"Paul Fremantle" <[EMAIL PROTECTED]> wrote on 08/02/2007 06:01:14 AM:
> I recently found some parts of DOOM that weren't implemented (I got > TODO exceptions). > > I believe these were in implementations of SOAPEnvelope/SOAPBody. I > also had a problem where DOOM couldn't cope with an empty Namespace. > > Anyway, my point is this. Surely we shouldn't have two implementations > of the SOAP* interfaces - one for DOOM and one for LLOM? Won't the > code be exactly the same in both cases? The SOAP implementation objects extend the DOOM or LLOM objects. Thus> two implementation hierarchies are needed (one for DOOM and one for LLOM). An alternative approach is to have a single SOAP implementation hierarchy that delegates to an OM tree. Then you would have a single SOAP implementation. However this has a different set of challenges. The main challenge is that you may have two objects for each element (e.g. A lightweight SOAPEnvelope object that delegates to an OMElement). If you took this approach, I also suggest splitting out the SAAJ code and using a similar approach.). SAAJ objects extend (versus delegate) to LLOM objects. Is there any benefit to moving the SAAJ model to Axiom. The only downside is that Axis2 would still need to implement SOAPConnection...but the rest of the code is basically modelling code. > > If not, I'm sure we could refactor so it is. > > Paul > > -- > Paul Fremantle > Co-Founder and VP of Technical Sales, WSO2 > OASIS WS-RX TC Co-chair > > blog: http://pzf.fremantle.org > [EMAIL PROTECTED] > > "Oxygenating the Web Service Platform", www.wso2.com > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > Rich Scheuerle IBM Web Services Apache Axis2 ([EMAIL PROTECTED]) 512-838-5115 (IBM TL 678-5115)
