[pylucene-dev] Re: using JCC

Andi Vajda vajda at osafoundation.org
Wed Feb 20 14:41:21 PST 2008


On Wed, 20 Feb 2008, Bill Janssen wrote:

>>    find / -name 'jni.h' -print
>
> That process is still running...  Lots of disk on that machine.
>
> I have a hypothesis: if you specify "--package java.lang", header
> files are generated, which are found before the gcj header files, so
> things work.  If you don't, the bad gcj header files are found.  I'll
> try it and report back.

It is scary to think that inspite of your include flags in setup.py being 
correct, you're still picking up the wrong header files because gcj is the 
default java VM. Once you solve that, maybe the next problem is going to be 
that you're linking against the wrong libraries.

The fix should be simple: do not have your system configured with gcj as the 
default VM. By configuring it right, you'll have the VM of your choice be 
all in the right location, header files and libraries. This is how it works 
on Ubuntu, at least.

Taking a closer look at the error you sent me earlier, it seems to me that 
the correct jni.h file is picking up the wrong array.h file. So the bug is 
clearly not in JCC but either in your system's setup or in jdk 1.6's jni.h 
file.

That big find statement you're running is probably a waste of time. Get the 
gcj stuff out of the way, that is, out of the gcc innards. I'm hoping there 
is a command like the one on Ubuntu, update-java-alternative, that does this 
cleanly. Otherwise, you might want to consider reinstalling gcc by excluding 
gcj.

Andi..


More information about the pylucene-dev mailing list