Apache OpenOffice (AOO) Bugzilla – Issue 16294
UTIL: gcc 3.3.1 compile error
Last modified: 2003-07-31 02:28:47 UTC
gcc is tightening the noose on standards compliance once again: -o ../../unxlngi4.pro/slo/accessiblewrapper.o /data2/office/comphelper/source/misc/accessiblewrapper.cxx /data2/office/comphelper/source/misc/accessiblewrapper.cxx: In member function `virtual comphelper::OAccessibleContextWrapper* comphelper::OAccessibleWrapper::createAccessibleContext(const com::sun::star::uno::Reference<com::sun::star::accessibility::XAccessibleContext>&)': /data2/office/comphelper/source/misc/accessiblewrapper.cxx:373: error: cannot call member function `const com::sun::star::uno::Reference<com::sun::star::lang::XMultiServiceFactory>& comphelper::OProxyAggregation::getORB()' without object /data2/office/comphelper/source/misc/accessiblewrapper.cxx: In constructor `comphelper::OAccessibleContextWrapperHelper::OAccessibleContextWrapperHelper(const com::sun::star::uno::Reference<com::sun::star::lang::XMultiServiceFactory>&, cppu::OBroadcastHelper&, const com::sun::star::uno::Reference<com::sun::star::accessibility::XAccessibleContext>&, const com::sun::star::uno::Reference<com::sun::star::accessibility::XAccessible>&, const com::sun::star::uno::Reference<com::sun::star::accessibility::XAccessible>&)': /data2/office/comphelper/source/misc/accessiblewrapper.cxx:413: error: cannot call member function `const com::sun::star::uno::Reference<com::sun::star::lang::XMultiServiceFactory>& comphelper::OProxyAggregation::getORB()' without object dmake: Error code 1, while making '../../unxlngi4.pro/slo/accessiblewrapper.obj' ---* TG_SLO.MK *--- ERROR: Error 65280 occurred while making /data2/office/comphelper/source/misc dmake: Error code 1, while making 'build_all'
Reassigned.
It's your class.
Created attachment 7534 [details] perhaps a fix
Ken, as I do not have access to a gcc 3.3 environment, could you please try if the attached diff file, applied to comphelper/inc/comphelper/proxyaggregation.hxx, fixes the problem?
That patch worked. Making: ../../unxlngi4.pro/slo/accessiblewrapper.obj ccache g++-3.3 -fmessage-length=0 -c -I. -I. -I../inc -I../../inc -I../../unx/inc -I../../unxlngi4.pro/inc -I. -I/data2/office/solver/645/unxlngi4.pro/inc/stl -I/data2/office/solver/645/unxlngi4.pro/inc/external -I/data2/office/solver/645/unxlngi4.pro/inc -I/data2/office/solenv/unxlngi4/inc -I/data2/office/solenv/inc -I/data2/office/res -I/data2/office/solver/645/unxlngi4.pro/inc/stl -I/data2/office/solenv/inc/Xp31 -I/opt/jdk/include -I/opt/jdk/include/linux -I/opt/jdk/include/native_threads/include -I/usr/X11R6/include -I. -I../../res -I. -O3 -pipe -mcpu=pentiumpro -fno-for-scope -fpermissive -fno-rtti -include preinclude.h -fexceptions -fno-enforce-eh-specs -fpic -DLINUX -DUNX -DVCL -DGCC -DC300 -DINTEL -DCVER=C300 -D_USE_NAMESPACE -DGLIBC=2 -DX86 -D_PTHREADS -D_REENTRANT -DNEW_SOLAR -D_USE_NAMESPACE=1 -DSTLPORT_VERSION=400 -D__DMAKE -DUNIX -DCPPU_ENV=gcc3 -DSUPD=645 -DBUILD=8639 -DPRODUCT -DNDEBUG -DPRODUCT_FULL -DOSL_DEBUG_LEVEL=0 -DOPTIMIZE -DEXCEPTIONS_ON -DCUI -DSOLAR_JAVA -DSRX645 -DSHAREDLIB -D_DLL_ -DMULTITHREAD -o ../../unxlngi4.pro/slo/accessiblewrapper.o /data2/office/comphelper/source/misc/accessiblewrapper.cxx if ( -e ../../unxlngi4.pro/slo/accessiblewrapper.o) touch ../../unxlngi4.pro/slo/accessiblewrapper.obj
Next break after that: Making: ../../unxlngi4.pro/slo/builddata.obj ccache g++-3.3 -fmessage-length=0 -c -I. -I. -I../inc -I../../source/inc -I../../inc -I../../unx/inc -I../../unxlngi4.pro/inc -I. -I/data2/office/solver/645/unxlngi4.pro/inc/stl -I/data2/office/solver/645/unxlngi4.pro/inc/external -I/data2/office/solver/645/unxlngi4.pro/inc -I/data2/office/solenv/unxlngi4/inc -I/data2/office/solenv/inc -I/data2/office/res -I/data2/office/solver/645/unxlngi4.pro/inc/stl -I/data2/office/solenv/inc/Xp31 -I/opt/jdk/include -I/opt/jdk/include/linux -I/opt/jdk/include/native_threads/include -I/usr/X11R6/include -I. -I../../res -I. -O3 -pipe -mcpu=pentiumpro -fno-for-scope -fpermissive -fno-rtti -include preinclude.h -fexceptions -fno-enforce-eh-specs -fpic -DLINUX -DUNX -DVCL -DGCC -DC300 -DINTEL -DCVER=C300 -D_USE_NAMESPACE -DGLIBC=2 -DX86 -D_PTHREADS -D_REENTRANT -DNEW_SOLAR -D_USE_NAMESPACE=1 -DSTLPORT_VERSION=400 -D__DMAKE -DUNIX -DCPPU_ENV=gcc3 -DSUPD=645 -DBUILD=8639 -DPRODUCT -DNDEBUG -DPRODUCT_FULL -DOSL_DEBUG_LEVEL=0 -DOPTIMIZE -DEXCEPTIONS_ON -DCUI -DSOLAR_JAVA -DSRX645 -DSHAREDLIB -D_DLL_ -DMULTITHREAD -o ../../unxlngi4.pro/slo/builddata.o /data2/office/configmgr/source/tree/builddata.cxx /data2/office/configmgr/source/tree/builddata.cxx: In member function `configmgr::data::TreeAddress configmgr::data::TreeNodeBuilder::allocTreeFragment(configmgr::memory::UpdateAccessor&)': /data2/office/configmgr/source/tree/builddata.cxx:656: error: invalid static_cast from type `Address' to type `configmgr::memory::Pointer' dmake: Error code 1, while making '../../unxlngi4.pro/slo/builddata.obj' ---* TG_SLO.MK *--- ERROR: Error 65280 occurred while making /data2/office/configmgr/source/tree
Ken, if you want to take ownership of the issue, I can supply a patch. If you want code owners to fix their code for each breakage please create a separate issue and assign to me. You might then consider to create an umbrella issue for all 3.3.1 problems. BTW: It appears that this is the same problem that has already occurred in autodoc. That would mean it is (IMHO) a gcc regression. Is 3.3.1 released ? Do we really want to take it upon ourselves to build with all new gcc releases, even if they have such regressions ? In the past we have rejected some versions of some compilers, if they had regressions that broke the build in <too many> places. BTW2: If you want to discover more breakages early, you could try to see how far you get with 'build -i' (simply ignoring errors as long as possible). If you have a prefilled (by an earlier gcc) solver you might even succeed and can still grep for errors in your build log.
Ken, glad to hear that the fix for comphelper worked. I have to agree to Joerg that it would be best to have separate issues for different breaks, this makes tracking and working much easier. I assign this one back to you, and suggest that you mark it as FIXED, plus submit a new issue for the configmgr problem. Thanks.
Patch verified, not applied to RC3 as far as I can see. Separate issue rasied for other problem.
This patch should no longer be required. Awaiting fresh version of gcc to be uploaded to debian unstable to test properly. http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11568
Compiler bug, now resolved.
Closing