Hi Peter,
In order to make the collection itself nillable you'd need to add a
wrapper element for the collection (such as by using
name=CategoryLists on the collection element). That would give you a
way of saying the collection is not present in the XML representation.
As it is now, you
Hi Stephane,
The tutorial section on collections gives some examples that should be
relevant: http://jibx.sourceforge.net/tutorial/binding-collects.html For
what you want, a simple binding like:
collection field=list name=list
structure name=item type=MyItem
value name=field
how to bind an array within a class, but not an array of
class. Is it possible?
Thanks,
Stephane
On 11/9/06, *Dennis Sosnoski* [EMAIL PROTECTED]
mailto:[EMAIL PROTECTED] wrote:
Hi Stephane,
The tutorial section on collections gives some examples that should be
relevant
Hi Alexey,
There's a set of 1.1.2 jars in the Maven repository at
http://www.jibx.org/maven/jibx/jars, but I haven't done a formal
release. I'm tracking one more problem that mainly effects JiBX use for
Web services, and once I have that fixed I'll do a formal 1.1.3 release
which will include
for external developers.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Dennis
Sosnoski
Sent: Wednesday, November 22, 2006 5:19 PM
To: JiBX users
Subject: Re: [jibx-users] Approximated date of JiBX 1.1.2 release
Hi Alexey,
There's a set of 1.1.2 jars
The problem here is that JiBX for some reason things the B class mapping
does not extend the Ref class mapping. Given your binding, this should
work, so I'm not sure where the problem is arising.
I suggest you first double-check to make sure that all your classes are
up to date (so clean all
Hi Jan,
There's no way to do this at present, though it's on the list for the
JiBX 2.0 changes
(http://www.sosnoski.com/jibx-wiki/space/issues-2.0/binding-definition).
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Training and Consulting
http://www.sosnoski.com -
Hi Yann,
I had to do some digging to find the original message you were referring
to, but will reply to that one as well. I think Stefano's issue is
different from yours.
You're correct that reusing the same element name on a mapping should
work fine as long as the namespaces are different.
Hi Johan,
The unmarshalling error basically means that the code thought it had
processed everything in the owner element content, but hadn't found
anything to do with the collectiontypeone element. That's very
strange, especially since you show the contained structure is required
(meaning
Hi Stefano,
This was one of the clean-up steps I took because the JiBX bindings were
too loosely enforced. Duplicate names were obviously never officially
supported in bindings used for unmarshalling - there's no meaningful way
to do so, since there's no way of knowing which of the conflicting
Hi Brian,
Thanks for helping out on the users list!
Just as an improvement to your recommendation, though, since about 1.1
JiBX has supported a create-type=... attribute for collection
elements (along with mapping and structure, for that matter). This
allows you to avoid the nuisance of
Hi Luis,
Can you add a Jira for this with your code attached (preferably with an
Ant build to demonstrate the problem)? I didn't knowingly change
anything that would cause this, but there have been a lot of code
changes since 1.1.
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Hi Marcelo,
You should be able to do this by duplicating some information. Basically
just remove the value definitions from the base class (Vehicle) and
use them directly within the mapping for the subclasses (Truck and
Bus). So the mapping for Truck would look like:
mapping abstract=true
Hi Hao,
You're missing some details in your email. Are you using JibxSoap and
getting this problem? The JibxSoap distribution is seriously
out-of-date, so you get problems if you try to use it with the current
JiBX distribution. I'll try to get it updated by this weekend. If you're
not using
Hi Stefano,
Xsd2Jibx is seriously outdated (though still usable with simple
schemas). That aside, it doesn't know anything about Hibernate so could
not generate anything to control Hibernate persistence.
I am actively working on the replacement for Xsd2Jibx, and I'm hoping to
make that
The easiest way of doing this is by defining a custom
marshaller/unmarshaller for the Message class, with normal mapping
definitions for the Header and Body classes. Your custom unmarshaller
can unmarshal the Header and then check the resulting values to see if
the Body needs to be
You need to be a little more specific about what you want to do. Looking
over the binding tutorial section on Structure Mapping may get you
started: http://jibx.sourceforge.net/tutorial/binding-structures.html#intro
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Training and
Hi Per-Ivar,
You don't need choice=true for a collection of items - just say
ordered=false and flexible=true on the collection element itself,
then list the different alternatives you want to handle as child
structures within the collection. So something along these lines
should work:
Hi Jörg,
I can't really help you with this, aside from saying that JiBX has no
known legal restrictions on distribution and is freely available from
anywhere in the world through SourceForge distribution.
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Training and Consulting
Hi Rohan,
The Xsd2Jibx tool is very outdated and no longer maintained. You'll need
to download an old version of JiBX (try jibx-1.0RC1, since that's from
the same timeframe as the last Xsd2Jibx release) to make this work.
I am actively working on a replacement, and hope to have it available
Hi Joe,
There's a technique I'm using for the unit tests of new components
(including the schema model). I've posted that on the wiki at
http://www.sosnoski.com/jibx-wiki/space/usage-questions/junit-test-bindings
No ideas of how this would work out with Maven2, but it should be okay
if you
Hi Ian,
The order is given in the XML Summary page description of the binding
element, where the list of child elements is: namespace, format,
include, mapping (at least 1 mapping required)
It's also in the schema in the /docs directory of the distribution:
xs:element name=binding
Looks good, Joshua, though if the type=... value is constant you can
just say:
value style=attribute name=type constant=ns4:Vector/
rather than having to set a field.
This is kind of a kludge with the namespace handling. To really do this
correctly there's a QName class added to
Yes, marshallDocument() is documented as closing the output
stream/writer (indirectly - it says it's the same as calling a sequence
ending with endDocument(), and endDocument() says it closes the output.
To avoid this, you can just marshal your object directly (by casting the
object to
Hi Orjan,
Not sure why this has cropped up now (was I using a BufferedOutputStream
before at some point, and stopped? I'm too lazy to check...), but I've
made the change in my working copy and binding is substantially faster.
Thanks for pointing this out!
- Dennis
Dennis M. Sosnoski
SOA
Hi Varghese,
I'd appreciate it if you'd enter a Jira bug report for this issue. That
way I'll make sure I don't forget about it.
Thanks,
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Training and Consulting
http://www.sosnoski.com - http://www.sosnoski.co.nz
Seattle, WA
Hi Unmesh,
I'm not sure exactly what you want here - are the request and response
samples the way you want the input and output to look, or is the
response what you're currently getting from JiBX, and you want to
instead have it match the request? Either way, it should be easy to
handle this
Hi Varghese,
You can just add style=element on the value name=data/ element.
The value-style=... setting on a container just changes the default
for all the child value elements; you can always override this on the
individual value.
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Hi Beet,
I'm not aware of any bugs of this type, but would be very interested in
learning more. If you're able to supply more details (hopefully
including a working test case) please add a problem report in Jira.
Offhand, it's hard for me to understand how this could occur. If the
application
Hi Michael,
I haven't tried this, but since EnumSet is just a specialized collection
you should be able to do something along the following lines:
collection name=privileges get-method=getObjectNames
set-method=setObjectNames
value name=name type=...AbstractPrivilege$ObjectNames/
SourceVO having different amounts of child-SourceVOs, for
instance.
Is this possible with JiBX?
Regards,
Johannes
Original-Nachricht
Datum: Mon, 05 Mar 2007 19:23:19 +1300
Von: Dennis Sosnoski [EMAIL PROTECTED]
An: JiBX users jibx-users@lists.sourceforge.net
CC:
Betreff
,
neither shall be the nesting level depth of the Java SourceVOs.
Is this possible? Or is the flat XML structure you proposed the only way of
dealing with this situation?
Regards,
Johannes
Original-Nachricht
Datum: Wed, 07 Mar 2007 23:24:03 +1300
Von: Dennis Sosnoski
Hi Johannes,
There's no way to ignore the value during unmarshalling. For version 2
I've planned a split component of the binding that allows separate
input and output branches, which would make this type of situation
easier to handle.
You can make the added get/set methods you need for this
Hi Josh,
The format attribute is only used when you're using a non-default
conversion for the value type. The section of the binding tutorial on
serializer/deserializers mentions this:
http://jibx.sourceforge.net/tutorial/binding-extend.html#serdeser (The
third way of defining custom
Already answered on the Axis2 list, but for the benefit of others:
If you're using unwrapped you shouldn't need to map this at all. If
you're using wrapped handling you'd just have a marker class with an
empty mapping:
mapping class=UpdateResponse name=updateResponse/
Wrapped handling
Hi Javier,
That looks very strange to me. From the error messages, it appears that
your binding is not being read correctly. I don't know what could cause
this. What version of the main jibx-binding.jar are you using with the
schema generator (either located in the same directory as the
Hi Konrad,
Yes, this was a design decision. The point is that the deserializer can
then decide what value to be returned for a missing input. But in
retrospect, I think this may have been a bad call - if the user wants a
default value, they can specify it on the value or format element.
I'm
Hi Dan,
It looks like you've got mismatched jar versions. You should update to
the Axis2 1.1.1 release and the JiBX jars included in that release (or
the more recent JiBX 1.1.3 release jars - they should be compatible with
Axis2 1.1.1). Axis2 1.0.1 was pretty bad in any case, so going to Axis2
Hi Ronald,
I removed this restriction in the current 1.1.3 release code (though I
never did update the web site to match). You can do exactly what you
want, have a binding with only format definitions which you then
include into other bindings.
- Dennis
Dennis M. Sosnoski
SOA and Web
Hi Usama,
You can use two separate binding definitions for this purpose. All the
common stuff can go in one binding, which the other bindings then include:
common-binding.xml:
binding
mapping name=common...
/mapping
/binding
message1-binding.xml:
binding
include
Hi Brian,
Yes, I agree that this is an error - or at least an ambiguity - in the
JiBX handling. There are unfortunately several issues around this type
of nested structure binding. I've been planning to get this cleaned up
completely in the JiBX 2.0 code. The only hitch is that I have no
The reason for the jsr173_api.jar reference is because xpp3.jar wasn't
found. JiBX uses xpp3.jar if it's present on the classpath, but
otherwise looks for a StAX parser.
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Training and Consulting
http://www.sosnoski.com -
data type (ie
java.util.Date) and try to automatically convert an xsd:dateTime to it?
Joshua
On 3/27/07, *Dennis Sosnoski* [EMAIL PROTECTED]
mailto:[EMAIL PROTECTED] wrote:
Hi Josh,
The format attribute is only used when you're using a non-default
conversion
Javier Diaz wrote:
It's very a good to know that you will be releasing a new version for
the generator soon. Will it compatible with the latest jibx version?
Yes, the new code requires the latest JiBX version, and generates
bindings using the latest features.
- Dennis
Hi Johannes,
This behavior was a design decision. The funny thing about design
decisions is that you make them (or at least I do) without necessarily
having a full understanding of the downstream implications. When I was
writing the code, it seemed natural to me to pass a null value to the
Yes, a post-set method will allow you to check for a null collection
reference (which will occur when the list element is missing from the
input XML) and replace it with an empty List instance.
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Training and Consulting
Hi Paul,
There's a public void setDocument(java.io.Reader rdr, java.lang.String
name, boolean nsa) method in the
org.jibx.runtime.impl.UnmarshallingContext that allows you to disable
namespace handling by the parser. I believe that if you disable the
namespaces you get the entire name,
Hi Serkan,
Xsd2Jibx is unfortunately very out of date and not maintained. I'm
working on a replacement, and hope to have that out in May. In the
meantime, the Axis2 page
http://ws.apache.org/axis2/1_1_1/userguide-creatingclients-jibx.html has
instructions for how you can use Xsd2Jibx to
Hi Chris,
You do need to have a particular class associated with your top-level
element, in this case the people element. If this people element
were instead embedded inside some other XML you could avoid this, by
just using a collection element in your binding definition that
references the
Hi Chris,
Yes, a nice solution to a somewhat unusual problem. Of course, it's also
possible to just define a separate binding that doesn't include the
values you want to suppress on output - but if you're suppressing
different combinations of attributes at different times this would
require a
Yes, remember that the schema representation of dateTime values and
variations only represents values in GMT (or in an unspecified zone,
which is essentially useless for most programming purposes). But if you
just print out the Time value resulting from unmarshalling you'll use
the default
Hi Javier,
The problem here is that the generated xsd doesn't match the binding.
The binding namespace element has default=all, which is rarely what
you want to use in practice. Change to using default=elements and your
binding should match the schema definition.
XMLSpy does a pretty
I haven't tried out the XFire support for JiBX yet personally, but from
what I've seen the performance should be about the same. I did try
digging out the JibxSoap code recently, and found that it's
substantially faster than using JiBX with Axis2. I'll try to get this
cleaned up to the point
Hi Steven,
I don't see anything obviously wrong with what you're doing. Here's a
sample from the JiBX build test cases:
mapping name=myelement class=simple.MyClass3
structure ordered=false choice=true
value name=a field=a usage=optional/
value name=b field=b usage=optional/
Hi Rohit,
Assuming the element you want to process this way is embedded inside
some other XML element (and the object to be unmarshalled is a property
of some containing object), you should be able to just specify your
custom unmarshaller for the object at that point in the binding. Then
you
of the application. the
Dispatcher would then look at the root element and using this get the
required bind factory.
If you have any thoughts on these ideas or alternatives please let me know
Regards,
Tarek
- Original Message -
From: Dennis Sosnoski [EMAIL PROTECTED]
To: JiBX users jibx-users
Hi Johannes,
The problem is actually at the parser level, not at the JiBX level. When
you're parsing an XML document there's no way to tell the parser that
you just want to get the content of a particular element as a string,
rather than as XML (or HTML) components. So if you include
Steven E. Harris wrote:
Dennis Sosnoski [EMAIL PROTECTED] writes:
Here's a sample from the JiBX build test cases:
mapping name=myelement class=simple.MyClass3
structure ordered=false choice=true
value name=a field=a usage=optional/
value name=b field=b usage=optional
of the choice handling to do what users
would expect.
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Training and Consulting
http://www.sosnoski.com - http://www.sosnoski.co.nz
Seattle, WA +1-425-939-0576 - Wellington, NZ +64-4-298-6117
Steven E. Harris wrote:
Dennis Sosnoski [EMAIL
Hi Joanna,
It looks like you're getting a somewhat unintended side-effect of the
abstract mapping in your second binding, with the collection as a
separate mapping. Abstract mappings aren't really intended to be used
directly as collection item types, and I should probably have a
validation
.wsdl.codegen.extension.JiBXExtension.engage(JiBXExtensi
on.java:74)
at
org.apache.axis2.wsdl.codegen.CodeGenerationEngine.generate(CodeGenerati
onEngine.java:177)
... 2 more
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Dennis
Sosnoski
Sent: Friday, April 20, 2007 1
Hi Glenard,
I don't know of any errors of this type. Can you give some specifics on
where you're seeing the Namespace URI has not been declared exception?
That probably means some sort of state conflict, where either the StAX
writer or the JiBX code is somehow confused about the state of the
Xsd2Jibx is very out dated, and you're probably going to need to do some
major work on the output binding in order to make it usable with Axis2.
That said, the particular problem you're seeing below should be fixed in
the 1.1.4 JiBX jars, which you can access from maven repository at
Hi Holger,
You'd need to read the DOCTYPE information from the parser yourself
before you turn control over to the JiBX unmarshalling. See the
Controlling JiBX with front-end code example at
http://jibx.sourceforge.net/tutorial/binding-custom.html#frontend for an
example of how this type of
Hi Rajan,
The Xsd2Jibx code is outdated, and does not generate either code or
bindings that make use of the features added to JiBX over the last 2-3
years. JiBX does include choice support, using a choice='true' attribute
on the mapping and structure elements, but as of now there's no way
to
Hi Rajan,
The exception message is rather strange, and I suspect there's something
unexpected in your binding to cause this. Can you show the portion of
the binding that includes your com.mediaslate.og.schema.xml.mdp.Request
class?
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Hi Eugeny,
Do you need to use both versions (both namespaces) within a single
document? Your easiest solution is to use two separate binding
definitions, which works fine as long as you're not mixing the two
within a single document. Otherwise it gets more complicated.
- Dennis
Dennis M.
Hi Javier,
The JibxSoap code has been left in an alpha form for some time, as I'm
sure you're aware. Despite this alpha state, it is usable for web
services that don't need WS-Security or other WS-* layers on top of SOAP.
In the current state of the code I believe you're correct that the
Hi Andrey,
It looks like there's a problem in the simpleType reference handling
with the Axis2 JiBX unwrapped code generation. This is something I
haven't tried before, but I'll look into it. In the meantime, it'd be
good if you could report this as an issue in the Axis2 Jira at
I'm now scheduling JiBX and/or Axis2 training classes for the U.S. for
June, and for Australia in July. The current Axis2 in Action course
outline is at http://www.sosnoski.com/companys/axis2class.html with a
separate SOA for Developers course outline at
Hi Joanna,
The extends=... attribute is only needed for a mapping when you're
implementing the equivalent of schema substitution groups. It's not
intended for use where you're just doing the equivalent of schema type
extension. The tutorial at
Hi Yuvraj,
Xsd2Jibx is badly outdated and no longer maintained, but can still be
used. You'll need to download an old version of JiBX (try jibx-1.0RC1,
since that's from the same timeframe as the last Xsd2Jibx release).
I'll at least recompile the bindings and post an updated Xsd2Jibx that
- http://www.sosnoski.co.nz
Seattle, WA +1-425-939-0576 - Wellington, NZ +64-4-298-6117
Dennis Sosnoski wrote:
Hi Glenard,
I think I've gotten this corrected - but only for the Axis2 1.2 release
code, and there's not yet a Rampart build available for that Axis2. I'll
try to backport
Muench wrote:
I just tried the axis2-1.2b.jar you have posted on the wiki. Any idea why
wsdl2java no longer produces a service stub?
Joanna
Dennis Sosnoski wrote:
Ok, I've backported the fix to Axis2 1.1.1 and posted that at
http://www.sosnoski.com/jibx-wiki/space/axis2-jibx I've also
it improves the
quality of the product.
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Training and Consulting
http://www.sosnoski.com - http://www.sosnoski.co.nz
Seattle, WA +1-425-939-0576 - Wellington, NZ +64-4-298-6117
Dennis Sosnoski wrote:
I'm not seeing any problems with this. So
Hi Gerry,
This is basically just a filler I used in cases where I either was
unsure some particular situation could occur or didn't want to deal with
the complication at the time I was writing the code.
Since you're running into this I'd appreciate it if you'd email me
(directly, not via the
--
Dennis M. Sosnoski
SOA and Web Services in Java
Training and Consulting
http://www.sosnoski.com - http://www.sosnoski.co.nz
Seattle, WA +1-425-939-0576 - Wellington, NZ +64-4-298-6117
Jérôme BERNARD wrote:
On 5/22/07, Dennis Sosnoski [EMAIL PROTECTED] wrote:
Hi Jérôme,
Hi Dennis
and Consulting
http://www.sosnoski.com - http://www.sosnoski.co.nz
Seattle, WA +1-425-939-0576 - Wellington, NZ +64-4-298-6117
Jérôme BERNARD wrote:
Hi Dennis,
On 5/24/07, Dennis Sosnoski [EMAIL PROTECTED] wrote:
The jar files have been in the maven repository since last night, so
Hi Todd,
The source code for the plugin is available from the SourceForge
repository, with the module name jibxeclipse. See
http://sourceforge.net/cvs/?group_id=69358 for details of how to use
this. It has been dropped from active distribution because there was no
one to maintain it.
-
Hi Holger,
Maybe I'm missing something, but I don't see why using java.lang.Object
wouldn't work for this case. ID/IDREF are an untyped linkage mechanism,
so there shouldn't be any constraints on using the same form of
references to different types of elements.
- Dennis
Dennis M. Sosnoski
Hi Todd,
Yes, it's definitely possible to map polymorphic collections in a couple
of different ways. But to do so, you need to make use of the class
structure. Your binding references the (private) fields of the abstract
base class directly from within the mapping of the child class, so the
Hi Tom,
You need to show the actual binding definition you're using, which I
assume you're probably generating with Xsd2Jibx.
The current Xsd2Jibx code is not very reliable, though I'd expect it to
handle such a simple schema structure without problems.
- Dennis
Dennis M. Sosnoski
SOA and
Hi Eugeny,
I don't know any way to avoid this, since AFAIK Ant has no way of
checking for recompiled classes.
On the bright side, the binding compiler should generally be smart
enough to avoid modifying the class files which already contain the JiBX
bytecode enhancements (the exception would
unmarshalling from Xml
I got the same error when I didn't have the xml elements in the same
order as they were listed in the binding definition. You might check that.
Tammy
Dennis Sosnoski wrote:
Hi Tom,
You need to show the actual binding definition you're using, which I
assume you're
Hi Pavel,
I've got an updated version I'll post tomorrow. The changes include
handling for interfaces as well as subclasses, and clean up several
problems that occurred with the original version. The source is in the
SourceForge CVS for JiBX, but I haven't committed my changes recently -
I'll
Hi Sowjanya,
I don't see anything wrong with this offhand, but it's possible that
something is going wrong in the generated code. Are you using the 1.1.5
release of JiBX? If not, please try with that. If you *are* using 1.1.5,
please enter a Jira at
Hi Surjit,
You have several choices here. One is to just treat the guid simpleType
as though it were a plain string in schema terms (since it's a
restriction type based on string). In this case you just have a value
element in your binding that links the element using the guid type to a
I suspect the problem EMMA is having is in the bytecode added by JiBX.
Since there's no source code corresponding to this bytecode there's also
no debug information.
The best thing I can suggest is that you run EMMA first, for offline
instrumentation, and then the JiBX binding compiler after
Hi Peter,
This is not officially supported in the 1.X code, but you should be able
to gracefully close out the XML that you've generated so far if you're
comfortable digging into the code a bit. The key would be to create your
own org.jibx.runtime.IXMLWriter which tracks the state information
-6117
Dennis Sosnoski wrote:
Hi Surjit,
You have several choices here. One is to just treat the guid
simpleType as though it were a plain string in schema terms (since
it's a restriction type based on string). In this case you just have a
value element in your binding that links
Hi Sowjanya,
flexible=true is supported, and there are several test cases under
/build/test/simple that verify the operation. Of course, you may be
running into a problem in something that doesn't match the test cases. I
see your later posting seems to be giving more details on this, and as I
Hi Eugene,
The item-type attribute on a collection only works when you have a
mapping defined for the item-type, not with a format (the difference
being that the former defines an element, while the latter just gives a
text value - and you need elements for the items in the collection). I
see
Hi Jennifer,
My bad on this - it needs to use a built-in custom
serializer/deserializer. I'll fix this for 1.1.6.
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Training and Consulting
http://www.sosnoski.com - http://www.sosnoski.co.nz
Seattle, WA +1-425-939-0576 - Wellington, NZ
Hi Maciek,
I don't know enough about the maven plugin to make any suggestions on
this. Can any maven mavens comment?
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Training and Consulting
http://www.sosnoski.com - http://www.sosnoski.co.nz
Seattle, WA +1-425-939-0576 - Wellington,
The easiest way to handle this is probably by using a using a custom
marshaller/unmarshaller, as Varghese suggested. If you define abstract
mappings for the Person and Business objects, your custom
marshaller/unmarshaller can basically just handle the ns:Party wrapper
element and xsi:type
I've posted an update of the Jibx2Wsdl code at
http://www.sosnoski.com/jibx-wiki/space/axis2-jibx/jibx2wsdl The changes
in this version include:
1. Fixed handling of -b parameter to set root binding definition file
name.
2. Added support for using interfaces as equivalent to
Hi Pavel,
Sure looks like a bug to me. :-) Can you send me (directly) your
project, so I can investigate both this and the other issue for myself?
Thanks,
- Dennis
Dennis M. Sosnoski
SOA and Web Services in Java
Training and Consulting
http://www.sosnoski.com - http://www.sosnoski.co.nz
Hi Pavel,
You need to use the -p parameter once for *each* class path parameter -
there's currently no scanning for path separator characters, though I
suppose I could add that. Meanwhile, I'll make this clearer on the wiki
page for Jibx2Wsdl.
- Dennis
Dennis M. Sosnoski
SOA and Web
xsd:element ref=tns:IProjectDTOComponent minOccurs=0
maxOccurs=unbounded/
Pavel
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Dennis
Sosnoski
Sent: Tuesday, July 24, 2007 1:19 AM
To: JiBX users
Subject: Re: [jibx-users] jibx2wsdl: wrong schema generated
Hi Linus,
Some thoughts and ideas...
Is it possible you're somehow getting two copies of the
edu.iris.ws.core.Event class in your server classpath (perhaps with an
old jar still left in place)? ClassCastException always confuses me, but
it looks like it's saying that the the error is
401 - 500 of 757 matches
Mail list logo