Saxon XSLT and XQuery Processor

I am using Saxon 765a and using the <result-document/> to write data =
out.
This works fine as long as I do it to the local drives, but the moment
I try to write it to a drive letter that has been mapped from another =
machine,
I get the following error:

Transformer Error: Failed to create output file :///X:/VRU/TestApp.xml

I have all the write permissions on to this other machine's drive.

Is there something wrong I am doing here ?


Raja

I'd like to give users in my app an option to cancel a long-running query.

1. What, if anything is likely to go wrong with QueryProcessor if I 
forcibly stop a thread in the middle of processing by invoking 
thread.stop()? Would it be necessary to set up a new query processor for 
future queries?

2. Is there any way to find out how far along the query has gotten? I 
tend to doubt it. This is probably equialenet to the halting problem and 
may also require knowing how big the source documents are; but just in 
case I missed something I thought I'd ask.

--
Elliotte Rusty Harold

Hi -- I'm new here... :-)

Does Saxon support XSL's "tunnel parameters"?

I ask because when I add 'tunnel=3D"yes"' to both the calling =
xsl:with-param and the receiving xsl:param elements, I get the error: =
"Attribute tunnel is not allowed on this element" at both locations.

I am using Saxon 7.4.  I think I'm following the XSL spec for tunnel =
parameters properly.  The actual code is:

[...]
   <xsl:call-template name=3D"dataroot">
    <xsl:with-param name=3D"year" select=3D"$year" tunnel=3D"yes"/>
   </xsl:call-template>
[...]
 <xsl:template match=3D"/" name=3D"dataroot">
  <xsl:param name=3D"year" tunnel=3D"yes"/>
[...]

It works properly if I remove the tunnel attributes.  What am I doing =
wrong?

  Thanks,
- Chip Whitmer

This is a crash in the new code introduced in 7.8 to run transforms
directly on a DOM, rather than stripping the DOM down and building a
Saxon tree.

I'll probably need a copy of your source document and stylesheet to
reproduce it. If you built the DOM dynamically then I may need a copy of
the program that built it. Meanwhile I'll take a look at the stack
trace, it's just possible I can find the problem without a repro, though
it's always more comfortable to have a test case to check any fix.

Michael Kay

> -----Original Message-----
> From: saxon-help-admin@... 
> [mailto:saxon-help-admin@...] On Behalf Of 
> Geis, Matt
> Sent: 20 November 2003 18:50
> To: 'saxon-help@...'
> Subject: [saxon] NullPointerException in 7.8, related to NamePool
> 
> 
> Hi,
> I'm getting a NPE on a transform using Saxon 7.8.  The 
> transformation worked well in 7.6.5. (it fails in 7.7 for a 
> different reason, below).
> 
> It appears to be pretty low-level.  Any suggestions as to 
> what it might be, or how I can get more information that 
> would be useful in helping me fix the error if it's on my 
> side, or to isolate the bug, if it is in fact a bug (sorry 
> for not already knowing how to do that, I'm pretty new to Saxon).
> 
> java.lang.NullPointerException
>         at net.sf.saxon.om.NamePool.getPrefixIndex(NamePool.java:324)
>         at net.sf.saxon.om.NamePool.allocate(NamePool.java:385)
>         at net.sf.saxon.om.NamePool.allocate(NamePool.java:370)
>         at 
> net.sf.saxon.dom.NodeWrapper.getNameCode(NodeWrapper.java:304)
>         at 
> net.sf.saxon.dom.NodeWrapper.getFingerprint(NodeWrapper.java:353)
>         at 
> net.sf.saxon.om.Navigator$AxisFilter.advance(Navigator.java:604)
>         at 
> net.sf.saxon.om.Navigator$AxisFilter.<init>(Navigator.java:598)
>         at 
> net.sf.saxon.dom.NodeWrapper.iterateAxis(NodeWrapper.java:572)
>         at 
> net.sf.saxon.expr.AxisExpression.iterate(AxisExpression.java:183)
>         at net.sf.saxon.expr.Atomizer.iterate(Atomizer.java:77)
>         at 
> net.sf.saxon.expr.UntypedAtomicConverter.iterate(UntypedAtomic
> Converter.java
> :83)
>         at net.sf.saxon.expr.ItemChecker.iterate(ItemChecker.java:80)
>         at
> net.sf.saxon.expr.CardinalityChecker.iterate(CardinalityChecke
> r.java:78)
>         at
> net.sf.saxon.expr.ExpressionTool.eagerEvaluate(ExpressionTool.
> java:149)
>         at
> net.sf.saxon.expr.ExpressionTool.lazyEvaluate(ExpressionTool.java:120)
>         at
> net.sf.saxon.expr.UserFunctionCall.callFunction(UserFunctionCa
> ll.java:205)
>         at
> net.sf.saxon.expr.UserFunctionCall.iterate(UserFunctionCall.java:197)
>         at net.sf.saxon.expr.Atomizer.iterate(Atomizer.java:77)
>         at 
> net.sf.saxon.expr.AtomicSequenceConverter.iterate(AtomicSequen
> ceConverter.ja
> va:82)
>         at
> net.sf.saxon.functions.StringJoin.evaluateItem(StringJoin.java:39)
>         at 
> net.sf.saxon.expr.ComputedExpression.evaluateAsString(Computed
> Expression.jav
> a:283)
>         at 
> net.sf.saxon.instruct.ValueOf.processLeavingTail(ValueOf.java:88)
>         at 
> net.sf.saxon.instruct.Instruction.processChildrenLeavingTail(I
> nstruction.jav
> a:261)
>         at
> net.sf.saxon.instruct.Sequence.processLeavingTail(Sequence.java:144)
>         at net.sf.saxon.instruct.Template.expand(Template.java:105)
>         at
> net.sf.saxon.instruct.Template.processLeavingTail(Template.java:74)
>         at net.sf.saxon.Controller.applyTemplates(Controller.java:326)
>         at
> net.sf.saxon.instruct.ApplyTemplates.apply(ApplyTemplates.java:140)
>         at
> net.sf.saxon.instruct.ApplyTemplates.processLeavingTail(ApplyT
> emplates.java:
> 80)
>         at 
> net.sf.saxon.instruct.Instruction.processChildrenLeavingTail(I
> nstruction.jav
> a:261)
>         at 
> net.sf.saxon.instruct.Block.processLeavingTail(Block.java:92)
>         at 
> net.sf.saxon.instruct.Choose.processLeavingTail(Choose.java:61)
>         at 
> net.sf.saxon.instruct.Instruction.processChildrenLeavingTail(I
> nstruction.jav
> a:261)
>         at
> net.sf.saxon.instruct.Sequence.processLeavingTail(Sequence.java:144)
>         at net.sf.saxon.instruct.Template.expand(Template.java:105)
>         at
> net.sf.saxon.instruct.Template.processLeavingTail(Template.java:74)
>         at net.sf.saxon.Controller.applyTemplates(Controller.java:329)
>         at net.sf.saxon.Controller.defaultAction(Controller.java:354)
>         at net.sf.saxon.Controller.applyTemplates(Controller.java:321)
>         at 
> net.sf.saxon.instruct.ApplyTemplates$ApplyTemplatesPackage.pro
> cessLeavingTai
> l(ApplyTemplates.java:180)
>         at net.sf.saxon.Controller.run(Controller.java:215)
>         at 
> net.sf.saxon.Controller.transformDocument(Controller.java:1599)
>         at net.sf.saxon.Controller.transform(Controller.java:1366)
>         at 
> schwab.rrbus.tools.integration.CtlGenerator.generateCtl(CtlGen
> erator.java:60
> )
> 
> 
> In 7.7, I get the following...
> 
> javax.xml.transform.TransformerException: DOMSender: no start 
> node defined
>         at net.sf.saxon.event.DOMSender.send(DOMSender.java:75)
>         at net.sf.saxon.event.Sender.sendDOMSource(Sender.java:108)
>         at net.sf.saxon.event.Sender.send(Sender.java:61)
>         at net.sf.saxon.Controller.transform(Controller.java:1377)
>         at 
> schwab.rrbus.tools.integration.CtlGenerator.generateCtl(CtlGen
> erator.java:60
> )
> 
> I can include the XML/XSLT in a follow-up email if anyone has 
> any ideas.
> 
> Thanks,
> Matt
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: SF.net Giveback Program. 
> Does SourceForge.net help you be more productive?  Does it 
> help you create better code?  SHARE THE LOVE, and help us 
> help YOU!  Click Here: http://sourceforge.net/donate/ 
> _______________________________________________
> saxon-help mailing list
> saxon-help@... 
> https://lists.sourceforge.net/lists/listinfo/s> axon-help
>

At 9:44 AM -0800 11/20/03, Michael Burbidge wrote:
>The following query fails with the message:
>
>Failed to load document file://Users/mburbidg/fdrake-01/build/mac/prj/test.xml
>
><target-names>
>	{
>		for $target in 
>doc("file://Users/mburbidg/fdrake-01/build/mac/prj/test.xml")/TARGETLIST/TARGET
>			return
>				<name>{ $target/NAME/text() }</name>
>	}
></target-names>
>
>Does saxon support file based URIs? Is mine the correct format for a 
>file base URI?
>

I've never seen a double slash after the file. try a single slash 
and/or a triple slash. That is,

file:///Users/mburbidg/fdrake-01/build/mac/prj/test.xml

or

file:/Users/mburbidg/fdrake-01/build/mac/prj/test.xml

-- 

   Elliotte Rusty Harold
   elharo@...
   Effective XML (Addison-Wesley, 2003)
   http://www.cafeconleche.org/books/effectivexml
   http://www.amazon.com/exec/obidos/ISBN%3D0321150406/ref%3Dnosim/cafeaulaitA

Jan van der Lee wrote:
> Interesting: a relatively simple XSL (1.0) transformation on a
> relatively large XML file (several Mb) is about 1.8 times faster with
> Saxon 7.7 than with Xalan 2.5.1. Saxon is exactly 5 times faster than
> 4xslt (based on python).

Which version of 4Suite did you test with? It can make a huge difference. 

Saxon will probably still "win", but I'm willing to bet it won't be by a
factor of 5. I doubt people tend to choose based on speed alone, though.
Comparing processors that aren't really usable in the same environment is
is an academic exercise, since the average Java application developer is
not going to consider a Python-based processor to be an option.

FYI, 4XSLT is mostly Python, but parts are implemented in C, for speed: the
XML parser (Expat), the source & result tree implementation, parts of the
XML/HTML serializers, the XPath expression parser, and a couple other odds
& ends. Also note that Python, like Java, has interpreter overhead, and
garbage collection is not aggressive, so your benchmark methodology may
need some adjustment.

If speed is all you care about (hopefully it isn't), be sure to try the
venerable XT, or the major C-based processors: MSXML, Sablotron, and libxslt.

Anyway, if you send the XML and XSLT files (zipped), I'd like to test and make
sure we didn't miss a major bottleneck before the upcoming 1.0 beta release.
Email me off-list, please.

TIA,
 
 - Mike (one of the 4Suite developers)