From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 9821 invoked by alias); 13 Mar 2010 14:24:39 -0000 Received: (qmail 9812 invoked by uid 22791); 13 Mar 2010 14:24:38 -0000 X-SWARE-Spam-Status: No, hits=-3.2 required=5.0 tests=AWL,BAYES_00,RCVD_IN_DNSWL_LOW X-Spam-Check-By: sourceware.org Received: from adelie.canonical.com (HELO adelie.canonical.com) (91.189.90.139) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Sat, 13 Mar 2010 14:24:32 +0000 Received: from hutte.canonical.com ([91.189.90.181]) by adelie.canonical.com with esmtp (Exim 4.69 #1 (Debian)) id 1NqSGT-0005AL-Vt for ; Sat, 13 Mar 2010 14:24:30 +0000 Received: from dslb-088-074-004-223.pools.arcor-ip.net ([88.74.4.223] helo=[192.168.42.17]) by hutte.canonical.com with esmtpsa (TLS-1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.69) (envelope-from ) id 1NqSGT-0005xc-SS for java@gcc.gnu.org; Sat, 13 Mar 2010 14:24:29 +0000 Message-ID: <4B9BA018.5000205@ubuntu.com> Date: Sat, 13 Mar 2010 14:24:00 -0000 From: Matthias Klose User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1.9pre) Gecko/20100217 Shredder/3.0.3pre MIME-Version: 1.0 To: java@gcc.gnu.org Subject: Re: unable to regenerate the pregenerated libjava header files on the trunk References: <4B894E9A.4000903@ubuntu.com> <4B8A4453.4030606@redhat.com> In-Reply-To: <4B8A4453.4030606@redhat.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: 2010-03/txt/msg00014.txt.bz2 On 28.02.2010 11:24, Andrew Haley wrote: > On 02/27/2010 04:55 PM, Matthias Klose wrote: >> this is http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42676, now >> downgraded to P4. At least for Debian, this would mean not shipping gcj >> anymore in the main section of the archive (cannot be built from >> source). Is this a restriction for the upstream release as well? Can >> this be reproduced by others? > > I think you're bootstrapping with an obsolete version of gjavah. > > To see if I'm right, generate the header file for > javax/xml/xpath/XPathFunctionResolver. Like this: > > $ gjavah -cni -force -classpath/x86_64-unknown-linux-gnu/libjava/libgcj-4.5.0.jar javax/xml/xpath/XPathFunctionResolver > > With the new version of gjavah, you'll get > > namespace xml > { > namespace namespace$ > > With the old version you'll get > > namespace xml > { > namespace namespace > > Andrew. with a current build, I didn't see these changes anymore. A few differences in the header files for inner classes are still present, of the form --- javax/net/ssl/TrustManagerFactory$1.h (revision 157407) +++ javax/net/ssl/TrustManagerFactory$1.h (working copy) @@ -27,7 +27,7 @@ public: // actually package-private TrustManagerFactory$1(); public: - ::java::lang::Object * run(); + virtual ::java::lang::Object * run(); static ::java::lang::Class class$; }; These are seen with the updated ecj-latest.jar (based on eclipse-3.5.1). 2010-03-13 Matthias Klose PR libgcj/42676 * Regenerate .h files for anonymous inner classes. Checking in, approved on irc by Andrew. Complete diff in the bug report. Matthias