use this tokenize function in an XPath query ......
> if yes, where do i register the xalan namespace in the evaluation
> context....i am evaluating the xpath expression using
>
> XPathAPI.eval(doc,xpath);
> If i give the following xpath expression //book[.=xalan:tokenize(hello ho)],
> it gives an error -
>
> Prefix must resolve to a namespace : xalan
>
> ----- Original Message -----
> From: Gary L Peskin <ga...@firstech.com>
> To: <xa...@xml.apache.org>
> Sent: Wednesday, July 04, 2001 11:23 PM
> Subject: Re: Addition of a new function
>
> > Shruti --
> >
> > I think you'd be better off by implementing an XSLT extension function
> > rather than changing the core of XalanJ.
> >
> > Have a look at http://xml.apache.org/xalan-j/extensions.html and at the
> > XSLT Recommendation itself.
> >
> > You'd also find some good examples in org.apache.xalan.lib.Extensions.
> > It seems like you might be able to use the tokenize extension
> > (http://xml.apache.org/xalan-j/extensionslib.html#tokenize), already
> > supplied by XalanJ, to accomplish what you want:
> >
> > file://book[. = xalan:tokenize("hello ho")]
> >
> > or something like that.
> >
> > Gary
> >
> > > shruti wrote:
> > >
> > > Hello,
> > >
> > > i have to add a new function to the xalan's xpath list of functions. i
> > > am adding a function for searching a list of words .
> > > It would be an extension to the xalan's contains function ...
> > >
> > > file://book[contains(./,"hello")]
> > > file://book[contains-words(./,"hello hi","any")] - this would search for
> > > words hello and hi and returns all the nodes which have either hello
> > > or hi in their text.
> > >
> > > For this i had to add a class (FunctionContainsWords.java) in the
> > > org/apache/xpath/functions package and make an appropriate entry for
> > > this function in the function table and keywords class
> > >
> > > Could their be any alternative approach possible by which i could add
> > > this function by not changing the existing code . But by just deriving
> > > my own classes from the existing ones..
> > >
> > > Please suggest...
> > >
> > > thanx
> > > Shruti.
> > >
> > >
Re: Addition of a new function
Posted by shruti <sh...@noida.hcltech.com>.
Can i use this tokenize function in an XPath query ......
if yes, where do i register the xalan namespace in the evaluation
context....i am evaluating the xpath expression using
XPathAPI.eval(doc,xpath);
If i give the following xpath expression //book[.=xalan:tokenize(hello ho)],
it gives an error -
Prefix must resolve to a namespace : xalan
----- Original Message -----
From: Gary L Peskin <ga...@firstech.com>
To: <xa...@xml.apache.org>
Sent: Wednesday, July 04, 2001 11:23 PM
Subject: Re: Addition of a new function
> Shruti --
>
> I think you'd be better off by implementing an XSLT extension function
> rather than changing the core of XalanJ.
>
> Have a look at http://xml.apache.org/xalan-j/extensions.html and at the
> XSLT Recommendation itself.
>
> You'd also find some good examples in org.apache.xalan.lib.Extensions.
> It seems like you might be able to use the tokenize extension
> (http://xml.apache.org/xalan-j/extensionslib.html#tokenize), already
> supplied by XalanJ, to accomplish what you want:
>
> file://book[. = xalan:tokenize("hello ho")]
>
> or something like that.
>
> Gary
>
> > shruti wrote:
> >
> > Hello,
> >
> > i have to add a new function to the xalan's xpath list of functions. i
> > am adding a function for searching a list of words .
> > It would be an extension to the xalan's contains function ...
> >
> > file://book[contains(./,"hello")]
> > file://book[contains-words(./,"hello hi","any")] - this would search for
> > words hello and hi and returns all the nodes which have either hello
> > or hi in their text.
> >
> > For this i had to add a class (FunctionContainsWords.java) in the
> > org/apache/xpath/functions package and make an appropriate entry for
> > this function in the function table and keywords class
> >
> > Could their be any alternative approach possible by which i could add
> > this function by not changing the existing code . But by just deriving
> > my own classes from the existing ones..
> >
> > Please suggest...
> >
> > thanx
> > Shruti.
> >
> >