Hi, I've started trying to use various examples about JEE5/JEE6 which all seem to use mappedname as Glassfish, Weblogic and recently Websphere is using it (as a name that can be used at both declaration and resource lookup time within the same module) I know that mappendname is left unspecified in the spec, but it seems most other JEE vendors have aligned (or at least seem to have aligned). >From what I can understand and see JBoss EJB3 have had different implementations of this over time and in AS 7 have moved to simply completely ignore the mappedname attribute. Does there exist any forum threads and documentation (beyond the that outlines the reasons why JBoss EJB3 isn't aligning with what seems to be the consensus across the various servers ? I'm asking since it would make not only users but also tool developers life much easier if one could actually write and generate almost portable code ;) /max http://about.me/maxandersen _______________________________________________ jboss-as7-dev mailing list jboss-as7-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/jboss-as7-dev

... With AS7-900 i'm now even more confused what will actually work/not work in AS 7 :)

/max > > On Tue, May 31, 2011 at 1:17 PM, Max Rydahl Andersen<max.andersen(a)redhat.com> wrote: > Hi, > > I've started trying to use various examples about JEE5/JEE6 which all seem to use mappedname as Glassfish, Weblogic and recently Websphere is using it (as a name that can be used at both declaration and resource lookup time within the same module) > > I know that mappendname is left unspecified in the spec, but it seems most other JEE vendors have aligned (or at least seem to have aligned). > > > From what I can understand and see JBoss EJB3 have had different implementations of this over time and in AS 7 have moved to simply completely ignore the mappedname attribute. > > Does there exist any forum threads and documentation (beyond the that outlines > the reasons why JBoss EJB3 isn't aligning with what seems to be the consensus across the various servers ? > > I'm asking since it would make not only users but also tool developers life much easier if one could actually write and generate almost portable code ;) > > /max > http://about.me/maxandersen > > > > > _______________________________________________ > jboss-as7-dev mailing list > jboss-as7-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/jboss-as7-dev > > > > -- > --- > Nik /max http://about.me/maxandersen _______________________________________________ jboss-as7-dev mailing list jboss-as7-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/jboss-as7-dev

> From what I can understand and see JBoss EJB3 have had different implementations of this over time and in AS 7 have moved to simply completely ignore the mappedname attribute.

Does there exist any forum threads and documentation (beyond the that outlines the reasons why JBoss EJB3 isn't aligning with what seems to be the consensus across the various servers ? I'm asking since it would make not only users but also tool developers life much easier if one could actually write and generate almost portable code ;) /max http://about.me/maxandersen _______________________________________________ jboss-as7-dev mailing list jboss-as7-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/jboss-as7-dev

The test is wrong. mappedName is a bit of a bummer. Basically if I specify @Stateless(mappedName="A") and @EJB(mappedName="A") they should match up via the interface of the @EJB as view. The registry that is used for mappedName is left unspecified. Carlo On 05/31/2011 12:24 PM, Jaikiran Pai wrote: > On Tuesday 31 May 2011 03:47 PM, Max Rydahl Andersen wrote: >> >>> From what I can understand and see JBoss EJB3 have had different implementations of this over time and in AS 7 have moved to simply completely ignore the mappedname attribute. > mappedName on @Stateless, @Stateful and @MessageDriven has always been > ignored by JBoss EJB3 and isn't something new in AS7. > > mappedName on @Resource and @EJB is expected to work in AS7 (and > previous versions). If this is currently broken in AS7 upstream, then it > is a bug. I don't think it's broken because we have a testcase for > exactly this specific attribute > https://github.com/jbossas/jboss-as/tree/master/testsuite2/spec/src/test/.... > If there's some specific application running into a problem, then please > point me to that. > > -Jaikiran >> Does there exist any forum threads and documentation (beyond the that outlines >> the reasons why JBoss EJB3 isn't aligning with what seems to be the consensus across the various servers ? >> >> I'm asking since it would make not only users but also tool developers life much easier if one could actually write and generate almost portable code ;) >> >> /max >> http://about.me/maxandersen >> >> >> >> >> _______________________________________________ >> jboss-as7-dev mailing list >> jboss-as7-dev(a)lists.jboss.org >> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev > _______________________________________________ > jboss-as7-dev mailing list > jboss-as7-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/jboss-as7-dev _______________________________________________ jboss-as7-dev mailing list jboss-as7-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/jboss-as7-dev

> The point of the story here is don't use mappedName. It's not portable, > and it's essentially deprecated. Tell that to all the examples, tutorials and IDE's out there ;) /max > > On 5/31/11 9:04 AM, Carlo de Wolf wrote: >> The test is wrong. mappedName is a bit of a bummer. >> >> Basically if I specify @Stateless(mappedName="A") and >> @EJB(mappedName="A") they should match up via the interface of the @EJB >> as view. The registry that is used for mappedName is left unspecified. >> >> Carlo >> >> On 05/31/2011 12:24 PM, Jaikiran Pai wrote: >>> On Tuesday 31 May 2011 03:47 PM, Max Rydahl Andersen wrote: >>>> >>>>> From what I can understand and see JBoss EJB3 have had different implementations of this over time and in AS 7 have moved to simply completely ignore the mappedname attribute. >>> mappedName on @Stateless, @Stateful and @MessageDriven has always been >>> ignored by JBoss EJB3 and isn't something new in AS7. >>> >>> mappedName on @Resource and @EJB is expected to work in AS7 (and >>> previous versions). If this is currently broken in AS7 upstream, then it >>> is a bug. I don't think it's broken because we have a testcase for >>> exactly this specific attribute >>> https://github.com/jbossas/jboss-as/tree/master/testsuite2/spec/src/test/.... >>> If there's some specific application running into a problem, then please >>> point me to that. >>> >>> -Jaikiran >>>> Does there exist any forum threads and documentation (beyond the that outlines >>>> the reasons why JBoss EJB3 isn't aligning with what seems to be the consensus across the various servers ? >>>> >>>> I'm asking since it would make not only users but also tool developers life much easier if one could actually write and generate almost portable code ;) >>>> >>>> /max >>>> http://about.me/maxandersen >>>> >>>> >>>> >>>> >>>> _______________________________________________ >>>> jboss-as7-dev mailing list >>>> jboss-as7-dev(a)lists.jboss.org >>>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev >>> _______________________________________________ >>> jboss-as7-dev mailing list >>> jboss-as7-dev(a)lists.jboss.org >>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev >> >> _______________________________________________ >> jboss-as7-dev mailing list >> jboss-as7-dev(a)lists.jboss.org >> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev > > > -- > Jason T. Greene > JBoss, a division of Red Hat > _______________________________________________ > jboss-as7-dev mailing list > jboss-as7-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/jboss-as7-dev /max http://about.me/maxandersen

Hell even the javadoc spells this out: "A product specific name(for example, global JNDI name) that this session bean should be mapped to. Application servers are not required to support any particular form or type of mapped name, nor the ability to use mapped names. The mapped name is product-dependent and often installation-dependent. No use of a mapped name is portable." On 5/31/11 10:25 AM, Jason T. Greene wrote: > I can't do anything about that. No matter what we implement it will > still not be portable. > > On 5/31/11 10:23 AM, Max Rydahl Andersen wrote: >> >>> The point of the story here is don't use mappedName. It's not portable, >>> and it's essentially deprecated. >> >> Tell that to all the examples, tutorials and IDE's out there ;) >> >> /max >> >>> >>> On 5/31/11 9:04 AM, Carlo de Wolf wrote: >>>> The test is wrong. mappedName is a bit of a bummer. >>>> >>>> Basically if I specify @Stateless(mappedName="A") and >>>> @EJB(mappedName="A") they should match up via the interface of the @EJB >>>> as view. The registry that is used for mappedName is left unspecified. >>>> >>>> Carlo >>>> >>>> On 05/31/2011 12:24 PM, Jaikiran Pai wrote: >>>>> On Tuesday 31 May 2011 03:47 PM, Max Rydahl Andersen wrote: >>>>>> >>>>>>> From what I can understand and see JBoss EJB3 have had different implementations of this over time and in AS 7 have moved to simply completely ignore the mappedname attribute. >>>>> mappedName on @Stateless, @Stateful and @MessageDriven has always been >>>>> ignored by JBoss EJB3 and isn't something new in AS7. >>>>> >>>>> mappedName on @Resource and @EJB is expected to work in AS7 (and >>>>> previous versions). If this is currently broken in AS7 upstream, then it >>>>> is a bug. I don't think it's broken because we have a testcase for >>>>> exactly this specific attribute >>>>> https://github.com/jbossas/jboss-as/tree/master/testsuite2/spec/src/test/.... >>>>> If there's some specific application running into a problem, then please >>>>> point me to that. >>>>> >>>>> -Jaikiran >>>>>> Does there exist any forum threads and documentation (beyond the that outlines >>>>>> the reasons why JBoss EJB3 isn't aligning with what seems to be the consensus across the various servers ? >>>>>> >>>>>> I'm asking since it would make not only users but also tool developers life much easier if one could actually write and generate almost portable code ;) >>>>>> >>>>>> /max >>>>>> http://about.me/maxandersen >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> jboss-as7-dev mailing list >>>>>> jboss-as7-dev(a)lists.jboss.org >>>>>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev >>>>> _______________________________________________ >>>>> jboss-as7-dev mailing list >>>>> jboss-as7-dev(a)lists.jboss.org >>>>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev >>>> >>>> _______________________________________________ >>>> jboss-as7-dev mailing list >>>> jboss-as7-dev(a)lists.jboss.org >>>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev >>> >>> >>> -- >>> Jason T. Greene >>> JBoss, a division of Red Hat >>> _______________________________________________ >>> jboss-as7-dev mailing list >>> jboss-as7-dev(a)lists.jboss.org >>> https://lists.jboss.org/mailman/listinfo/jboss-as7-dev >> >> /max >> http://about.me/maxandersen >> >> >> > > -- Jason T. Greene JBoss, a division of Red Hat

The test is wrong. mappedName is a bit of a bummer. Basically if I specify @Stateless(mappedName="A") and @EJB(mappedName="A") they should match up via the interface of the @EJB as view. The registry that is used for mappedName is left unspecified.