[Libreoffice-commits] core.git: Branch 'libreoffice-5-2' - sal/osl
sal/osl/unx/process.cxx |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
New commits:
commit b09eb27b7a52d2abd9bdbf01b4532f23279c6edc
Author: Richard PALO <rich...@netbsd.org>
Date: Sun May 29 08:16:28 2016 +0200
tdf#100021 - format argument issue in process.cxx
avoid warning whether pid_t is long or int
Change-Id: I533175b53166cb233169902d277906f5ce979421
Signed-off-by: Richard PALO <rich...@netbsd.org>
Reviewed-on: https://gerrit.libreoffice.org/25593
Reviewed-by: Michael Stahl <mst...@redhat.com>
Tested-by: Michael Stahl <mst...@redhat.com>
(cherry picked from commit acbbb01940588604de965c4fcef64b2e17473fe6)
Reviewed-on: https://gerrit.libreoffice.org/25979
diff --git a/sal/osl/unx/process.cxx b/sal/osl/unx/process.cxx
index 4d372a3..9a9ba28 100644
--- a/sal/osl/unx/process.cxx
+++ b/sal/osl/unx/process.cxx
@@ -1023,7 +1023,7 @@ oslProcessError SAL_CALL osl_getProcessInfo(oslProcess
Process, oslProcessData F
int fd;
sal_Char name[PATH_MAX + 1];
-snprintf(name, sizeof(name), "/proc/%u", pid);
+snprintf(name, sizeof(name), "/proc/%ld", (long)pid);
if ((fd = open(name, O_RDONLY)) >= 0)
{
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] core.git: sal/osl
sal/osl/unx/process.cxx |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
New commits:
commit acbbb01940588604de965c4fcef64b2e17473fe6
Author: Richard PALO <rich...@netbsd.org>
Date: Sun May 29 08:16:28 2016 +0200
tdf#100021 - format argument issue in process.cxx
avoid warning whether pid_t is long or int
Change-Id: I533175b53166cb233169902d277906f5ce979421
Signed-off-by: Richard PALO <rich...@netbsd.org>
Reviewed-on: https://gerrit.libreoffice.org/25593
Reviewed-by: Michael Stahl <mst...@redhat.com>
Tested-by: Michael Stahl <mst...@redhat.com>
diff --git a/sal/osl/unx/process.cxx b/sal/osl/unx/process.cxx
index 4d372a3..9a9ba28 100644
--- a/sal/osl/unx/process.cxx
+++ b/sal/osl/unx/process.cxx
@@ -1023,7 +1023,7 @@ oslProcessError SAL_CALL osl_getProcessInfo(oslProcess
Process, oslProcessData F
int fd;
sal_Char name[PATH_MAX + 1];
-snprintf(name, sizeof(name), "/proc/%u", pid);
+snprintf(name, sizeof(name), "/proc/%ld", (long)pid);
if ((fd = open(name, O_RDONLY)) >= 0)
{
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: [libreoffice-l10n] Invalid Entry in Po Files
Le 24/05/16 12:37, Christian Lohmaier a écrit : >>> 41 invalid nplurals value >>> 41 invalid plural expression > > Indeed. > > 41 "Plural-Forms: nplurals=INTEGER; plural=EXPRESSION;\n" > (all in be project) > >>> 82 n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2);\n" > > that line seems to be missing something, likely bug in your parser. > Anyway, LO doesn't really make use of plural forms.. > > $ git grep msgid_plural > $ >... one full example > translations/source/be/shell/source/win32/shlxthandler/res.po:3: invalid > nplurals value > Try using > the following, valid for Belarusian: > > "Plural-Forms: nplurals=3; plural=(n%10==1 && n%100!=11 ? 0 : n%10>=2 && > n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2);\n" > translations/source/be/shell/source/win32/shlxthandler/res.po:3: invalid > plural expression > Try using > the following, valid for Belarusian: > > "Plural-Forms: nplurals=3; plural=(n%10==1 && n%100!=11 ? 0 : n%10>=2 && > n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2);\n" though 'win32' is probably not actually parsed in my build > > so no idea where you get the 82 lines only - it has a nonmatched ), so > that line is incomplete, but see above - LO doesn't make use of plural > forms... > I think it stems from 82 = 41 * 2 (for invalid nplurals value + invalid plural expression) > well - nothing really broken, e xcept maybe the plural expression, but > for that one needs the actual bug line > $ git grep -h 'n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : > 2);\\n"' |sort |uniq -c >1466 "Plural-Forms: nplurals=3; plural=(n%10==1 && n%100!=11 ? 0 : > n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2);\n" > 351 "Plural-Forms: nplurals=3; plural=(n==1 ? 0 : n%10>=2 && > n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2);\n" > > so no idea where you get the 82 lines only - it has a nonmatched ), so > that line is incomplete, but see above - LO doesn't make use of plural > forms... see above. so if I understand correctly, there is nothing to be done outside of intervention by the translation teams (or individuals) about these errors/warnings.. okay. thanks anyway for taking the time to explain. BTW, also noticed the following during the build: > Warning : Can't open > /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.3.2/translations/source/af/dictionaries/ca.po > Warning : Can't open > /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.3.2/translations/source/af/dictionaries/cs_CZ.po > Warning : Can't open > /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.3.2/translations/source/af/dictionaries/gd_GB.po they seem to be non-existent in git too. cheers -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [libreoffice-l10n] Invalid Entry in Po Files
Le 28/07/13 10:11, Yaron Shahrabani a écrit :
I'm noticing with 5.1.3.2 lots and lots of warnings (building with all
languages) such as:
> [build SRS] dbaccess/source/ui/misc/WizardPages.src
> Warning :
> /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.3.2/translations/source/da/dbaccess/source/ui/dlg.po
> contains invalid entry
> Warning :
> /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.3.2/translations/source/el/dbaccess/source/ui/dlg.po
> contains invalid entry
> Warning :
> /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.3.2/translations/source/eo/dbaccess/source/ui/dlg.po
> contains invalid entry
>...
For grins, tried this [g]msgfmt invocation to see what was up, and I see:
> $ LANG=C gfind translations/source -name '*.po' -exec gmsgfmt -vco /dev/null
> {} \; >& msgfmtlog.txt
> $ cat msgfmtlog.txt |grep -Ev 'translated|fatal error' |cut -d: -f3- |sort
> |uniq -c
> 125
>3 'msgid' and 'msgstr' entries do not both begin with '\n'
> 207 'msgid' and 'msgstr' entries do not both end with '\n'
> 41 invalid nplurals value
> 41 invalid plural expression
> 82 n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2);\n"
> 26640 warning: header field 'Language-Team' still has the initial default
> value
> 157 warning: header field 'Language' missing in header
>8 warning: header field 'Language' still has the initial default value
> 3745 warning: header field 'Last-Translator' still has the initial default
> value
> 3748 warning: header field 'PO-Revision-Date' still has the initial default
> value
> 27629 warning: header field 'Project-Id-Version' still has the initial
> default value
> 43 warning: PO file header fuzzy
hoping something can be done to get over the bulk of these.
--
Richard PALO
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Py_hash_t on python27
Le 22/01/16 16:47, Rene Engelhard a écrit : > Hi, > > On Fri, Jan 22, 2016 at 03:58:22PM +0100, Richard PALO wrote: >> Le 21/01/16 20:11, Richard PALO a écrit : >>>> Isn't this a python3 feature? >>>> >> >> looks like it was 3.2 according to the python website (in the release notes >> for 3.2) >> so I use instead: >>> -#if PY_VERSION_HEX < 0x0207 >>> +#if PY_VERSION_HEX < 0x0302 > > We need python >= 3.3 when using python3 anyway (see configure; > and that is because of python3-internal changes), so that is a > given. > ?? Python 3.2 and later define Py_hash_t already... This is, I believe, exclusively for usage with prior versions. -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Py_hash_t on python27
Le 22/01/16 17:35, Rene Engelhard a écrit :
> On Fri, Jan 22, 2016 at 05:26:05PM +0100, Richard PALO wrote:
>>>>> -#if PY_VERSION_HEX < 0x0207
>>>>> +#if PY_VERSION_HEX < 0x0302
>>>
>>> We need python >= 3.3 when using python3 anyway (see configure;
>>> and that is because of python3-internal changes), so that is a
>>> given.
>>>
>> ?? Python 3.2 and later define Py_hash_t already... This is, I believe,
>> exclusively for usage with prior versions.
>
> Whose are not supported. As said, PyUNO needs python >= 3.3 for python 3.
> See http://cgit.freedesktop.org/libreoffice/core/tree/configure.ac#n8071
>
> No idea whether python 2.7 is supposed to be still supported, don't think
> with 5.1s improved pyUNO.. I stopped building the extra pyUNO package for
> python 2 longish ago.
>
> Regards,
>
> Rene
>
>
We patched that long ago, and things seem to work okay (is there something in
particular to watch out for?)
because LO always seemed to pick up the wrong python...!
> --- configure.ac.orig 2016-01-13 01:56:20.0 +
> +++ configure.ac
> @@ -7997,7 +7997,7 @@ no|disable)
> else
> # Unset variables set by the above AM_PATH_PYTHON so that
> # we actually do check anew.
> -unset PYTHON am_cv_pathless_PYTHON ac_cv_path_PYTHON
> am_cv_python_version am_cv_python_platform am_cv_python_pythondir
> am_cv_python_pyexecdir
> +#unset PYTHON am_cv_pathless_PYTHON ac_cv_path_PYTHON
> am_cv_python_version am_cv_python_platform am_cv_python_pythondir
> am_cv_python_pyexecdir
> AM_PATH_PYTHON([3.3],, [:])
> if test "$PYTHON" = ":"; then
> if test -z "$PYTHON_FOR_BUILD"; then
> @@ -8050,9 +8050,9 @@ if test $enable_python = system; then
> elif test "$cross_compiling" != yes; then
> # Unset variables set by the above AM_PATH_PYTHON so that
> # we actually do check anew.
> -unset PYTHON am_cv_pathless_PYTHON ac_cv_path_PYTHON
> am_cv_python_version am_cv_python_platform am_cv_python_pythondir
> am_cv_python_pyexecdir
> +#unset PYTHON am_cv_pathless_PYTHON ac_cv_path_PYTHON
> am_cv_python_version am_cv_python_platform am_cv_python_pythondir
> am_cv_python_pyexecdir
> # This causes an error if no python command is found
> -AM_PATH_PYTHON([3.3])
> +AM_PATH_PYTHON([2.6])
> python_include=`$PYTHON -c "import distutils.sysconfig;
> print(distutils.sysconfig.get_config_var('INCLUDEPY'));"`
> python_version=`$PYTHON -c "import distutils.sysconfig;
> print(distutils.sysconfig.get_config_var('VERSION'));"`
> python_libs=`$PYTHON -c "import distutils.sysconfig;
> print(distutils.sysconfig.get_config_var('LIBS'));"`
--
Richard PALO
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Py_hash_t on python27
Le 21/01/16 20:11, Richard PALO a écrit : >> Isn't this a python3 feature? >> looks like it was 3.2 according to the python website (in the release notes for 3.2) so I use instead: > -#if PY_VERSION_HEX < 0x0207 > +#if PY_VERSION_HEX < 0x0302 -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Py_hash_t on python27
HI, came across some issues on SunOS with python2.7. > [build LNK] Library/libbasegfxlo.so > /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.0.2/pyuno/source/module/pyuno.cxx:387:1: > error: 'Py_hash_t' does not name a type > Py_hash_t PyUNO_hash( PyObject *self ) > ^ > /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.0.2/pyuno/source/module/pyuno.cxx:1660:5: > error: 'PyUNO_hash' was not declared in this scope > PyUNO_hash, > ^ The following seems to get over that: > --- > /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.0.2/pyuno/source/module/pyuno_impl.hxx.orig > 2016-01-13 01:56:20.0 + > +++ > /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.0.2/pyuno/source/module/pyuno_impl.hxx > @@ -26,7 +26,7 @@ > > #include > > -#if PY_VERSION_HEX < 0x0207 > +#if PY_VERSION_HEX <= 0x02071100 > typedef long Py_hash_t; > #endif Isn't this a python3 feature? -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Py_hash_t on python27
Le 21/01/16 19:18, Richard PALO a écrit : > HI, came across some issues on SunOS with python2.7. > >> [build LNK] Library/libbasegfxlo.so >> /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.0.2/pyuno/source/module/pyuno.cxx:387:1: >> error: 'Py_hash_t' does not name a type >> Py_hash_t PyUNO_hash( PyObject *self ) >> ^ >> /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.0.2/pyuno/source/module/pyuno.cxx:1660:5: >> error: 'PyUNO_hash' was not declared in this scope >> PyUNO_hash, >> ^ > > The following seems to get over that: >> --- >> /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.0.2/pyuno/source/module/pyuno_impl.hxx.orig >> 2016-01-13 01:56:20.0 + >> +++ >> /tmp/pkgsrc/misc/libreoffice/work/libreoffice-5.1.0.2/pyuno/source/module/pyuno_impl.hxx >> @@ -26,7 +26,7 @@ >> >> #include >> >> -#if PY_VERSION_HEX < 0x0207 >> +#if PY_VERSION_HEX <= 0x02071100 >> typedef long Py_hash_t; >> #endif > > Isn't this a python3 feature? > Guess I omitted saying this was 5.1.0.2 -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] core.git: bin/distro-install-desktop-integration
bin/distro-install-desktop-integration |5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
New commits:
commit b147a77f2e179f474051a4e50fa7ea300cdec8ee
Author: Richard PALO <rich...@netbsd.org>
Date: Tue Sep 15 14:44:31 2015 +0200
Use portable ordering for mktemp
Add omitted APPDATA_SOURCE_DIR for *.appdata.xml files
Avoid gnu rmdir by using portable invocation
Change-Id: I15a507f3d181f27c86b5332d58a9d76b31eb2fd1
Reviewed-on: https://gerrit.libreoffice.org/18588
Tested-by: Jenkins <c...@libreoffice.org>
Reviewed-by: Michael Stahl <mst...@redhat.com>
diff --git a/bin/distro-install-desktop-integration
b/bin/distro-install-desktop-integration
index 4cf652d..ed31d5b 100755
--- a/bin/distro-install-desktop-integration
+++ b/bin/distro-install-desktop-integration
@@ -94,7 +94,7 @@ add_wrapper()
}
# install desktop integration from plain packages
-sysui_temp=`mktemp -t -d distro-pack-desktop-integration-XX`
+sysui_temp=`mktemp -d -t distro-pack-desktop-integration-XX`
cp -a workdir/CustomTarget/sysui/share/libreoffice/* "$sysui_temp"
cp -a sysui/desktop/share/create_tree.sh "$sysui_temp"
builddir=`pwd`
@@ -116,6 +116,7 @@ sed -i \
export PREFIXDIR=${PREFIXDIR}
export GNOMEDIR=$PREFIXDIR
export GNOME_MIME_THEME=hicolor
+export APPDATA_SOURCE_DIR=$builddir/sysui/desktop/appstream-appdata
bash ./create_tree.sh
)
cd -
@@ -124,7 +125,7 @@ rm -rf $sysui_temp
# we do not want some stuff from the plain packages
if test -d $DESTDIR/opt ; then
rm -f $DESTDIR/opt/$INSTALLDIRNAME
-rmdir --ignore-fail-on-non-empty $DESTDIR/opt
+rmdir $DESTDIR/opt 2>/dev/null || true
fi
# we want non-versioned desktop files
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: OS checks with configuration time DLOPEN_NEEDS_LIBDL
Le 19/10/15 06:37, Norbert Thiebaud a écrit : On Sun, Oct 18, 2015 at 10:54 PM, Richard PALO <rich...@netbsd.org> wrote: I'd like to advance, one way or another on this topic. Sure, but the patch still need to not break Windows. The updated patchset seems to pass okay, in any event configure sets the correct value. It appears to be an unrelated problem causing the latest windows failure: # tarball names # does use some of the variables defined above include C:/cygwin/home/tdf/lode/jenkins/workspace/lo_gerrit_master/Gerrit/Gerrit/Platform/Windows/download.lst jenkins_build_client make /home/tdf/lode/bin/jenkins_build_client: connect: Connection refused /home/tdf/lode/bin/jenkins_build_client: line 2: /dev/tcp/localhost/2628: Connection refused Build step 'Execute shell' marked build as failure Finished: FAILURE ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: OS checks with configuration time DLOPEN_NEEDS_LIBDL
Le 19/10/15 15:40, Norbert Thiebaud a écrit : > On Mon, Oct 19, 2015 at 3:13 AM, Richard PALO <rich...@netbsd.org> wrote: >> Le 19/10/15 06:37, Norbert Thiebaud a écrit : >>> >>> On Sun, Oct 18, 2015 at 10:54 PM, Richard PALO <rich...@netbsd.org> wrote: >>>> >>>> >>>> I'd like to advance, one way or another on this topic. >>> With the latest revision working on gerrit, I'm curious whether it would be preferable to pass directly the library, if needed. That is, instead of 'DLOPEN_NEEDS_LIBDL' being '' or 'TRUE', something like 'DLOPEN_LIBS' being either '' or '-ldl'. Then the usages would look like: > $(eval $(call gb_Library_add_libs,sofficeapp,\ > -$(if $(filter $(OS),LINUX), \ > --ldl \ > +$(if $(filter LINUX %BSD SOLARIS, $(OS)), \ > +$(DLOPEN_LIBS) \ > -lpthread \ > ) \ instead of > $(eval $(call gb_Library_add_libs,sofficeapp,\ > -$(if $(filter $(OS),LINUX), \ > --ldl \ > +$(if $(filter LINUX %BSD SOLARIS, $(OS)), \ > +$(if $(DLOPEN_NEEDS_LIBDL), -ldl) \ > -lpthread \ > ) \ much simpler and a bit more like some other prerequisite library needs. -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: OS checks with configuration time DLOPEN_NEEDS_LIBDL
> Change subject: Add DLOPEN_LIBS to configure for some cases where -ldl is > needed use std::sqrt in vcldemo.cxx > .. > > > Patch Set 9: Verified+1 > > Build Successful > > http://ci.libreoffice.org/job/lo_gerrit_master/7482/ : SUCCESS > -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: OS checks with configuration time DLOPEN_NEEDS_LIBDL
Le 20/10/15 08:37, Norbert Thiebaud a écrit : > On Tue, Oct 20, 2015 at 1:11 AM, Richard PALO <rich...@netbsd.org> wrote: >> Le 19/10/15 15:40, Norbert Thiebaud a écrit : >>> On Mon, Oct 19, 2015 at 3:13 AM, Richard PALO <rich...@netbsd.org> wrote: >>>> Le 19/10/15 06:37, Norbert Thiebaud a écrit : >>>>> >>>>> On Sun, Oct 18, 2015 at 10:54 PM, Richard PALO <rich...@netbsd.org> wrote: >>>>>> >>>>>> >>>>>> I'd like to advance, one way or another on this topic. >>>>> >> >> With the latest revision working on gerrit, I'm curious whether it would be >> preferable >> to pass directly the library, if needed. >> >> That is, instead of 'DLOPEN_NEEDS_LIBDL' being '' or 'TRUE', something like >> 'DLOPEN_LIBS' being either '' or '-ldl'. >> >> Then the usages would look like: >>> $(eval $(call gb_Library_add_libs,sofficeapp,\ >>> -$(if $(filter $(OS),LINUX), \ >>> --ldl \ >>> +$(if $(filter LINUX %BSD SOLARIS, $(OS)), \ >>> +$(DLOPEN_LIBS) \ > > yeah.. but then it does not need to be in a $(if does it ? > > Norbert > > I believe, then, we would need to tackle '-lpthread' at the same time. Perhaps it is worth it? -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] core.git: config_host.mk.in configure.ac desktop/Library_sofficeapp.mk include/LibreOfficeKit libreofficekit/Executable_gtktiledviewer.mk libreofficekit/Executable_tilebench.mk l
Repository.mk |8 config_host.mk.in |1 + configure.ac|9 + desktop/Library_sofficeapp.mk | 12 +--- include/LibreOfficeKit/LibreOfficeKitInit.h |4 +++- libreofficekit/Executable_gtktiledviewer.mk |5 ++--- libreofficekit/Executable_tilebench.mk |3 +-- libreofficekit/Library_libreofficekitgtk.mk |5 ++--- libreofficekit/Module_libreofficekit.mk |4 ++-- svx/Executable_gengal.mk|6 +++--- vcl/Executable_icontest.mk |5 ++--- vcl/Executable_mtfdemo.mk |5 ++--- vcl/Executable_ui-previewer.mk |5 ++--- vcl/Executable_vcldemo.mk |5 ++--- vcl/Library_desktop_detector.mk |5 ++--- vcl/Library_vcl.mk |7 +++ vcl/Module_vcl.mk |2 +- vcl/workben/vcldemo.cxx |6 +++--- 18 files changed, 49 insertions(+), 48 deletions(-) New commits: commit fd612a144c0028972513a18167a13d29326a1798 Author: Richard PALO <rich...@netbsd.org> Date: Tue Sep 15 15:58:51 2015 +0200 Add DLOPEN_LIBS to configure for some cases where -ldl is needed use std::sqrt in vcldemo.cxx Change-Id: I24d8ba15ee267d0cad3b063df9b7cfd8d284f4ee Reviewed-on: https://gerrit.libreoffice.org/18591 Tested-by: Jenkins <c...@libreoffice.org> Reviewed-by: Norbert Thiebaud <nthieb...@gmail.com> diff --git a/Repository.mk b/Repository.mk index 925d109..c8649fb 100644 --- a/Repository.mk +++ b/Repository.mk @@ -68,13 +68,13 @@ $(eval $(call gb_Helper_register_executables,NONE, \ svptest \ svpclient \ pixelctl ) \ - $(if $(and $(ENABLE_GTK), $(filter LINUX,$(OS))), tilebench) \ - $(if $(filter LINUX MACOSX WNT,$(OS)),icontest \ + $(if $(and $(ENABLE_GTK), $(filter LINUX %BSD SOLARIS,$(OS))), tilebench) \ + $(if $(filter LINUX MACOSX SOLARIS WNT %BSD,$(OS)),icontest \ outdevgrind) \ vcldemo \ tiledrendering \ mtfdemo \ - $(if $(and $(ENABLE_GTK), $(filter LINUX,$(OS))), gtktiledviewer) \ + $(if $(and $(ENABLE_GTK), $(filter LINUX %BSD SOLARIS,$(OS))), gtktiledviewer) \ )) $(eval $(call gb_Helper_register_executables_for_install,SDK,sdk, \ @@ -582,7 +582,7 @@ $(eval $(call gb_Helper_register_libraries_for_install,PLAINLIBS_OOO,ooo, \ i18nlangtag \ i18nutil \ index_data \ - $(if $(and $(ENABLE_GTK3), $(filter LINUX,$(OS))), libreofficekitgtk) \ + $(if $(and $(ENABLE_GTK3), $(filter LINUX %BSD SOLARIS,$(OS))), libreofficekitgtk) \ localedata_en \ localedata_es \ localedata_euro \ diff --git a/config_host.mk.in b/config_host.mk.in index 947bbd4..372b647 100644 --- a/config_host.mk.in +++ b/config_host.mk.in @@ -100,6 +100,7 @@ export DISABLE_DYNLOADING=@DISABLE_DYNLOADING@ export DISABLE_EXPORT=@DISABLE_EXPORT@ export DISABLE_OPENSSL=@DISABLE_OPENSSL@ export DISABLE_PYTHON=@DISABLE_PYTHON@ +export DLOPEN_LIBS=@DLOPEN_LIBS@ export DLLTOOL=@DLLTOOL@ export DOCDIR=@DOCDIR@ export DOXYGEN=@DOXYGEN@ diff --git a/configure.ac b/configure.ac index e5350de..be21c49 100644 --- a/configure.ac +++ b/configure.ac @@ -881,6 +881,15 @@ AC_SUBST(WITH_MINGW) AC_SUBST(PTHREAD_CFLAGS) AC_SUBST(PTHREAD_LIBS) +if test $_os != "WINNT"; then +save_LIBS="$LIBS" +AC_SEARCH_LIBS([dlopen], [dl], +[case "$ac_cv_search_dlopen" in -l*) DLOPEN_LIBS="$ac_cv_search_dlopen";; esac], +[AC_MSG_ERROR([dlopen not found in either libc nor libdl])]) +LIBS="$save_LIBS" +fi +AC_SUBST(DLOPEN_LIBS) + ### # Extensions switches --enable/--disable ### diff --git a/desktop/Library_sofficeapp.mk b/desktop/Library_sofficeapp.mk index df5cfb4..1b8c5f7 100644 --- a/desktop/Library_sofficeapp.mk +++ b/desktop/Library_sofficeapp.mk @@ -17,8 +17,8 @@ $(eval $(call gb_Library_set_include,sofficeapp,\ )) $(eval $(call gb_Library_add_libs,sofficeapp,\ -$(if $(filter $(OS),LINUX), \ --ldl \ +$(if $(filter LINUX %BSD SOLARIS, $(OS)), \ +$(DLOPEN_LIBS) \ -lpthread \ ) \ )) @@ -99,19 +99,17 @@ $(eval $(call gb_Library_add_exception_objects,sofficeapp,\ ifeq ($(ENABLE_HEADLESS),TRUE) $(eval $(call gb_Library_add_libs,sofficeapp,\ - -lm \ - -ldl \ + -lm $(DLOPEN_LIBS) \ -lpthread \ )) else -ifeq ($(OS),LINUX) +ifeq ($(OS), $(filter LINUX %BSD SOLARIS, $(OS))) $(eval $(call gb_Library_use_static_libraries,sofficeapp,\ glxtest \ )) $(eval $(call gb_Library_add_libs,sofficeapp,\ - -lm \ - -ldl \ + -lm $(DL
Re: OS checks with configuration time DLOPEN_NEEDS_LIBDL
Le 30/09/15 16:38, Richard PALO a écrit : > Le 22/09/15 18:51, Norbert Thiebaud a écrit : >> On Tue, Sep 22, 2015 at 6:31 AM, Richard PALO <rich...@netbsd.org> wrote: >>> Since GUIBASE was replaced with the equivalent OS checks, there seems >>> to be a bit of abandon in keeping multiple OSes up to snuff. >>> >>> in https://gerrit.libreoffice.org/#/c/18591/ I took our base in pkgsrc and >>> given the biggest difference in unix systems is the need for -ldl or not >>> for dlopen, >>> >>> I added a configuration check for DLOPEN_NEEDS_LIBDL and updated a number >>> of the existing >>> OS checks to check simultaneously for, among others, %BSD and SOLARIS in >>> addition to LINUX. >>> >>> It would be nice to get some review on this approach. >> >> Jenkins got us a first review. the patch break on windows because it seems >> that >> AC_SEARCH_LIBS([dlopen], >> trigger a search of gcc... >> >> is there any chance that is relevant for windows ? if not the 'if >> test' that out of configure ofr cygwin case (and prolly for macosx >> case as at bes it will work , but it can pick up accidentally some >> crap if the mac has some extra stuff installed) >> >> that being said couldn't this be done without putting all that logic >> in the 'client' makefile. >> >> and just have $(if $(DLOPEN_NEEDS_LIBDL), -ldl) logic in gbuild and >> define a gb_LIBDL defined to -ldl or empty (most likely in >> RepositoryExternal.mk or in platform/* not sure on top of my head >> which is the cleanest. >> >> and then have for example >> >> $(eval $(call gb_Executable_add_libs,gengal,\ >> $(gb_LIBDL) \ >> -lpthread \ >> >> (and yes that could be extended to the few place that do the same kind >> of exercise with lpthread... and there is maybe an even better way but >> one step at the time...) >> >> >> also you may want to split the part of the patch that touch >> Repositoty.mk as that seems somewhat orthogonal with the -ldl stuff >> >> Norbert > > Well, personally I'm of the opinion that configure is the typical place for > this type > of check, in order to leave more pertinent things in platform... > > I added an `if test $_OS != "WINNT"` guard, I believe that should be ok. > As far as -lpthread, I guess I agree... I can add that to this now as well, > or it could be a follow up patch. > > gerrit updated, if it's possible to launch a jenkins run. > I'd like to advance, one way or another on this topic. I can add an addition guard to avoid MACOSX if need be but I doubt the use of a non-configuration parameter buried in platform settings. It would be a pity to leave things as they are as most BSD and SunOS currently need to seriously patch things in their ports. -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: errors building wikisend.xhp
Le 02/06/15 07:48, Richard PALO a écrit :
> I'd like to sollicit any hints to understand this new breakage:
>> /tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2/workdir/Extension/wiki-publisher/help/qtz/com.sun.wiki-publisher/wikisend.xhp:35:
>> parser error : Opening and ending tag mismatch: ahelp line 35 and paragraph
>> r document. Click Add to add a new server to the
>> list./ahelp
>>
>> ^
>> /tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2/workdir/Extension/wiki-publisher/help/qtz/com.sun.wiki-publisher/wikisend.xhp:37:
>> parser error : Opening and ending tag mismatch: ahelp line 37 and paragraph
>> e, your upload will overwrite the existing Wiki
>> entry./ahelp
>>
>> ^
>> /tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2/workdir/Extension/wiki-publisher/help/qtz/com.sun.wiki-publisher/wikisend.xhp:39:
>> parser error : Opening and ending tag mismatch: ahelp line 39 and paragraph
>> ot;http://meta.wikimedia.org/wiki/Help:Edit_summary/link;.>
>> ^
>> /tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2/workdir/Extension/wiki-publisher/help/qtz/com.sun.wiki-publisher/wikisend.xhp:41:
>> parser error : Opening and ending tag mismatch: emph line 41 and paragraph
>> edit of the already existing page with the same
>> title./ahelp>
>> ^
>> /tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2/workdir/Extension/wiki-publisher/help/qtz/com.sun.wiki-publisher/wikisend.xhp:43:
>> parser error : Opening and ending tag mismatch: emph line 43 and paragraph
>> our system web browser and show the uploaded Wiki
>> page./ahelp>
>> ^
>> /tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2/workdir/Extension/wiki-publisher/help/qtz/com.sun.wiki-publisher/wikisend.xhp:47:
>> parser error : Opening and ending tag mismatch: paragraph line 43 and body
>>
>>^
>> /tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2/workdir/Extension/wiki-publisher/help/qtz/com.sun.wiki-publisher/wikisend.xhp:48:
>> parser error : Opening and ending tag mismatch: paragraph line 41 and
>> helpdocument
>>
>>^
>> /tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2/workdir/Extension/wiki-publisher/help/qtz/com.sun.wiki-publisher/wikisend.xhp:49:
>> parser error : Premature end of data in tag paragraph line 39
>>
>> ^
>> /tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2/workdir/Extension/wiki-publisher/help/qtz/com.sun.wiki-publisher/wikisend.xhp:49:
>> parser error : Premature end of data in tag paragraph line 37
>>
>> ^
>> /tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2/workdir/Extension/wiki-publisher/help/qtz/com.sun.wiki-publisher/wikisend.xhp:49:
>> parser error : Premature end of data in tag paragraph line 35
>>
>> ^
>> /tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2/workdir/Extension/wiki-publisher/help/qtz/com.sun.wiki-publisher/wikisend.xhp:49:
>> parser error : Premature end of data in tag body line 30
>>
>> ^
>> /tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2/workdir/Extension/wiki-publisher/help/qtz/com.sun.wiki-publisher/wikisend.xhp:49:
>> parser error : Premature end of data in tag helpdocument line 19
>>
>> ^
>> ERROR: file not existing:
>> /tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2/workdir/Extension/wiki-publisher/help/qtz/com.sun.wiki-publisher/wikisend.xhp
>>
>
> curious if it is internal or external (or both) to libreoffice..
>
> For what it is worth, I also get earlier the following type of messages:
>> S=/tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.4.3.2 && I=$S/instdir &&
>> W=$S/workdir && mkdir -p
>> $W/Extension/wiki-publisher/help/af/com.sun.wiki-publisher/ &&
>> MERGEINPUT=`mktemp -t gbuild.XX` && echo
>> $S/translations/source/af/swext/mediawiki/help.po > ${MERGEINPUT} &&
>> LD_LIBRARY_PATH=${LD_LIBRARY_PATH:+$LD_LIBRARY_PATH:}"$I/ure/lib:$I/program"
>> $W/LinkTarget/Executable/helpex -i $S/swext/mediawiki/help/wikiaccount.xhp
>> -o
>> $W/Extension/wiki-publisher/help/af/com.sun.wiki-publisher/wikiaccount.xhp
>> -l af -m ${MERGEINPUT
Re: OS checks with configuration time DLOPEN_NEEDS_LIBDL
Le 22/09/15 18:51, Norbert Thiebaud a écrit : > On Tue, Sep 22, 2015 at 6:31 AM, Richard PALO <rich...@netbsd.org> wrote: >> Since GUIBASE was replaced with the equivalent OS checks, there seems >> to be a bit of abandon in keeping multiple OSes up to snuff. >> >> in https://gerrit.libreoffice.org/#/c/18591/ I took our base in pkgsrc and >> given the biggest difference in unix systems is the need for -ldl or not for >> dlopen, >> >> I added a configuration check for DLOPEN_NEEDS_LIBDL and updated a number of >> the existing >> OS checks to check simultaneously for, among others, %BSD and SOLARIS in >> addition to LINUX. >> >> It would be nice to get some review on this approach. > > Jenkins got us a first review. the patch break on windows because it seems > that > AC_SEARCH_LIBS([dlopen], > trigger a search of gcc... > > is there any chance that is relevant for windows ? if not the 'if > test' that out of configure ofr cygwin case (and prolly for macosx > case as at bes it will work , but it can pick up accidentally some > crap if the mac has some extra stuff installed) > > that being said couldn't this be done without putting all that logic > in the 'client' makefile. > > and just have $(if $(DLOPEN_NEEDS_LIBDL), -ldl) logic in gbuild and > define a gb_LIBDL defined to -ldl or empty (most likely in > RepositoryExternal.mk or in platform/* not sure on top of my head > which is the cleanest. > > and then have for example > > $(eval $(call gb_Executable_add_libs,gengal,\ > $(gb_LIBDL) \ > -lpthread \ > > (and yes that could be extended to the few place that do the same kind > of exercise with lpthread... and there is maybe an even better way but > one step at the time...) > > > also you may want to split the part of the patch that touch > Repositoty.mk as that seems somewhat orthogonal with the -ldl stuff > > Norbert Well, personally I'm of the opinion that configure is the typical place for this type of check, in order to leave more pertinent things in platform... I added an `if test $_OS != "WINNT"` guard, I believe that should be ok. As far as -lpthread, I guess I agree... I can add that to this now as well, or it could be a follow up patch. gerrit updated, if it's possible to launch a jenkins run. Also, I wonder if it is worthwhile to already fix all (or most of) the cases : > richard@omnis:/home/richard/src/libreoffice$ git grep -w '\-ldl\>' -- '*.mk' > |grep -v DLOPEN_NEED > bridges/Library_cpp_uno.mk: -ldl \ > chart2/Library_chartcore.mk: -ldl \ > chart2/Library_chartopengl.mk:-ldl \ > clew/Library_clew.mk: -ldl \ > connectivity/Library_postgresql-sdbc-impl.mk: $(if $(filter-out > MACOSX,$(OS)),-ldl) \ > extensions/Executable_pluginapp.bin.mk: -ldl \ > extensions/Executable_pluginapp.bin.mk: -ldl \ > extensions/Library_scn.mk:-ldl \ > external/clucene/Library_clucene.mk: -ldl \ > external/libxslt/ExternalProject_xslt.mk: $(if > $(SYSBASE),$(if $(filter SOLARIS LINUX,$(OS)),-L$(SYSBASE)/lib > -L$(SYSBASE)/usr/lib -lpthread -ldl))" \ > external/mysqlcppconn/Library_mysqlcppconn.mk:$(if $(filter-out > MACOSX,$(OS)),-ldl) \ > external/neon/Library_neon.mk:-ldl \ > external/redland/ExternalProject_raptor.mk: $(if > $(SYSBASE),$(if $(filter LINUX SOLARIS,$(OS)),-L$(SYSBASE)/lib > -L$(SYSBASE)/usr/lib -lpthread -ldl))" \ > external/redland/ExternalProject_rasqal.mk: $(if > $(SYSBASE),$(if $(filter LINUX SOLARIS,$(OS)),-L$(SYSBASE)/lib > -L$(SYSBASE)/usr/lib -lpthread -ldl))" \ > external/redland/ExternalProject_redland.mk: $(if > $(SYSBASE),$(if $(filter LINUX SOLARIS,$(OS)),-L$(SYSBASE)/lib > -L$(SYSBASE)/usr/lib -lpthread -ldl))" \ > libreofficekit/CppunitTest_libreofficekit_tiledrendering.mk:-ldl \ > odk/Executable_unoapploader.mk:$(if $(filter-out FREEBSD NETBSD > DRAGONFLY,$(OS)),-ldl) \ > pyuno/Library_pyuno_wrapper.mk: -ldl \ > sal/Library_sal.mk: -ldl \ > setup_native/Library_getuid.mk: -ldl \ > smoketest/Executable_libtest.mk: -ldl \ > ucb/Library_ucpdav1.mk: -ldl \ > vcl/Executable_xid_fullscreen_on_all_monitors.mk:-ldl \ > vcl/Library_vclplug_gen.mk: -ldl \ > vcl/Library_vclplug_gtk.mk: -ldl \ > vcl/Library_vclplug_gtk3.mk: -ldl \ > vcl/Library_vclplug_kde4.mk: -ldl \ > vcl/Library_vclplug_tde.mk: -ldl \ > vcl/StaticLibrary_glxtest.mk: -ldl \ > xmlsecurity/Library_xsec_xmlsec.mk: -ldl \ -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
OS checks with configuration time DLOPEN_NEEDS_LIBDL
Since GUIBASE was replaced with the equivalent OS checks, there seems to be a bit of abandon in keeping multiple OSes up to snuff. in https://gerrit.libreoffice.org/#/c/18591/ I took our base in pkgsrc and given the biggest difference in unix systems is the need for -ldl or not for dlopen, I added a configuration check for DLOPEN_NEEDS_LIBDL and updated a number of the existing OS checks to check simultaneously for, among others, %BSD and SOLARIS in addition to LINUX. It would be nice to get some review on this approach. -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: use of xsltproc without '--nonet'
Le 25/08/15 15:08, David Tardon a écrit : Hi, On Tue, Aug 25, 2015 at 08:53:41AM +0200, Richard PALO wrote: I noticed (by chance) some uses of xsltproc without the '--nonet' option in the following files: extras/CustomTarget_autocorr.mk extras/CustomTarget_autotextshare.mk extras/CustomTarget_autotextuser.mk solenv/gbuild/UIConfig.mk solenv/gbuild/platform/macosx.mk Is there any particular reason these cannot be forced to build locally like the other uses of xsltproc? Why do you think the option has been left out intentionally? Or, even better, why do you think it has been left out? A more plausible explanation is that the author of the code (I in all cases, IIRC) forgot to use it... D. No idea as to why it was left out, wherefore my inquiry. Perhaps worthwhile a build with these cases updated to use '--nonet' is in order. cheers -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
use of xsltproc without '--nonet'
I noticed (by chance) some uses of xsltproc without the '--nonet' option in the following files: extras/CustomTarget_autocorr.mk extras/CustomTarget_autotextshare.mk extras/CustomTarget_autotextuser.mk solenv/gbuild/UIConfig.mk solenv/gbuild/platform/macosx.mk Is there any particular reason these cannot be forced to build locally like the other uses of xsltproc? -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: use of xsltproc without '--nonet'
Le 25/08/15 12:55, Christian Lohmaier a écrit : Hi Richard, On Tue, Aug 25, 2015 at 8:53 AM, Richard PALO rich...@netbsd.org wrote: I noticed (by chance) some uses of xsltproc without the '--nonet' option in the following files: extras/CustomTarget_autocorr.mk extras/CustomTarget_autotextshare.mk extras/CustomTarget_autotextuser.mk solenv/gbuild/UIConfig.mk solenv/gbuild/platform/macosx.mk Is there any particular reason these cannot be forced to build locally like the other uses of xsltproc? What do you mean? nonet means that xlstproc should not look for the schemata/DTD/Entities definition at the URLs defined in the xml. ciao Christian exactly. Builds should [strive to] be local; consequently any dtds should be available locally, *especially* when --disable-fetch-external is specified. Otherwise the results may differ when connected or not. -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
errors building wikisend.xhp
LID= TYP=help Entity: line 1: parser error : Opening and ending tag mismatch: ahelp line 1 and root ntik, sal jou oplaai die bestaande wiki-inskrywing vervang.lt;/ahelpgt;/root ^ Entity: line 1: parser error : Premature end of data in tag root line 1 ntik, sal jou oplaai die bestaande wiki-inskrywing vervang.lt;/ahelpgt;/root ^ warn:l10ntools:135002:1:l10ntools/source/helpmerge.cxx:268: Can't find GID=par_id2794885 LID= TYP=help Entity: line 1: parser error : Opening and ending tag mismatch: ahelp line 1 and root yquot;gt;http://meta.wikimedia.org/wiki/Help:Edit_summarylt;/linkgt;./root ^ Entity: line 1: parser error : Premature end of data in tag root line 1 yquot;gt;http://meta.wikimedia.org/wiki/Help:Edit_summarylt;/linkgt;./root ^ warn:l10ntools:135002:1:l10ntools/source/helpmerge.cxx:268: Can't find GID=par_id2486342 LID= TYP=help Entity: line 1: parser error : Opening and ending tag mismatch: emph line 1 and root te merk van 'n reeds bestaande bladsy met dieselfde titel.lt;/ahelpgt;/root ^ Entity: line 1: parser error : Premature end of data in tag root line 1 te merk van 'n reeds bestaande bladsy met dieselfde titel.lt;/ahelpgt;/root ^ warn:l10ntools:135002:1:l10ntools/source/helpmerge.cxx:268: Can't find GID=par_id823999 LID= TYP=help Entity: line 1: parser error : Opening and ending tag mismatch: emph line 1 and root naar se blaaier te open en die opgelaaie wikibladsy te wys.lt;/ahelpgt;/root ^ Entity: line 1: parser error : Premature end of data in tag root line 1 naar se blaaier te open en die opgelaaie wikibladsy te wys.lt;/ahelpgt;/root ^ warn:l10ntools:135002:1:l10ntools/source/helpmerge.cxx:268: Can't find GID=par_id6592913 LID= TYP=help -- Richard PALO ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] core.git: Branch 'libreoffice-4-4' - include/vcl
include/vcl/window.hxx |3 +++
1 file changed, 3 insertions(+)
New commits:
commit 31c2a2230e6debd0fa53656f6a2159918d2d67d8
Author: Richard PALO rich...@netbsd.org
Date: Fri Apr 17 08:34:40 2015 +0200
Avoid redefine of TRANSPARENT on solaris after including stream.h
Change-Id: I90611a4855320148f1d4b6e61a420ada47308c28
Reviewed-by: Michael Stahl mst...@redhat.com
(cherry picked from commit e4f10cb0e12e7965df5e469cc22e6893044cabc3)
Signed-off-by: Michael Stahl mst...@redhat.com
diff --git a/include/vcl/window.hxx b/include/vcl/window.hxx
index ce43cd4..bc75618 100644
--- a/include/vcl/window.hxx
+++ b/include/vcl/window.hxx
@@ -228,6 +228,9 @@ namespace svt { class PopupWindowControllerImpl; }
#define AUTOSCROLL_VERT ((sal_uInt16)0x0001)
#define AUTOSCROLL_HORZ ((sal_uInt16)0x0002)
+#ifdef TRANSPARENT
+#undef TRANSPARENT
+#endif
// Flags for StateChanged()
enum class StateChangedType : sal_uInt16
{
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] core.git: 3 commits - starmath/inc starmath/source sysui/desktop vcl/Library_vcl.mk
starmath/inc/unomodel.hxx |4
starmath/source/unomodel.cxx |9 -
starmath/source/view.cxx | 23
sysui/desktop/apparmor/program.oosplash| 31 +
sysui/desktop/apparmor/program.open-url| 30 +
sysui/desktop/apparmor/program.senddoc | 35 ++
sysui/desktop/apparmor/program.soffice.bin | 157 +
sysui/desktop/apparmor/program.xpdfimport | 25
sysui/desktop/share/apparmor.sh| 44
vcl/Library_vcl.mk | 12 --
10 files changed, 332 insertions(+), 38 deletions(-)
New commits:
commit bd969ed90302655d41743e67124aef7101c99763
Author: Richard PALO rich...@netbsd.org
Date: Wed Mar 4 22:25:44 2015 +0100
remove /usr/sfw/lib rpath for solaris in vcl
can be patched back in by any distros still using sfw.
Change-Id: I813cc734642c0dc6c1f7e3094955cf94daa14a65
Reviewed-on: https://gerrit.libreoffice.org/15360
Tested-by: Jenkins c...@libreoffice.org
Reviewed-by: Caolán McNamara caol...@redhat.com
Tested-by: Caolán McNamara caol...@redhat.com
diff --git a/vcl/Library_vcl.mk b/vcl/Library_vcl.mk
index 89ed3bb..b3b8d02 100644
--- a/vcl/Library_vcl.mk
+++ b/vcl/Library_vcl.mk
@@ -724,18 +724,6 @@ $(eval $(call gb_Library_add_exception_objects,vcl,\
))
endif
-ifeq ($(OS),SOLARIS)
-ifeq ($(CPUNAME),SPARC64)
-$(eval $(call gb_Library_add_ldflags,vcl,\
--R/usr/sfw/lib/64 \
-))
-else
-$(eval $(call gb_Library_add_ldflags,vcl,\
--R/usr/sfw/lib \
-))
-endif
-endif
-
# Runtime dependency for unit-tests
$(eval $(call gb_Library_use_restarget,vcl,vcl))
commit 2dc69b55f4d349cc851d1b31e5634819bed64339
Author: Bryan Quigley gqu...@gmail.com
Date: Mon Apr 20 14:01:54 2015 -0400
Initial AppArmor profiles
Profiles and a script to help generate them for
different environments. Not part of the build system,
yet.
Change-Id: I502705631494b124f26a71d57a2c9a9eae130fb4
Reviewed-on: https://gerrit.libreoffice.org/15452
Reviewed-by: Caolán McNamara caol...@redhat.com
Tested-by: Caolán McNamara caol...@redhat.com
diff --git a/sysui/desktop/apparmor/program.oosplash
b/sysui/desktop/apparmor/program.oosplash
new file mode 100644
index 000..80d6f58
--- /dev/null
+++ b/sysui/desktop/apparmor/program.oosplash
@@ -0,0 +1,31 @@
+# --
+#
+#Copyright (C) 2015 Canonical Ltd.
+#
+#This Source Code Form is subject to the terms of the Mozilla Public
+#License, v. 2.0. If a copy of the MPL was not distributed with this
+#file, You can obtain one at http://mozilla.org/MPL/2.0/.
+#
+#Author: Bryan Quigley bryan.quig...@canonical.com
+#
+# --
+
+#include tunables/global
+
+profile libreoffice-oopslash INSTDIR-program/oosplash {
+ #include abstractions/base
+
+ /etc/libreoffice/ r,
+ /etc/libreoffice/** r,
+ /etc/passwd r,
+ /etc/nsswitch.confr,
+ /run/nscd/passwd r,
+ /usr/lib{,32,64}/ure/bin/javaldx Cx,
+ /usr/share/libreoffice/program/* r,
+ INSTDIR-program/soffice.bin rmPUx,
+ INSTDIR-ure/bin/javaldx rmPUx,
+ owner @{HOME}/.Xauthority r,
+ owner @{HOME}/.config/libreoffice{,dev}/?/user/uno_packages/cache/log.txt rw,
+ unix peer=(addr=@/tmp/.ICE-unix/* label=unconfined),
+ unix peer=(addr=@/tmp/.X11-unix/* label=unconfined),
+}
diff --git a/sysui/desktop/apparmor/program.open-url
b/sysui/desktop/apparmor/program.open-url
new file mode 100644
index 000..1d3d466
--- /dev/null
+++ b/sysui/desktop/apparmor/program.open-url
@@ -0,0 +1,30 @@
+# --
+#
+#Copyright (C) 2015 Canonical Ltd.
+#
+#This Source Code Form is subject to the terms of the Mozilla Public
+#License, v. 2.0. If a copy of the MPL was not distributed with this
+#file, You can obtain one at http://mozilla.org/MPL/2.0/.
+#
+#Author: Bryan Quigley bryan.quig...@canonical.com
+#
+# --
+
+#include tunables/global
+
+profile libreoffice-open-url INSTDIR-program/open-url {
+ #include abstractions/base
+ #include abstractions/ubuntu-helpers
+
+ owner /tmp/lu** rw,#makes files like luR.tmp/lub.tmp where
R is random
+ #Note, usually it's lub or luc, don't know why.
+ /bin/sh rmix,
+ /bin/dash rmix,
+ /bin/bash rmix,
+ /bin/sed rmix,
+ /bin/grep rmix,
+ /usr/bin/xdg-open Cxr - sanitized_helper,
+ /dev/null rw,
+ /usr/share/libreoffice/share/config/*r,
+ owner @{HOME}/.config/libreoffice{,dev}/?/user/uno_packages/cache/log.txt
rw,
+}
diff --git a/sysui
[Libreoffice-commits] core.git: solenv/gbuild
solenv/gbuild/platform/solaris.mk |2 ++ 1 file changed, 2 insertions(+) New commits: commit f3cffeb3aeb67e979d43bb4b8e970cc466b14d45 Author: Richard PALO rich...@netbsd.org Date: Sun Feb 22 09:23:32 2015 +0100 Reintegrate portion of a patch originally submitted by Michael Stahl m...@openoffice.org Mar 30 2011 solaris11: #i117606#: solaris.mk: add -Bdirect -z defs to LDFLAGS Change-Id: Iab16960010810185a3b1f7c617c44db3d9822fb5 Reviewed-on: https://gerrit.libreoffice.org/14579 Tested-by: Jenkins c...@libreoffice.org Tested-by: Michael Stahl mst...@redhat.com Reviewed-by: Michael Stahl mst...@redhat.com diff --git a/solenv/gbuild/platform/solaris.mk b/solenv/gbuild/platform/solaris.mk index eef6826..d66345e 100644 --- a/solenv/gbuild/platform/solaris.mk +++ b/solenv/gbuild/platform/solaris.mk @@ -67,6 +67,8 @@ endif gb_LinkTarget_LDFLAGS += \ -L$(SYSBASE)/lib \ -L$(SYSBASE)/usr/lib \ + -Wl,-B,direct \ + -Wl,-z,defs \ -Wl,-z,combreloc \ ifeq ($(HAVE_LD_HASH_STYLE),TRUE) ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: cppunit errors in vcl_app_test, on illumos/xstreamos
Le 23/02/15 09:33, Gabriele Bulfon a écrit : Oh yes, I found this in Library_vcl.mk: ifeq ($(CPUNAME),SPARC64) $(eval $(call gb_Library_add_ldflags,vcl,\ -R/usr/sfw/lib/64 \ )) else $(eval $(call gb_Library_add_ldflags,vcl,\ -R/usr/sfw/lib \ )) endif My system fall into the else section...and this is breaking my build. I will try commenting out this and rebuild from scratch. There should be a better way to patch this. Gabriele. Are you building this on opensolaris or OI/hipster? I use pkgsrc on omnios so there's not much at all in /usr/sfw/lib that can get in the way there. It would be very useful to know if there is *anybody* building on a legacy SunOS platform that really *needs* something from /usr/sfw/lib* (that is, which isn't actually found in [/usr]/lib* because of symlinks) Please speak up! Otherwise, I'm all for blasting this section. ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] core.git: Branch 'distro/collabora/cp-4.3' - 26 commits - connectivity/source dbaccess/source embeddedobj/source formula/source include/connectivity include/toolkit sc/inc sc/qa
connectivity/source/parse/sqlnode.cxx |5 -
dbaccess/source/ui/browser/sbagrid.cxx |1
dbaccess/source/ui/dlg/tablespage.cxx |1
dbaccess/source/ui/querydesign/QueryDesignView.cxx | 15
embeddedobj/source/commonembedding/persistence.cxx |5 +
formula/source/core/api/FormulaCompiler.cxx|2
include/connectivity/sqlparse.hxx |5 +
include/toolkit/helper/listenermultiplexer.hxx | 10 +--
include/toolkit/helper/macros.hxx | 10 +--
sc/inc/compiler.hxx|2
sc/inc/scmatrix.hxx|9 +-
sc/qa/unit/ucalc.hxx |2
sc/qa/unit/ucalc_formula.cxx | 17 +
sc/source/core/tool/compiler.cxx | 23 ++
sc/source/core/tool/formularesult.cxx |2
sc/source/core/tool/interpr1.cxx |6 +
sc/source/core/tool/interpr4.cxx | 15
sc/source/core/tool/scmatrix.cxx | 23 +-
sc/source/filter/oox/sheetdatabuffer.cxx |2
sc/source/ui/dbgui/PivotLayoutDialog.cxx |9 ++
sc/source/ui/docshell/docsh4.cxx |7 +-
sc/source/ui/inc/PivotLayoutDialog.hxx |1
solenv/gbuild/platform/solaris.mk |5 -
svx/source/dialog/langbox.cxx | 24 ---
sw/source/core/crsr/findtxt.cxx|4 -
sw/source/core/layout/tabfrm.cxx |2
sw/source/core/txtnode/ndtxt.cxx | 28 ++--
sw/source/core/undo/unnum.cxx |4 -
sw/source/core/undo/untbl.cxx | 25 +++
sw/source/filter/html/htmlgrin.cxx | 70 -
sw/source/filter/ww8/rtfsdrexport.cxx |3
sw/source/ui/chrdlg/drpcps.cxx |4 -
sw/source/ui/fldui/javaedit.cxx|9 ++
toolkit/source/helper/listenermultiplexer.cxx |9 ++
vcl/generic/glyphs/graphite_serverfont.cxx |4 -
vcl/inc/graphite_layout.hxx| 18 +++--
36 files changed, 271 insertions(+), 110 deletions(-)
New commits:
commit b5c6bae95eed4df45617ec4cadda59088261a3a1
Author: Richard PALO rich...@netbsd.org
Date: Fri Feb 13 19:17:54 2015 +0100
Sun ld doesn't support -O1 optimize flag
reported by Gabriele Bulfon (gbulfon@sonicle)
Change-Id: Ie0ff8fe296054a7a266b4c73c30938bb2b7353ea
Reviewed-on: https://gerrit.libreoffice.org/14484
Tested-by: Michael Stahl mst...@redhat.com
Reviewed-by: Michael Stahl mst...@redhat.com
(cherry picked from commit 34a167e1bf9b478069c290dbf977c68639149852)
Signed-off-by: Michael Stahl mst...@redhat.com
(cherry picked from commit f081a415221bd8665cceebe91665e2253781b516)
diff --git a/solenv/gbuild/platform/solaris.mk
b/solenv/gbuild/platform/solaris.mk
index 8256620..e4bc23b 100644
--- a/solenv/gbuild/platform/solaris.mk
+++ b/solenv/gbuild/platform/solaris.mk
@@ -131,11 +131,8 @@ gb_LinkTarget_LDFLAGS += \
endif
-ifneq ($(gb_DEBUGLEVEL),0)
+# sun ld doesn't understand -O1 optimize flag
gb_LINKEROPTFLAGS :=
-else
-gb_LINKEROPTFLAGS := -Wl,-O1
-endif
ifeq ($(gb_SYMBOL),$(true))
gb_LINKERSTRIPDEBUGFLAGS :=
commit c5420c816f3977c5f4002d66561eef8111faecbe
Author: Laurent Godard lgodard.li...@laposte.net
Date: Thu Feb 19 10:45:26 2015 +0100
tdf#89460 test pChangeTrack before using it
Reviewed-on: https://gerrit.libreoffice.org/14553
Reviewed-by: Eike Rathke er...@redhat.com
Tested-by: Eike Rathke er...@redhat.com
(cherry picked from commit bf5c5f359cf4eb36a0a9a1110ec2aa3393bb87df)
Conflicts:
sc/source/ui/docshell/docsh4.cxx
Change-Id: Ic02351c448929daa7d4c5a9a00df9842c2c5aa91
Reviewed-on: https://gerrit.libreoffice.org/14558
Reviewed-by: Eike Rathke er...@redhat.com
Tested-by: Eike Rathke er...@redhat.com
(cherry picked from commit bf57bbef91bc40090561d614e4c2070c01aef7cb)
diff --git a/sc/source/ui/docshell/docsh4.cxx b/sc/source/ui/docshell/docsh4.cxx
index dd9eaf8..13855e6 100644
--- a/sc/source/ui/docshell/docsh4.cxx
+++ b/sc/source/ui/docshell/docsh4.cxx
@@ -590,8 +590,11 @@ void ScDocShell::Execute( SfxRequest rReq )
if ( bDo )
{
-if ( pChangeTrack-IsProtected() )
-bDo = ExecuteChangeProtectionDialog( NULL );
+if (pChangeTrack)
+{
+if ( pChangeTrack-IsProtected() )
+bDo = ExecuteChangeProtectionDialog( NULL
);
+}
if ( bDo )
{
pDoc
Re: error during build of mork_helper, on illumos/xstreamos
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Le 21/02/15 08:29, Richard PALO a écrit : In pkgsrc this is patched as follows: $NetBSD: patch-vcl_Library__vcl.mk,v 1.1 2015/02/04 18:19:34 ryoon Exp $ --- vcl/Library_vcl.mk.orig 2015-01-22 20:05:28.0 + +++ vcl/Library_vcl.mk @@ -703,7 +703,7 @@ endif endif endif -ifeq ($(OS),LINUX) +ifeq ($(GUIBASE),unx) $(eval $(call gb_Library_add_libs,vcl,\ -lm \ -ldl \ If this works for you, I'll prepare to upstream this patch. The author of this patch responded to me that he'll upstream this himself as well as a few others of the same sort (in pkgsrc) soon. As to the '-Wl,-Bdirect' and '-Wl,-zdefs' in solaris.mk I'm testing if this passes, if so I'll upstream that as well.. (builds/runs here without it, at least on gcc49) As hoped, I had no fallout from this so will submit the reintegration of this bit to gerritt today. - -- Richard PALO -BEGIN PGP SIGNATURE- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJU6Y+aAAoJECAB22fHtp27oOYIAIBtDwPr7U2HfCrbFNNSOsbU zZZj+qfLv4DbLUm3ZdCS5zwYUwl/agEkltsO+tolzhTT6YFIMXqxNffMbFrITh9z TmMu6PgR8JtXhnfzyWhT9gxNucNphKiwc2wwvWNDyxe5bOPYWIZTDtkGKWNbPjuE +sLGGIZ7YBaiFazRww/3cFKy+YwmOyXBFCsS3LhHEw8bgn/UnwdA0kmgBsSkYmKw YTwPd4iqm0kKis54jnr8CElzUb/Unzv6Sjz6JmNdbZc+OpSsQhFrela7IZJWSs7N oNe2TmDZrNv8XGLTlsvipA4lZNCSWJ5fiPOp19NkX0YezrEFTFmP8uL3uL2w3hA= =SRUy -END PGP SIGNATURE- ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: error during build of mork_helper, on illumos/xstreamos
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Le 20/02/15 09:52, Gabriele Bulfon a écrit : Great Michael, I had to double patch, both to have LDFLAGS with -Wl,-z,defs and to enable some X11 linking in vcl/Library_vcl.mk. Specifically I borrowed the ifeq ($(OS),LINUX) various linkings, and added them into the ifeq ($(OS),SOLARIS), before checking for SPARC or not. Now, I get this stupid single one...: Undefined first referenced symbol in file main/usr/lib/libcups.so In pkgsrc this is patched as follows: $NetBSD: patch-vcl_Library__vcl.mk,v 1.1 2015/02/04 18:19:34 ryoon Exp $ --- vcl/Library_vcl.mk.orig 2015-01-22 20:05:28.0 + +++ vcl/Library_vcl.mk @@ -703,7 +703,7 @@ endif endif endif -ifeq ($(OS),LINUX) +ifeq ($(GUIBASE),unx) $(eval $(call gb_Library_add_libs,vcl,\ -lm \ -ldl \ If this works for you, I'll prepare to upstream this patch. As to the '-Wl,-Bdirect' and '-Wl,-zdefs' in solaris.mk I'm testing if this passes, if so I'll upstream that as well.. (builds/runs here without it, at least on gcc49) BTW I had far fewer difficulties to get a working LO with gcc=4.8 on SunOS. - -- Richard PALO -BEGIN PGP SIGNATURE- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJU6DPiAAoJECAB22fHtp27OX4H/Rpx7tt6VaTCHBCzBJAfF+nD xuI0o7kNL1dudDyIaNbWIhXr+/JaJdOhvU8ttSyUrOeQgvGChqGBJfKzHPzzbfaZ fIC0L3ejEQQRV59lSC15JjDAvQOQRKUfScoIcdeI6SqRGy2SebkanJsXtTmXHDy8 Y79l44NSYvzjVYBQx/zYprVoQfBlQeXrzRSJbz0osjsdeCls+HvYGlF5zKg6Ezk0 FMWKPJ+AzgcehXbAnNeDLHklpja7brkGa6MytLV7zDDOPfjwFQZLm06PqqkATvR1 rQIqGzHs2gp6u4anig8M+1BCAO7mlUy5HPMTj6SaS2gO8gdQW9mKVUMl/voKGv4= =MVjx -END PGP SIGNATURE- ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] core.git: Branch 'libreoffice-4-4' - solenv/gbuild
solenv/gbuild/platform/solaris.mk |5 + 1 file changed, 1 insertion(+), 4 deletions(-) New commits: commit 326bbd521c12a6b85b12396c2424126d9218d561 Author: Richard PALO rich...@netbsd.org Date: Fri Feb 13 19:17:54 2015 +0100 Sun ld doesn't support -O1 optimize flag reported by Gabriele Bulfon (gbulfon@sonicle) Change-Id: Ie0ff8fe296054a7a266b4c73c30938bb2b7353ea Reviewed-on: https://gerrit.libreoffice.org/14484 Tested-by: Michael Stahl mst...@redhat.com Reviewed-by: Michael Stahl mst...@redhat.com (cherry picked from commit 34a167e1bf9b478069c290dbf977c68639149852) Signed-off-by: Michael Stahl mst...@redhat.com diff --git a/solenv/gbuild/platform/solaris.mk b/solenv/gbuild/platform/solaris.mk index c2147d4..cc3df57 100644 --- a/solenv/gbuild/platform/solaris.mk +++ b/solenv/gbuild/platform/solaris.mk @@ -83,11 +83,8 @@ gb_LinkTarget_LDFLAGS += \ endif -ifneq ($(gb_DEBUGLEVEL),0) +# sun ld doesn't understand -O1 optimize flag gb_LINKEROPTFLAGS := -else -gb_LINKEROPTFLAGS := -Wl,-O1 -endif ifeq ($(gb_SYMBOL),$(true)) gb_LINKERSTRIPDEBUGFLAGS := ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] core.git: solenv/gbuild
solenv/gbuild/platform/solaris.mk |5 + 1 file changed, 1 insertion(+), 4 deletions(-) New commits: commit 34a167e1bf9b478069c290dbf977c68639149852 Author: Richard PALO rich...@netbsd.org Date: Fri Feb 13 19:17:54 2015 +0100 Sun ld doesn't support -O1 optimize flag reported by Gabriele Bulfon (gbulfon@sonicle) Change-Id: Ie0ff8fe296054a7a266b4c73c30938bb2b7353ea Reviewed-on: https://gerrit.libreoffice.org/14484 Tested-by: Michael Stahl mst...@redhat.com Reviewed-by: Michael Stahl mst...@redhat.com diff --git a/solenv/gbuild/platform/solaris.mk b/solenv/gbuild/platform/solaris.mk index 78698db..eef6826 100644 --- a/solenv/gbuild/platform/solaris.mk +++ b/solenv/gbuild/platform/solaris.mk @@ -83,11 +83,8 @@ gb_LinkTarget_LDFLAGS += \ endif -ifneq ($(gb_DEBUGLEVEL),0) +# sun ld doesn't understand -O1 optimize flag gb_LINKEROPTFLAGS := -else -gb_LINKEROPTFLAGS := -Wl,-O1 -endif ifeq ($(gb_SYMBOL),$(true)) gb_LINKERSTRIPDEBUGFLAGS := ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] core.git: Branch 'libreoffice-4-3' - solenv/gbuild
solenv/gbuild/platform/solaris.mk |5 + 1 file changed, 1 insertion(+), 4 deletions(-) New commits: commit f081a415221bd8665cceebe91665e2253781b516 Author: Richard PALO rich...@netbsd.org Date: Fri Feb 13 19:17:54 2015 +0100 Sun ld doesn't support -O1 optimize flag reported by Gabriele Bulfon (gbulfon@sonicle) Change-Id: Ie0ff8fe296054a7a266b4c73c30938bb2b7353ea Reviewed-on: https://gerrit.libreoffice.org/14484 Tested-by: Michael Stahl mst...@redhat.com Reviewed-by: Michael Stahl mst...@redhat.com (cherry picked from commit 34a167e1bf9b478069c290dbf977c68639149852) Signed-off-by: Michael Stahl mst...@redhat.com diff --git a/solenv/gbuild/platform/solaris.mk b/solenv/gbuild/platform/solaris.mk index 8256620..e4bc23b 100644 --- a/solenv/gbuild/platform/solaris.mk +++ b/solenv/gbuild/platform/solaris.mk @@ -131,11 +131,8 @@ gb_LinkTarget_LDFLAGS += \ endif -ifneq ($(gb_DEBUGLEVEL),0) +# sun ld doesn't understand -O1 optimize flag gb_LINKEROPTFLAGS := -else -gb_LINKEROPTFLAGS := -Wl,-O1 -endif ifeq ($(gb_SYMBOL),$(true)) gb_LINKERSTRIPDEBUGFLAGS := ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] libmspub.git: src/lib
src/lib/MSPUBMetaData.cpp |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) New commits: commit db5a1e5b20463fc5704c53aa679e17e875314da2 Author: Richard PALO rich...@netbsd.org Date: Fri Feb 13 17:16:59 2015 +0100 use std::pow in order to avoid compile time errors like: error: call of overloaded 'pow(int, uint16_t)' is ambiguous as seen on SunOS 5.11 Change-Id: I0e0bcdf4c50f9c8d1e8f088193eac82247178e80 Reviewed-on: https://gerrit.libreoffice.org/14478 Tested-by: David Tardon dtar...@redhat.com Reviewed-by: David Tardon dtar...@redhat.com diff --git a/src/lib/MSPUBMetaData.cpp b/src/lib/MSPUBMetaData.cpp index 7e5888e..1234fe2 100644 --- a/src/lib/MSPUBMetaData.cpp +++ b/src/lib/MSPUBMetaData.cpp @@ -194,7 +194,7 @@ bool libmspub::MSPUBMetaData::parseTimes(librevenge::RVNGInputStream *input) uint32_t firstDirSectorLocation = readU32(input); // Seek to the Root Directory Entry - size_t sectorSize = pow(2, sectorShift); + size_t sectorSize = std::pow(2, sectorShift); input-seek((firstDirSectorLocation + 1) * sectorSize, librevenge::RVNG_SEEK_SET); // DirectoryEntryName: 64 bytes // DirectoryEntryNameLength: 2 bytes ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] libvisio.git: src/lib
src/lib/VSDMetaData.cpp |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) New commits: commit 9686192d63bd208df1c6b53c892ba5a680cdc4b6 Author: Richard PALO rich...@netbsd.org Date: Fri Feb 13 17:58:07 2015 +0100 use std::pow in order to avoid compile time errors like: error: call of overloaded 'pow(int, uint16_t)' is ambiguous as seen on SunOS 5.11 Change-Id: I2ac60caa2b0c3731064cdea6150523a00647036e Reviewed-on: https://gerrit.libreoffice.org/14481 Tested-by: David Tardon dtar...@redhat.com Reviewed-by: David Tardon dtar...@redhat.com diff --git a/src/lib/VSDMetaData.cpp b/src/lib/VSDMetaData.cpp index dac4c97..4235b56 100644 --- a/src/lib/VSDMetaData.cpp +++ b/src/lib/VSDMetaData.cpp @@ -210,7 +210,7 @@ bool libvisio::VSDMetaData::parseTimes(librevenge::RVNGInputStream *input) uint32_t firstDirSectorLocation = readU32(input); // Seek to the Root Directory Entry - size_t sectorSize = pow(2, sectorShift); + size_t sectorSize = std::pow(2, sectorShift); input-seek((firstDirSectorLocation + 1) * sectorSize, librevenge::RVNG_SEEK_SET); // DirectoryEntryName: 64 bytes // DirectoryEntryNameLength: 2 bytes ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: patch for alloca on xstreamos/illumos
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Le 23/01/15 10:01, Gabriele Bulfon a écrit : Hi, going on building, I found a couple of vcl sources using alloca, requiring an include on solaris/illumos. So I had to patch both toolkit/source/awt/vclxgraphics.cxx and toolkit/source/awt/vclxfont.cxx to add this: #ifdef __sun__ #include alloca.h #endif and now it goes on ;) Gabriele. == richard@omnis:/home/richard/src/libreoffice$ git grep 'alloca.h' include/sal/alloca.h:#include alloca.h If you look there you see a guard for solaris... Therefore, perhaps consider simply including sal/alloca.h and submit a patch if not already in master. Hopefully there are not too many of these left... -BEGIN PGP SIGNATURE- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJUwhu4AAoJECAB22fHtp27wrAH/1efW9CF4RLXdw/HdDuENzj5 spgnZ7nzCdBhojiFDHlPYW0CRfjgrX7BKcNTRiFGQWGgQeXfAscGtepl5+Xr4kLa v35uKh3YSStVOTCaXMEoxhrA4/WF1XBzcwAv9xhGveWNerw+4n4rMLBUkmWgCZmw 8KrDTUQAsAYq+nKUmNxlIb5vmIRfUli1imzg/M7SuRKoNxhQj9WPIA+6fKS7fRdM O+rkC7mBc084GfKdAQPnDnmE7SmTvhi6hMAOvZsxAYcuJsY4vGzl4WpOifNhIXon OoGCD4xxkz7saCtGedB4A0UijBDpBoDRbugeMg7UIhcMDGbU1fqI0VSVzpkZa7U= =PzG3 -END PGP SIGNATURE- ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: error building vclxbitmap.cxx on xstreamos/illumos
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Le 22/01/15 19:11, Markus Mohrhard a écrit : Yah, but... shouldn't the enum be in a namespace in the first place? Or are there c-only source files looking for it... That doesn't matter here as it is a define and therefore the preprocessor will replace the string in the definition. We have a number of similar cases for other defines. Regards, Markus Okay, but by observation it does seem to be a rather overloaded term though, missing a prefix or namespace.. richard@omnis:/home/richard/src/libreoffice$ git grep '\TRANSPARENT\[[:space:]]*=' include/vcl/window.hxx:TRANSPARENT= 12, offapi/com/sun/star/awt/InvalidateStyle.idl:const short TRANSPARENT = 16; offapi/type_reference/offapi.idl: const short TRANSPARENT = 16; reportbuilder/java/org/libreoffice/report/pentaho/output/spreadsheet/SpreadsheetRawReportTarget.java: private static final String TRANSPARENT = transparent; vcl/source/filter/wmf/winmtf.hxx:TRANSPARENT = 1, wizards/com/sun/star/wizards/ui/ImageList.py:TRANSPARENT = -1 -BEGIN PGP SIGNATURE- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJUwW0SAAoJECAB22fHtp27S4MH/11QhQ0UAxiQiWj1VNNy+vIv FsURIbpZ4uQWiyTrfETXnsWQRm6hnfU6WMC0Gq0KpdQbnWel71Z4bZfNjzjZoNHW LIKR/tmLY5nx76YYJA9GeLjrMREHEFRI8X5fKRpIN+tM60ejedUwNKBwmhLS4QaQ 2isksKl6qGeH0G6esn9p9ye2d2iBjOVICwm2ornI7uxYYWbp4IHX1IeV/FAon30g LPZ8xya7Pvdw2dj70XiVIyCIa4ZEdXJU7fg/+9Y9WYNS6/qF7AZFaW24ebd49fGa GiA0icOoYP+6Wg3EGyhJ+Jp2FMk+z0yT3YgmpUlGOsQZebe3Tli7EO6C+b2YROA= =IfD1 -END PGP SIGNATURE- ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: error building vclxbitmap.cxx on xstreamos/illumos
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Le 22/01/15 15:20, Stephan Bergmann a écrit : If line 245 of include/vcl/window.hxx reads TRANSPARENT= 12, for you, then most likely you happen to include some definition of a macro named TRANSPARENT. ___ on SunOS: $ grep TRANSPARENT /usr/include/sys/stream.h #define TRANSPARENT (unsigned int)(-1) -BEGIN PGP SIGNATURE- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJUwS9dAAoJECAB22fHtp27me4H/j2akRFE9XW1V76jGnNd0RRX f5nQnVtLqd/yruPaoRyADaaD8fzlKfhYA05pQ0n64WoAV+P2wQba7UIbiZGyLHL9 6yzx7thG8v8cNgkVFXssWPJFnKCHKFvvVofPtJiz2RfNN1UicnszQZTAxvkicISP Fw5IJfGfyShof16o6sGIg8v+a/lv7h2lxcUfq0wf7aC0yMIBJ/ilFdBgsCo0EzKM OIdaLyfnDIkZqlunEDHdBDenbXfHC2g7DAaIMN63S80s3RKxEmADnkNe9NenlJS5 2fjFfheUuKjJIghShtaVyRCnKIEZW2iE6JqBUShpcgocCyBcsgjst2JB/ta5NW0= =kETx -END PGP SIGNATURE- ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: error building vclxbitmap.cxx on xstreamos/illumos
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Le 22/01/15 18:35, Gabriele Bulfon a écrit : yes, I added a patch to add this just before the enum, and it goes on : #ifdef TRANSPARENT # undef TRANSPARENT #endif going on ;) thanks! Yah, but... shouldn't the enum be in a namespace in the first place? Or are there c-only source files looking for it... -BEGIN PGP SIGNATURE- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJUwTkGAAoJECAB22fHtp27QIwH/1/3twfLCAURyxgBgzE5059Y zjO7mjwYevxrWASI1/jNwKcniwef/8xazacIapz8EccvZVbYnc5/94Dkl0ej2cus caiXYh3oslDF6VY1IAU15RLqFXe++UfaIgeLILw4LXOpu+Pumvc/hlh3HtMOa0nq vx0HoIo3HNotFkogs3DyX3F8hwdR0nFzMOTmJQs18yCsIduWNr1Kiid/o2R+3Ryj 25GIF4llzWQkziD+4neBccIwDJr7gk8oqMpVFEDTzXZKycBtSbyHTzDEdlJfJ2aC uHi/5Fu4M6sb8gCkdluCaLWkS2VHD6AGhDpfkX0jGADSDMmSMye4k//5DyE2t8U= =vm0l -END PGP SIGNATURE- ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Solved: 4.4.0.1 build error on libstdc++ on solaris/illumos
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Le 20/01/15 18:07, Gabriele Bulfon a écrit : I thought the same. But maybe g++ thinks it's going to run gnu ld, while it's going to run sun ld. Secondly, I surely had to add /usr/gcc/4.7/lib to the link path, but maybe I can avoid specifying -lstdc++, maybe... I'll try and send you feedback :) These last mods seem a bit overdone... Using pkgsrc, gcc is in (my case) /opt/local/gcc49... builds have most everything in the path or passed in. there is/are perhaps other things steering things awry. what is your $PATH ( and $(SYSBASE) in LO )? I haven't seen this type of problem (at least not with 4.3.5). -BEGIN PGP SIGNATURE- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJUvpfCAAoJECAB22fHtp27bxYIAKt2j7IPo3JespCXTdstBgE+ 9hcS3cKMMDMSeSlk/q4MaLFLII2++a2pSix62Z4EQhsfTIgm5/3aQ3Vuap9ntEzH /6Tf27oaaul4HXb5zBYb37j9A7L7CX62eesGCZkUlzUV4OSEi6riGiPk6ssrQM30 eIU4rgoCNGl0JLV7o7mUtFT517A1Mc8nI7WoGBkI6yDvjowTfCumMYPZJA4zqXBx UIu5j9Pj7lEHt9jpoDmsfz7UnCg8/F3p6ECssPFjuPcVB/WTDEpFa1RWFoAmlz6u HMAIEaKylYi7D2UShC/2rtlctTdfCfdS7Ayz47miXo9lTnmZmHLX4JodLDswCJE= =cqr9 -END PGP SIGNATURE- ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: 4.4.0.1 build error on sal/types.h on solaris/illumos
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Le 18/01/15 21:33, Gabriele Bulfon a écrit : Ok, I commented out these on Solaris.mk: #ifneq ($(gb_DEBUGLEVEL),0) #gb_LINKEROPTFLAGS := #else #gb_LINKEROPTFLAGS := -Wl,-O1 #endif will let you know :) -- Da: Norbert Thiebaud A: gbul...@sonicle.com Cc: Michael Stahl libreoffice Richard PALO Data: 18 gennaio 2015 20.15.22 CET Oggetto: Re: 4.4.0.1 build error on sal/types.h on solaris/illumos On Sun, Jan 18, 2015 at 12:18 PM, Gabriele Bulfon wrote: Ok, I checked and it looks fine, so solaris.mk should be taken. Also config.log shows correct variables for OS, CPU and COM. What actually happens is during make, after it has downloaded and extracted various stuff. The last one is translations file. Then it goes on building concat-deps, and linking fails: [build C ] solenv/bin/concat-deps.c [build LNK] Executable/concat-deps ld: fatal: unrecognized option '-O' ld: fatal: unrecognized option '-1' ld: fatal: use the -z help option for usage information collect2: error: ld returned 1 exit status Maybe just concat-deps have problems linking with wrong options on Solaris? Yes it is quite possible... concat-deps is a small utility I wrote to speed up the original perl-based one... and I'm quite sure I never tried to link it on Solaris. Otoh concat-deps is built using the standard gbuild mechanism for that: see solenv/Executable_concat-deps.mk so it is more likely that the problem is generic and that concat-deps just happen to be the first one to be linked. BTW, is there any way to issue gmake and let it show what command LNK is doing? It tried forcing a -n but I did not get the command debugged verbose=t make Norbert Yeah, this should go as not valid for solaris ld. Apparently comes from here: commit 3c4cd1deaf71d0d800957b3580d426c721bf7844 Author: Jonathan Adams t12nsloo...@gmail.com Date: Fri Mar 16 21:50:37 2012 +0100 gbuild: switch solaris.mk to GCC -BEGIN PGP SIGNATURE- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJUvCwiAAoJECAB22fHtp27MBMH/2IYZRT44q8six0NjeXYkG6B vDe0AJund3FTcN/t9zT0bpFx9mPCCdVxWbqqdv0CFKYQrfSLhPRaksuWrAnK1mOI Nl5vcDxI8D0ls6N1c40ZIAcpI3mOeY1Xye6rpKwYJSjWxqOZHxgWEMMvIrzd9Jxb wnChiTHrDj4ljX1QCPWLxJQB0+e3gXrFmcLvQSEhIPslGPou4i14ur3tf2vVwk0a FZzCoiYfCnUil7vQJHA8vUtbVASMxX4sJ6iYOfD0K/Z+11c4PwF4xZZtZCtG6Rek qLd/Yz/OhqIg2POqJeaDL8S9fYOfrWUgfKuwcFhqBgqDQMknQH8DSeJt6IPAea0= =ZdNh -END PGP SIGNATURE- ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] core.git: solenv/gbuild
solenv/gbuild/Deliver.mk |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) New commits: commit fc4a339d83652613dbbc4c2cd7b206d6b1115a71 Author: Richard PALO rich...@netbsd.org Date: Fri Dec 26 17:47:20 2014 +0100 use special $(TOUCH) for touch -hr Change-Id: Ifb95dc6bb4bc70f7c4fcd765d96b9aeb3cbdeb38 Reviewed-on: https://gerrit.libreoffice.org/13664 Reviewed-by: Caolán McNamara caol...@redhat.com Tested-by: Caolán McNamara caol...@redhat.com diff --git a/solenv/gbuild/Deliver.mk b/solenv/gbuild/Deliver.mk index 00bb460..ca609d9 100644 --- a/solenv/gbuild/Deliver.mk +++ b/solenv/gbuild/Deliver.mk @@ -51,7 +51,7 @@ endif endef define gb_Deliver__deliver -$(if $(gb_Deliver_CLEARONDELIVER),rm -f $(2) ) $(if $(gb_Deliver_HARDLINK),ln,cp -P -f) $(1) $(2) touch -hr $(1) $(2) +$(if $(gb_Deliver_CLEARONDELIVER),rm -f $(2) ) $(if $(gb_Deliver_HARDLINK),ln,cp -P -f) $(1) $(2) $(TOUCH) -hr $(1) $(2) endef ifneq ($(strip $(gb_Deliver_GNUCOPY)),) ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: libi18nlangtag crash on SunOS 5.11
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Le 21/12/14 10:11, Richard PALO a écrit : This seems to be liblangtag issue as its testsuite has problems as well. I'll need to get those resolved first. thx ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice As a quick workaround I needed to configure liblangtag with '--disable-modules' and that seems to get over the test-suite's check-tag failures and consequently allows format-cells to function correctly in libreoffice. Since LO is seemingly the only thing I have currently running liblangtag, is there any imperative need for the extension modules now disabled? In any event, the problem is reported in bitbucket for liblangtag. -BEGIN PGP SIGNATURE- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJUnVTEAAoJECAB22fHtp27U/MH/2YI7TaOsowTlPq9Pt01XwP1 1niyzZOXoFS1B9612jHfd+mv4hRqOFoVYPdb4ZgnLFCh3Y7XrDzpV6JxkdS6rgZx 6tfOIqzVOMmCPdVhLAR7egG1buKBRzhqVs2i/YVvHNGb/i8hb+4nb7CsY12lwIUI aRjF8t+gbQCN9qTgXii0hhrZSDVnBTo6pvPxgF4hvI06IQ/QgjgGyKQpq+dYQztv /pPG7UI9Emosv8qyxy7yFjNVgCb+xMeHr2Ym+5M45kyR9vJn+bFy6UnpFk8GIBTg M0PIwzDjStSemXNZo7WbVrv2V/TA043t55TPuGDIO1DvVsoVtEQYsn3BmbUFpnY= =j+iR -END PGP SIGNATURE- ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: libi18nlangtag crash on SunOS 5.11
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 This seems to be liblangtag issue as its testsuite has problems as well. I'll need to get those resolved first. thx -BEGIN PGP SIGNATURE- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJUlo7FAAoJECAB22fHtp275RMH+wSSMSde5HrZ4IEghLICF7V5 X2QkRK8OrkpJ33v7LXvJH7TO0HGDxM+QeOdKQxyIo8mOHFWpMnB7e9LgfdelP7IS JopdQkKkvJvMpKTyYpZQJKcoGNCU9ikc5f+k56+urgeyPIrF5QaUZe2TCaH5Qf/l +JFsSsDq1lNR2KbSWn9tPHGLYJJh8EkpyVkc/zIwvUVWbSih11An3c+6guzHV1Hc hGPRjffnSV7DNa2psStiATFwTpCK0SMSlyYMZLQjIctuurulMCQIA5NsWP2PDvfs U25tQ8q8ovKPLVV/R+1JjYJ2EnAIV2IJKFpOgMagAGjwVmSKyzTRkDTvI3xKjVk= =+95j -END PGP SIGNATURE- ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
In glyph fallback throwing away the language property of en because the detected script for '0x9f3' is Bengali and that language doesn't make sense....
Yeah, sort of surprised myself. Running scalc 'LibreOfficeDev 4.3.4.1 430m0(Build:1)' I figured out that I can easily force a core by right clicking on any cell and selecting 'format cell' I get the following in the terminal window: richard@omnis:/home/richard$ soffice test.ods warn:sfx.control:101172:1:sfx2/source/control/dispatch.cxx:1505: Childwindow slot missing: 10365 right-click format cell here! warn:vcl:101172:1:vcl/generic/fontmanager/fontconfig.cxx:868: In glyph fallback throwing away the language property of en because the detected script for '0x9f 3' is Bengali and that language doesn't make sense. Autodetecting instead. My locale is fr_FR.UTF-8 on SunOS 5.11 under pkgsrc. it's coring with: sigmask = 0xbefc,0x,0x01ff cursig = SIGSEGV Coredump indicates: 08045fb8 libi18nlangtag.so`LiblantagDataRef::setup+0x4c(fe5b5bdc, fe5b5bdc, fe5b5a1c, 804605c) 08046128 libi18nlangtag.so`LanguageTagImpl::canonicalize+0x784(e34a900, 0, fe59b58b, fe5b28bc) 08046158 libi18nlangtag.so`_ZN15LanguageTagImpl15synCanonicalizeEv.part.179+0xf( fed8d824, 0, 8046198, feca4a02, 18, e33df90) 08046278 libi18nlangtag.so`LanguageTag::registerImpl+0x1a50(80462a0, 8046458, f8f52a40, fed1086b, fed8d830, 0) 080462b8 libi18nlangtag.so`LanguageTag::getImpl+0x50(80462e8, 8046458, 80462e8, fed8d000, fed8d824, 8046458) 08046308 libi18nlangtag.so`LanguageTag::getLanguage+0x24(804639c, 8046458, 8046338, fee83cba, e346860, f3) 08046358 libi18nlangtag.so`LanguageTag::getLanguageAndScript+0x21(804639c, 8046458, 80463d8, fcb1bb2e, fed8d824, 80464fc) 080463d8 libvcllo.so`_GLOBAL__N_1::mapToFontConfigLangTag+0x10b(f8f52a40, fed1086b, fcb819f0, e2e9e18, 804651c, 8046528) 080465c8 libvcllo.so`psp::PrintFontManager::Substitute+0x98b(fcc71b80, 80465f8, 804676c, fcb0325a, 8046684, 0) 08046678 libvcllo.so`FcGlyphFallbackSubstititution::FindFontSubstitute+0x140( fcc67480, 8046788, 804676c, 80467ac) 080466c8 libvcllo.so`PhysicalFontCollection::GetGlyphFallbackFont+0xbc(ce997c8, 8046788, 804676c, 0, 10, 8046770) 08046718 libvcllo.so`ImplFontCache::GetGlyphFallbackFont+0x37(ce91d78, ce997c8, 8046788, 1, 804676c, 8046774) 08046808 libvcllo.so`OutputDevice::ImplGlyphFallbackLayout+0x367(e31fc08, e33f138, 8046840, fed8d000, fed8d824, 8046840) 080468b8 libvcllo.so`OutputDevice::ImplLayout+0x12d(e31fc08, e346780, 0, 20, 804690c, 0) 080469d8 libvcllo.so`OutputDevice::GetTextArray+0x8c(e31fc08, e346780, 0, 0, , 0) 08046a08 libvcllo.so`OutputDevice::GetTextWidth+0x25(e31fc08, e346780, 0, , fed8d824, 0) 08046a78 libvcllo.so`ImplListBoxWindow::ImplUpdateEntryMetrics+0x285(e31fc08, e346780, e346780, 0) 08046aa8 libvcllo.so`_ZN17ImplListBoxWindow11InsertEntryElP13ImplEntryType.part.119+0x64( e346780, 481, e31bc18, a) 08046ae8 libvcllo.so`ImplListBox::InsertEntry+0x6e(e31fad8, 7fff, e3447f4, 548, f0ef6864, d4c1700) 08046b18 libvcllo.so`ListBox::InsertEntry+0x33(e31bc18, e3447f4, 7fff, fdfec5aa) 08046b68 libcuilo.so`SvxNumberFormatTabPage::FillCurrencyBox+0x74(e2f8f28, 8046bc4, 26, 0, f12c5d08, 0) 08046be8 libcuilo.so`SvxNumberFormatTabPage::Reset+0x342(e2f8f28, e2e6298, e2f8f28, fdef2c4c, 0, fdd8e2e4) 08046cb8 libsfxlo.so`SfxTabDialog::ActivatePageHdl+0x4ff(e2f50e0, e2f7f00, 8046cf0, 0, fed8d824, 0) 08046d28 libsfxlo.so`SfxTabDialog::Start_Impl+0xd5(e2f50e0, e2f50e0, 8046d68, f11f589e, f450d87c, e2e42a8) 08046d48 libsfxlo.so`SfxTabDialog::Execute+0x34(e2f50e0, c001, f11f586b, f450d87c) 08046d68 libscuilo.so`ScAbstractTabDialog_Impl::Execute+0x15(e2e42a8, d4d7508, d66e4c0, e2e6298, 0, 0) 08046e78 libsclo.so`ScTabViewShell::ExecuteCellFormatDlg+0x3bc(d66b430, 80472c0 , 8046efc, fed8d000, f8f52a40, fed8d824) 080470e8 libsclo.so`ScCellShell::Execute+0x2064(d612100, 80472c0, fe76c38b, fe13e51c) 08047108 libsclo.so`SfxStubScCellShellExecute+0x1d(d612100, 80472c0, fdda2330, 0 ) 08047178 libsfxlo.so`SfxDispatcher::Call_Impl+0x27a(d30a760, d612100, f46eadc4, 80472c0, 1, 0) 080471e8 libsfxlo.so`SfxBindings::Execute_Impl+0x66(d57c3a8, 80472c0, f46eadc4, d612100, 8068b80, 0) 08047368 libsfxlo.so`SfxDispatchController_Impl::dispatch+0xeef(e2f0798, e2f5658 , 80473f0, 8047394, fcbb6654, 8068a60) 080473b8 libsfxlo.so`SfxOfficeDispatch::dispatch+0xfc(e2f0760, e2f5658, 80473f0 , fed11d78, feee69e0, 0) 08047488 libsfxlo.so`SfxAsyncExec_Impl::TimerHdl+0x6b(e2f5658, e2f5688, 108, 137f0c7f) 080474a8 libvcllo.so`Timer::Timeout+0x19(e2f5688, fcbb6654, 80474e8, fca5b961, e2f2c94, 3e8) 080474e8 libvcllo.so`Timer::ImplTimerCallbackProc+0x191(fc539418, ce5d920, f8c891bb, fc538290, e1af058, e2f2c60) 08047508 libvclplug_gtklo.so`sal_gtk_timeout_dispatch+0x5a(e2f2c60, fc484bf0, 0 , 8047548, 912faf1e, 8047548) 08047568 libglib-2.0.so.0.4200.1`g_main_context_dispatch+0x129(ce5d920, 12c, d1e5860, 2) 080475b8 libglib-2.0.so.0.4200.1`g_main_context_iterate.isra.29+0x209(1, 0,
Re: libi18nlangtag crash on SunOS 5.11
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Le 19/12/14 16:02, Eike Rathke a écrit :
Hi Richard,
On Friday, 2014-12-19 14:18:36 +0100, Richard PALO wrote:
warn:vcl:101172:1:vcl/generic/fontmanager/fontconfig.cxx:868: In glyph
fallback
throwing away the language property of en because the detected script for
'0x9f
3' is Bengali and that language doesn't make sense. Autodetecting instead.
That warning is only indirectly related because something may trigger
initialization of liblangtag for an unknown locale value, I've seen it
without any crash. Adapted subject..
My locale is fr_FR.UTF-8 on SunOS 5.11 under pkgsrc.
it's coring with:
sigmask = 0xbefc,0x,0x01ff
cursig = SIGSEGV
Coredump indicates:
08045fb8 libi18nlangtag.so`LiblantagDataRef::setup+0x4c(fe5b5bdc, fe5b5bdc,
fe5b5a1c, 804605c)
08046128 libi18nlangtag.so`LanguageTagImpl::canonicalize+0x784(e34a900, 0,
fe59b58b, fe5b28bc)
That is the actual top of the backtrace?
There isn't much in i18nlangtag/source/languagetag/languagetag.cxx
LiblantagDataRef::setup() that could go wrong. The only thing that comes
to mind is that LiblantagDataRef is a static instance initialized with
namespace {
struct theDataRef : public rtl::Static LiblantagDataRef, theDataRef {};
}
which for some reason may not work on SunOS / your compiler combination.
As probably no one is using that here, you'd have to break in
LanguageTagImpl::canonicalize() line 1204 where
theDataRef::get().incRef() is called (or in the
LiblantagDataRef::incRef() inline) and investigate what actually happens
the very first time when the incRef() calls setup()
Other than that maybe the lt_db_initialize() call misbehaves, but since
that is not in the backtrace..
However, only a step-through could provide insights.
Eike
Okay, thanks. I'm building now an upgrade to 4.3.5.2 and will attempt to see
more in
the suggested areas. (I too noticed I'd be wasting my time in fontconfig, hence
my post:)
For info, build is with gcc4.9.2 and the back trace comes from 'mdb' on the
core file
with c++ demangling turned on.
cheers
-BEGIN PGP SIGNATURE-
Version: GnuPG v1
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
iQEcBAEBAgAGBQJUlEHYAAoJECAB22fHtp27V90H+gIPADtnj4ZPmEIyT00SLjJL
xob8DpRcllcNzk3nd9blPqC0EMF5cIz86d4iz/Ryeek+HEEYkwDQrNAdnNTSizzK
o1XG3cX+wpvE5j4upvivs+Wa4cTNYT5jwAmJIjRtNGFG50oYWvYouvQzWZyO0COP
jT2ZLszVIwH+BFafztsOJD6SbZl+d/CJk1PAGZaVmNzSYaZTGCflzbMRm0QFsm8Z
55n1HAL3ImI5P6r2yzOFlTRqXaw5nPsmDJCPHLGrtuc+F3OXQZS0II/v65zw1JFr
DvQRaIdXevVloYyvFJZtPmbhjEWRhsf584ejrkgoOnvqBKp1R46LEIrrtWdC1Dw=
=se6a
-END PGP SIGNATURE-
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice
Childwindow slot missing
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Starting up 'swriter', 'scalc' or such I notice the following warnings systematically : | richard@omnis:/home/richard$ swriter | warn:sfx.control:101391:1:sfx2/source/control/dispatch.cxx:1505: | Childwindow slot missing: 10365 | warn:sfx.control:101391:1:sfx2/source/control/dispatch.cxx:1505: | Childwindow slot missing: 10365 | warn:sfx.control:101391:1:sfx2/source/control/dispatch.cxx:1505: | Childwindow slot missing: 10365 Searching bugzilla it seems I'm not alone, although it doesn't seem to be the primary topic. Is this a concern needing a specific issue to be filed? -BEGIN PGP SIGNATURE- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJUZZZQAAoJECAB22fHtp27pVoIAI4SruXFMqhGXuW4I2uVbR7t XuKkPGv23sm1TZQUEFrV8mgCmUqus6NnvxrrBPmtDwXooTvQ5eiGBfUC9b8RcEwy JgSpmz2nDlqN2F3A6upVKQWw5F9JhVP7Y0t7sU+x/lbDW0mOaBiNQKc3MHbFCjug QHx7tqguDV2cE0il2SWenOaXN88UmqErYgQtS2BtC1c0fO0ciMNdD5H8qDuYrlZ2 ZjsLj8pLKphi82tsvIejQrjPseUNaFvfZWwKWiHHdyjiNTgTqfza6Y0k+HF3HL6c 3BJNHKfjvuttapFpFKeClntdz9yodt0iAp8wZTjTo/hVfOe7exvzK5NetStB7tU= =5rFL -END PGP SIGNATURE- ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] core.git: vcl/source
vcl/source/font/PhysicalFontCollection.cxx |4
vcl/source/outdev/text.cxx |4
2 files changed, 8 insertions(+)
New commits:
commit c22480ada409325a1c9a67dd06ce95b1028cb07e
Author: Richard PALO rich...@netbsd.org
Date: Mon Nov 10 17:46:43 2014 +0100
alloca needs special include on SOLARIS
Change-Id: I72415dbaf1350913938950eb1c0487a78cd67935
Signed-off-by: Michael Stahl mst...@redhat.com
diff --git a/vcl/source/font/PhysicalFontCollection.cxx
b/vcl/source/font/PhysicalFontCollection.cxx
index 56a8d74..dbff2c3 100644
--- a/vcl/source/font/PhysicalFontCollection.cxx
+++ b/vcl/source/font/PhysicalFontCollection.cxx
@@ -36,6 +36,10 @@
#include PhysicalFontCollection.hxx
+#ifdef SOLARIS
+#include alloca.h
+#endif
+
static unsigned lcl_IsCJKFont( const OUString rFontName )
{
// Test, if Fontname includes CJK characters -- In this case we
diff --git a/vcl/source/outdev/text.cxx b/vcl/source/outdev/text.cxx
index 2dd7859..151d888 100644
--- a/vcl/source/outdev/text.cxx
+++ b/vcl/source/outdev/text.cxx
@@ -50,6 +50,10 @@
#include graphite_features.hxx
#endif
+#ifdef SOLARIS
+#include alloca.h
+#endif
+
#define TEXT_DRAW_ELLIPSIS (TEXT_DRAW_ENDELLIPSIS | TEXT_DRAW_PATHELLIPSIS |
TEXT_DRAW_NEWSELLIPSIS)
ImplMultiTextLineInfo::ImplMultiTextLineInfo()
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] core.git: Branch 'libreoffice-4-3' - vcl/source
vcl/source/font/PhysicalFontCollection.cxx |4
vcl/source/outdev/text.cxx |4
2 files changed, 8 insertions(+)
New commits:
commit f0025f96439c8738b045329468fe3ce34a498f47
Author: Richard PALO rich...@netbsd.org
Date: Mon Nov 10 17:46:43 2014 +0100
alloca needs special include on SOLARIS
Signed-off-by: Michael Stahl mst...@redhat.com
(cherry picked from commit c22480ada409325a1c9a67dd06ce95b1028cb07e)
Change-Id: I72415dbaf1350913938950eb1c0487a78cd67935
diff --git a/vcl/source/font/PhysicalFontCollection.cxx
b/vcl/source/font/PhysicalFontCollection.cxx
index e195065..0bf7904 100644
--- a/vcl/source/font/PhysicalFontCollection.cxx
+++ b/vcl/source/font/PhysicalFontCollection.cxx
@@ -36,6 +36,10 @@
#include PhysicalFontCollection.hxx
+#ifdef SOLARIS
+#include alloca.h
+#endif
+
static OUString lcl_stripCharSetFromName(const OUString _aName)
{
// I worry that someone will have a font which *does* have
diff --git a/vcl/source/outdev/text.cxx b/vcl/source/outdev/text.cxx
index e426368..bb6369d 100644
--- a/vcl/source/outdev/text.cxx
+++ b/vcl/source/outdev/text.cxx
@@ -46,6 +46,10 @@
#include graphite_features.hxx
#endif
+#ifdef SOLARIS
+#include alloca.h
+#endif
+
#define TEXT_DRAW_ELLIPSIS (TEXT_DRAW_ENDELLIPSIS | TEXT_DRAW_PATHELLIPSIS |
TEXT_DRAW_NEWSELLIPSIS)
ImplMultiTextLineInfo::ImplMultiTextLineInfo()
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] core.git: Branch 'libreoffice-4-3' - solenv/gbuild
solenv/gbuild/platform/solaris.mk | 74 +++--- 1 file changed, 45 insertions(+), 29 deletions(-) New commits: commit d6afd367167ed406cd97192da358b776dbf5158e Author: Richard PALO rich...@netbsd.org Date: Mon Nov 10 17:36:30 2014 +0100 in general, SOLARIS should use /usr/bin/ld. Make equivalent to unxgcc.mk (with gld). NB: use $(READELF) updated in configure.ac and config_host.mk.in Change-Id: Iaf175fb7db095c567e9837eac8456c0c3809e334 Signed-off-by: Michael Stahl mst...@redhat.com diff --git a/solenv/gbuild/platform/solaris.mk b/solenv/gbuild/platform/solaris.mk index a781a6c..8256620 100644 --- a/solenv/gbuild/platform/solaris.mk +++ b/solenv/gbuild/platform/solaris.mk @@ -25,15 +25,6 @@ ifneq ($(origin AR),default) gb_AR := $(AR) endif -# use CC/CXX if they are nondefaults -ifneq ($(origin CC),default) -gb_CC := $(CC) -gb_GCCP := $(CC) -endif -ifneq ($(origin CXX),default) -gb_CXX := $(CXX) -endif - ifeq ($(CPUNAME),SPARC) gb_CPUDEFS := -D__sparcv8plus endif @@ -61,18 +52,21 @@ gb_CXXFLAGS := \ -fPIC \ -Wshadow \ -Woverloaded-virtual \ - -Wno-non-virtual-dtor \ - -# enable debug STL -ifeq ($(gb_ENABLE_DBGUTIL),$(true)) -gb_COMPILERDEFS += \ - -D_GLIBCXX_DEBUG \ - -endif ifeq ($(COM_GCC_IS_CLANG),) gb_GccLess460 := $(shell expr $(GCC_VERSION) \ 406) +# Only GCC 4.6 has a fix for http://gcc.gnu.org/bugzilla/show_bug.cgi?id=7302 +# -Wnon-virtual-dtor should't complain of protected dtor and supports #pragma +# GCC diagnostic push/pop required e.g. in cppuhelper/propertysetmixin.hxx to +# silence warnings about a protected, non-virtual dtor in a class with virtual +# members and friends: +ifeq ($(gb_GccLess460),1) +gb_CXXFLAGS += -Wno-non-virtual-dtor +else +gb_CXXFLAGS += -Wnon-virtual-dtor +endif + #At least SLED 10.2 gcc 4.3 overly aggressively optimizes uno::Sequence into #junk, so only strict-alias on = 4.6.0 gb_StrictAliasingUnsafe := $(gb_GccLess460) @@ -81,12 +75,23 @@ ifeq ($(gb_StrictAliasingUnsafe),1) gb_CFLAGS += -fno-strict-aliasing gb_CXXFLAGS += -fno-strict-aliasing endif + +else # Clang +gb_CXXFLAGS += -Wnon-virtual-dtor +endif + + +# enable debug STL +ifeq ($(gb_ENABLE_DBGUTIL),$(true)) +gb_COMPILERDEFS += \ + -D_GLIBCXX_DEBUG \ + endif ifeq ($(HAVE_CXX11),TRUE) #Currently, as well as for its own merits, c++11/c++0x mode allows use to use #a template for SAL_N_ELEMENTS to detect at compiler time its misuse -gb_CXXFLAGS += -std=c++0x +gb_CXXFLAGS += $(CXXFLAGS_CXX11) #We have so many std::auto_ptr uses that we need to be able to disable #warnings for those so that -Werror continues to be useful, seeing as moving @@ -107,8 +112,6 @@ gb_LinkTarget_LDFLAGS += \ -Wl,--sysroot=$(SYSBASE) endif -#JAD# -Wl,-rpath-link,$(SYSBASE)/lib:$(SYSBASE)/usr/lib \ - gb_LinkTarget_LDFLAGS += \ -L$(SYSBASE)/lib \ -L$(SYSBASE)/usr/lib \ @@ -134,6 +137,12 @@ else gb_LINKEROPTFLAGS := -Wl,-O1 endif +ifeq ($(gb_SYMBOL),$(true)) +gb_LINKERSTRIPDEBUGFLAGS := +else +gb_LINKERSTRIPDEBUGFLAGS := -Wl,-zredlocsym -Wl,-znoldynsym +endif + # LinkTarget class define gb_LinkTarget__get_rpath_for_layer @@ -152,18 +161,20 @@ gb_LinkTarget_CFLAGS := $(gb_CFLAGS) $(gb_CFLAGS_WERROR) gb_LinkTarget_CXXFLAGS := $(gb_CXXFLAGS) $(gb_CFLAGS_WERROR) ifeq ($(gb_SYMBOL),$(true)) -gb_LinkTarget_CXXFLAGS += -ggdb2 -gb_LinkTarget_CFLAGS += -ggdb2 +gb_LinkTarget_CXXFLAGS += $(GGDB2) +gb_LinkTarget_CFLAGS += $(GGDB2) endif # note that `cat $(extraobjectlist)` is needed to build with older gcc versions, e.g. 4.1.2 on SLED10 # we want to use @$(extraobjectlist) in the long run +# link with C compiler if there are no C++ files (pyuno_wrapper depends on this) define gb_LinkTarget__command_dynamiclink $(call gb_Helper_abbreviate_dirs,\ $(if $(CXXOBJECTS)$(GENCXXOBJECTS)$(EXTRAOBJECTLISTS),$(gb_CXX),$(gb_CC)) \ $(if $(filter Library CppunitTest,$(TARGETTYPE)),$(gb_Library_TARGETTYPEFLAGS)) \ - $(if $(SOVERSIONSCRIPT),-Wl$(COMMA)--soname=$(notdir $(1)) \ - -Wl$(COMMA)--version-script=$(SOVERSIONSCRIPT)) \ + $(gb_LTOFLAGS) \ + $(if $(SOVERSIONSCRIPT),-Wl$(COMMA)-soname=$(notdir $(1)) \ + ) \ $(subst \d,$$,$(RPATH)) \ $(T_LDFLAGS) \ $(foreach object,$(COBJECTS),$(call gb_CObject_get_target,$(object))) \ @@ -177,12 +188,12 @@ $(call gb_Helper_abbreviate_dirs,\ $(call gb_StaticLibrary_get_target,$(lib))) \ $(T_LIBS) \ -Wl$(COMMA)--end-group \ - -Wl$(COMMA)--no-as-needed \ + -Wl$(COMMA)-zrecord \ $(patsubst lib%.a,-l%,$(patsubst lib%.so,-l%,$(patsubst %.$(gb_Library_UDK_MAJORVER),%,$(foreach lib,$(LINKED_LIBS),$(call gb_Library_get_filename,$(lib)) \ -o $(1) \ $(if $(SOVERSIONSCRIPT
[Libreoffice-commits] core.git: solenv/gbuild
solenv/gbuild/platform/solaris.mk | 44 +++--- 1 file changed, 23 insertions(+), 21 deletions(-) New commits: commit 573506c778410fc0eb6afc0c07bdd849613503a8 Author: Richard PALO rich...@netbsd.org Date: Mon Nov 10 17:36:30 2014 +0100 in general, SOLARIS should use /usr/bin/ld. Make equivalent to unxgcc.mk (with gld). NB: use $(READELF) updated in configure.ac and config_host.mk.in Signed-off-by: Michael Stahl mst...@redhat.com (cherry picked from commit d6afd367167ed406cd97192da358b776dbf5158e) Conflicts: solenv/gbuild/platform/solaris.mk note: while cherry-picking to master, removed all mention of toolchains that are unsupported on master due to lack of C++11 support (GCC 4.6). Change-Id: Iaf175fb7db095c567e9837eac8456c0c3809e334 Reviewed-on: https://gerrit.libreoffice.org/12384 Reviewed-by: Michael Stahl mst...@redhat.com Tested-by: Michael Stahl mst...@redhat.com diff --git a/solenv/gbuild/platform/solaris.mk b/solenv/gbuild/platform/solaris.mk index 1d3c256..3b44e8c 100644 --- a/solenv/gbuild/platform/solaris.mk +++ b/solenv/gbuild/platform/solaris.mk @@ -25,15 +25,6 @@ ifneq ($(origin AR),default) gb_AR := $(AR) endif -# use CC/CXX if they are nondefaults -ifneq ($(origin CC),default) -gb_CC := $(CC) -gb_GCCP := $(CC) -endif -ifneq ($(origin CXX),default) -gb_CXX := $(CXX) -endif - ifeq ($(CPUNAME),SPARC) gb_CPUDEFS := -D__sparcv8plus endif @@ -61,7 +52,7 @@ gb_CXXFLAGS := \ -fPIC \ -Wshadow \ -Woverloaded-virtual \ - -std=c++0x \ + $(CXXFLAGS_CXX11) \ # enable debug STL ifeq ($(gb_ENABLE_DBGUTIL),$(true)) @@ -77,8 +68,6 @@ gb_LinkTarget_LDFLAGS += \ -Wl,--sysroot=$(SYSBASE) endif -#JAD# -Wl,-rpath-link,$(SYSBASE)/lib:$(SYSBASE)/usr/lib \ - gb_LinkTarget_LDFLAGS += \ -L$(SYSBASE)/lib \ -L$(SYSBASE)/usr/lib \ @@ -104,6 +93,12 @@ else gb_LINKEROPTFLAGS := -Wl,-O1 endif +ifeq ($(gb_SYMBOL),$(true)) +gb_LINKERSTRIPDEBUGFLAGS := +else +gb_LINKERSTRIPDEBUGFLAGS := -Wl,-zredlocsym -Wl,-znoldynsym +endif + # LinkTarget class define gb_LinkTarget__get_rpath_for_layer @@ -122,18 +117,20 @@ gb_LinkTarget_CFLAGS := $(gb_CFLAGS) $(gb_CFLAGS_WERROR) gb_LinkTarget_CXXFLAGS := $(gb_CXXFLAGS) $(gb_CFLAGS_WERROR) ifeq ($(gb_SYMBOL),$(true)) -gb_LinkTarget_CXXFLAGS += -ggdb2 -gb_LinkTarget_CFLAGS += -ggdb2 +gb_LinkTarget_CXXFLAGS += $(GGDB2) +gb_LinkTarget_CFLAGS += $(GGDB2) endif # note that `cat $(extraobjectlist)` is needed to build with older gcc versions, e.g. 4.1.2 on SLED10 # we want to use @$(extraobjectlist) in the long run +# link with C compiler if there are no C++ files (pyuno_wrapper depends on this) define gb_LinkTarget__command_dynamiclink $(call gb_Helper_abbreviate_dirs,\ $(if $(CXXOBJECTS)$(GENCXXOBJECTS)$(EXTRAOBJECTLISTS),$(gb_CXX),$(gb_CC)) \ $(if $(filter Library CppunitTest,$(TARGETTYPE)),$(gb_Library_TARGETTYPEFLAGS)) \ - $(if $(SOVERSIONSCRIPT),-Wl$(COMMA)--soname=$(notdir $(1)) \ - -Wl$(COMMA)--version-script=$(SOVERSIONSCRIPT)) \ + $(gb_LTOFLAGS) \ + $(if $(SOVERSIONSCRIPT),-Wl$(COMMA)-soname=$(notdir $(1)) \ + ) \ $(subst \d,$$,$(RPATH)) \ $(T_LDFLAGS) \ $(foreach object,$(COBJECTS),$(call gb_CObject_get_target,$(object))) \ @@ -147,12 +144,12 @@ $(call gb_Helper_abbreviate_dirs,\ $(call gb_StaticLibrary_get_target,$(lib))) \ $(T_LIBS) \ -Wl$(COMMA)--end-group \ - -Wl$(COMMA)--no-as-needed \ + -Wl$(COMMA)-zrecord \ $(patsubst lib%.a,-l%,$(patsubst lib%.so,-l%,$(patsubst %.$(gb_Library_UDK_MAJORVER),%,$(foreach lib,$(LINKED_LIBS),$(call gb_Library_get_filename,$(lib)) \ -o $(1) \ $(if $(SOVERSIONSCRIPT), ln -sf ../../ure-link/lib/$(notdir $(1)) $(ILIBTARGET))) $(if $(filter Library,$(TARGETTYPE)), $(call gb_Helper_abbreviate_dirs,\ - readelf -d $(1) | grep SONAME $(WORKDIR)/LinkTarget/$(2).exports.tmp; \ + $(READELF) -d $(1) | grep SONAME $(WORKDIR)/LinkTarget/$(2).exports.tmp; \ $(NM) --dynamic --extern-only --defined-only --format=posix $(1) \ | cut -d' ' -f1-2 \ $(WORKDIR)/LinkTarget/$(2).exports.tmp \ @@ -164,6 +161,7 @@ define gb_LinkTarget__command_staticlink $(call gb_Helper_abbreviate_dirs,\ rm -f $(1) \ $(gb_AR) -rsu $(1) \ + $(if $(LD_PLUGIN),--plugin $(LD_PLUGIN)) \ $(foreach object,$(COBJECTS),$(call gb_CObject_get_target,$(object))) \ $(foreach object,$(CXXOBJECTS),$(call gb_CxxObject_get_target,$(object))) \ $(foreach object,$(ASMOBJECTS),$(call gb_AsmObject_get_target,$(object
[Libreoffice-commits] core.git: 7 commits - config_host.mk.in configure.ac extensions/Executable_pluginapp.bin.mk extensions/source sal/osl shell/Module_shell.mk shell/source sysui/CustomTarget_solari
config_host.mk.in |1 + configure.ac|4 +++- extensions/Executable_pluginapp.bin.mk | 17 + extensions/source/plugin/inc/plugin/unx/plugcon.hxx |4 sal/osl/unx/file.cxx|2 +- sal/osl/unx/system.h|2 +- shell/Module_shell.mk |8 shell/source/unix/sysshell/recently_used_file.cxx |2 +- sysui/CustomTarget_solaris.mk |6 +++--- 9 files changed, 19 insertions(+), 27 deletions(-) New commits: commit 88a40e70fe921680ed303b8ab69ca866ae67078e Author: Richard PALO rich...@netbsd.org Date: Sun Nov 9 18:15:45 2014 +0100 O_RDWR and O_CREAT are defined in fcntl.h Change-Id: I568ee2c3f10320f6c39340a5915b10aeebb4 Signed-off-by: Michael Stahl mst...@redhat.com diff --git a/shell/source/unix/sysshell/recently_used_file.cxx b/shell/source/unix/sysshell/recently_used_file.cxx index 2189d0b..08f96a3 100644 --- a/shell/source/unix/sysshell/recently_used_file.cxx +++ b/shell/source/unix/sysshell/recently_used_file.cxx @@ -27,7 +27,7 @@ #include sys/file.h #include sys/types.h #include sys/stat.h - +#include fcntl.h #include unistd.h const OUString RECENTLY_USED_FILE_NAME(.recently-used); commit 457105c7d9ede7a4d73d69d1c87306e88d57c0ac Author: Richard PALO rich...@netbsd.org Date: Mon Nov 10 17:45:39 2014 +0100 update solaris directories being referenced also borrowed from Gabriele Bulfon gabriele.bulfon at sonicle.com Change-Id: I17787fbbf12f1bc8b3e78b8dd2e8b33679bab5b1 Signed-off-by: Michael Stahl mst...@redhat.com diff --git a/sysui/CustomTarget_solaris.mk b/sysui/CustomTarget_solaris.mk index 9f671a8..ed264fe 100644 --- a/sysui/CustomTarget_solaris.mk +++ b/sysui/CustomTarget_solaris.mk @@ -45,11 +45,11 @@ $(solaris_WORKDIR)/%/postinstall: $(solaris_SRCDIR)/postinstall $(solaris_WORKDIR)/%/postremove: $(solaris_SRCDIR)/postremove cat $ | tr -d \015 $@ -$(solaris_WORKDIR)/%/mailcap: $(solaris_SRCDIR)/prototype +$(solaris_WORKDIR)/%/prototype: $(solaris_SRCDIR)/prototype cat $ | tr -d \015 | sed -e s/%PREFIX/$(UNIXFILENAME.$*)/g -e s_%SOURCE_$(call gb_CustomTarget_get_workdir,sysui/share)/$*_g -e s/%ICONPREFIX/$(UNIXFILENAME.$*)/g $@ -$(solaris_WORKDIR)/%-desktop-integration.tar.gz: $(solaris_WORKDIR)/copyright $(solaris_WORKDIR)/pkginfo $(solaris_WORKDIR)/depend $(solaris_WORKDIR)/mailcap $(solaris_WORKDIR)/postinstall $(solaris_WORKDIR)/postremove $(solaris_WORKDIR)/prototype $(call gb_CustomTarget_get_workdir,sysui/share)/%/openoffice.org.xml - pkgmk -l 1073741824 -r $(solaris_WORKDIR) -f $(solaris_WORKDIR)/$*/prototype -o -d $(solaris_WORKDIR) ARCH=all VERSION=$(PKGVERSION.$*) +$(solaris_WORKDIR)/%-desktop-integration.tar.gz: $(solaris_WORKDIR)/%/copyright $(solaris_WORKDIR)/%/pkginfo $(solaris_WORKDIR)/%/depend $(solaris_WORKDIR)/%/mailcap $(solaris_WORKDIR)/%/postinstall $(solaris_WORKDIR)/%/postremove $(solaris_WORKDIR)/%/prototype $(call gb_CustomTarget_get_workdir,sysui/share)/%/openoffice.org.xml +# pkgmk -l 1073741824 -r $(solaris_WORKDIR) -f $(solaris_WORKDIR)/$*/prototype -o -d $(solaris_WORKDIR) ARCH=all VERSION=$(PKGVERSION.$*) $(GNUTAR) -cf - -C $(solaris_WORKDIR) $*$(LIBO_MAJOR) -desktop-int | gzip $@ # vim: set noet sw=4 ts=4: commit a395a395e52af67e943eaa80fc003954fe44efae Author: Richard PALO rich...@netbsd.org Date: Mon Nov 10 17:33:47 2014 +0100 Do as for Linux, borrowed from Gabriele Bulfon gabriele.bulfon at sonicle.com Change-Id: Id15c4b355591678023ae97b1183f8f3a01823084 Signed-off-by: Michael Stahl mst...@redhat.com diff --git a/shell/Module_shell.mk b/shell/Module_shell.mk index 48c7cc3..46f081d 100644 --- a/shell/Module_shell.mk +++ b/shell/Module_shell.mk @@ -22,6 +22,14 @@ $(eval $(call gb_Module_add_targets,shell,\ endif endif +ifeq ($(OS),SOLARIS) +ifeq ($(ENABLE_GIO),TRUE) +$(eval $(call gb_Module_add_targets,shell,\ +Library_losessioninstall \ +)) +endif +endif + ifeq ($(ENABLE_GCONF),TRUE) $(eval $(call gb_Module_add_targets,shell,\ Library_gconfbe \ commit 5f35b74bb1367a07b1f2d85c6665b31c906b9538 Author: Richard PALO rich...@netbsd.org Date: Mon Nov 10 17:30:44 2014 +0100 improve SOLARIS guard Change-Id: Ic5edb7c0977b6c589329cd80616aa320d6cc929b Signed-off-by: Michael Stahl mst...@redhat.com diff --git a/sal/osl/unx/system.h b/sal/osl/unx/system.h index d952518..b3162f1 100644 --- a/sal/osl/unx/system.h +++ b/sal/osl/unx/system.h @@ -43,7 +43,7 @@ #include sys/types.h /* Make sockets of type AF_UNIX use underlying FS rights */ -#ifdef SOLARIS +#if defined(SOLARIS) !defined(_XOPEN_SOURCE) # define _XOPEN_SOURCE 500 # include sys/socket.h # undef _XOPEN_SOURCE commit 92c5bc7d6bc9dec429114a0f72f6f5989d70f731 Author: Richard PALO rich
[Libreoffice-commits] core.git: Branch 'libreoffice-4-3' - 8 commits - bridges/Library_cpp_uno.mk config_host.mk.in configure.ac extensions/Executable_pluginapp.bin.mk extensions/source sal/osl shell/
bridges/Library_cpp_uno.mk |8 +++- config_host.mk.in |1 + configure.ac|4 +++- extensions/Executable_pluginapp.bin.mk | 17 + extensions/source/plugin/inc/plugin/unx/plugcon.hxx |4 sal/osl/unx/file.cxx|2 +- sal/osl/unx/system.h|2 +- shell/Module_shell.mk |8 shell/source/unix/sysshell/recently_used_file.cxx |2 +- sysui/CustomTarget_solaris.mk |6 +++--- 10 files changed, 22 insertions(+), 32 deletions(-) New commits: commit d2a617aac6466e246de6209731791c1ce835ee62 Author: Richard PALO rich...@netbsd.org Date: Sun Nov 9 18:15:45 2014 +0100 O_RDWR and O_CREAT are defined in fcntl.h Change-Id: I568ee2c3f10320f6c39340a5915b10aeebb4 Signed-off-by: Michael Stahl mst...@redhat.com (cherry picked from commit 88a40e70fe921680ed303b8ab69ca866ae67078e) diff --git a/shell/source/unix/sysshell/recently_used_file.cxx b/shell/source/unix/sysshell/recently_used_file.cxx index 2189d0b..08f96a3 100644 --- a/shell/source/unix/sysshell/recently_used_file.cxx +++ b/shell/source/unix/sysshell/recently_used_file.cxx @@ -27,7 +27,7 @@ #include sys/file.h #include sys/types.h #include sys/stat.h - +#include fcntl.h #include unistd.h const OUString RECENTLY_USED_FILE_NAME(.recently-used); commit 7997129c449b6cbf4c5e7dd4e3a982dc23da7720 Author: Richard PALO rich...@netbsd.org Date: Mon Nov 10 17:45:39 2014 +0100 update solaris directories being referenced also borrowed from Gabriele Bulfon gabriele.bulfon at sonicle.com Change-Id: I17787fbbf12f1bc8b3e78b8dd2e8b33679bab5b1 Signed-off-by: Michael Stahl mst...@redhat.com (cherry picked from commit 457105c7d9ede7a4d73d69d1c87306e88d57c0ac) diff --git a/sysui/CustomTarget_solaris.mk b/sysui/CustomTarget_solaris.mk index 9f671a8..ed264fe 100644 --- a/sysui/CustomTarget_solaris.mk +++ b/sysui/CustomTarget_solaris.mk @@ -45,11 +45,11 @@ $(solaris_WORKDIR)/%/postinstall: $(solaris_SRCDIR)/postinstall $(solaris_WORKDIR)/%/postremove: $(solaris_SRCDIR)/postremove cat $ | tr -d \015 $@ -$(solaris_WORKDIR)/%/mailcap: $(solaris_SRCDIR)/prototype +$(solaris_WORKDIR)/%/prototype: $(solaris_SRCDIR)/prototype cat $ | tr -d \015 | sed -e s/%PREFIX/$(UNIXFILENAME.$*)/g -e s_%SOURCE_$(call gb_CustomTarget_get_workdir,sysui/share)/$*_g -e s/%ICONPREFIX/$(UNIXFILENAME.$*)/g $@ -$(solaris_WORKDIR)/%-desktop-integration.tar.gz: $(solaris_WORKDIR)/copyright $(solaris_WORKDIR)/pkginfo $(solaris_WORKDIR)/depend $(solaris_WORKDIR)/mailcap $(solaris_WORKDIR)/postinstall $(solaris_WORKDIR)/postremove $(solaris_WORKDIR)/prototype $(call gb_CustomTarget_get_workdir,sysui/share)/%/openoffice.org.xml - pkgmk -l 1073741824 -r $(solaris_WORKDIR) -f $(solaris_WORKDIR)/$*/prototype -o -d $(solaris_WORKDIR) ARCH=all VERSION=$(PKGVERSION.$*) +$(solaris_WORKDIR)/%-desktop-integration.tar.gz: $(solaris_WORKDIR)/%/copyright $(solaris_WORKDIR)/%/pkginfo $(solaris_WORKDIR)/%/depend $(solaris_WORKDIR)/%/mailcap $(solaris_WORKDIR)/%/postinstall $(solaris_WORKDIR)/%/postremove $(solaris_WORKDIR)/%/prototype $(call gb_CustomTarget_get_workdir,sysui/share)/%/openoffice.org.xml +# pkgmk -l 1073741824 -r $(solaris_WORKDIR) -f $(solaris_WORKDIR)/$*/prototype -o -d $(solaris_WORKDIR) ARCH=all VERSION=$(PKGVERSION.$*) $(GNUTAR) -cf - -C $(solaris_WORKDIR) $*$(LIBO_MAJOR) -desktop-int | gzip $@ # vim: set noet sw=4 ts=4: commit 329429cacf5e1117645bcfed52a15cbd720090a9 Author: Richard PALO rich...@netbsd.org Date: Mon Nov 10 17:33:47 2014 +0100 Do as for Linux, borrowed from Gabriele Bulfon gabriele.bulfon at sonicle.com Change-Id: Id15c4b355591678023ae97b1183f8f3a01823084 Signed-off-by: Michael Stahl mst...@redhat.com (cherry picked from commit a395a395e52af67e943eaa80fc003954fe44efae) diff --git a/shell/Module_shell.mk b/shell/Module_shell.mk index 3c51964..27158e4 100644 --- a/shell/Module_shell.mk +++ b/shell/Module_shell.mk @@ -22,6 +22,14 @@ $(eval $(call gb_Module_add_targets,shell,\ endif endif +ifeq ($(OS),SOLARIS) +ifeq ($(ENABLE_GIO),TRUE) +$(eval $(call gb_Module_add_targets,shell,\ +Library_losessioninstall \ +)) +endif +endif + ifeq ($(ENABLE_GCONF),TRUE) $(eval $(call gb_Module_add_targets,shell,\ Library_gconfbe \ commit 46fe02b89b2a7f93930f16a399d48efaae0232a4 Author: Richard PALO rich...@netbsd.org Date: Mon Nov 10 17:30:44 2014 +0100 improve SOLARIS guard Change-Id: Ic5edb7c0977b6c589329cd80616aa320d6cc929b Signed-off-by: Michael Stahl mst...@redhat.com (cherry picked from commit 5f35b74bb1367a07b1f2d85c6665b31c906b9538) diff --git a/sal/osl/unx/system.h b/sal/osl/unx/system.h index 3a0babd..2cdf7fd 100644 --- a/sal/osl
Richard PALO license statement
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 All of my past future contributions to LibreOffice may be licensed under the MPLv2/LGPLv3+ dual license. -BEGIN PGP SIGNATURE- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJUX1aIAAoJECAB22fHtp27NtIIAKa7Cq3XcrnlSaUva0BneApT jun4n+iuIc4GRHRINoVkSpUrDqnR2csvG0Qc+CfyTZr7/FPGZmMFKpxRpA5zivEx sd17eAqLWnI50WEyoPYD08KLDcj7S2/WqGhhBcabYARPrzD1wbv+DiuAFx40igsl 9IJsn5Yz+uj2Ip0Uj5u8LQN1HhwkM1UfLgvten0mGQwWWvrvQioOFUwFLyFYnRjL oGTtuTMlhjJMivZ/XyFt6VyrTBWU5qnq34P/YDoz993hVbRGpj5HlkWNMaBmg4yp wGhT+kgmotK4mdGXpsR4kJu/VCj5Gn1GVXlXumt/lBIeZmFwnIXD4cxI/4pwN0Q= =fpGy -END PGP SIGNATURE- ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: ModuleUIConfigurationManager
Le 30/05/14 12:47, Richard PALO a écrit : Le 30/05/14 10:45, Stephan Bergmann a écrit : ... For libfwklo.so, the relevant symbol that needs to be exported is fwk_component_getFactory (denoting a C function). Stephan Hmm, that seems to be okay: richard@devzone:/opt/local/libreoffice4-4.2.4.2/program$ elfdump -s libfwklo.so |grep fwk_component_getFactory [957] 0x001e3ab0 0x2408 FUNC GLOB D1 .text fwk_component_getFactory [23029] 0x001e3ab0 0x2408 FUNC GLOB D0 .text fwk_component_getFactory !indexvalue size type bind oth ver shndx name In relooking at services.rdb, thought I'd try seeing if the many references to $LO_LIB_DIR was a necessary external env variable... (is this documented?) This is now the failure: richard@devzone:~$ export LO_LIB_DIR=/opt/local/libreoffice4-4.2.4.2/ure-link/lib richard@devzone:~$ /opt/local/bin/soffice terminate called after throwing an instance of 'com::sun::star::loader::CannotActivateFactoryException' Guess I'll have to build a debug kit later this week... ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: ModuleUIConfigurationManager
Le 03/06/14 22:33, Richard PALO a écrit : Le 30/05/14 12:47, Richard PALO a écrit : Le 30/05/14 10:45, Stephan Bergmann a écrit : ... For libfwklo.so, the relevant symbol that needs to be exported is fwk_component_getFactory (denoting a C function). Stephan Hmm, that seems to be okay: richard@devzone:/opt/local/libreoffice4-4.2.4.2/program$ elfdump -s libfwklo.so |grep fwk_component_getFactory [957] 0x001e3ab0 0x2408 FUNC GLOB D1 .text fwk_component_getFactory [23029] 0x001e3ab0 0x2408 FUNC GLOB D0 .text fwk_component_getFactory !indexvalue size type bind oth ver shndx name In relooking at services.rdb, thought I'd try seeing if the many references to $LO_LIB_DIR was a necessary external env variable... (is this documented?) This is now the failure: richard@devzone:~$ export LO_LIB_DIR=/opt/local/libreoffice4-4.2.4.2/ure-link/lib richard@devzone:~$ /opt/local/bin/soffice terminate called after throwing an instance of 'com::sun::star::loader::CannotActivateFactoryException' Guess I'll have to build a debug kit later this week... I notice after closer inspection that LO_LIB_DIR looks for libraries in the 'program' directory and not in 'ure-link/lib'... unfortunately setting these variable(s) (also a LO_JAVA_DIR one) seem to do nothing (actually, the result is the same as the attachment in my original post) so back to the initial square. ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: ModuleUIConfigurationManager
Le 30/05/14 09:42, Stephan Bergmann a écrit : On 05/29/2014 06:05 PM, Richard PALO wrote: Hi, guess I'm looking for hints on how to track down a startup problem (attached is the screenshot) Is this a local LO build of your own? If yes, how did you configure it? What platform is this? This is SunOS ABI=32 (illumos) using pkgsrc current and gcc 4.7 = $ ./configure --enable-ext-wiki-publisher --with-java --with-jdk-home=/opt/loc al/java/openjdk7 --with-ant-home= --enable-scripting-beanshell --enable-scriptin g-javascript --with-system-hsqldb --with-hsqldb-jar=/opt/local/lib/java/hsqldb18 /hsqldb.jar --enable-debug --enable-selective-debuginfo=all --disable-gtk3 --dis able-kde4 --prefix=/opt/local/libreoffice4-4.2.4.2 --disable-ccache --enable-dbu s --disable-epm --disable-fetch-external --disable-firebird-sdbc --disable-gconf --disable-gnome-vfs --disable-odk --disable-opengl --disable-randr-link --disab le-systray --disable-crashdump --disable-gio --enable-gstreamer --enable-lockdow n --enable-python=system --enable-ext-wiki-publisher --enable-ext-nlpsolver --wi th-alloc=system --with-compat-oowrappers --with-external-tar=/export/pub/pkgsrc/ distfiles/libreoffice --with-fonts --with-lang=ALL --with-myspell-dicts --with-p arallelism=6 --with-boost-libdir=/opt/local/lib --with-system-libs --with-system -dicts --enable-scripting-beanshell --without-system-beanshell --enable-scriptin g-javascript --without-system-vigra --with-vendor=pkgsrc --without-doxygen --wit hout-junit --without-krb5 --without-gssapi --without-system-apache-commons --wit h-system-cppunit --without-system-jfreereport --without-system-npapi-headers --w ithout-system-sane ac_cv_path_SYSTEM_GENCCODE=/opt/local/sbin/genccode ac_cv_pat h_SYSTEM_GENCMN=/opt/local/sbin/gencmn --with-gnu-patch=/opt/local/bin/gpatch -- with-gnu-cp=/opt/local/bin/gcp --srcdir=/var/tmp/pkgsrc/misc/libreoffice4/work/l ibreoffice-4.2.4.2 --enable-option-checking=fatal === Can't launch application. component context fails to supply service ...ModuleUIConfigurationManager of type ...XModuleUIConfigurationManager2: an error occured during file opening. program/services/services.rdb should record that UNO service com.sun.star.ui.ModuleUIConfigurationManager is provided by the fwk library. Relevant places in the code are loadSharedLibComponentFactory (cppuhelper/source/shlib.cxx) that would load the fwk library (but this is called a lot, for all the different libs; uri would contain the file URL of the fwk lib) and ModuleUIConfigurationManager::ModuleUIConfigurationManager (framework/source/uiconfiguration/moduleuiconfigurationmanager.cxx) where the C++ object implementing the UNO service is instantiated. anybody already come across this java problem? What makes you assume that it is a Java problem? Sorry, perhaps a hasty assumption. guess 'com.sun.star.ui...' threw me. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: ModuleUIConfigurationManager
Le 30/05/14 10:02, Richard PALO a écrit : Le 30/05/14 09:42, Stephan Bergmann a écrit : On 05/29/2014 06:05 PM, Richard PALO wrote: Hi, guess I'm looking for hints on how to track down a startup problem (attached is the screenshot) Is this a local LO build of your own? If yes, how did you configure it? What platform is this? This is SunOS ABI=32 (illumos) using pkgsrc current and gcc 4.7 = $ ./configure --enable-ext-wiki-publisher --with-java --with-jdk-home=/opt/loc al/java/openjdk7 --with-ant-home= --enable-scripting-beanshell --enable-scriptin g-javascript --with-system-hsqldb --with-hsqldb-jar=/opt/local/lib/java/hsqldb18 /hsqldb.jar --enable-debug --enable-selective-debuginfo=all --disable-gtk3 --dis able-kde4 --prefix=/opt/local/libreoffice4-4.2.4.2 --disable-ccache --enable-dbu s --disable-epm --disable-fetch-external --disable-firebird-sdbc --disable-gconf --disable-gnome-vfs --disable-odk --disable-opengl --disable-randr-link --disab le-systray --disable-crashdump --disable-gio --enable-gstreamer --enable-lockdow n --enable-python=system --enable-ext-wiki-publisher --enable-ext-nlpsolver --wi th-alloc=system --with-compat-oowrappers --with-external-tar=/export/pub/pkgsrc/ distfiles/libreoffice --with-fonts --with-lang=ALL --with-myspell-dicts --with-p arallelism=6 --with-boost-libdir=/opt/local/lib --with-system-libs --with-system -dicts --enable-scripting-beanshell --without-system-beanshell --enable-scriptin g-javascript --without-system-vigra --with-vendor=pkgsrc --without-doxygen --wit hout-junit --without-krb5 --without-gssapi --without-system-apache-commons --wit h-system-cppunit --without-system-jfreereport --without-system-npapi-headers --w ithout-system-sane ac_cv_path_SYSTEM_GENCCODE=/opt/local/sbin/genccode ac_cv_pat h_SYSTEM_GENCMN=/opt/local/sbin/gencmn --with-gnu-patch=/opt/local/bin/gpatch -- with-gnu-cp=/opt/local/bin/gcp --srcdir=/var/tmp/pkgsrc/misc/libreoffice4/work/l ibreoffice-4.2.4.2 --enable-option-checking=fatal === Can't launch application. component context fails to supply service ...ModuleUIConfigurationManager of type ...XModuleUIConfigurationManager2: an error occured during file opening. program/services/services.rdb should record that UNO service com.sun.star.ui.ModuleUIConfigurationManager is provided by the fwk library. Relevant places in the code are loadSharedLibComponentFactory (cppuhelper/source/shlib.cxx) that would load the fwk library (but this is called a lot, for all the different libs; uri would contain the file URL of the fwk lib) and ModuleUIConfigurationManager::ModuleUIConfigurationManager (framework/source/uiconfiguration/moduleuiconfigurationmanager.cxx) where the C++ object implementing the UNO service is instantiated. anybody already come across this java problem? What makes you assume that it is a Java problem? Sorry, perhaps a hasty assumption. guess 'com.sun.star.ui...' threw me. Stephan after objdumping libfwklo.so, I believe this may be perhaps related to my dropping the mapfile given it's invalid on solaris (with sun ld) because of the inappropriate wildcarding syntax used. Is there any work advancing on fixing these? I may need to come up with alternative mapfiles where necessary. (too bad) ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: ModuleUIConfigurationManager
Le 30/05/14 10:45, Stephan Bergmann a écrit : ... For libfwklo.so, the relevant symbol that needs to be exported is fwk_component_getFactory (denoting a C function). Stephan Hmm, that seems to be okay: richard@devzone:/opt/local/libreoffice4-4.2.4.2/program$ elfdump -s libfwklo.so |grep fwk_component_getFactory [957] 0x001e3ab0 0x2408 FUNC GLOB D1 .text fwk_component_getFactory [23029] 0x001e3ab0 0x2408 FUNC GLOB D0 .text fwk_component_getFactory !indexvalue size type bind oth ver shndx name ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
ModuleUIConfigurationManager
Hi, guess I'm looking for hints on how to track down a startup problem (attached is the screenshot) Can't launch application. component context fails to supply service ...ModuleUIConfigurationManager of type ...XModuleUIConfigurationManager2: an error occured during file opening. anybody already come across this java problem? ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: suggestion for adding sdk/lib to pluginapp.bin or libuno_sal.so symlink in ure/lib
Le 21/05/14 10:56, Stephan Bergmann a écrit : On 05/21/2014 07:13 AM, Richard PALO wrote: S=/var/tmp/pkgsrc/misc/libreoffice4/work/libreoffice-4.2.4.2 I=$S/instdir W=$S/workdir g++-Wl,-z,origin '-Wl,-rpath,$ORIGIN:$ORIGIN/../ure-link/lib' -L$I/ure/lib -L$I/program -z nodefs $W/CxxObject/extensions/source/plugin/unx/npwrap.o $W/CxxObject/extensions/source/plugin/unx/npnapi.o -Wl,--start-group $W/LinkTarget/StaticLibrary/libplugcon.a -lm -lnsl -lsocket -lXt -lXext -lX11 -ldl -L/opt/local/lib -Wl,-R/opt/local/lib -lgthread-2.0 -lpthread -lthread -Wl,-R/opt/local/lib -lglib-2.0 -lintl -L/opt/local/lib -Wl,-R/opt/local/lib -lgtk-x11-2.0 -Wl,-R/opt/local/lib -lgdk-x11-2.0 -Wl,-R/opt/local/lib -lpangocairo-1.0 -Wl,-R/opt/local/lib -latk-1.0 -Wl,-R/opt/local/lib -lcairo -Wl,-R/opt/local/lib -lgio-2.0 -Wl,-R/opt/local/lib -lpangoft2-1.0 -Wl,-R/opt/local/lib -lpango-1.0 -Wl,-R/opt/local/lib -lfontconfig -Wl,-R/opt/local/lib -lfreetype -Wl,-R/opt/local/lib -lgdk_pixbuf_xlib-2.0 -Wl,-R/opt/local/lib -lgmodule-2.0 -lpthread -lthread -Wl,-R/opt/local/lib -lgdk_pixbuf-2.0 -Wl,-R/opt/local/lib -lgobject-2.0 -Wl,-R/opt/local/lib -lglib-2.0 -lintl -L/opt/local/lib -Wl,-R/opt/local/lib -lgtk-x11-2.0 -Wl,-R/opt/local/lib -lgdk-x11-2.0 -Wl,-R/opt/local/lib -lpangocairo-1.0 -Wl,-R/opt/local/lib -latk-1.0 -Wl,-R/opt/local/lib -lcairo -Wl,-R/opt/local/lib -lgdk_pixbuf-2.0 -Wl,-R/opt/local/lib -lgio-2.0 -Wl,-R/opt/local/lib -lpangoft2-1.0 -Wl,-R/opt/local/lib -lpango-1.0 -Wl,-R/opt/local/lib -lgobject-2.0 -Wl,-R/opt/local/lib -lglib-2.0 -lintl -Wl,-R/opt/local/lib -lfontconfig -Wl,-R/opt/local/lib -lfreetype -Wl,--end-group -Wl,--no-as-needed -luno_sal -o $I/program/pluginapp.bin ld: fatal: library -luno_sal: not found = Either 'sdk/lib' needs to be added as a runpath-link (which works fine manually) or a symlink to libuno_sal.so.3 would need to be created in 'ure/lib' as that path is already used. -L$I/sdk/lib should be on the above command line (and is on Linux) via -L$(INSTDIR)/$(SDKDIRNAME)/lib in the definition of gb_LinkTarget_get_linksearchpath_for_layer in solenv/gbuild/platform/com_GCC_class.mk. No idea why that does not work for you on SunOS. Stephan Hi, Thanks for the pointer, and yes this file is included in solaris.mk as it is in unxgcc.mk... So, I was curious to notice that nsplugin seemed to work, and in comparing the two mk files the following extract from Executable_pluginapp.bin.mk is the only pertinent difference: # the orignal dmakefile said: don't ask, it's ugly ifeq ($(OS),SOLARIS) $(eval $(call gb_Executable_set_ldflags,pluginapp.bin,\ -z nodefs \ )) endif when I look into gb_Executable_set_ldflags, the problem is in LinkTarget.mk: # call gb_LinkTarget_add_ldflags,linktarget,ldflags define gb_LinkTarget_add_ldflags $(call gb_LinkTarget_get_target,$(1)) : T_LDFLAGS += $(2) endef # real use in RepositoryExternal.mk # call gb_LinkTarget_set_ldflags,linktarget,ldflags define gb_LinkTarget_set_ldflags $(call gb_LinkTarget_get_target,$(1)) : T_LDFLAGS := $(2) ^^ Earlier in LinkTarget.mk your function is evaluted thus: $(call gb_LinkTarget_get_target,$(1)) : T_LDFLAGS := $$(gb_LinkTarget_LDFLAGS) $(call gb_LinkTarget_get_linksearchpath_for_layer,$(3)) $(call gb_LinkTarget__get_ldflags,$(2)) therefore overridden by the subsequent set_ldflags call. In summary, the following patch to Executable_pluginapp.bin.mk seems to build for now (the first part is because motif is, well...:-): $ pkgdiff Executable_pluginapp.bin.mk $NetBSD$ --- Executable_pluginapp.bin.mk.orig2014-04-30 19:49:45.0 + +++ Executable_pluginapp.bin.mk @@ -33,15 +33,7 @@ $(eval $(call gb_Executable_use_librarie sal \ )) -ifeq ($(OS),SOLARIS) -$(eval $(call gb_Executable_add_libs,pluginapp.bin,\ - -lXm \ - -lXt \ - -lXext \ - -lX11 \ - -ldl \ -)) -else ifeq ($(filter-out FREEBSD NETBSD OPENBSD DRAGONFLY,$(OS)),) +ifeq ($(filter-out FREEBSD NETBSD OPENBSD DRAGONFLY,$(OS)),) $(eval $(call gb_Executable_add_libs,pluginapp.bin,\ -lXt \ -lXext \ @@ -73,7 +65,7 @@ $(eval $(call gb_Executable_use_external )) # the orignal dmakefile said: don't ask, it's ugly -ifeq ($(OS),SOLARIS) +ifeq ($(OS),XSOLARIS) $(eval $(call gb_Executable_set_ldflags,pluginapp.bin,\ -z nodefs \ )) BTW, for the moment, I don't have an issue with omitting the -znodefs perhaps an old problem since resolved... will keep an eye out. cheers ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Solaris build
Le 23/05/14 12:34, Michael Stahl a écrit : plugcon.hxx: #if defined SOLARIS #define USE_MOTIF #endif yes, this is already patched out quite a while ago in my test tree as well as the first part of the pluginapp.bin patch.. cheers ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
