[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Contributions of work on Jikes RVM with GNU Classpath
From: |
Tom Tromey |
Subject: |
Re: Contributions of work on Jikes RVM with GNU Classpath |
Date: |
21 Nov 2002 18:50:42 -0700 |
User-agent: |
Gnus/5.09 (Gnus v5.9.0) Emacs/21.2 |
>>>>> "Mark" == Mark Wielaard <address@hidden> writes:
>> On Thu, 2002-11-21 at 17:19, Julian Dolby wrote:
Mark> Impressive! I really look forward to running Eclipse using a
Mark> free VM and the GNU Classpath libraries.
Me too. Now we just have to shake out the libgcj bugs to get it to
run there :-)
>> 3) The one change we needed to make that we cannot---for legal
>> reasons---easily contribute is to javax.naming; while running
>> Eclipse we came across incompletenesses in the GNU Classpath code
>> base, which we worked around by merging the javax.naming code from
>> GCJ into the GNU Classpath code base.
This is perfectly fine.
>> We do not want to suck the GCJ code base into IBM and then send
>> it back to GNU Classpath. This would be highly problematic for us
>> legally. Ideally, someone at GNU Classpath can merge these code
>> bases independently for us.
I don't understand the legal problem here. In any case it doesn't
matter. The "merge" in this case is just a `cp'; anybody with commit
access can do it. I haven't since I've been hoping Mark would write
the javadoc, as he points out...
>> We would be happy to contribute our changes in any manner people at
>> GNU Classpath find convenient. We can send you diffs in any format or
>> simply a tarball of the GNU Classpath tree we have here. Please let
>> me know what you would find most useful.
Mark> It is probably a good idea to make a new release of GNU
Mark> Classpath first (say in the next two weeks). So that we have a
Mark> clear point from which to start merging in the changes.
Are the changes made against a snapshot of the tree without the CVS
directories? I think the very best way to get the changes would be a
sequence of `cvs diff's, one per logical change.
Mark> The best way to do the actual merge/patches is by trying to
Mark> identify independent changes and post them on the classpath
Mark> mailinglist so people can discuss them. Just sending a big
Mark> tarball or one gaint patch/diff makes it hard to analyse what
Mark> changed and for what reason.
Yup.
Mark> Brian Jones can also give one of you CVS write access to the GNU
Mark> Classpath tree if you want.
I'd like to see this too. That way once a patch has been discussed
someone from the JRVM team can just check it in.
Tom