From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 14225 invoked by alias); 12 Sep 2011 00:54:06 -0000 Received: (qmail 14215 invoked by uid 22791); 12 Sep 2011 00:54:04 -0000 X-SWARE-Spam-Status: No, hits=-1.7 required=5.0 tests=AWL,BAYES_00,TW_GC X-Spam-Check-By: sourceware.org Received: from bfs011.mtl1.boxfabric.com (HELO bfs011.mtl1.boxfabric.com) (74.117.141.19) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Mon, 12 Sep 2011 00:53:49 +0000 Received: (qmail 9005 invoked by uid 399); 12 Sep 2011 00:53:48 -0000 Received: from unknown (HELO ?173.35.101.49?) (mark@mark.mielke.cc@173.35.101.49) by bfs011.mtl1.boxfabric.com with ESMTPAM; 12 Sep 2011 00:53:48 -0000 X-Sender: mark@mark.mielke.cc Message-ID: <4E6D581B.5050801@mark.mielke.cc> Date: Mon, 12 Sep 2011 00:54:00 -0000 From: Mark Mielke User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:6.0.2) Gecko/20110906 Thunderbird/6.0.2 MIME-Version: 1.0 To: Alan Eliasen CC: java@gcc.gnu.org Subject: Re: Why GCJ won't load my plugin when compiling bytecode? References: <4E6D571C.8000903@mindspring.com> In-Reply-To: <4E6D571C.8000903@mindspring.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-IsSubscribed: yes Mailing-List: contact java-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: java-owner@gcc.gnu.org X-SW-Source: 2011-09/txt/msg00007.txt.bz2 On 09/11/2011 08:49 PM, Alan Eliasen wrote: > On 09/09/2011 04:24 PM, Bryce McKinlay wrote: >> Modern versions of gcj only compile .class files to object code. gcj >> doesn't directly compile Java source code, nor generate bytecode. >> (early versions of gcj did do these things, but it was changed years >> ago) > > Is this a permanent decision? Is there any intention to make > gcj compile from java source code again? > > Can anyone give me pointers to where this was decided upon and > announced? I'm curious to see the discussion. > Doesn't seem like years but maybe... Java 5. Support for generics and such. Continue enhancing an alternative compiler with limited contributors, or use of that is basically guaranteed to be supported (Eclipse's compiler). I wasn't part of the decision - only an observer - but it seemed like an excellent decision when it was made, and I don't see why it isn't an excellent decision still. Cheers, -- Mark Mielke