Hello,
I have setup CXF Webservice using HTTP Servlet
(org.apache.cxf.jaxws.servlet.CXFServlet) following the guidelines on the
site.
http://cwiki.apache.org/CXF20DOC/servlet-transport.html
I am using Eclipse Europa 3.3.2, CXF 2.1, Tomcat 6.0.16, on a WSDL-first
project with maven build.
Thanks Dan for your help.
I started from scratch with the Hello world sample (wsdl-first) approach,
and it turned out that the instructions for Setting up Apache Tomcat 6.0.6
on the CXF site
(http://cwiki.apache.org/CXF20DOC/cxfeclipseplugininstructions.html) were
the source of this problem.
1
I suggest creating the various services as 'clients' in separate projects,
and just reference those artifacts in the project that is consuming them as
Maven dependencies. Keeps things pretty clean and an easy way to manage
them individually when they change.
--
View this message in context:
htt
I'm getting started on a project using CXF where I'm building a web service
acting as a proxy for calling another (more complex) web service. I've
played around with CXF/Eclipse/Tomcat and things became complicated very
quickly.
I'd appreciate hearing from anyone who could suggest some references
I'm struggling with the same/similar problem using a GSOAP generated WSDL
file with any of the popular Java SOAP stacks (Metro, Axis2, XFire, CXF,
etc...). I have a very simple interface that attempts to retrieve a "chunk"
of a file from a service by specifying a filename, file offset, and length
This error is pretty different. Does the message have headers in
it?
I THINK I may have this fixed now. I'm going to deploy a new
snapshot tonight. If you could retry with that snapshot tomorrow
morning, that would be great.
Dan
On May 18, 2008, at 1:51 PM, XeN0s wrote:
This is most likely some sort of wacky classloader issue.The
ClientBehaviorPolicyType is trying to be cast to a
TExtensibilityElementImpl and is failing, but
TExtensibilityElementImpl is the superclass of
ClientBehaviorPolicyType so it should work. Normally, that means
TExtensibil
On May 20, 2008, at 3:26 PM, Cord Awtry wrote:
Thanks Dan, that's kind of what I thought.
So, the question is, if I override dateTime to be Calendar instead of
XMLGregorianCalendar, could that cause problems with those asm
generated
classes?
Possibly. I'm not really sure though. :-(
I
Thanks Dan, that's kind of what I thought.
So, the question is, if I override dateTime to be Calendar instead of
XMLGregorianCalendar, could that cause problems with those asm generated
classes?
I'm trying to create a small test service which'll expose the problem. While
debugging, I did see both
Any indication when to move toward a 0.7 or 0.8 JAX-WS version?
Also browsing through the code, I have noticed that the StringProvider does
not support primitive and generic types whereas other JAX-WS implementations do.
Thanks
Olivier
- Original Message
From: Sergey Beryozkin <[EMAIL
Thanks for pointing this out, you are right, maven will go get what it needs
and cache it in .m2. For a number of reasons however, I need to reference
an "off-line" repository on the local network (I don't really want to debate
the merits of doing this ...)
dkulp was nice enough to tell me what
OK. I fixed this in the tooling. Next snapshot will have the fix.
Dan
On May 20, 2008, at 1:11 PM, Daniel Kulp wrote:
This is technically not a valid wsdl. I'm not sure why the
wsdlvalidator isn't flagging it as invalid.
Basically, you have:
but the import for the wsa schema i
Perfect - exactly what I was hoping for - thanks!
--dan
dkulp wrote:
>
>
> On May 20, 2008, at 1:30 PM, burki wrote:
>
>>
>> I'm finally switching form ant to maven - maven is new to me, so
>> please bear
>> with me if the question below sounds idiotic ;-)
>>
>> I need to build an off-line
On May 20, 2008, at 1:22 PM, Cord Awtry wrote:
Hmmm. I'm fairly certain I have it set up right. And those
WrapperTypeHelper
classes are no-where to be found on my system. Do they get created
on the
fly by cxf during the processing?
Yes, they do. We use asm to generate some helper classes
On May 20, 2008, at 1:30 PM, burki wrote:
I'm finally switching form ant to maven - maven is new to me, so
please bear
with me if the question below sounds idiotic ;-)
I need to build an off-line maven repository for the cxf modules.
As a
first step, I used the assembly task to build a
2008-05-20 burki wrote:
> I'm finally switching form ant to maven - maven is new to me, so please bear
> with me if the question below sounds idiotic ;-)
>
> I need to build an off-line maven repository for the cxf modules. As a
> first step, I used the assembly task to build a repository packag
I'm finally switching form ant to maven - maven is new to me, so please bear
with me if the question below sounds idiotic ;-)
I need to build an off-line maven repository for the cxf modules. As a
first step, I used the assembly task to build a repository packaged as a
jar.
Most of the cxf th
Hmmm. I'm fairly certain I have it set up right. And those WrapperTypeHelper
classes are no-where to be found on my system. Do they get created on the
fly by cxf during the processing?
On Tue, May 20, 2008 at 8:24 AM, <[EMAIL PROTECTED]>
wrote:
> Hello,
>
> well, according to my sources, it'll re
This is technically not a valid wsdl. I'm not sure why the
wsdlvalidator isn't flagging it as invalid.
Basically, you have:
but the import for the wsa schema is commented out. Thus, it would
not be able to fine the ReplyTo element in the schemas and thus would
not be able to properl
Is there any way you can create a small sample project that
demonstrates this?
This might be a bug in the DataTypeAdapter code somehow, but I'm not
really sure how that's possible as the code there is VERY simple.
It looks like jaxb is sending a null into
public static String printDateTi
I just committed a possible fix for this to trunk. Is there any
chance you could checkout the code and give it a quick try?
A very simple query to strikeiron returned OK, but I'm not a "power
user" type thing so if you have more complex cases, I'd love to make
sure they work as well.
I'm encountering the following exception using CXF 2.1 with JAXB-impl
2.1.6:
Caused by: java.lang.NullPointerException
at java.util.Calendar.setTime(Calendar.java:1032)
at
org.apache.cxf.tools.common.DataTypeAdapter.printDateTime(DataTypeAdapte
r.java:52)
at gen.jdnet
Ah... strikeiron. :-(
I signed up this morning and have the NFL thing up and running in the
debugger now. I'm digging into it to figure out what's happening.
Dan
On May 20, 2008, at 10:54 AM, JSP wrote:
I just ran through the debugger and the problem is the same as when
I was
usi
I just ran through the debugger and the problem is the same as when I was
using 2.0.5. Everything I mentioned previously is still the same - that is,
the MessagePartInfo.getTypeClass() method returns null which causes the
JAXBEncoderDecoder to attempt to unmarshal a Node without a class. Then it
Sergey,
Thanks. Would be nice to treat output the same way than the WS stack in
CXF, using Aegis or JAXB for instance, no. One could have cntrol with
JAXB annotation on the output of a specific object at the
attribute/element level.
Handling primitive types and Collections would be great!
Ol
Hi
No primitive types (or their object equivalents) are supported on the output
by default, this is per JAX-RS spec, but nothing prevents a user from
creating a custom writer, have a look please at a sample code for a Long
writer at http://cwiki.apache.org/CXF20DOC/jax-rs-jsr-311.html.
The reaso
Dan, thank you for your quick reply.
You nailed down the problem really well, I was indeed using jax-ws services
(well, REST but using http bindings) and I should have used the @WebResult
annotation directly on the SEI instead of annotating the types.
I fixed it and it works perfectly now. My S
Well, unfortunately, that's a pretty useless stack trace.That NPE
is occuring while trying to create the Fault to represent the real
error. Thus, the real error from JAXB is lost. :-(
I've committed a fix to the JAXBEncoderDecoder that will hopefully fix
that so we can see what the
Hi jano,
I saw the question replied to the similar email - again sorry for the
doublepost - but I have checked my XSD schemas thoroughly and I can't locate
any cyclic references.
I attached the WSDL file and the XSD schemas (located in
WEB-INF/servicedescriptions).
http://www.nabble.com/file/p17
Hello,
well, according to my sources, it'll return java.util.GregorianCalendar
(but internally during it's creation also XMLGregorianCalendar is used)
didn't you mixed something?
after i used CalendarAdapter (instead of setting parse/print methods),
alll strange classes like foo.bar.SetOrderCal
Hmm, that's a very good question.
If I eliminate the AOP proxy stuff below (the
quotaAnmodningHistorikSamlingHent bean), by pointing the endpoint
implementor directly at the implementation class (the
quotaAnmodningHistorikSamlingHentTarget bean), then I get the following
stack overflow:
Stack tr
Jano,
That did the trick, thanks. I'm curious if you or anyone else has any
insight on a follow-up question. It would seem the XMLGregorianCalendar is
still in use, at least partially, after using the custom bindings file to
convert them all to java.util.Calendar objects. Below is a
ClassCastExc
Yes ,
thank you , you're wright.
There was a fault in some bindings, as you say. After fixing it, i've seen,
that the second binding.xjb was not used.
After your reply, i test the first way, like Dan said and now it works.
My fault ,sorry.
Thanks a lot
Bronchito
jan.minaroviech wrote
Hello Bronchito,
it's not correct !
Error you had was produced by jaxb, because some binding was incorrect
(maybe only with combination with others).
for the same wsdl it's correct as Dan wrote. I'm using this way also 5
binding files
best regards
jano
bronchito <[EMAIL PROTECTED]>
05/20/
34 matches
Mail list logo
