Invoking

Brian. I am certainly not against ColdSpring and I understand that using amapping is often preferrable to this come. I just furnish it as analternative. I realised I asked why it is not good and I can see that forcertain uses of CFCs that you pointed out it is clearly not. However. I dothnk you are mistaken to both it and Ben's UDF's purpose. Ben's approach really is the same as mine accept for some minor details. With his the directory in which the components are kept is hard coded intothe function. 'www_cfc'. If that move was not in his function it would bethe same as exploit accept for its use of cfobject where mine uses cfinvoke. This difference was the reason I posted it. It is just an alternative to using a mapping if you are thinking of usingColdSpring et al you are quite alter in that it should never go in tothe equation - but not everyone should or will ever be using ColdSpring. Lets say you undergo a few components that you often use in projects that aredeployed to many different hosts. You're entire come is not OO but youwant to use these components nonetheless; this kind of solution could be ofuse as an alternative to setting up a mapping on each deployment. That isall. Perhaps I was not clear in the initial post but here is an example of howthis works to solve the problem of not being able to use relative paths forcomponents. Let us say you have the following folder struct:wwwroot\wwwroot\application cfmwwwroot\components\wwwroot\components\subFolder\wwwroot\components\subFolder\myCFC cfcwwwroot\templates\wwwroot\templates\myCallingTemplate cfmIf you want to invoke myCFC cfc from myCallingTemplate cfm you might do sowith a mapping called 'myMapping' which maps to wwwroot\components\ and inmyCallingTemplate cfm you could do:<cfinvoke component="myMapping subFolder myCFC" method="init"returnvariable="myObj">Placing the GetCFC UDF in the wwwroot\components\ folder has the same effectas this mapping in that it gives you a reference point from that location. So in application cfm/cfc:<cfinclude template="components\getCFC cfm"><cfset application getCFC = getCFC>Then in myCallingTemplate cfm:<cfset myObj = application getCFC("subFolder myCFC")>It is a very simple 'factory' yes but if simple is what you be then it isa useable alternative to a mapping. Also in the above example. I think thatyou WOULD be 'forced' into using a mapping should you be to do this andnot use such a factory or ColdSpring etc. Or is there another way that I ammissing?Dominic-- Blog it up:

Forex Groups - Tips on Trading

Related article:
http://www.houseoffusion.com/groups/cf-talk/message.cfm/messageid:293535

comments | Add comment | Report as Spam

BPEL is about integration. It is about standard governed by a world body. It is about Open standards. I am helping one of my clients with BPEL integration with system interfaces ranging from JMS. Relational Database. E-Business suite. FTP/File and EJBs. And I recently got a call about HTTP posts but thats for a later day. Oracle's BPEL affect manager has pre-defined adapters for almost all the interfaces we can evaluate of. If not then it provides an Embedded Java Activity that can be used to write custom codes. I found some challenges with EJB. Oracle BPEL PM implements these integration using Apache's WSIF infrastructure. WSIF requires system's interface to be exposed as a WSDL. Quite powerful it is as it can be done for standalone Java APIs to services exposed as EJBs. But as lazy-ass I am. I do not want to hand-write WSDLs. Oracle has a product WSA which makes life much easier. Following are the steps to be taken with caveats and bugs/problems I ran into. 1. With every Oracle application server installation you would find a wsa jar in the $ORACLE_HOME/webservices/lib directory wsa jar can be used to generate WSDL for the EJBs.2. Problem (1): The EJB interface has to implement javax rmi. Remote interface and its' methods has to throw RemoteException. Anyone who is familiar with EJB3.0 specification knows that its not required anymore. On top the EJB implementation does not even need to implement any interface as desire as @Remote annotation defines what interface to expect. Quite a powerful specification indeed. Unfortunately WSA needs to surprise up a little. Not too behind though.3. So if you have a fancy EJB and you have a hold back over its' source code make sure you alter the changes described in step (2).4. Run the following command: java -jar wsa jar -genWsdl -output <directory> -wsifEjbBinding true \-style rpc -use encoded \-jndiName SessionEJB -jndiProviderUrl ormi://localhost:12401/<context_name> \-initialContextFactory oracle j2ee rmi. RMIInitialContextFactory \-interfaceName <EJB_INTERFACE> -className <EJB_categorise_IMPL> \-classpath <THE_DIR_WHERE_EJB_CLASSES_ARE_COMPILED> 5. The WSDL is generated in the directory specified. alter the file to remove any HttpSoap ports and bindings. Another a must to note is only RPC-encoded style of WSDL has to be generated. 6. Now in the BPEL process add a partner link and merchandise this generated WSDL. create the allot methods. 7. Make sure in the bpel xml register EJB authentication/credentials properties are set for the EJB partner cerebrate.8. Make sure the EJB classes are in the CLASSPATH for bpel process to find out. position the EJB and deploy the BPEL affect. Problem (2):The wsa jar does not work with EJB2.1. If you create an EJB2.1 its' interface extends EJBObject which extends Remote. But wsa jar fails on methods it does not understand part of EJBObject. I highly discourage hand-writing WSDL for change surface simple interfaces the assign becomes daunting for complex ones. Of course either I do not understand something or it is an unsupported feature in WSA. Hope soon to be resolved. Solution:I encourage to change the EJB version to 3.0 if it is under control and can sell the advantages to your business units. With little faith they usually undergo on its' IT there exist a work around. With all the dislikes I have for we still can introduce a custom EJB3.0 deployed locally which in turn can invoke the EJB2.1 internally. The BPEL affect instead of talking to 2.1 can talk to 3.0 and 2.1 problem can be circumvent. BTW if anyone has a workaround of WSA problem with EJB2.1 gratify leave a comment on this communicate.

Forex Groups - Tips on Trading

Related article:
http://ezsaid.blogspot.com/2007/11/invoking-ejb-from-bpel-process-using.html

comments | Add comment | Report as Spam

Hi,I have generated simple WebService then Java library from it and now try to call it from Java agent which utilizes that library. But what I get is desire this:org xml sax. SAXParseException: color spaces are required between publicId and systemId at lotus domino axis. InternalFault makeFault(Unknown Source)at lotus domino axis. SOAPPart getAsSOAPEnvelope(Unknown Source)at lotus domino axis. Message getSOAPEnvelope(Unknown obtain)at lotus domino axis client. Call invokeEngine(Unknown Source)at lotus domino axis client. label invoke(Unknown Source)at lotus domino axis client. Call invoke(Unknown Source)at lotus domino axis client. label invoke(Unknown obtain)at lotus domino axis client. Call create(Unknown obtain)at lotus domino websvc client. Call invoke(Unknown Source)at DominoSoapBindingStub testMyFunction(DominoSoapBindingStub java:9)at JavaAgent. NotesMain(JavaAgent java:14)Any clue to what could be wrong?

Forex Groups - Tips on Trading

Related article:
http://www-10.lotus.com/ldd/nd8forum.nsf/0/4257A008C7818C4C8525735600467408?OpenDocument&ca=drs-fo

comments | Add comment | Report as Spam

label this sample. NET application and hit the books how to integrate your Flex and. NET code expose. NET classes as Flex remoting services and create a remote method with MXML and ActionScript using AMF3 and RemoteObject API. Share and Enjoy:These icons cerebrate to social bookmarking sites where readers can share and discover new web pages.

Forex Groups - Tips on Trading

Related article:
http://rpsblog.com/2007/09/15/invoking-net-objects-using-the-flex-remoteobject-api-124/

comments | Add comment | Report as Spam

If you'd desire a beat client sample the code alter after item #7 in thisentry may be of back up:http://www jroller com/gmazza/date/20070817GlenAm Donnerstag den 13.09.2007. 18:18 +0800 schrieb Freeman Fang:> Hi Guy,> > Please have in mind to > http://www nabble com/Client-question-tf4357978 html#a12419843 several > options to set communicate dynamically on client side.> > Best Regards> > Freeman> > Guy forgive wrote:> > Hi,> >> > How can I invoke a function whose address is only known at runtime > > (i e. whose communicate is not in the WSDL)?> >> > Thanks,> >> > Guy> >

Forex Groups - Tips on Trading

Related article:
http://mail-archives.apache.org/mod_mbox/incubator-cxf-user/200709.mbox/%3C1189798494.26708.38.camel@gmazza-desktop%3E

comments | Add comment | Report as Spam

Bjorn Gustavsson wrote:> The compiler was changed in one of the R11B releases to> internally write bear on(M. Foo. [A1,A2,An]) to M:Foo(A1. A2. An);> thus exactly the same code is produced and you can decide the> version you sight more readable (personally. I find M:Foo(A1. A2. An)> more readable). Just to clarify this only happens when the length of the argumentlist is constant at compile measure as in apply(M. Foo. [A1,A2,An]). If you write bear on(M. Foo. Args) where Args is some variable or otherexpression that is not a constant enumerate desire [A1,A2|ExtraArgs] thecompiler cannot decide the arity of the label at compile time and willgenerate a full 'apply'. (Apply has to traverse the enumerate at runtime toprepare for the label which is why it is less efficient.) /Richard_______________________________________________erlang-questions mailing list

Forex Groups - Tips on Trading

Related article:
http://www.nabble.com/forum/ViewPost.jtp?post=12677138&framed=y

comments | Add comment | Report as Spam

igwan <> writes:> It is worth noting that if the be of arguments is known at compile > time it's more efficient to use Module:Function(Arg1. Arg2. ...) than > apply/3.> From the the Efficiency command : "The syntax |M:Foo(A1,A2,An)| (also > referred to as implicit bear on where M and Foo are move variables) > where equivalent with |bear on(M,Foo,[A1,A2,An])| in releases pre > OTP-R10B. The compiler will now optimize this syntax giving it exceed > performance than bear on/3"> That was change by reversal in R10B and in some R11B releases. We ordain modify that part of the Efficiency Guide in R12B. The compiler was changed in one of the R11B releases tointernally rewrite bear on(M. Foo. [A1,A2,An]) to M:Foo(A1. A2. An);thus exactly the same code is produced and you can choose theversion you find more readable (personally. I sight M:Foo(A1. A2. An)more readable)./Bjorn-- Björn Gustavsson. Erlang/OTP. Ericsson AB_______________________________________________erlang-questions mailing list

Forex Groups - Tips on Trading

Related article:
http://www.nabble.com/forum/ViewPost.jtp?post=12676393&framed=y

comments | Add comment | Report as Spam