I have to agree to Srinath on this. If we are going to move the transports out, it *must* be done via a major release and no question about it. And I see no harm in putting your own transport - even if it wraps a similar transport the core carries right now.
Ajith On 10/4/07, Srinath Perera <[EMAIL PROTECTED]> wrote: > IMO defining your own transport is the right way to do it, when Axis2 > is being designed, it is designed to be pluggable so we can do this > kind of things. > > If we need to move tranports out of Axis2 kernel, I belive it is a > seperate discussion and I do not belive above use case is > justification for doing that (As it can easily achived via > configuration). > > On the other hand it must be done in a major relase, if it is to be done. > > Thanks > Srinath > > On 10/4/07, Ruwan Linton <[EMAIL PROTECTED]> wrote: > > Hhmmm... > > > > I don't think it is impossible, but at the same time, IMO it is not the > > right thing to do? (although that will solve the problem). > > > > I think it is nice to abstract out the transports from the kernel, even for > > axis2. That will modularize the things, so that one can write an axis2 > > transport and plug it with the axis2 kernel (I mean as a replacement of the > > existing axis2 transport, I know the same applies here as well and you can > > change the axis2.xml to pick that :D) > > > > If this is a hard task for axis2, then we will try to get rid of this (may > > be with renaming the transport classes and putting that in to the axis2.xml) > > > > Asankha, WDYT? > > > > Thanks, > > Ruwan > > > > > > > > On 10/4/07, Rajith Attapattu <[EMAIL PROTECTED]> wrote: > > > Ruwan, Asankha, > > > > > > Actually I was also wondering about the same? > > > Isn't that possible ?? > > > > > > Regards, > > > > > > Rajith > > > > > > > > > > > > On 10/4/07, Srinath Perera <[EMAIL PROTECTED]> wrote: > > > > Can't we do this? > > > > > > > > Write a new http transport (Http transport edited for new Http core, > > > > impl class name should be different), and change the axis2.xml to > > > > provide new Http implementation as implementation for http transport, > > > > and ship synapse with new axis2.xml (make changing the class name for > > > > new http transport and removing old http jar from dependencies for > > > > synapse you should be able to get away from class loading problems) > > > > > > > > May be you already considered this and I am missing something > > > > Thanks > > > > Srinath > > > > > > > > On 10/4/07, Asankha C. Perera <[EMAIL PROTECTED]> wrote: > > > > > > > > > > Sorry to drop in late on this one.. > > > > > > > > > > What would be ideal is just a separation of the Axis2 transports code > > into > > > > > a separate module within Axis2 - i.e. outside of the Kernel. > > > > > > > > > > thanks > > > > > asankha > > > > > > > > > > > > > > > Ruwan Linton wrote: > > > > > Hi Chinthaka, > > > > > > > > > > It does have API level problems and that is why we are running in to > > issues > > > > > in the cases where the transport impl is picked from the > > axis2-kernel's > > > > > classes. For example; > > > > > > > > > > The method > > org.apache.http.RequestLine.getHttpVersion () > > > > > in http-core-alpha5 has been changed to > > > > > org.apache.http.RequestLine.getProtocolVersion() in > > the > > > > > http-core-alpha6-SNAPSHOT > > > > > > > > > > and because we have the http-core-alpha6-SNAPSHOT jar in synapse, if > > the > > > > > axis2 class is loaded rather than the synapse transport class, then we > > run > > > > > in to runtime exception saying the method getHttpVersion is not found. > > > > > > > > > > We can not remove the axis2-kernel dependency (obvious) at the same > > time we > > > > > need to go with the http-core-alpha6 with its perf improvements. > > > > > > > > > > Thanks, > > > > > Ruwan > > > > > > > > > > > > > > > On 10/4/07, Eran Chinthaka <[EMAIL PROTECTED] > wrote: > > > > > > -----BEGIN PGP SIGNED MESSAGE----- > > > > > > Hash: SHA1 > > > > > > > > > > > > Just out of curiosity, if you take out the http transport out of the > > > > > > kernel, won't that change the packages of it which will affect the > > > > > > transport listing in axis2.xml which in turn will be visible to > > users. > > > > > > That means, we might have to do a major release with this change. > > > > > > > > > > > > Also what will happen if you make Axis2 to run with the new kernel > > jars, > > > > > > that you want to have synapse? In other words can you remove the > > > > > > existing version of http-core and run Axis2 with the latest version > > of > > > > > > http-core? Is there a package level problem? > > > > > > > > > > > > Thanks, > > > > > > Chinthaka > > > > > > > > > > > > Ruwan Linton wrote: > > > > > > > Hi axis-devs, > > > > > > > > > > > > > > We are getting ready for the Synapse 1.1 release and we faced to a > > > > > > > problem with the transports. Synapse is going to ship with the > > http-core > > > > > > > transport version 4.0-alpha6 and we have changed synapse transport > > > > > > > module with the improvements for that version ( > > 4.0-alpha6-SNAPSHOT) > > > > > > > which is not compatible with the current http-core version of > > axis2. > > > > > > > But, because of the fact that axis2-kernel carries the same > > classes > > > > > > > inside the kernel jar some times we are running in to class > > loading > > > > > issues. > > > > > > > > > > > > > > Can we get axis2 transports out of the axis2 kernel module and get > > a > > > > > > > 1.3.0.1 <http://1.3.0.1 > (or any point) release of axis2, so that > > we can > > > > > > > depend on the axis2-kernel without transports. Is this possible? > > > > > > > > > > > > > > Thanks, > > > > > > > Ruwan > > > > > > > > > > > > > > -- > > > > > > > Ruwan Linton > > > > > > > http://www.wso2.org - "Oxygenating the Web Services Platform" > > > > > > > > > > > > -----BEGIN PGP SIGNATURE----- > > > > > > Version: GnuPG v1.4.6 (GNU/Linux) > > > > > > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org > > > > > > > > > > > > > > > > > > > iD8DBQFHBGC3jON2uBzUhh8RAqLJAKCZipLBYZt1YKIN9YNs9TUX4pA0mgCgjIQs > > > > > > nmLxNEzpmXI1EDt12oJv9JQ= > > > > > > =JPr5 > > > > > > -----END PGP SIGNATURE----- > > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------- > > > > > > To unsubscribe, e-mail: > > > > > [EMAIL PROTECTED] > > > > > > For additional commands, e-mail: [EMAIL PROTECTED] > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > > Ruwan Linton > > > > > http://www.wso2.org - "Oxygenating the Web Services Platform" > > > > > > > > > > > > -- > > > > ============================ > > > > Srinath Perera: > > > > Indiana University, Bloomington > > > > http://www.cs.indiana.edu/~hperera/ > > > > http://www.bloglines.com/blog/hemapani > > > > > > > > Beauty of style and harmony and grace and good rhythm depend on > > > > simplicity -- Plato > > > > > > > > > > --------------------------------------------------------------------- > > > > To unsubscribe, e-mail: > > [EMAIL PROTECTED] > > > > For additional commands, e-mail: [EMAIL PROTECTED] > > > > > > > > > > > > > > > > > > > > > > -- > > > > Ruwan Linton > > http://www.wso2.org - "Oxygenating the Web Services Platform" > > > -- > ============================ > Srinath Perera: > Indiana University, Bloomington > http://www.cs.indiana.edu/~hperera/ > http://www.bloglines.com/blog/hemapani > > Beauty of style and harmony and grace and good rhythm depend on > simplicity -- Plato > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > -- Ajith Ranabahu Reading, after a certain age, diverts the mind too much from its creative pursuits. Any man who reads too much and uses his own brain too little falls into lazy habits of thinking - Albert Einstein --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
