cxf-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Daniel Kulp <dk...@apache.org>
Subject Re: CachedOutputStreamCallback question
Date Wed, 19 Aug 2009 21:18:45 GMT
On Mon August 17 2009 1:46:03 pm Eamonn Dwyer wrote:
> Hi
> I may be missin git but I can't find documentation around the use of a
> CachedOutputStreamCallback so I would like some clarification on its use. 
> From debugging, it looks like it would typically be used to as a mechanism
> to do some last bits of work just before an invocation is made on the
> service. Is that a valid understanding? That seems to be the case for SOAP
> over HTTP (at least) but doesn't seem to be the case for SOAP/CORBA. I'm
> wondering if its a bug in the CORBA implementation or just my
> misunderstanding of it.

Basically, it allows the stream to really know "at the very last possible 
moment" that it can modify things like headers and such.  That's what http 
does with it.   

The one in CORBA is interesting and I didn't even know it did this.   
Basically, there are a couple interceptors and such that assume an 
OutputStream exists.   For CORBA, the OutputStream would never be written to.   
It's mostly just there to satisfy those couple of interceptors as well as get 
the "close" event that would trigger the request/response.   Ideally, this 
would get completely eliminated.   The interceptors that require and 
OutputStream would be updated to be smarter and the "close" trigger could also 
trigger off the XMLOutputStream.close() which the CORBA version of the 
XMLOutputStream could implement properly.



-- 
Daniel Kulp
dkulp@apache.org
http://www.dankulp.com/blog

Mime
View raw message