well it is exactly the expected case (well guessing but AFAIK it should): the archive was not complete and the app was expecting the finder to leak outside the archive.
I expect to handle it in 4.1 if needed (ie the app can't create the archive correctly cause it ignores totally what it scans) Romain Manni-Bucau Twitter: @rmannibucau Blog: http://rmannibucau.wordpress.com/ LinkedIn: http://fr.linkedin.com/in/rmannibucau Github: https://github.com/rmannibucau 2014-08-23 18:40 GMT+02:00 Mark Struberg <[email protected]>: > Is there an easy way to fix those OSGi issues? If so +0, if there is no > _easy_ way then +1 and let's fix this later. > > LieGrue, > strub > > > On Friday, 22 August 2014, 15:03, Jean-Baptiste Onofré <[email protected]> > wrote: > > > > +0 (non binding) > > I have some issue on OSGi (but I think it was expected ;)). > > Regards > JB > > On 08/20/2014 07:23 PM, Romain Manni-Bucau wrote: >> Hi, >> >> As said in a mail last week I'm starting a vote for xbean 4.0 release. >> >> The main changes are: >> 1) force AnnotationFinder to respect archive in its scanning (don't >> leak in external classes) >> 1bis) skip java.* classes since we'll not get their bytecode for sure >> (protected method if needed) >> 2) Remove asynchronous finder which was really specific and can lead >> to some classloading issue pretty quickly on recent JVMs >> >> Binaries: >> https://repository.apache.org/content/repositories/orgapachegeronimo-1003/ >> >> Tag: >> http://svn.apache.org/repos/asf/geronimo/xbean/tags/xbean-4.0/ >> >> Vote open 72h. >> >> [ ] +1 release this >> [ ] 0 don't care >> [ ] -1 don't release this (please explain) >> >> >> Here is my +1 >> >> >> >> Romain Manni-Bucau >> Twitter: @rmannibucau >> Blog: http://rmannibucau.wordpress.com/ >> LinkedIn: http://fr.linkedin.com/in/rmannibucau >> Github: https://github.com/rmannibucau > >> > > -- > Jean-Baptiste Onofré > [email protected] > http://blog.nanthrax.net > Talend - http://www.talend.com >
