Re: findjava requirement (was: 4. RfD for a new debian java policy)
--- Jan Schulz <jasc.usenet@gmx.de> wrote:
> >> The 'findjava' algo was described in one of my mail.
> >Hmm, I remember seeing it but I don't remember anyone suggesting that it
> >be made compulsory.
>
> Dalibor argued quire havily for it :)
Yes, I did. I think if debian had a java runtime selection system that
considered the interestst of administrators, users, developers and packagers
and worked in a way that would be acceptable to all here, then it might as well
be made compulsory, so you'd be able to pick your java runtime the same for all
debian java packages. This relies on packages using findjava to find their VMs,
of course. If a package uses its own scheme, then I can set environment
veriables till I'm blue in the face and nothing will happen ;)
> BTW, the last open topic I still have is 'include' and
> 'include/something'. How should that dealted with? Put everything into
> one dir, as IBM does and patch around? Use a java-config setting?
Hm, this is another 'stupid by design' issue. Sun has specified jni.h but
hasn't specified the contents of jni_md.h. So it shouldn't be needed to build
JNI apps. But Sun's jni.h apparently includes jni_md.h without specifying its
search path. Unfortunately, jni_md.h seems to be stuck in a OS-specific
directory, so Sun wants you to stuck the OS-specific directory to your
compiler's header file seach path.
So I wouldn't mandate that a VM has a include/something dir. Instead the policy
shoudl advise programs building with JNI to include both $JAVA_HOME/include/
and $JAVA_HOME/include/linux if the build process is supported with sun's JDK.
cheers,
dalibor topic
__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.yahoo.com
Reply to: