I just wanted to contribute my changes to the blackdown-jdk package. I'll attach some files to the bug report...
Created attachment 77577 [details, diff] attach the patch for the latest development version of the jdk These are the lines I added to the ebuild script...
Created attachment 77578 [details] rmiregistry init script This is the init script that goes in /etc/init.d/rmiregistry
Created attachment 77579 [details] rmiregistry init conf This is the init conf for /etc/init.d/rmiregistry which goes in /etc/conf.d/rmiregistry
That's it for now. I may add an "example" rmi startup script later. Perhaps we could install it in /etc/init.d/rmi.example
Oh, and the rmiregistry conf should perhaps set the user to "rmi". I don't know how ebuilds can add users and groups to the system, so I never bothered for now, I just run it as myself. Perhaps someone else might want to give it a try, and I can test it?
I think this would actually make sense as a separate ebuild, considering that more than one JDK provide rmiregistry. I would imagine it would simply RDEPEND on virtual/jdk, and just install the init script and env.d file.
That's what I was going to do originally, but then I was wondering if it might be better to include it in java because all JDKs have the rmiregistry. But, why don't all the guys responsible for java on gentoo discuss it, and let me know. If you want it as another package, then either create it yourself, or let me know, and I can do it for you, and submit it back. Thanks. (In reply to comment #6) > I think this would actually make sense as a separate ebuild, considering that > more than one JDK provide rmiregistry. > > I would imagine it would simply RDEPEND on virtual/jdk, and just install the > init script and env.d file. >
Ok going to be harsh with this. 1) No application has needed this yet almost 5 years. 2) I believe that rmiregistry really should be managed per application. 3) Don't even want to consider the possibilities of different jvm versions using a system rmiregisty.
