On Wed, Jul 10, 2013 at 9:41 AM, Sebastien Blanc <scm.blanc(a)gmail.com&gt;wrote: > Hi Folks ! > As you know the current Java Sender API comes today with 2 different > implementations : > > - Resteasy client (used by default) > - AsyncHttpClient > > I discussed a bit with Matzew and we agreed that we should probably > remove the AsyncHttpClient implementation and just keep the resteasy > client, wdyt ? > yep, I'd prefer one client (based on RestEasy). Also that means we don't need all the JavaSender, DefaultSender etc...

We just have one client, and need to name it "reasonably".

I'd also argue that we should worry that the client can be injected, using CDI. Makes sense when being used on JavaEE backends

> > Seb > > > _______________________________________________ > aerogear-dev mailing list > aerogear-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/aerogear-dev > -- Matthias Wessendorf blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf twitter: http://twitter.com/mwessendorf _______________________________________________ aerogear-dev mailing list aerogear-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/aerogear-dev

On Wednesday, July 10, 2013, Sebastien Blanc wrote: > > > > On Wed, Jul 10, 2013 at 9:52 AM, Matthias Wessendorf <matzew(a)apache.org&gt;wrote: > >> >> >> >> On Wed, Jul 10, 2013 at 9:41 AM, Sebastien Blanc <scm.blanc(a)gmail.com&gt;wrote: >> >>> Hi Folks ! >>> As you know the current Java Sender API comes today with 2 different >>> implementations : >>> >>> - Resteasy client (used by default) >>> - AsyncHttpClient >>> >>> I discussed a bit with Matzew and we agreed that we should probably >>> remove the AsyncHttpClient implementation and just keep the resteasy >>> client, wdyt ? >>> >> >> >> yep, I'd prefer one client (based on RestEasy). Also that means we don't >> need all the JavaSender, DefaultSender etc... >> > +1 > >> >> We just have one client, and need to name it "reasonably". >> > Yea how should we name it ? > >> >> I'd also argue that we should worry that the client can be injected, >> using CDI. Makes sense when being used on JavaEE backends >> > isn't that the responsability of the backend app using the sender or is > there anything to do on the sender side ? (in the latest case a jira would > be cool) > it is possible to write the class in a way, that CDI container can not create a proxy See last "bug" from Christos > > >> >> >> >>> >>> Seb >>> >>> >>> _______________________________________________ >>> aerogear-dev mailing list >>> aerogear-dev(a)lists.jboss.org >>> https://lists.jboss.org/mailman/listinfo/aerogear-dev >>> >> >> >> >> -- >> Matthias Wessendorf >> >> blog: http://matthiaswessendorf.wordpress.com/ >> sessions: http://www.slideshare.net/mwessendorf >> twitter: http://twitter.com/mwessendorf >> >> _______________________________________________ >> aerogear-dev mailing list >> aerogear-dev(a)lists.jboss.org >> https://lists.jboss.org/mailman/listinfo/aerogear-dev >> > > -- Matthias Wessendorf blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf twitter: http://twitter.com/mwessendorf _______________________________________________ aerogear-dev mailing list aerogear-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/aerogear-dev

there is a bug on that :-) On Wednesday, July 10, 2013, Sebastien Blanc wrote: Okay, I will open a jira for this and seems like smart to do this before working on the fluent Message API. Seb On Wed, Jul 10, 2013 at 10:14 AM, Matthias Wessendorf <matzew(a)apache.org&gt;wrote: On Wednesday, July 10, 2013, Sebastien Blanc wrote: On Wed, Jul 10, 2013 at 9:52 AM, Matthias Wessendorf <matzew(a)apache.org&gt;wrote: On Wed, Jul 10, 2013 at 9:41 AM, Sebastien Blanc <scm.blanc(a)gmail.com&gt;wrote: Hi Folks ! As you know the current Java Sender API comes today with 2 different implementations : - Resteasy client (used by default) - AsyncHttpClient I discussed a bit with Matzew and we agreed that we should probably remove the AsyncHttpClient implementation and just keep the resteasy client, wdyt ? yep, I'd prefer one client (based on RestEasy). Also that means we don't need all the JavaSender, DefaultSender etc... +1 We just have one client, and need to name it "reasonably". Yea how should we name it ? I'd also argue that we should worry that the client can be injected, using CDI. Makes sense when being used on JavaEE backends isn't that the responsability of the backend app using the sender or is there anything to do on the sender side ? (in the latest case a jira would be cool) it is possible to write the class in a way, that CDI container can not create a proxy See last "bug" from Christos Seb _______________________________________________ aerogear-dev mailing list aerogear-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/aerogear-dev -- Matthias Wessendorf blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf twitter: http://twitter.com/mwessendorf _______________________________________________ aerogear-dev mailing list aerogear-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/aerogear-dev -- Matthias Wessendorf blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf twitter: http://twitter.com/mwessendorf _______________________________________________ aerogear-dev mailing list -- Matthias Wessendorf blog: http://matthiaswessendorf.word<http://matthiaswessendorf.wordpress.com...

On Wed, Jul 10, 2013 at 09:41:48AM +0200, Sebastien Blanc wrote: > I discussed a bit with Matzew and we agreed that we should probably remove > the AsyncHttpClient implementation and just keep the resteasy client, wdyt > ? Will resteasy client work fine outside of EAP/Wildfly?

-- qmx _______________________________________________ aerogear-dev mailing list aerogear-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/aerogear-dev

Ok just an update : RestEasy client seems to be a mess when beeing used inside an app deployed on a JBoss AS app , pretty lame :/ See this thread : http://sourceforge.net/mailarchive/forum.php?thread_name=51DE9678.9060103... Having to modify your module settings on AS level just to have to use the JAva Sender API is for me is total no go .... For this reason I plan to swtich back to the Http Async Client impl, wdyt ? Seb On Wed, Jul 10, 2013 at 5:00 PM, Matthias Wessendorf <matzew(a)apache.org&gt;wrote: > > > > On Wed, Jul 10, 2013 at 4:45 PM, Douglas Campos <qmx(a)qmx.me&gt; wrote: > >> On Wed, Jul 10, 2013 at 09:41:48AM +0200, Sebastien Blanc wrote: >> > I discussed a bit with Matzew and we agreed that we should probably >> remove >> > the AsyncHttpClient implementation and just keep the resteasy client, >> wdyt >> > ? >> Will resteasy client work fine outside of EAP/Wildfly? >> > > yep - works on JavaSE. It's basically a wrapper around Apache Http Client. > > On the Unit test (for the java sender) I was using it to send messages > from there (before we added the mocking) > > >> >> -- >> qmx >> _______________________________________________ >> aerogear-dev mailing list >> aerogear-dev(a)lists.jboss.org >> https://lists.jboss.org/mailman/listinfo/aerogear-dev >> > > > > -- > Matthias Wessendorf > > blog: http://matthiaswessendorf.wordpress.com/ > sessions: http://www.slideshare.net/mwessendorf > twitter: http://twitter.com/mwessendorf > > _______________________________________________ > aerogear-dev mailing list > aerogear-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/aerogear-dev > _______________________________________________ aerogear-dev mailing list aerogear-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/aerogear-dev