Re: Solarmutex Deadlock when Closing Calc which contains functions with XVolatileResult return values
On 08/10/2012 03:35 PM, anwen wrote: Stephan Bergmann-2 wrote I think you should execute your shutdown activies on notifyTermination, not queryTermination. The latter can be veto'ed by some other listener, in which case Desktop would cancel the termination and continue running (and notify you with queryTermination eventually followed by notifyTermination again later on). Thank you for your suggestion. Yes, queryTermination should be a place to set up some status changes. And real stop or shutdown a process should be done in notifyTermination. BTW, I saw that there is a XTerminationListener2 where cancelTermination() is called when the master environment (e.g., desktop) was canceled in it's terminate request. Is it good to use this interface? As long as you only listen on notifyTermination, you don't need to support the derived XTerminationListener2. That queryTermination is called with SolarMutex locked is a bug. (Is notifyTermination also called with SolarMutex locked? Would need to check, in framework/source/services/desktop.cxx.) However, it might be a bug not easy to fix without opening any number of pandora's boxes... Is there a bug report number regarding this issue? I would like to check its progress in the future. In addition, since notifyTermination is also protected by SolarMutex, I am wondering if you see any workaround to deal with such a deadlock issue related to multithread shutdown. Thank you. I'm not aware of any specific bug for this. I see no workaround short of a proper fix. And a proper fix would mean to (a) identify the place up the call stack from Desktop::terminate where SolarMutex is locked (I assume that TransactionGuard is not SolarMutex, but yet another locking for Desktop's private parts, but I may be wrong), (b) redesign that code path so that Desktop::terminate can call notifyTermination without SolarMutex, (c) see if any of the existing listeners on notifyTermination exploited the fact that they used to be called with SolarMutex locked and need to be adapted now. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[REVIEW 3-6 (but only 3.6.2?)] No more prereg, no more unopkg sync
Quoting http://cgit.freedesktop.org/libreoffice/core/commit/?id=6dcb3d4ef46312729bb6f16c473b433474863f68 Related fdo#51252: No more prereg, no more unopkg sync: Now that 5c47e5f63a79a9e72ec4a100786b1bbf65137ed4 'fdo#51252 Disable copying share/prereg/bundled to avoid startup crashes' removed the use of share/prereg, there is no longer need to generate it in the first place (by calling 'unopkg sync' at build or installation time), and so no need for the 'unopkg sync' sub-command, either. This also allows to simplify some of the jvmfwk code that was only there so that 'unopkg sync' (which can require a JVM) can work in 'hostile' environments (during build and installation). This is a somewhat large patch, removing stuff all over the code base. I checked it on all of Linux, Mac OS X, and Windows, but might well have missed one or the other nuance, esp. in terra incognita like setup_native/source/win32/customactions. Hence, I would suggest to not jeopardize LO 3.6.1 with this clean up, but only backport it to libreoffice-3-6 after branch-off of libreoffice-3-6-1 early next week. (This clean up would nicely remove the root for https://bugs.freedesktop.org/show_bug.cgi?id=44628 unopkg.exe does not find VC runtime during install (extension registration), but it might be better to address that with yet another targeted band-aid fix for LO 3.6.1, like by just not calling unopkg sync from the msi install scripts.) Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: i18npool/Module_i18npool.mk
i18npool/Module_i18npool.mk |1 - 1 file changed, 1 deletion(-) New commits: commit 293151c8fc10c5a0ba84c5c5a796a3c72f573654 Author: Stephan Bergmann sberg...@redhat.com Date: Fri Aug 10 18:54:27 2012 +0200 Disabling CppunitTest_i18npool_test_languagetag for now ... as it causes problems on Windows: file:///C:/lo/core/solver/wntmsci12.pro/share/liblangtag%5Clanguage-subtag-registry.xml:709: parser error : Input is not proper UTF-8, indicate encoding ! Bytes: 0xE5 0x6C 0x3C 0x2F descriptionNorwegian Bokmål/description Apart from the encoding issue (looks like that language-subtag-registry.xml delivered from module liblangtag is generated as MS-1252 rather than UTF-8 on Windows), what also looks worrying in the error message is the backslash encoded as %5C (and the entity resolving that URL apparently decoding it somewhat naively). Change-Id: I756f911a92b144e3732f63668d36b4ae941f975a diff --git a/i18npool/Module_i18npool.mk b/i18npool/Module_i18npool.mk index 9214d00..bed175a 100644 --- a/i18npool/Module_i18npool.mk +++ b/i18npool/Module_i18npool.mk @@ -63,7 +63,6 @@ endif $(eval $(call gb_Module_add_check_targets,i18npool,\ CppunitTest_i18npool_test_breakiterator \ CppunitTest_i18npool_test_characterclassification \ - CppunitTest_i18npool_test_languagetag \ )) # vim: set noet sw=4 ts=4: ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: desktop/source
desktop/source/app/app.cxx |7 ++- 1 file changed, 6 insertions(+), 1 deletion(-) New commits: commit 0ab5ce54c441b12ea6c12a2cedb9069f36cfd0f4 Author: Stephan Bergmann sberg...@redhat.com Date: Fri Aug 10 19:36:08 2012 +0200 fdo#53006: Remove directories too, not just plain files ... that was an oversight in 5300f6f711c2167931d45248c1b72dbce3a7df38; thanks Petr for spotting it. Change-Id: I4d0179165bc3b29d698cd776f7e1aa51fa73ebe3 diff --git a/desktop/source/app/app.cxx b/desktop/source/app/app.cxx index c0bc0ed..960a6e4 100644 --- a/desktop/source/app/app.cxx +++ b/desktop/source/app/app.cxx @@ -186,11 +186,16 @@ void removeTree(OUString const url) { } else { if (osl::File::remove(stat.getFileURL()) != osl::FileBase::E_None) { throw css::uno::RuntimeException( -cannot remove + stat.getFileURL(), +cannot remove file + stat.getFileURL(), css::uno::Reference css::uno::XInterface ()); } } } +if (osl::Directory::remove(url) != osl::FileBase::E_None) { +throw css::uno::RuntimeException( +cannot remove directory + url, +css::uno::Reference css::uno::XInterface ()); +} } // Remove any existing UserInstallation's user/extensions/bundled cache ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[REVIEW 3-6] fdo#53006: Remove directories too, not just plain files
On 08/10/2012 05:29 PM, Petr Mladek wrote: Just to be sure. removeTree(url) removes only files and not directories. Is it by intention? I think that both variants would be fine. Oh, good catch. No, it was an oversight to not remove directories. So http://cgit.freedesktop.org/libreoffice/core/commit/?id=0ab5ce54c441b12ea6c12a2cedb9069f36cfd0f4 fdo#53006: Remove directories too, not just plain files fixes that. I think it is best to backport it to libreoffice-3-6, too. (Seen working as intended on both Linux and Windows master.) Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Build LO for Android on Mac OS 10.8
On 08/13/2012 08:38 AM, James Chen wrote: ld: warning: ignoring file /opt/local/lib/libiconv.dylib, file was built for unsupported file forma ld: warning: ignoring file /opt/local/lib/libz.dylib, file was built for unsupported file format ld: in /opt/local/lib/libiconv.2.dylib, file was built for unsupported file format Hmm, what is this /opt/local/lib? In general the LO build for MacOSX will break if it tries to use some existing add-on builds of common 3rd-party libraries, like MacPorts etc. At least that is what everybody says, I don't have personal experience as I don't have MacPorts... Try to make sure this /opt/local stuff is not found when building LibreOffice, removing /opt/local/bin from PATH (if you have it there) probably helps. --tml Thanks for pointing out and the hints. Removing /opt/local/bin from PATH does not work for me. Even not with a clean rebuild from scratch afterwards? I'd assume some mis-configuration (likely caused by picking up MacPort's /opt/local/bin/pkg-check) was still manifest in your build. Stephan I turn out to reinstall the lib as universal binary via MacPort as others suggest to similar issue. E.g.: sudo port upgrade --enforce-variants libiconv +universal I probably should override the linker path to prevent the build to link against the MacPort version of libs? Anyway, it's on the track now. :P ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: connectivity/registry connectivity/source connectivity/workben
connectivity/registry/postgresql/org/openoffice/Office/DataAccess/Drivers.xcu |2 +- connectivity/source/drivers/postgresql/postgresql-sdbc-impl.component |2 +- connectivity/source/drivers/postgresql/postgresql-sdbc.component |2 +- connectivity/source/drivers/postgresql/pq_connection.cxx |2 +- connectivity/source/drivers/postgresql/pq_driver.cxx |4 ++-- connectivity/workben/postgresql/metadata.py |2 +- connectivity/workben/postgresql/preparedstatement.py |2 +- connectivity/workben/postgresql/sdbcx.py |2 +- connectivity/workben/postgresql/statement.py |2 +- 9 files changed, 10 insertions(+), 10 deletions(-) New commits: commit f50a41437ee99aee499dbabd1d36ef8197db5559 Author: Stephan Bergmann sberg...@redhat.com Date: Mon Aug 13 09:52:43 2012 +0200 fdo#53246 Fresh postgresql-sdbc implementation names The descriptions for fdo#53246 Libo36 does not start, InvalidRegistryException from program/services/postgresql-sdbc.rdb make it look like the user had installed the functionality that comes bundled in postgresql-sdbc.rdb also as an extension installed per-user or shared. http://wiki.documentfoundation.org/PostgreSQL-SDBC indicates that what is now bundled due to c97d74b5f0608866bad3f7ae943811bdff4f7986 fdo#45090: Changed postgresql-sdbc from bundled oxt to proper optional part indeed was once available as a freestanding postgresql-sdbc-*.oxt that could have been installed per-user or shared. To prevent problems with such scenarios, rename the relevant implementation names by adding a .noext suffix. Change-Id: I43414a64461ba643f0ca0b2749450099cb6c8a4d diff --git a/connectivity/registry/postgresql/org/openoffice/Office/DataAccess/Drivers.xcu b/connectivity/registry/postgresql/org/openoffice/Office/DataAccess/Drivers.xcu index 8958123..75328ef 100644 --- a/connectivity/registry/postgresql/org/openoffice/Office/DataAccess/Drivers.xcu +++ b/connectivity/registry/postgresql/org/openoffice/Office/DataAccess/Drivers.xcu @@ -56,7 +56,7 @@ node oor:name=Installed install:module=postgresql node oor:name=sdbc:postgresql:* oor:op=replace prop oor:name=Driver -valueorg.openoffice.comp.connectivity.pq.Driver/value +valueorg.openoffice.comp.connectivity.pq.Driver.noext/value /prop prop oor:name=DriverTypeDisplayName oor:type=xs:string value xml:lang=en-USPostgreSQL/value diff --git a/connectivity/source/drivers/postgresql/postgresql-sdbc-impl.component b/connectivity/source/drivers/postgresql/postgresql-sdbc-impl.component index bd3c38c..407a0e6 100644 --- a/connectivity/source/drivers/postgresql/postgresql-sdbc-impl.component +++ b/connectivity/source/drivers/postgresql/postgresql-sdbc-impl.component @@ -29,7 +29,7 @@ component loader=com.sun.star.loader.SharedLibrary xmlns=http://openoffice.org/2010/uno-components; - implementation name=org.openoffice.comp.connectivity.pq.Connection + implementation name=org.openoffice.comp.connectivity.pq.Connection.noext service name=com.sun.star.sdbc.Connection/ /implementation /component diff --git a/connectivity/source/drivers/postgresql/postgresql-sdbc.component b/connectivity/source/drivers/postgresql/postgresql-sdbc.component index 016029e..f691963 100644 --- a/connectivity/source/drivers/postgresql/postgresql-sdbc.component +++ b/connectivity/source/drivers/postgresql/postgresql-sdbc.component @@ -29,7 +29,7 @@ component loader=com.sun.star.loader.SharedLibrary xmlns=http://openoffice.org/2010/uno-components; - implementation name=org.openoffice.comp.connectivity.pq.Driver + implementation name=org.openoffice.comp.connectivity.pq.Driver.noext service name=com.sun.star.sdbc.Driver/ /implementation /component diff --git a/connectivity/source/drivers/postgresql/pq_connection.cxx b/connectivity/source/drivers/postgresql/pq_connection.cxx index e1e3754..209ca09 100644 --- a/connectivity/source/drivers/postgresql/pq_connection.cxx +++ b/connectivity/source/drivers/postgresql/pq_connection.cxx @@ -160,7 +160,7 @@ public: OUStringConnectionGetImplementationName() { -return OUString( RTL_CONSTASCII_USTRINGPARAM( org.openoffice.comp.connectivity.pq.Connection ) ); +return OUString( RTL_CONSTASCII_USTRINGPARAM( org.openoffice.comp.connectivity.pq.Connection.noext ) ); } com::sun::star::uno::Sequencertl::OUString ConnectionGetSupportedServiceNames(void) { diff --git a/connectivity/source/drivers/postgresql/pq_driver.cxx b/connectivity/source/drivers/postgresql/pq_driver.cxx index 98ee8ea..ed8480c 100644 --- a/connectivity/source/drivers/postgresql/pq_driver.cxx +++ b/connectivity/source/drivers/postgresql/pq_driver.cxx @@ -106,7 +106,7 @@ OUString
[REVIEW 3-6] fdo#53246 Fresh postgresql-sdbc implementation names
Besides the general problem of LO 3.6 not starting due to duplicate UNO implementations caused by the bundled extension prereg mechanism, there are also reports for a more specific problem of not starting due to duplicate UNO implementations of org.openoffice.comp.connectivity.pq.Driver, https://bugs.freedesktop.org/show_bug.cgi?id=53246 Libo36 does not start, InvalidRegistryException from program/services/postgresql-sdbc.rdb. This appears to be caused by an old postgresql-sdbc-*.oxt installed per-user or shared (http://wiki.documentfoundation.org/PostgreSQL-SDBC) while the same functionality is included as a non-extension today (http://cgit.freedesktop.org/libreoffice/core/commit/?id=c97d74b5f0608866bad3f7ae943811bdff4f7986 fdo#45090: Changed postgresql-sdbc from bundled oxt to proper optional part.). http://cgit.freedesktop.org/libreoffice/core/commit/?id=f50a41437ee99aee499dbabd1d36ef8197db5559 fdo#53246 Fresh postgresql-sdbc implementation names works around that on master, with a rather trivial patch of choosing new names for the relevant UNO implementations. I verified that it does not break things with the instructions from http://lists.freedesktop.org/archives/libreoffice/2012-January/024333.html [Libreoffice] [PATCH] rpath / ... for extensions (and the case of PostgreSQL-SDBC). Please review and consider for backporting to libreoffice-3-6. Thanks, Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: 2 commits - RepositoryExternal.mk RepositoryFixes.mk
RepositoryExternal.mk |2 +- RepositoryFixes.mk|1 + 2 files changed, 2 insertions(+), 1 deletion(-) New commits: commit fc2b42ed182c8e7d20a8626712b8a416d7a5bae5 Author: Stephan Bergmann sberg...@redhat.com Date: Mon Aug 13 11:46:50 2012 +0200 Proper sorting Change-Id: Ibf8303cd065ebd4c02b696077d402bf9221012b0 diff --git a/RepositoryExternal.mk b/RepositoryExternal.mk index 5c0dd05..0d6c660 100644 --- a/RepositoryExternal.mk +++ b/RepositoryExternal.mk @@ -1918,9 +1918,9 @@ else $(eval $(call gb_Helper_register_libraries,PLAINLIBS_OOO,\ nspr4 \ -smime3 \ nss3 \ plc4 \ +smime3 \ )) define gb_LinkTarget__use_nss3 commit 066cf169e843719db02cc1171be162e75cd8c7d7 Author: Stephan Bergmann sberg...@redhat.com Date: Mon Aug 13 11:46:21 2012 +0200 Missing ismime3.lib on Windows Change-Id: I289a4da4473d58736ecdf3846864ccde9a1a03d1 diff --git a/RepositoryFixes.mk b/RepositoryFixes.mk index b5ecb1c..829967f 100644 --- a/RepositoryFixes.mk +++ b/RepositoryFixes.mk @@ -139,6 +139,7 @@ gb_Library_NOILIBFILENAMES += \ nspr4 \ nss3 \ shlwapi \ +smime3 \ uwinapi \ version \ ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: [REVIEW 3-6 ] No more prereg, no more unopkg sync
On 08/13/2012 01:48 PM, Michael Meeks wrote: IMHO I'd prefer to have an RC3 for 3.6.1 than the potential for regression by including this in 3.6.2 - and having the state-of-the-art code that is shared between master and -3-6 for testing seems sensible (to me). In the end, I'd be fine either way. An additional review might make sense when integrating it now. The only gotcha I spotted was the: @@ -222,18 +211,6 @@ NodeJava::NodeJava(Layer layer): constructor; but it seems we initialize m_layer earlier already so just remove a duplicate initialization. Yes, the removed else-branch was redundant. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: desktop/source
desktop/source/app/app.cxx| 75 ++ desktop/source/deployment/registry/component/dp_component.cxx |7 2 files changed, 55 insertions(+), 27 deletions(-) New commits: commit e9631f40ee49a244b6c6816e5cfb36814b9bafba Author: Stephan Bergmann sberg...@redhat.com Date: Mon Aug 13 17:48:25 2012 +0200 fdo#53006: And detect LO downgrading scenarios... ...that reinstall potentially broken prereg data but do not touch the buildid file. Change-Id: I1b02d2c265523dd751d8ef7f2dccb59ed048f568 diff --git a/desktop/source/app/app.cxx b/desktop/source/app/app.cxx index 960a6e4..87c34e8 100644 --- a/desktop/source/app/app.cxx +++ b/desktop/source/app/app.cxx @@ -216,6 +216,18 @@ void removeTree(OUString const url) { // detected here and lead to a removal of user/extensions/bundled, so that // Desktop::SynchronizeExtensionRepositories will then definitely resync // share/extensions.) +// As a special case, if you create a UserInstallation with LO = 3.6.1, then +// run an old LO = 3.5.x using share/prereg/bundled on the same +// UserInstallation (so that it partially overwrites user/extensions/bundled, +// potentially duplicating component information, but not touching +// user/extensions/bundled/buildid), and then run the new LO = 3.6.1 on the +// same UserInstallation again, it can fail to start (due to the duplicated +// component information). Even though such downgrading scenarios at best work +// by luck in general, the special token LIBO_NON_PREREG_BUNDLED_EXTENSIONS=TRUE +// is used to detect and fix that problem: The assumption is that if an old LO +// = 3.5.x messed with user/extensions/bundled in the meantime, then it would +// have rewritten the unorc (dropping the token), and LO = 3.6.1 can detect +// that. void refreshBundledExtensionsDir() { OUString buildId( ${$BRAND_BASE_DIR/program/ SAL_CONFIGFILE(version) :buildid}); @@ -223,36 +235,44 @@ void refreshBundledExtensionsDir() { OUString dir($BUNDLED_EXTENSIONS_USER); rtl::Bootstrap::expandMacros(dir); //TODO: detect failure OUString url(dir + /buildid); -osl::File f(url); -switch (f.open(osl_File_OpenFlag_Read)) { -case osl::FileBase::E_None: -{ -rtl::ByteSequence s1; -osl::FileBase::RC rc = f.readLine(s1); -if (f.close() != osl::FileBase::E_None) { -SAL_WARN(desktop, cannot close + url + after reading); -} -if (rc != osl::FileBase::E_None) { -throw css::uno::RuntimeException( -cannot read from + url, -css::uno::Reference css::uno::XInterface ()); -} -OUString s2( -reinterpret_cast char const * (s1.getConstArray()), -s1.getLength(), RTL_TEXTENCODING_ISO_8859_1); -// using ISO 8859-1 avoids any and all conversion errors; the -// content should only be a subset of ASCII, anyway -if (s2 == buildId) { -return; +OUString nonPrereg( +${$BUNDLED_EXTENSIONS_USER/registry/ +com.sun.star.comp.deployment.component.PackageRegistryBackend/unorc: +LIBO_NON_PREREG_BUNDLED_EXTENSIONS}); +rtl::Bootstrap::expandMacros(nonPrereg); +if (nonPrereg == TRUE) { +osl::File f(url); +switch (f.open(osl_File_OpenFlag_Read)) { +case osl::FileBase::E_None: +{ +rtl::ByteSequence s1; +osl::FileBase::RC rc = f.readLine(s1); +if (f.close() != osl::FileBase::E_None) { +SAL_WARN( +desktop, cannot close + url + after reading); +} +if (rc != osl::FileBase::E_None) { +throw css::uno::RuntimeException( +cannot read from + url, +css::uno::Reference css::uno::XInterface ()); +} +OUString s2( +reinterpret_cast char const * (s1.getConstArray()), +s1.getLength(), RTL_TEXTENCODING_ISO_8859_1); +// using ISO 8859-1 avoids any and all conversion errors; +// the content should only be a subset of ASCII, anyway +if (s2 == buildId) { +return; +} +break; } +case osl::FileBase::E_NOENT: break; +default: +throw css::uno::RuntimeException( +cannot open + url + for reading, +css::uno::Reference css::uno::XInterface ()); } -case osl::FileBase::E_NOENT: -break; -default: -throw css::uno::RuntimeException( -cannot open + url + for reading, -css::uno::Reference css::uno::XInterface ()); } removeTree(dir); switch (osl::Directory
[REVIEW 3-6] fdo#53006: And detect LO downgrading scenarios
As Petr found out, if you first run LO 3.6.1, then some old LO 3.5.x (making use of share/prereg/bundled), then LO 3.6.1 again, all on the same UserInstallation, it can still happen that LO 3.6.1 refuses to start (as LO 3.5.x copied potentially broken data to user/extensions/bundled but did not touch user/extensions/bundled/buildid, so that LO 3.6.1 would not detect that it needs to remove the potentially poisonous data again). http://cgit.freedesktop.org/libreoffice/core/commit/?id=e9631f40ee49a244b6c6816e5cfb36814b9bafba fdo#53006: And detect LO downgrading scenarios... addresses that by adding yet another pirouette (even though one could argue that such downgrading scenarios never worked by design, anyway). Between Petr and me, we have seen it to be effective for both Linux and Windows (master, at least). Please review and consider for backporting to libreoffice-3-6. Thanks, Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: minutes of ESC call ...
On 08/13/2012 08:10 PM, Bjoern Michaelsen wrote: On Mon, Aug 13, 2012 at 10:26:57AM -0700, bfo wrote: I think many bugs introduced are an effect of omitting one (or more) steps from that list by the developer. nope, I think you are just underestimating the size of the codebase (and the range of our target platforms). Suggested reading: Michael Feathers, Working Effectively with Legacy Code ... and suggested action: writing a dozen fixes for LO. There is no shortage of wisdom around here how to do things well in theory, I think. Sorry for sounding rude... Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Solarmutex Deadlock when Closing Calc which contains functions with XVolatileResult return values
On 08/14/2012 12:35 AM, anwen wrote: I implemented the above code with c++ . However, it did not work. Anything wrong with the following code: m_xCC is a given ReferenceXComponentContext ReferenceXEventBroadcaster xBrd(m_xCC-getServiceManager()-createInstanceWithContext(OUString(RTL_CONSTASCII_USTRINGPARAM(com.sun.star.frame.GlobalEventbroadcaster)), it's GlobalEventBroadcaster (with a capital B) Stephan m_xCC), UNO_QUERY); //xBrd is still empty, not sure why? ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: hunspell/makefile.mk
hunspell/makefile.mk |3 +++ 1 file changed, 3 insertions(+) New commits: commit d3bec41a1b92cc8ca9f4750c5f9740da52879085 Author: Stephan Bergmann sberg...@redhat.com Date: Tue Aug 14 10:53:06 2012 +0200 Patches reported upstream Change-Id: I1275b73d4f18633ba03831b6973b801b0467cf0d diff --git a/hunspell/makefile.mk b/hunspell/makefile.mk index 920a703..156cbbe 100644 --- a/hunspell/makefile.mk +++ b/hunspell/makefile.mk @@ -40,6 +40,9 @@ TARFILE_NAME=hunspell-1.3.2 TARFILE_MD5=3121aaf3e13e5d88dfff13fb4a5f1ab8 ADDITIONAL_FILES+=config.h +# hunspell-1.3.2-nullptr.patch, hunspell-1.3.2-literal.patch: reported upstream +# as https://sourceforge.net/tracker/?func=detailaid=3557292group_id=143754; +# atid=756395 Some C++11 related fixes PATCH_FILES=\ hunspell-twoaffixcompound.patch \ hunspell-static.patch \ ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: 3 commits - offapi/com offapi/type_reference offapi/UnoApi_offapi.mk
offapi/UnoApi_offapi.mk | 11 +++ offapi/com/sun/star/auth/SSOExceptions.idl| 33 -- offapi/com/sun/star/auth/SSOManagerFactory.idl| 11 ++- offapi/com/sun/star/auth/SSOPasswordCache.idl | 11 ++- offapi/com/sun/star/auth/XSSOAcceptorContext.idl |5 ++- offapi/com/sun/star/auth/XSSOInitiatorContext.idl |5 ++- offapi/com/sun/star/auth/XSSOManager.idl |4 ++ offapi/com/sun/star/auth/XSSOPasswordCache.idl|3 +- offapi/type_reference/typelibrary_history.txt |2 - offapi/type_reference/types.rdb |binary offapi/type_reference/update-rdb.sh | 32 + 11 files changed, 59 insertions(+), 58 deletions(-) New commits: commit 629dd9dae17b902b53839842929fcff4b836b47a Author: Stephan Bergmann sberg...@redhat.com Date: Tue Aug 14 14:40:39 2012 +0200 Remove SSOExceptions.idl Change-Id: I930603bfcee8084eae0a086d3ce02182b8b13541 diff --git a/offapi/UnoApi_offapi.mk b/offapi/UnoApi_offapi.mk index 152358c..8322f90 100644 --- a/offapi/UnoApi_offapi.mk +++ b/offapi/UnoApi_offapi.mk @@ -299,9 +299,6 @@ $(eval $(call gb_UnoApi_add_idlfiles_noheader,offapi,offapi/com/sun/star/accessi Accessible \ AccessibleContext \ )) -$(eval $(call gb_UnoApi_add_idlfiles_noheader,offapi,offapi/com/sun/star/auth,\ - SSOExceptions \ -)) $(eval $(call gb_UnoApi_add_idlfiles_noheader,offapi,offapi/com/sun/star/awt,\ AccessibleButton \ AccessibleCheckBox \ diff --git a/offapi/com/sun/star/auth/SSOExceptions.idl b/offapi/com/sun/star/auth/SSOExceptions.idl deleted file mode 100644 index 097a7c8..000 --- a/offapi/com/sun/star/auth/SSOExceptions.idl +++ /dev/null @@ -1,33 +0,0 @@ -/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */ -/* - * This file is part of the LibreOffice project. - * - * 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/. - * - * This file incorporates work covered by the following license notice: - * - * Licensed to the Apache Software Foundation (ASF) under one or more - * contributor license agreements. See the NOTICE file distributed - * with this work for additional information regarding copyright - * ownership. The ASF licenses this file to you under the Apache - * License, Version 2.0 (the License); you may not use this file - * except in compliance with the License. You may obtain a copy of - * the License at http://www.apache.org/licenses/LICENSE-2.0 . - */ - -#ifndef __com_sun_star_auth_SSOExceptions_idl__ -#define __com_sun_star_auth_SSOExceptions_idl__ - -#include com/sun/star/auth/AuthenticationFailedException.idl -#include com/sun/star/auth/InvalidArgumentException.idl -#include com/sun/star/auth/InvalidContextException.idl -#include com/sun/star/auth/InvalidCredentialException.idl -#include com/sun/star/auth/InvalidPrincipalException.idl -#include com/sun/star/auth/PersistenceFailureException.idl -#include com/sun/star/auth/UnsupportedException.idl - -#endif - -/* vim:set shiftwidth=4 softtabstop=4 expandtab: */ diff --git a/offapi/com/sun/star/auth/XSSOAcceptorContext.idl b/offapi/com/sun/star/auth/XSSOAcceptorContext.idl index af9de07..d8eb2c2 100644 --- a/offapi/com/sun/star/auth/XSSOAcceptorContext.idl +++ b/offapi/com/sun/star/auth/XSSOAcceptorContext.idl @@ -29,8 +29,11 @@ #ifndef __com_sun_star_auth_XSSOAcceptorContext_idl__ #define __com_sun_star_auth_XSSOAcceptorContext_idl__ +#include com/sun/star/auth/AuthenticationFailedException.idl +#include com/sun/star/auth/InvalidArgumentException.idl +#include com/sun/star/auth/InvalidContextException.idl +#include com/sun/star/auth/InvalidCredentialException.idl #include com/sun/star/auth/XSSOContext.idl -#include com/sun/star/auth/SSOExceptions.idl module com { module sun { module star { module auth { diff --git a/offapi/com/sun/star/auth/XSSOInitiatorContext.idl b/offapi/com/sun/star/auth/XSSOInitiatorContext.idl index 7b4408b..a1b69a3 100644 --- a/offapi/com/sun/star/auth/XSSOInitiatorContext.idl +++ b/offapi/com/sun/star/auth/XSSOInitiatorContext.idl @@ -29,8 +29,11 @@ #ifndef __com_sun_star_auth_XSSOInitiatorContext_idl__ #define __com_sun_star_auth_XSSOInitiatorContext_idl__ +#include com/sun/star/auth/AuthenticationFailedException.idl +#include com/sun/star/auth/InvalidArgumentException.idl +#include com/sun/star/auth/InvalidContextException.idl +#include com/sun/star/auth/InvalidCredentialException.idl #include com/sun/star/auth/XSSOContext.idl -#include com/sun/star/auth/SSOExceptions.idl module com { module sun { module star { module auth { diff --git a/offapi/com/sun/star/auth/XSSOManager.idl b/offapi/com/sun/star/auth/XSSOManager.idl index acb2eb2..0a68a32 100644 --- a/offapi/com/sun/star/auth
[Libreoffice-commits] .: xmlsecurity/Module_xmlsecurity.mk
xmlsecurity/Module_xmlsecurity.mk |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) New commits: commit d72568fc7d2680107c8f86f777bc6f55e5f09ae7 Author: Stephan Bergmann sberg...@redhat.com Date: Tue Aug 14 14:47:20 2012 +0200 Fix condition Change-Id: I746778a0d0f229c65dca37bcf3fdae9dcebd4340 diff --git a/xmlsecurity/Module_xmlsecurity.mk b/xmlsecurity/Module_xmlsecurity.mk index 46646b0..752f625 100644 --- a/xmlsecurity/Module_xmlsecurity.mk +++ b/xmlsecurity/Module_xmlsecurity.mk @@ -32,7 +32,7 @@ $(eval $(call gb_Module_add_targets,xmlsecurity,\ AllLangResTarget_xsec \ Library_xmlsecurity \ Library_xsec_fw \ -$(ifneq ($(filter-out IOS ANDROID,$(OS)),),Library_xsec_xmlsec) \ +$(if $(filter-out IOS ANDROID,$(OS)),Library_xsec_xmlsec) \ )) #FIXME: ^^^, get nsslibxmlsec building on ios and android ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: solenv/gbuild
solenv/gbuild/UnoApiTarget.mk |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) New commits: commit 3d199e410bda1ba148eed26176e1003df21eaa75 Author: Stephan Bergmann sberg...@redhat.com Date: Fri May 25 10:27:30 2012 +0200 Removed documentation from type rdbs Before gbuild'ification, solenv/inc/target.mk carefully produced two variants of each type rdb, a plain one from calling idlc w/o -C and a *_doc variant from calling idlc w/ -C (generate complete type information, including documentation). After gbuild'ification, solenv/gbuild/UnoApiTarget.mk unconditionally only produced complete type rdbs from calling idlc w/ -C. It is unclear to me whether the old *_doc variants had actually been used for anything (what got packaged into installation sets apparently were the plain, slim variants; and autodoc apparently does not need the *_doc variants, either, as it produces UNOIDL documentation directly from .idl files). It is also unclear to me whether the gbuild switch to effectively package complete, fat rdbs was deliberate or not. (The only client-visible change I see is that low- level C/C++ typereg_reader_getDocumentation could now report something. The reflective UNO services at com.sun.star.reflection would not offer access to the documenation anyway, however.) The benefit of no longer including documentation in the packaged type rdbs is size; the URE types.rdb shrinks from 1.2 MiB to 819 KiB, and offapi.rdb shrinks from 11 MiB to 6.5 MiB. Change-Id: Ib278f74fc3b22169e00a09d778807f8cf58520c4 diff --git a/solenv/gbuild/UnoApiTarget.mk b/solenv/gbuild/UnoApiTarget.mk index 42584ce..206bbb9 100644 --- a/solenv/gbuild/UnoApiTarget.mk +++ b/solenv/gbuild/UnoApiTarget.mk @@ -58,7 +58,7 @@ define gb_UnoApiPartTarget__command RESPONSEFILE=$(call var2file,$(shell $(gb_MKTEMP)),500,\ $(INCLUDE) \ -M $(basename $(call gb_UnoApiPartTarget_get_dep_target,$(dir $(2 \ - -O $(call gb_UnoApiPartTarget_get_target,$(dir $(2))) -verbose -C \ + -O $(call gb_UnoApiPartTarget_get_target,$(dir $(2))) -verbose \ $(sort $(patsubst $(call gb_UnoApiPartTarget_get_target,%.urd),$(SRCDIR)/%.idl,$(3 \ $(gb_UnoApiPartTarget_IDLCCOMMAND) @$${RESPONSEFILE} /dev/null \ rm -f $${RESPONSEFILE} \ ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: [Libreoffice-commits] .: Removed documentation from type rdbs
See below commit for shrinking size of type rdbs by removing the unused(?) UNOIDL documentation contained within. I'm really not sure there was a reason to include it in the first place (in the *_doc variants before gbuild'ification, and unconditionally afterwards). If anybody knows about a reason, please do speak up. (Jürgen, do you remember any?) Otherwise, it might be fine to backport this master commit to libreoffice-3-5, too. (It slightly reduces installation set size. Whether it also improves performance I did not bother to measure---I doubt it makes a noticeable difference on contemporary hardware.) Stephan On 05/25/2012 10:54 AM, Stephan Bergmann wrote: solenv/gbuild/UnoApiTarget.mk |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) New commits: commit 3d199e410bda1ba148eed26176e1003df21eaa75 Author: Stephan Bergmannsberg...@redhat.com Date: Fri May 25 10:27:30 2012 +0200 Removed documentation from type rdbs Before gbuild'ification, solenv/inc/target.mk carefully produced two variants of each type rdb, a plain one from calling idlc w/o -C and a *_doc variant from calling idlc w/ -C (generate complete type information, including documentation). After gbuild'ification, solenv/gbuild/UnoApiTarget.mk unconditionally only produced complete type rdbs from calling idlc w/ -C. It is unclear to me whether the old *_doc variants had actually been used for anything (what got packaged into installation sets apparently were the plain, slim variants; and autodoc apparently does not need the *_doc variants, either, as it produces UNOIDL documentation directly from .idl files). It is also unclear to me whether the gbuild switch to effectively package complete, fat rdbs was deliberate or not. (The only client-visible change I see is that low- level C/C++ typereg_reader_getDocumentation could now report something. The reflective UNO services at com.sun.star.reflection would not offer access to the documenation anyway, however.) The benefit of no longer including documentation in the packaged type rdbs is size; the URE types.rdb shrinks from 1.2 MiB to 819 KiB, and offapi.rdb shrinks from 11 MiB to 6.5 MiB. Change-Id: Ib278f74fc3b22169e00a09d778807f8cf58520c4 diff --git a/solenv/gbuild/UnoApiTarget.mk b/solenv/gbuild/UnoApiTarget.mk index 42584ce..206bbb9 100644 --- a/solenv/gbuild/UnoApiTarget.mk +++ b/solenv/gbuild/UnoApiTarget.mk @@ -58,7 +58,7 @@ define gb_UnoApiPartTarget__command RESPONSEFILE=$(call var2file,$(shell $(gb_MKTEMP)),500,\ $(INCLUDE) \ -M $(basename $(call gb_UnoApiPartTarget_get_dep_target,$(dir $(2 \ - -O $(call gb_UnoApiPartTarget_get_target,$(dir $(2))) -verbose -C \ + -O $(call gb_UnoApiPartTarget_get_target,$(dir $(2))) -verbose \ $(sort $(patsubst $(call gb_UnoApiPartTarget_get_target,%.urd),$(SRCDIR)/%.idl,$(3 \ $(gb_UnoApiPartTarget_IDLCCOMMAND) @$${RESPONSEFILE} /dev/null \ rm -f $${RESPONSEFILE} \ ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: stoc/source
stoc/source/invocation_adapterfactory/iafactory.cxx | 16 ++-- 1 file changed, 2 insertions(+), 14 deletions(-) New commits: commit 3f250092747263fb14a3f13c4509ebd02ec2fb2a Author: Stephan Bergmann sberg...@redhat.com Date: Fri May 25 16:22:11 2012 +0200 Use simple createOneInstanceComponentFactory ...instead of static WeakReference and getGlobalMutex. The latter allowed for deadlock, when one thread is in __getTypeEntries (cppuhelper/source/implbase_ex.cxx), having locked getImplHelperInitMutex(), and from there calls some cppu_detail_getUnoType (from a cppumaker-generated header) that tries to lock getGloblaMutex, while another thread is in this FactoryImpl_create, having locked getGlobalMutex, and from there calls into __getTypeEntries. Change-Id: Iac68d28c1a5971049447cc79384b6e186f2a6d82 diff --git a/stoc/source/invocation_adapterfactory/iafactory.cxx b/stoc/source/invocation_adapterfactory/iafactory.cxx index 337bdb1..361e798 100644 --- a/stoc/source/invocation_adapterfactory/iafactory.cxx +++ b/stoc/source/invocation_adapterfactory/iafactory.cxx @@ -943,19 +943,7 @@ static Reference XInterface SAL_CALL FactoryImpl_create( const Reference XComponentContext xContext ) throw (Exception) { -Reference XInterface rRet; -{ -MutexGuard guard( Mutex::getGlobalMutex() ); -static WeakReference XInterface rwInstance; -rRet = rwInstance; - -if( ! rRet.is() ) -{ -rRet = (::cppu::OWeakObject *)new FactoryImpl( xContext ); -rwInstance = rRet; -} -} -return rRet; +return (::cppu::OWeakObject *)new FactoryImpl( xContext ); } } @@ -971,7 +959,7 @@ static struct ::cppu::ImplementationEntry g_entries[] = ::stoc_invadp::FactoryImpl_create, ::stoc_invadp::invadp_getImplementationName, ::stoc_invadp::invadp_getSupportedServiceNames, -::cppu::createSingleComponentFactory, +::cppu::createOneInstanceComponentFactory, ::stoc_invadp::g_moduleCount.modCnt , 0 }, { 0, 0, 0, 0, 0, 0 } ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: ciruclar dependencies (was: feature/gbuild_merge needs testing)
On 05/25/2012 01:28 PM, David Tardon wrote: At the end, an anecdotical evidence about how far we progressed into gbuild land: the move of gtk/kde file pickers from fpicker to vcl introduced a dependency loop, because the pickers in vcl depend on headers from fpicker, but fpicker already (indirectly) depends on vcl. What is our attitude towards circular module dependencies in an all-gbuild scenario? My understanding is that technically they do not pose a problem. But how about them from a hygiene point of view? (And if we want to avoid them, would there be technical measures to do so?) Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: shell/source
shell/source/backends/gconfbe/gconfaccess.cxx |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) New commits: commit b6ce391171e417199d33ae085d2f0768a3e952b3 Author: Stephan Bergmann sberg...@redhat.com Date: Tue May 29 09:01:32 2012 +0200 -Werror=deprecated-declarations g_strcasecmp - g_ascii_strcasecmp, hopefully the latter is already available in oldest relevant glib. Change-Id: Iac47ebd511672d4ca08d4dc998a536939d4a98c3 diff --git a/shell/source/backends/gconfbe/gconfaccess.cxx b/shell/source/backends/gconfbe/gconfaccess.cxx index 9771b5d..8b17824 100644 --- a/shell/source/backends/gconfbe/gconfaccess.cxx +++ b/shell/source/backends/gconfbe/gconfaccess.cxx @@ -401,7 +401,7 @@ sal_Bool SAL_CALL isDependencySatisfied( GConfClient* pClient, const Configurati if ( pGconfValue != NULL ) { -bool bOk = g_strcasecmp( manual, gconf_value_get_string( pGconfValue ) ) == 0; +bool bOk = g_ascii_strcasecmp( manual, gconf_value_get_string( pGconfValue ) ) == 0; gconf_value_free( pGconfValue ); if (bOk) return sal_True; } ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: connectivity/AllLangResTarget_hsqldb.mk
connectivity/AllLangResTarget_hsqldb.mk |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) New commits: commit e689eb6401e69c39c2e52a4659b49b7b98422914 Author: Stephan Bergmann sberg...@redhat.com Date: Tue May 29 10:41:31 2012 +0200 Typo Change-Id: I59cdfb17a384d124cc7857b73adf53a10795fd27 diff --git a/connectivity/AllLangResTarget_hsqldb.mk b/connectivity/AllLangResTarget_hsqldb.mk index 923833d..4bccc5e 100644 --- a/connectivity/AllLangResTarget_hsqldb.mk +++ b/connectivity/AllLangResTarget_hsqldb.mk @@ -30,7 +30,7 @@ $(eval $(call gb_AllLangResTarget_AllLangResTarget,hsqldb)) $(eval $(call gb_AllLangResTarget_set_reslocation,hsqldb,database)) $(eval $(call gb_AllLangResTarget_add_srs,hsqldb,\ - database \ + hsqldb/res \ )) $(eval $(call gb_SrsTarget_SrsTarget,hsqldb/res)) ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: connectivity/source
connectivity/source/drivers/kab/KPreparedStatement.hxx |3 +++ connectivity/source/drivers/kab/KResultSet.hxx |1 + connectivity/source/drivers/kab/KStatement.hxx |1 + 3 files changed, 5 insertions(+) New commits: commit f3d74f9df0548afe74bfe655c1268fa5f32c5cc4 Author: Stephan Bergmann sberg...@redhat.com Date: Tue May 29 10:53:49 2012 +0200 -Werror=overloaded-virtual Change-Id: Ic70ccbb9b9a3aeb226af116ce009937510cd8153 diff --git a/connectivity/source/drivers/kab/KPreparedStatement.hxx b/connectivity/source/drivers/kab/KPreparedStatement.hxx index dfaf040..44ccffd 100644 --- a/connectivity/source/drivers/kab/KPreparedStatement.hxx +++ b/connectivity/source/drivers/kab/KPreparedStatement.hxx @@ -82,6 +82,9 @@ namespace connectivity virtual sal_Int32 SAL_CALL executeUpdate( ) throw(::com::sun::star::sdbc::SQLException, ::com::sun::star::uno::RuntimeException); virtual sal_Bool SAL_CALL execute( ) throw(::com::sun::star::sdbc::SQLException, ::com::sun::star::uno::RuntimeException); virtual ::com::sun::star::uno::Reference ::com::sun::star::sdbc::XConnection SAL_CALL getConnection( ) throw(::com::sun::star::sdbc::SQLException, ::com::sun::star::uno::RuntimeException); +using KabCommonStatement::executeQuery; +using KabCommonStatement::executeUpdate; +using KabCommonStatement::execute; // XParameters virtual void SAL_CALL setNull( sal_Int32 parameterIndex, sal_Int32 sqlType ) throw(::com::sun::star::sdbc::SQLException, ::com::sun::star::uno::RuntimeException); diff --git a/connectivity/source/drivers/kab/KResultSet.hxx b/connectivity/source/drivers/kab/KResultSet.hxx index fc7e5f4..2792934 100644 --- a/connectivity/source/drivers/kab/KResultSet.hxx +++ b/connectivity/source/drivers/kab/KResultSet.hxx @@ -92,6 +92,7 @@ namespace connectivity virtual void SAL_CALL getFastPropertyValue( ::com::sun::star::uno::Any rValue, sal_Int32 nHandle) const; +using OPropertySetHelper::getFastPropertyValue; // you can't delete objects of this type virtual ~KabResultSet(); diff --git a/connectivity/source/drivers/kab/KStatement.hxx b/connectivity/source/drivers/kab/KStatement.hxx index dcc141f..3e1e3cd 100644 --- a/connectivity/source/drivers/kab/KStatement.hxx +++ b/connectivity/source/drivers/kab/KStatement.hxx @@ -93,6 +93,7 @@ namespace connectivity virtual void SAL_CALL getFastPropertyValue( ::com::sun::star::uno::Any rValue, sal_Int32 nHandle) const; +using OPropertySetHelper::getFastPropertyValue; virtual void resetParameters() const throw(::com::sun::star::sdbc::SQLException); virtual void getNextParameter(::rtl::OUString rParameter) const throw(::com::sun::star::sdbc::SQLException); ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: connectivity/Library_postgresql-sdbc-impl.mk
connectivity/Library_postgresql-sdbc-impl.mk |4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) New commits: commit a4849e9d2b409b399191efaa446d42b797cc3d1e Author: Stephan Bergmann sberg...@redhat.com Date: Tue May 29 11:01:35 2012 +0200 Unquoted $(COMMA) Change-Id: I6d15cd77896da73422ab8c92f62d6341f7e2561f diff --git a/connectivity/Library_postgresql-sdbc-impl.mk b/connectivity/Library_postgresql-sdbc-impl.mk index 62c4f2f..c31d225 100644 --- a/connectivity/Library_postgresql-sdbc-impl.mk +++ b/connectivity/Library_postgresql-sdbc-impl.mk @@ -62,10 +62,10 @@ ifneq ($(GUI)$(COM),WNTMSC) include $(OUTDIR_FOR_BUILD)/inc/postgresql/libpq-flags.mk $(eval $(call gb_Library_add_libs,postgresql-sdbc-impl,\ - $(if $(filter-out MACOSX,$(OS)),-Wl,--as-needed) \ + $(if $(filter-out MACOSX,$(OS)),-Wl$(COMMA)--as-needed) \ $(LIBPQ_DEP_LIBS) \ $(if $(filter-out MACOSX,$(OS)),-ldl) \ - $(if $(filter-out MACOSX,$(OS)),-Wl,--no-as-needed) \ + $(if $(filter-out MACOSX,$(OS)),-Wl$(COMMA)--no-as-needed) \ )) endif ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: connectivity/inc
connectivity/inc/connectivity/dbcharset.hxx |6 +- 1 file changed, 5 insertions(+), 1 deletion(-) New commits: commit 553455ae8a9b8adc477323043073068b8c7ecfc3 Author: Stephan Bergmann sberg...@redhat.com Date: Tue May 29 11:20:40 2012 +0200 -Werror=conversion Change-Id: I3d4055722e8f7c22c50f892e62b5b3246904ce50 diff --git a/connectivity/inc/connectivity/dbcharset.hxx b/connectivity/inc/connectivity/dbcharset.hxx index cb13fcd..cbaceb0 100644 --- a/connectivity/inc/connectivity/dbcharset.hxx +++ b/connectivity/inc/connectivity/dbcharset.hxx @@ -29,6 +29,10 @@ #ifndef _DBHELPER_DBCHARSET_HXX_ #define _DBHELPER_DBCHARSET_HXX_ +#include sal/config.h + +#include cstddef + #include comphelper/stl_types.hxx #include rtl/textenc.h #include rtl/tencinfo.h @@ -84,7 +88,7 @@ namespace dbtools */ CharsetIterator find(const ::rtl::OUString _rIanaName, const IANA) const; -sal_Int32 size() const { ensureConstructed( ); return m_aEncodings.size(); } +std::size_t size() const { ensureConstructed( ); return m_aEncodings.size(); } /// get access to the first element of the charset collection CharsetIterator begin() const; ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: scp2/source
scp2/source/ooo/file_library_ooo.scp | 17 +++-- 1 file changed, 3 insertions(+), 14 deletions(-) New commits: commit 79179df3d97e6f838fa07cfffbbeb2987ce66c17 Author: Stephan Bergmann sberg...@redhat.com Date: Tue May 29 14:49:57 2012 +0200 Adapted to changed lib names after gbuild'ification Change-Id: Ia1c75076254c0300b19d6e12879143f7966b8294 diff --git a/scp2/source/ooo/file_library_ooo.scp b/scp2/source/ooo/file_library_ooo.scp index b2a6cbb..f92ed29 100644 --- a/scp2/source/ooo/file_library_ooo.scp +++ b/scp2/source/ooo/file_library_ooo.scp @@ -449,19 +449,8 @@ End #endif #ifdef ENABLE_KAB -File gid_File_Lib_Kab_1 -LIB_FILE_BODY; -Styles = (PACKED); -Dir = SCP2_OOO_BIN_DIR; -Name = STRING(CONCAT2(libkab1,UNXSUFFIX)); -End - -File gid_File_Lib_Kab_Drv_1 -LIB_FILE_BODY; -Styles = (PACKED); -Dir = SCP2_OOO_BIN_DIR; -Name = STRING(CONCAT2(libkabdrv1,UNXSUFFIX)); -End +STD_LIB_FILE(gid_File_Lib_Kab_1, kab1); +STD_LIB_FILE(gid_File_Lib_Kab_Drv_1, kabdrv1); #endif #ifdef MACOSX @@ -895,7 +884,7 @@ End #if !defined(SYSTEM_MOZILLA) !defined(WITHOUT_MOZILLA) !defined(MACOSX) // #i91209# -STD_LIB_FILE(gid_File_Lib_Mozab,mozab) +STD_LIB_FILE(gid_File_Lib_Mozab,mozab2) STD_LIB_FILE(gid_File_Lib_Mozabdrv,mozabdrv) #else ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: scp2/source
scp2/source/ooo/file_library_ooo.scp |4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) New commits: commit e62b67648afe11b941257d1d6156dcf1e45610e3 Author: Stephan Bergmann sberg...@redhat.com Date: Tue May 29 16:34:57 2012 +0200 Removed spurious trailing semicolons Change-Id: I49680c13989c9031150bae110e498a02af2476c4 diff --git a/scp2/source/ooo/file_library_ooo.scp b/scp2/source/ooo/file_library_ooo.scp index f92ed29..3b09ee2 100644 --- a/scp2/source/ooo/file_library_ooo.scp +++ b/scp2/source/ooo/file_library_ooo.scp @@ -449,8 +449,8 @@ End #endif #ifdef ENABLE_KAB -STD_LIB_FILE(gid_File_Lib_Kab_1, kab1); -STD_LIB_FILE(gid_File_Lib_Kab_Drv_1, kabdrv1); +STD_LIB_FILE(gid_File_Lib_Kab_1, kab1) +STD_LIB_FILE(gid_File_Lib_Kab_Drv_1, kabdrv1) #endif #ifdef MACOSX ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: 2 commits - officecfg/registry qadevOOo/runner sc/source
officecfg/registry/schema/org/openoffice/Office/Calc.xcs |2 - qadevOOo/runner/lib/MultiMethodTest.java |1 qadevOOo/runner/lib/StatusException.java | 18 --- sc/source/core/tool/defaultsoptions.cxx |2 - 4 files changed, 4 insertions(+), 19 deletions(-) New commits: commit 3e07bdfce6ef700a94868ccc03268aa005c295b8 Author: Stephan Bergmann sberg...@redhat.com Date: Tue May 29 18:23:12 2012 +0200 Revert fdo#50183 Use one sheet as default in Calc This reverts commit 9a7d3d47d9b4d57b0c7380eadb294add4245ba3a. It caused various tests in sc/JunitTest_sc_complex and sc/JunitTest_sc_unoapi to fail, which apparently implicitly depend on three instead of one sheets per default in Calc. diff --git a/officecfg/registry/schema/org/openoffice/Office/Calc.xcs b/officecfg/registry/schema/org/openoffice/Office/Calc.xcs index 62f0ff0..0cd2ff3 100644 --- a/officecfg/registry/schema/org/openoffice/Office/Calc.xcs +++ b/officecfg/registry/schema/org/openoffice/Office/Calc.xcs @@ -1620,7 +1620,7 @@ authorAlbert Thuswaldner/author descOption to set initial number of sheet tabs/desc /info - value1/value + value3/value /prop prop oor:name=SheetPrefix oor:type=xs:string !-- UIHints: Tools - Options - Spreadsheet - Defaults -- diff --git a/sc/source/core/tool/defaultsoptions.cxx b/sc/source/core/tool/defaultsoptions.cxx index 9683a7a..034b585 100644 --- a/sc/source/core/tool/defaultsoptions.cxx +++ b/sc/source/core/tool/defaultsoptions.cxx @@ -61,7 +61,7 @@ ScDefaultsOptions::~ScDefaultsOptions() void ScDefaultsOptions::SetDefaults() { -nInitTabCount = 1; +nInitTabCount = 3; aInitTabPrefix = ScGlobal::GetRscString(STR_TABLE_DEF); // Default Prefix Sheet } commit 1b330990733bcffbbf485df7aaed8130c899e2b8 Author: Stephan Bergmann sberg...@redhat.com Date: Tue May 29 18:22:25 2012 +0200 Improved error reporting Change-Id: Ib852987c2040d47cd3de08c6aebc649a068378d6 diff --git a/qadevOOo/runner/lib/MultiMethodTest.java b/qadevOOo/runner/lib/MultiMethodTest.java index 901f499..a68461e 100644 --- a/qadevOOo/runner/lib/MultiMethodTest.java +++ b/qadevOOo/runner/lib/MultiMethodTest.java @@ -222,6 +222,7 @@ public class MultiMethodTest } catch (Exception e) { +e.printStackTrace(); setSubStates(e.toString()); return tRes; } diff --git a/qadevOOo/runner/lib/StatusException.java b/qadevOOo/runner/lib/StatusException.java index 4768a92..a413257 100644 --- a/qadevOOo/runner/lib/StatusException.java +++ b/qadevOOo/runner/lib/StatusException.java @@ -34,12 +34,6 @@ package lib; */ public class StatusException extends RuntimeException { /** - * Contains an exception if the StatusException was created with - * StatusException(String, Throwable) constructor. - */ -protected Throwable exceptionThrown; - -/** * The Status contained in the StatusException. */ protected Status status; @@ -51,8 +45,7 @@ public class StatusException extends RuntimeException { * @param t the exception of the exception Status */ public StatusException( String message, Throwable t ) { -super( message ); -exceptionThrown = t; +super( message, t ); status = Status.exception( t ); } @@ -65,17 +58,9 @@ public class StatusException extends RuntimeException { } /** - * @return an exception, if this represents an exception Status, - * ttfalse/tt otherwise. - */ -public Throwable getThrownException() { -return exceptionThrown; -} - -/** * @return a status contained in the StatusException. */ public Status getStatus() { return status; } -} \ No newline at end of file +} ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: Advice needed about some cppcheck reports
On 05/28/2012 06:03 PM, julien2412 wrote: [sal/osl/unx/file.cxx:1261] - [sal/osl/unx/file.cxx:1261]: (style) Same expression on both sides of '-'. 1257 if (nSize 0) 1258 { 1259 c^= pData[0]; 1260 pData += nSize; 1261 nSize -= nSize; 1262 } Just put nSize to 0 ? Same thing here : [sal/osl/w32/file.cxx:880] - [sal/osl/w32/file.cxx:880]: (style) Same expression on both sides of '-'. 876 if (nSize 0) 877 { 878 c ^= pData[0]; 879 pData += nSize; 880 nSize -= nSize; 881 } In both of the above cases, the updates of both pData and nSize can be removed completely from these overspill if-blocks, as the variables are not used any more afterwards. And here : [filter/source/graphicfilter/icgm/cgm.cxx:269] - [filter/source/graphicfilter/icgm/cgm.cxx:269]: (style) Same expression on both sides of '-'. 267 if ( pLong[ nSwitch ] 0 ) 268 { 269 nRetValue -= nRetValue; 270 } 271 nRetValue /= 65536; I'd second Lubos here, that from looking at just the code in CGM::ImplGetFloat (but not understanding any of the context), nRetValue = -nRetValue is most likely what is meant instead. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [PATCH] Unify OUTPATH values
On 05/26/2012 12:59 PM, Francois Tigeot wrote: I don't think there's any good reason to have wildly different OUTPATH names for each specific platform variant. The various names also lead to unnecessary work when packaging, just to correctly handle LD_LIBRARY_PATH. The attached patch reduces OUTPATH to two possible values: unx and wnt. Since it is potentially intrusive, I prefer to show it here for review instead of committing it directly. As Norbert already wrote, the best course of action is probably to get to a state where building with a read-only source tree works (i.e., where all the build artefacts are generated to some distinct place). Then, differentiating on platform and .pro/non-pro can go away completely (why stop at unx vs. win, anyway?), and only some differentiation based on build and host is needed for cross compilation. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: ciruclar dependencies
On 05/29/2012 01:33 PM, Michael Stahl wrote: On 29/05/12 13:17, Tor Lillqvist wrote: the fastest way to start is still to load only the code that is necessary to start. Isn't there something called demand paging that takes care of that? it is supposed to do this, but falls short when it comes to startup performance on rotating storage, which is why we have various platform specific hacks to do linear reads of all libraries required at startup in order to hide the seek latencies incurred by demand paging. It also falls short on avoiding the runtime loader's work on relocations in data segments, execution of .init sections, etc. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Work in progress to allow for editing of read-only documents
Hi all, Just a heads up on the attached 0001-Work-in-progress-to-allow-for-editing-of-read-only-d.patch. Its commit message gives a rationale for what I aim at with the patch, but also explains why its not yet ready for prime time. Nevertheless, I would appreciate it if people versed with the details of the various applications could give it a try, to see whether it breaks any obscure behaviour. (For example, there is reportedly already some support in Writer for editing form content of r/o documents. Does this patch affect that?) The patch looks rather short, but I struggled with it on-and-off over the course of several weeks, as every little change triggered unexpected changes in other, seemingly unrelated areas. I'll try to get the Calc (and other applications?) problems fixed as soon as possible (I only noticed that problem now that I thought I was mainly done, but had only ever tested my changes with Writer), but am reluctant to press this into master towards LO 3.6 so short before feature freeze. Experience won over the last weeks indicates that too many surprises lurk in this area. So I would really like to hear other people's thoughts. Thanks, Stephan From e215c5b5796194d5ba3b3f4b926d296b21e495cb Mon Sep 17 00:00:00 2001 From: Stephan Bergmann sberg...@redhat.com Date: Thu, 31 May 2012 09:25:10 +0200 Subject: [PATCH] Work in progress to allow for editing of read-only documents When e.g. viewing mail attachments (that have been stored r/o to some download directory by the mail application), it would be nice if the user could easily temporarily modify them (say, play around with a spreadsheet, changing some numbers and triggering recalculation of formulas) by clicking the Edit File button and not being asked to create a copy for editing. This patch tries to make editability of a view independent of the r/o status of the medium: * SID_EDITDOC (the Edit File button) now only toggles the r/o status of the view. It no longer asks to create a copy for editing if the underlying medium is r/o. * When a modified document is toggled to r/o via SID_EDITDOC, LO still asks the user to save or discard the changes. However, if the underlying medium is physically r/o (see next), saving the document opens the Save As dialog, instead of just doing a Save operation (which would fail on the r/o file). * A new state of IsOriginallyReadOnly needed to be added to the medium, to keep track whether the medium was originally opened r/o (and is thus assumed to be physically r/o), as toggling SID_EDITDOC in the view also changes the open mode of the underlying medium. Instead of trying to fully understand and disentangle that horrible mess, I just added yet another state to the mess... * The title of the document window now contains (read-only) if and only if either the view is r/o or the medium is originally r/o (or both). However, while this patch appears to work fine at least for simple cases of Writer documents, it does not work well yet e.g. for Calc documents. Opening a r/o .ods file shows it in r/o view mode, but nonetheless allows for modification of cell content etc. Change-Id: Ic81f625eb0494dbd90db2e27403e89d48b2c3f87 --- sfx2/inc/sfx2/docfile.hxx|5 + sfx2/inc/sfx2/objsh.hxx |1 + sfx2/source/doc/docfile.cxx | 20 sfx2/source/doc/objmisc.cxx | 12 sfx2/source/doc/objstor.cxx |2 +- sfx2/source/doc/objxtor.cxx |2 +- sfx2/source/view/viewfrm.cxx | 28 7 files changed, 48 insertions(+), 22 deletions(-) diff --git a/sfx2/inc/sfx2/docfile.hxx b/sfx2/inc/sfx2/docfile.hxx index f7f85d5..ca8ad49 100644 --- a/sfx2/inc/sfx2/docfile.hxx +++ b/sfx2/inc/sfx2/docfile.hxx @@ -213,6 +213,11 @@ public: GetVersionList( bool _bNoReload = false ); sal_BoolIsReadOnly(); +// Whether the medium had originally been opened r/o, independent of later +// changes via SetOpenMode; used to keep track of the true state of the +// medium across toggles via SID_EDITDOC (which do change SetOpenMode): +boolIsOriginallyReadOnly() const; + ::com::sun::star::uno::Reference ::com::sun::star::io::XInputStream GetInputStream(); voidCreateTempFile( sal_Bool bReplace = sal_True ); diff --git a/sfx2/inc/sfx2/objsh.hxx b/sfx2/inc/sfx2/objsh.hxx index 3c1837b..90acbb7 100644 --- a/sfx2/inc/sfx2/objsh.hxx +++ b/sfx2/inc/sfx2/objsh.hxx @@ -273,6 +273,7 @@ public: voidSetReadOnly(); sal_BoolIsReadOnly() const; sal_BoolIsReadOnlyMedium() const; +boolIsOriginallyReadOnlyMedium() const; voidSetReadOnlyUI( sal_Bool bReadOnly = sal_True ); sal_BoolIsReadOnlyUI() const; voidSetNoName(); diff --git a/sfx2/source/doc
[Libreoffice-commits] .: 2 commits - officecfg/registry qadevOOo/tests sc/source
officecfg/registry/schema/org/openoffice/Office/Calc.xcs |2 +- qadevOOo/tests/java/mod/_sc/ScModelObj.java |3 +++ sc/source/core/tool/defaultsoptions.cxx |2 +- 3 files changed, 5 insertions(+), 2 deletions(-) New commits: commit 71056d59de59561ee1d74a3c2d9d66a223422c7e Author: Stephan Bergmann sberg...@redhat.com Date: Thu May 31 11:36:17 2012 +0200 Revert Revert fdo#50183 Use one sheet as default in Calc This reverts commit 3e07bdfce6ef700a94868ccc03268aa005c295b8. The problems with the tests still assuming three sheets by default have now been addressed. diff --git a/officecfg/registry/schema/org/openoffice/Office/Calc.xcs b/officecfg/registry/schema/org/openoffice/Office/Calc.xcs index 245f2ff..1922c6c 100644 --- a/officecfg/registry/schema/org/openoffice/Office/Calc.xcs +++ b/officecfg/registry/schema/org/openoffice/Office/Calc.xcs @@ -1649,7 +1649,7 @@ authorAlbert Thuswaldner/author descOption to set initial number of sheet tabs/desc /info - value3/value + value1/value /prop prop oor:name=SheetPrefix oor:type=xs:string !-- UIHints: Tools - Options - Spreadsheet - Defaults -- diff --git a/sc/source/core/tool/defaultsoptions.cxx b/sc/source/core/tool/defaultsoptions.cxx index 034b585..9683a7a 100644 --- a/sc/source/core/tool/defaultsoptions.cxx +++ b/sc/source/core/tool/defaultsoptions.cxx @@ -61,7 +61,7 @@ ScDefaultsOptions::~ScDefaultsOptions() void ScDefaultsOptions::SetDefaults() { -nInitTabCount = 3; +nInitTabCount = 1; aInitTabPrefix = ScGlobal::GetRscString(STR_TABLE_DEF); // Default Prefix Sheet } commit 0210ec32f779bb2098657eb4ad69cca7a60835dc Author: Stephan Bergmann sberg...@redhat.com Date: Thu May 31 12:25:38 2012 +0200 Fix one more test that implicitly assumes more than one Calc sheet ...in the vein of e43966b29cc36ce36245cfb2c63c3e24a0eb523f Fix up unit tests that assumed two or more sheets per default Change-Id: I450f561721edeaa6855499d20c2196ceb8d2f546 diff --git a/qadevOOo/tests/java/mod/_sc/ScModelObj.java b/qadevOOo/tests/java/mod/_sc/ScModelObj.java index 6dd9394..c18e453 100644 --- a/qadevOOo/tests/java/mod/_sc/ScModelObj.java +++ b/qadevOOo/tests/java/mod/_sc/ScModelObj.java @@ -194,6 +194,9 @@ public class ScModelObj extends TestCase { oSheet.getCellByPosition(5, 5), oSheet.getCellByPosition(6, 5) }; + +// Make sure there are at least two sheets: +oSheets.insertNewByName(Some Sheet, (short) 1); } catch (com.sun.star.lang.WrappedTargetException e) { e.printStackTrace(log); throw new StatusException( ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: [PATCH] Re: 1 sheet by default [was Re: Tinderbox failure, Linux-RHEL6-x86_64@14-with-check, MASTER, last success: 2012-05-15 18:12:02]
On 05/30/2012 03:53 PM, Stefan Knorr (Astron) wrote: first: Stephan, I'm very sorry for breaking your tinderbox(es) for days and not noticing it. (I just assumed my 2-byte change of a commit was innocent enough. I'll do make check in the future as you advised.) No problem. And picking up using make check routinely is highly appreciated. With some dumbluck and [1], I've managed to get it down to 1 failing test. My approach so far was to just unconditionally add a sheet (in the case of the XMLStylesExporter and XMLSettingsExporter I just made sure these tests actually ran on Sheet 1 (=0), not Sheet 2 (=1)). Since I guess my approach is less than optimal, do tell me if I need to wrap stuff in try/catch or so. Or, if I should ask LibO how many sheets there are before creating a new one. Or something else. The approach looks rather sane to me (and apparently to Kendy also, who already pushed). In general, the qadevOOo based tests are rather a mess, and nobody loves them. However, they proved useful in finding regressions over and over again, so we should see to keep them working until newer, better tests replace their code coverage (and even go beyond). About the last failing test, sc_unoapi/ScModelObj, I am unsure where it needs the second sheet, so I didn't add anything yet. Advice appreciated. I just extended your approach of explicitly inserting an additional sheet to qadevOOo/tests/java/mod/_sc/ScModelObj.java (see http://cgit.freedesktop.org/libreoffice/core/commit/?id=0210ec32f779bb2098657eb4ad69cca7a60835dc Fix one more test that implicitly assumes more than one Calc sheet) and it appears to work well. So I also re-reverted the Use one sheet as default in Calc feature (see http://cgit.freedesktop.org/libreoffice/core/commit/?id=71056d59de59561ee1d74a3c2d9d66a223422c7e Revert 'Revert 'fdo#50183 Use one sheet as default in Calc''). Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: About fdo#47044 crashed when accessing Proxy etc
On 05/31/2012 01:52 PM, julien2412 wrote: Just noticed that fdo#49903 contained the same top lines (more than 10) on the bt. However, I wonder why this part of code (shell\source\backends\wininetbe\wininetbackend.cxx) is called. Looks like doc contains a http link to a graphic, so webdav ucp is called to load that, which asks configmgr about proxy settings, which (in default system proxy settings mode) asks WinInetBackend for the proxy setting values. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: cppuhelper/source
cppuhelper/source/defaultbootstrap.cxx | 433 +++-- 1 file changed, 314 insertions(+), 119 deletions(-) New commits: commit c16d0dd846e25df56feffe868a15eb33bb55e42a Author: Stephan Bergmann sberg...@redhat.com Date: Fri Jun 1 09:59:49 2012 +0200 fdo#49291 Postpone actual factory instantiation as long as possible ...see comment in ServiceManager::createContentEnumeration for a rationale. Splitting ImplementationInfo out of Implementation has become necessary to avoid circular references. Change-Id: I29aef81ce78b9ab71e18663f8c7e6ca913c6a650 diff --git a/cppuhelper/source/defaultbootstrap.cxx b/cppuhelper/source/defaultbootstrap.cxx index c903c2b..55a4f09 100644 --- a/cppuhelper/source/defaultbootstrap.cxx +++ b/cppuhelper/source/defaultbootstrap.cxx @@ -60,6 +60,7 @@ #include cppuhelper/compbase8.hxx #include cppuhelper/component_context.hxx #include cppuhelper/implbase1.hxx +#include cppuhelper/implbase3.hxx #include cppuhelper/shlib.hxx #include osl/file.hxx #include registry/registry.hxx @@ -124,36 +125,51 @@ void decodeRdbUri(rtl::OUString * uri, bool * optional, bool * directory) { } } +struct ImplementationInfo: private boost::noncopyable { +ImplementationInfo( +rtl::OUString const theName, rtl::OUString const theLoader, +rtl::OUString const theUri, rtl::OUString const thePrefix, +css::uno::Reference css::uno::XComponentContext const +theAlienContext): +name(theName), loader(theLoader), uri(theUri), prefix(thePrefix), +alienContext(theAlienContext) +{} + +explicit ImplementationInfo(rtl::OUString const theName): name(theName) {} + +rtl::OUString const name; +rtl::OUString const loader; +rtl::OUString const uri; +rtl::OUString const prefix; +css::uno::Reference css::uno::XComponentContext const alienContext; +std::vector rtl::OUString services; +std::vector rtl::OUString singletons; +}; + struct Implementation: private boost::noncopyable { Implementation( -rtl::OUString const theName, rtl::OUString const theLoader, -rtl::OUString const theUri, -rtl::OUString const thePrefix = rtl::OUString(), +rtl::OUString const name, rtl::OUString const loader, +rtl::OUString const uri, +rtl::OUString const prefix = rtl::OUString(), css::uno::Reference css::uno::XComponentContext const -theAlienContext +alienContext = css::uno::Reference css::uno::XComponentContext ()): -name(theName), loader(theLoader), uri(theUri), prefix(thePrefix), -alienContext(theAlienContext), loaded(false) +info(new ImplementationInfo(name, loader, uri, prefix, alienContext)), +loaded(false) {} Implementation( -rtl::OUString const theName, +rtl::OUString const name, css::uno::Reference css::lang::XSingleComponentFactory const theFactory1, css::uno::Reference css::lang::XSingleServiceFactory const theFactory2, css::uno::Reference css::lang::XComponent const theComponent): -name(theName), factory1(theFactory1), factory2(theFactory2), -component(theComponent), loaded(true) +info(new ImplementationInfo(name)), factory1(theFactory1), +factory2(theFactory2), component(theComponent), loaded(true) {} -rtl::OUString const name; -rtl::OUString const loader; -rtl::OUString const uri; -rtl::OUString const prefix; -css::uno::Reference css::uno::XComponentContext const alienContext; -std::vector rtl::OUString services; -std::vector rtl::OUString singletons; +boost::shared_ptr ImplementationInfo info; css::uno::Reference css::lang::XSingleComponentFactory factory1; css::uno::Reference css::lang::XSingleServiceFactory factory2; css::uno::Reference css::lang::XComponent component; @@ -463,13 +479,13 @@ void Parser::handleImplementation() { void Parser::handleService() { rtl::OUString name(getNameAttribute()); -implementation_-services.push_back(name); +implementation_-info-services.push_back(name); data_-services[name].push_back(implementation_); } void Parser::handleSingleton() { rtl::OUString name(getNameAttribute()); -implementation_-singletons.push_back(name); +implementation_-info-singletons.push_back(name); data_-singletons[name].push_back(implementation_); } @@ -585,8 +601,18 @@ public: context_ = context; } +css::uno::Reference css::uno::XComponentContext getContext() const { +assert(context_.is()); +return context_; +} + Data const getData() const { return data_; } +void loadImplementation( +boost::shared_ptr ImplementationInfo const info, +css::uno::Reference css::lang::XSingleComponentFactory * factory1, +css::uno
[Libreoffice-commits] .: 2 commits - cppuhelper/source
cppuhelper/source/defaultbootstrap.cxx | 26 +++--- 1 file changed, 11 insertions(+), 15 deletions(-) New commits: commit 5fd78feebe220f848c4a0d3beeb1ef0f04bbae6f Author: Stephan Bergmann sberg...@redhat.com Date: Fri Jun 1 10:10:07 2012 +0200 Revert bootstrapping should elide duplicate .rdb paths in the configuration This reverts commit b162aec6254ab535cc5eb990b249f46aa8e79153, which increased code complexity for no benefit (the dubious scenario it was introduced for concerned duplicate service rdbs rather than type rdbs, anyway). diff --git a/cppuhelper/source/defaultbootstrap.cxx b/cppuhelper/source/defaultbootstrap.cxx index dd7aeaf..1f119c9 100644 --- a/cppuhelper/source/defaultbootstrap.cxx +++ b/cppuhelper/source/defaultbootstrap.cxx @@ -2157,7 +2157,6 @@ css::uno::Reference css::registry::XSimpleRegistry createTypeRegistry( css::uno::Reference css::registry::XRegistryKey ()), css::uno::UNO_QUERY_THROW); css::uno::Reference css::registry::XSimpleRegistry reg; -std::vectorrtl::OUString loaded; for (sal_Int32 i = 0; i != -1;) { rtl::OUString uri(uris.getToken(0, ' ', i)); if (uri.isEmpty()) { @@ -2166,13 +2165,9 @@ css::uno::Reference css::registry::XSimpleRegistry createTypeRegistry( bool optional; bool directory; decodeRdbUri(uri, optional, directory); -if (::std::find (loaded.begin(), loaded.end(), uri) == loaded.end()) -{ -reg = directory -? readTypeRdbDirectory(uri, optional, reg, simpleRegs, nestedRegs) -: readTypeRdbFile(uri, optional, reg, simpleRegs, nestedRegs); -loaded.push_back(uri); -} +reg = directory +? readTypeRdbDirectory(uri, optional, reg, simpleRegs, nestedRegs) +: readTypeRdbFile(uri, optional, reg, simpleRegs, nestedRegs); } return reg; } commit 0951b76a00edd1cd89b8acf86c785d56fc776351 Author: Stephan Bergmann sberg...@redhat.com Date: Fri Jun 1 10:09:18 2012 +0200 Add virtual for consistency Change-Id: I8ecc2edde553edbd80ca04e2f3d541c31f516211 diff --git a/cppuhelper/source/defaultbootstrap.cxx b/cppuhelper/source/defaultbootstrap.cxx index 55a4f09..dd7aeaf 100644 --- a/cppuhelper/source/defaultbootstrap.cxx +++ b/cppuhelper/source/defaultbootstrap.cxx @@ -648,27 +648,28 @@ public: css::uno::Reference css::uno::XComponentContext const Context) throw (css::uno::Exception, css::uno::RuntimeException); -css::uno::Type SAL_CALL getElementType() throw (css::uno::RuntimeException); +virtual css::uno::Type SAL_CALL getElementType() +throw (css::uno::RuntimeException); -sal_Bool SAL_CALL hasElements() throw (css::uno::RuntimeException); +virtual sal_Bool SAL_CALL hasElements() throw (css::uno::RuntimeException); -css::uno::Reference css::container::XEnumeration SAL_CALL +virtual css::uno::Reference css::container::XEnumeration SAL_CALL createEnumeration() throw (css::uno::RuntimeException); -sal_Bool SAL_CALL has(css::uno::Any const aElement) +virtual sal_Bool SAL_CALL has(css::uno::Any const aElement) throw (css::uno::RuntimeException); -void SAL_CALL insert(css::uno::Any const aElement) +virtual void SAL_CALL insert(css::uno::Any const aElement) throw ( css::lang::IllegalArgumentException, css::container::ElementExistException, css::uno::RuntimeException); -void SAL_CALL remove(css::uno::Any const aElement) +virtual void SAL_CALL remove(css::uno::Any const aElement) throw ( css::lang::IllegalArgumentException, css::container::NoSuchElementException, css::uno::RuntimeException); -css::uno::Reference css::container::XEnumeration SAL_CALL +virtual css::uno::Reference css::container::XEnumeration SAL_CALL createContentEnumeration(rtl::OUString const aServiceName) throw (css::uno::RuntimeException); ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: About fdo#47044 crashed when accessing Proxy etc
On 05/30/2012 11:45 PM, julien2412 wrote: Quite recently, an interesting bt has been published by bfoman about this bug (see https://bugs.freedesktop.org/attachment.cgi?id=62055) , here are the top lines : sal3!rtl_uString_newFromAscii+0x15 wininetbe1_uno!rtl::OUString::createFromAscii+0x40 wininetbe1_uno!WinInetBackend::WinInetBackend+0x1a0 Searching about createFromAscii in wininetbackend files, I found them in wininetbackend.cxx The only lines containing createFromAscii are : 157 rtl::OUString aProxyList = rtl::OUString::createFromAscii( lpi-lpszProxy ); 158 rtl::OUString aProxyBypassList = rtl::OUString::createFromAscii( lpi-lpszProxyBypass ); Then I wonder what was lpi type and just read some lines above : 126 LPINTERNET_PROXY_INFO lpi = NULL; Then searching about LPINTERNET_PROXY_INFO gave this url : http://msdn.microsoft.com/en-us/library/windows/desktop/aa385148%28v=vs.85%29.aspx typedef struct { DWORD dwAccessType; LPCTSTR lpszProxy; LPCTSTR lpszProxyBypass; } INTERNET_PROXY_INFO, * LPINTERNET_PROXY_INFO; So kept on with attributes lpszProxy and lpszProxyBypass and its type LPCTSTR, it gave this link : http://msdn.microsoft.com/en-us/library/aa383751%28v=vs.85%29.aspx This type is declared in WinNT.h as follows: #ifdef UNICODE typedef LPCWSTR LPCTSTR; #else typedef LPCSTR LPCTSTR; #endif On the same page, we can read : for LPCSTR : A pointer to a constant null-terminated string of 8-bit Windows (ANSI) characters. For more information, see Character Sets Used By Fonts. This type is declared in WinNT.h as follows: typedef __nullterminated CONST CHAR *LPCSTR; and for LPCWSTR : A pointer to a constant null-terminated string of 16-bit Unicode characters. For more information, see Character Sets Used By Fonts. This type is declared in WinNT.h as follows: typedef CONST WCHAR *LPCWSTR; Then going back to createFromAscii to see if it could match with all this : sal/inc/rtl/ustring.hxx: static OUString createFromAscii( const sal_Char * value ) SAL_THROW(()) + sal/inc/sal/types.h : typedef char sal_Char; Now I wonder if it's ok to use createFromAscii to manage the attributes lpszProxy and lpszProxyBypass ? (no ironical question here, just a beginner question only :-)) wininetbackend.cxx explicitly calls the InternetQueryOptionA variant, so the returned lpszProxy and lpszProxyBypass are char-sized strings, so calling createFromAscii shall work. (Strictly speaking, it would cause confusion if the strings contained characters outside ASCII, but it would not lead to a crash within createFromAscii.) (And the version of LPINTERNET_PROXY_INFO seen in wininetbackend.cxx also is the one using LPCSTR, not LPCWSTR, as the call to createFromAscii would otherwise fail to compile; and again, char/wchar_t mismatch would not explain a crash within createFromAscii, anyway.) The crash within createFromAscii is near the start of rtl_uString_newFromAscii, from the crash information it looks like it appears at dereferencing the first byte of pCharStr (aka pTempStr), where pCharStr == 0x8fda7dbb. So it looks like InternetQueryOptionA returns with the lpszProxy and/or lpszProxyBypass pointers pointing to non-allocated memory. What one notices is that neither of the two calls to InternetQueryOptionA in WinInetBackend check the return value (the first shall return with FALSE and GetLastError()==ERROR_INSUFFICIENT_BUFFER, while the second shall return with TRUE). Maybe the call just fails and returns FALSE? (http://msdn.microsoft.com/en-us/library/windows/desktop/aa385328%28v=vs.85%29.aspx Option Flags in the description of INTERNET_OPTION_PROXY states that it is deprecated in favour of INTERNET_OPTION_PER_CONNECTION_OPTION.) Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Meaning of _SOLAR__PRIVATE
On 05/31/2012 05:51 PM, Eike Rathke wrote: Or rather if you state you're a friend of me I believe you ... wow, that would make _SOLAR__PRIVATE the forerunner of facebook ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: RepositoryModule_ooo.mk
RepositoryModule_ooo.mk |1 - 1 file changed, 1 deletion(-) New commits: commit da621dadb7d4e108d6d6187e5877f821ac6a1c0b Author: Stephan Bergmann sberg...@redhat.com Date: Fri Jun 1 11:58:35 2012 +0200 Remove duplicate xmlsecurity entry Change-Id: I700ba0d843f1ddd15f440826baa5f81285acabee diff --git a/RepositoryModule_ooo.mk b/RepositoryModule_ooo.mk index 5359f6e..1cd30dc 100644 --- a/RepositoryModule_ooo.mk +++ b/RepositoryModule_ooo.mk @@ -157,7 +157,6 @@ $(eval $(call gb_Module_add_moduledirs,ooo,\ x11_extensions \ ) \ xmerge \ -xmlsecurity \ xml2cmp \ xmlhelp \ xmloff \ ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: fdo#46808, Adapt UNO services to new style, Part 7, updating ::create
On 06/01/2012 03:27 PM, Noel Grandin wrote: I'm hoping Stephan can either patch up the problems or point me in the right direction - build error log attached. Will take a look. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: fdo#46808, Adapt UNO services to new style, Part 7, updating ::create
On 06/01/2012 04:53 PM, Riccardo Magliocchetti wrote: It's inside the zip file. Maybe your email firewall dropped it? Some of them do that. I see only the build error log attached Yep, patches.zip apparently got stripped from the mail as circulated via the mailing list (it is intact in the copy I got directly from Noel). Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: sc/source
sc/source/core/data/documen2.cxx |3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) New commits: commit 477c2229d2532a11ec8c81b9ef392ce49357f95c Author: Stephan Bergmann sberg...@redhat.com Date: Fri Jun 1 17:27:58 2012 +0200 Missing initialization ...which could cause cells in r/o Calc documents to be editable Change-Id: I2be0e23119cf8424123007a0440a6156c50b diff --git a/sc/source/core/data/documen2.cxx b/sc/source/core/data/documen2.cxx index 158c731..f072f87 100644 --- a/sc/source/core/data/documen2.cxx +++ b/sc/source/core/data/documen2.cxx @@ -218,7 +218,8 @@ ScDocument::ScDocument( ScDocumentMode eMode, mbExecuteLinkEnabled( true ), mbChangeReadOnlyEnabled( false ), mbStreamValidLocked( false ), -mnNamedRangesLockCount( 0 ) +mnNamedRangesLockCount( 0 ), +mbIsInTest( false ) { SetStorageGrammar( formula::FormulaGrammar::GRAM_STORAGE_DEFAULT); ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: sfx2/inc sfx2/source
sfx2/inc/sfx2/docfile.hxx|5 + sfx2/inc/sfx2/objsh.hxx |1 + sfx2/source/doc/docfile.cxx | 20 sfx2/source/doc/objmisc.cxx | 12 sfx2/source/doc/objstor.cxx |6 +++--- sfx2/source/doc/objxtor.cxx |2 +- sfx2/source/view/viewfrm.cxx | 28 7 files changed, 50 insertions(+), 24 deletions(-) New commits: commit b9ecec7c74687ed5a9470cffb7d02e0e6e83107e Author: Stephan Bergmann sberg...@redhat.com Date: Fri Jun 1 17:32:07 2012 +0200 Allow for editing of read-only documents When e.g. viewing mail attachments (that have been stored r/o to some download directory by the mail application), it would be nice if the user could easily temporarily modify them (say, play around with a spreadsheet, changing some numbers and triggering recalculation of formulas) by clicking the Edit File button and not being asked to create a copy for editing. This patch tries to make editability of a view independent of the r/o status of the medium: * SID_EDITDOC (the Edit File button) now only toggles the r/o status of the view. It no longer asks to create a copy for editing if the underlying medium is r/o. * When a modified document is toggled to r/o via SID_EDITDOC, LO still asks the user to save or discard the changes. However, if the underlying medium is physically r/o (see next), saving the document opens the Save As dialog, instead of just doing a Save operation (which would fail on the r/o file). * A new state of IsOriginallyReadOnly needed to be added to the medium, to keep track whether the medium was originally opened r/o (and is thus assumed to be physically r/o), as toggling SID_EDITDOC in the view also changes the open mode of the underlying medium. Instead of trying to fully understand and disentangle that horrible mess, I just added yet another state to the mess... * The title of the document window now contains (read-only) if and only if either the view is r/o or the medium is originally r/o (or both). Change-Id: I89d9c6adf0baab411e737a5f4e6f4e770e7a70be diff --git a/sfx2/inc/sfx2/docfile.hxx b/sfx2/inc/sfx2/docfile.hxx index f7f85d5..ca8ad49 100644 --- a/sfx2/inc/sfx2/docfile.hxx +++ b/sfx2/inc/sfx2/docfile.hxx @@ -213,6 +213,11 @@ public: GetVersionList( bool _bNoReload = false ); sal_BoolIsReadOnly(); +// Whether the medium had originally been opened r/o, independent of later +// changes via SetOpenMode; used to keep track of the true state of the +// medium across toggles via SID_EDITDOC (which do change SetOpenMode): +boolIsOriginallyReadOnly() const; + ::com::sun::star::uno::Reference ::com::sun::star::io::XInputStream GetInputStream(); voidCreateTempFile( sal_Bool bReplace = sal_True ); diff --git a/sfx2/inc/sfx2/objsh.hxx b/sfx2/inc/sfx2/objsh.hxx index 3c1837b..90acbb7 100644 --- a/sfx2/inc/sfx2/objsh.hxx +++ b/sfx2/inc/sfx2/objsh.hxx @@ -273,6 +273,7 @@ public: voidSetReadOnly(); sal_BoolIsReadOnly() const; sal_BoolIsReadOnlyMedium() const; +boolIsOriginallyReadOnlyMedium() const; voidSetReadOnlyUI( sal_Bool bReadOnly = sal_True ); sal_BoolIsReadOnlyUI() const; voidSetNoName(); diff --git a/sfx2/source/doc/docfile.cxx b/sfx2/source/doc/docfile.cxx index 4da1556..e77992e 100644 --- a/sfx2/source/doc/docfile.cxx +++ b/sfx2/source/doc/docfile.cxx @@ -312,6 +312,8 @@ public: uno::Reference logging::XSimpleLogRing m_xLogRing; +bool m_originallyReadOnly; + SfxMedium_Impl( SfxMedium* pAntiImplP ); ~SfxMedium_Impl(); }; @@ -339,7 +341,8 @@ SfxMedium_Impl::SfxMedium_Impl( SfxMedium* pAntiImplP ) pTempFile( NULL ), nLastStorageError( 0 ), m_bRemoveBackup( sal_False ), -m_nSignatureState( SIGNATURESTATE_NOSIGNATURES ) +m_nSignatureState( SIGNATURESTATE_NOSIGNATURES ), +m_originallyReadOnly(false) { aDoneLink.CreateMutex(); } @@ -1057,6 +1060,11 @@ sal_Bool SfxMedium::LockOrigFileOnDemand( sal_Bool bLoading, sal_Bool bNoUI ) bContentReadonly = IsReadonlyAccordingACL( aPhysPath.GetBuffer() ); } #endif + +if ( bContentReadonly ) +{ +pImp-m_originallyReadOnly = true; +} } // do further checks only if the file not readonly in fs @@ -2874,15 +2882,14 @@ SfxMedium::SfxMedium( const ::com::sun::star::uno::Sequence ::com::sun::star::b } } -sal_Bool bReadOnly = sal_False; SFX_ITEMSET_ARG( pSet, pReadOnlyItem, SfxBoolItem, SID_DOC_READONLY, sal_False
Re: Work in progress to allow for editing of read-only documents
On 05/31/2012 11:51 AM, Michael Meeks wrote: On Thu, 2012-05-31 at 10:09 +0200, Stephan Bergmann wrote: Nevertheless, I would appreciate it if people versed with the details of the various applications could give it a try, to see whether it breaks any obscure behaviour. (For example, there is reportedly already some support in Writer for editing form content of r/o documents. Does this patch affect that?) I'll send you a sample document with the wunder-interoperable fields in it. Hm, with both my changed master as well as a stock 3.5, I could only get that document to expose editable fields when the document was in r/w mode and non-editable fields when the document was in r/o mode. So not sure that document (or the way I inspected it) helps verify I broke something. I'll try to get the Calc (and other applications?) problems fixed as soon as possible (I only noticed that problem now that I thought I was mainly done, but had only ever tested my changes with Writer), but am Turned out the Calc thing was independent of my changes; http://cgit.freedesktop.org/libreoffice/core/commit/?id=477c2229d2532a11ec8c81b9ef392ce49357f95c Missing initialization fixes that. reluctant to press this into master towards LO 3.6 so short before feature freeze. Experience won over the last weeks indicates that too many surprises lurk in this area. So I would really like to hear other people's thoughts. How many unit tests did we get out of that experience ;- if there were some, I'd be well up for putting it into 3.6. If we find it causes Yeah, unit tests would have been nice. fill in lame excuse of choice here... significant problems, it looks small enough to back out reasonably easily (right?). Indeed. So I boldly pushed this now as http://cgit.freedesktop.org/libreoffice/core/commit/?id=b9ecec7c74687ed5a9470cffb7d02e0e6e83107e Allow for editing of read-only documents (with one small change compared to the earlier patch, so that r/o non-ODF documents correctly open as r/o, too). Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice-commits] .: 2 commits - chart2/source connectivity/source cui/source desktop/source editeng/source embeddedobj/source extensions/source forms/source framework/source oox/source sc/Cpp
On 06/04/2012 09:14 PM, Norbert Thiebaud wrote: On 6/4/12 09:58 , Stephan Bergmann wrote: ... sdext/Library_minimizer.mk |1 sdext/source/minimizer/impoptimizer.cxx |6 sdext/source/minimizer/informationdialog.cxx |5 54 files changed, 426 insertions(+), 490 deletions(-) New commits: commit b3c76dee6d44d07eae404b8d7341e6c88e6c4429 Author: Noel Grandinn...@peralex.com Date: Sat May 26 13:53:19 2012 +0200 fdo#46808, Adapt UNO services to new style, Part 7, updating ::create Update calls to factories to use new ::create methods Change-Id: I01d4417820f52718836c92faf3c2fae0dc96b30d Signed-off-by: Stephan Bergmannsberg...@redhat.com, added some tweaks. seems to make MacOSX unhappy: illegal combination shl/OXT/OOO at /Volumes/TBRAM/core/solenv/bin/macosx-change-install-names.pl line 70,IN line 3. make[2]: *** [/Volumes/TBRAM/core/workdir/unxmacxi.pro/LinkTarget/Library/SunPresentationMinimizer.uno.dylib] Error 9 make[2]: *** Deleting file `/Volumes/TBRAM/core/workdir/unxmacxi.pro/LinkTarget/Library/SunPresentationMinimizer.uno.dylib' argh, sure, an extension must not link against comphelper will take care Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: sdext/Library_minimizer.mk sdext/source
sdext/Library_minimizer.mk |1 - sdext/source/minimizer/impoptimizer.cxx |5 ++--- sdext/source/minimizer/informationdialog.cxx |3 +-- 3 files changed, 3 insertions(+), 6 deletions(-) New commits: commit 719618f28e003c545c20a8fc94d348b667c8376a Author: Stephan Bergmann sberg...@redhat.com Date: Tue Jun 5 09:15:05 2012 +0200 Remove bogus dependency of extension on comphelper Change-Id: I2a0d4bc968b9b1d285006f26603da4a98fbd870b diff --git a/sdext/Library_minimizer.mk b/sdext/Library_minimizer.mk index f19b541..5c918c3 100644 --- a/sdext/Library_minimizer.mk +++ b/sdext/Library_minimizer.mk @@ -35,7 +35,6 @@ $(eval $(call gb_Library_use_api,SunPresentationMinimizer,\ )) $(eval $(call gb_Library_use_libraries,SunPresentationMinimizer,\ -comphelper \ cppu \ cppuhelper \ sal \ diff --git a/sdext/source/minimizer/impoptimizer.cxx b/sdext/source/minimizer/impoptimizer.cxx index 699dc34..8afd16f 100644 --- a/sdext/source/minimizer/impoptimizer.cxx +++ b/sdext/source/minimizer/impoptimizer.cxx @@ -69,7 +69,6 @@ #include com/sun/star/io/XSeekable.hpp #include com/sun/star/frame/XComponentLoader.hpp #include com/sun/star/util/URL.hpp -#include comphelper/componentcontext.hxx using namespace ::std; using namespace ::rtl; @@ -351,7 +350,7 @@ Reference XGraphic ImpCompressGraphic( const Reference XComponentContext { Reference XStream xTempFile( rxMSF-getServiceManager()-createInstanceWithContext( OUString(com.sun.star.io.TempFile), rxMSF ), UNO_QUERY_THROW ); Reference XOutputStream xOutputStream( xTempFile-getOutputStream() ); -Reference XGraphicProvider xGraphicProvider( GraphicProvider::create(comphelper::ComponentContext(rxMSF).getUNOContext()) ); +Reference XGraphicProvider xGraphicProvider( GraphicProvider::create( rxMSF ) ); ImpCompressGraphic( xGraphicProvider, xGraphic, xOutputStream, aDestMimeType, aLogicalSize, rGraphicSettings.mnJPEGQuality, rGraphicSettings.mnImageResolution, bRemoveCropArea, aGraphicCropLogic ); Reference XInputStream xInputStream( xTempFile-getInputStream() ); @@ -371,7 +370,7 @@ Reference XGraphic ImpCompressGraphic( const Reference XComponentContext rtl::OUString aDestMimeType( aSourceMimeType ); Reference XStream xTempFile( rxMSF-getServiceManager()-createInstanceWithContext( OUString(com.sun.star.io.TempFile), rxMSF ), UNO_QUERY_THROW ); Reference XOutputStream xOutputStream( xTempFile-getOutputStream() ); -Reference XGraphicProvider xGraphicProvider( GraphicProvider::create( ::comphelper::ComponentContext(rxMSF).getUNOContext() ) ); +Reference XGraphicProvider xGraphicProvider( GraphicProvider::create( rxMSF ) ); ImpCompressGraphic( xGraphicProvider, xGraphic, xOutputStream, aDestMimeType, aLogicalSize, rGraphicSettings.mnJPEGQuality, rGraphicSettings.mnImageResolution, sal_False, aGraphicCropLogic ); Reference XInputStream xInputStream( xTempFile-getInputStream() ); Reference XSeekable xSeekable( xInputStream, UNO_QUERY_THROW ); diff --git a/sdext/source/minimizer/informationdialog.cxx b/sdext/source/minimizer/informationdialog.cxx index 7f510b6..c74599f 100644 --- a/sdext/source/minimizer/informationdialog.cxx +++ b/sdext/source/minimizer/informationdialog.cxx @@ -37,7 +37,6 @@ #include sal/macros.h #include com/sun/star/util/URL.hpp #include com/sun/star/util/XURLTransformer.hpp -#include comphelper/componentcontext.hxx #define DIALOG_WIDTH240 #define DIALOG_HEIGHT 80 @@ -220,7 +219,7 @@ OUString InformationDialog::ImpGetStandardImage( const OUString sPrivateURL ) Reference XOutputStream xOutputStream( mxTempFile-getOutputStream() ); if ( xOutputStream.is() xPropSet.is() ) { -Reference graphic::XGraphicProvider xGraphicProvider( graphic::GraphicProvider::create( ::comphelper::ComponentContext(mxMSF).getUNOContext() ) ); +Reference graphic::XGraphicProvider xGraphicProvider( graphic::GraphicProvider::create( mxMSF ) ); Sequence PropertyValue aArgs( 1 ); aArgs[ 0 ].Name = OUString(URL); aArgs[ 0 ].Value = sPrivateURL; ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: [Libreoffice-commits] .: 2 commits - chart2/source connectivity/source cui/source desktop/source editeng/source embeddedobj/source extensions/source forms/source framework/source oox/source sc/Cpp
On 06/05/2012 08:47 AM, Stephan Bergmann wrote: argh, sure, an extension must not link against comphelper will take care Fixed now as http://cgit.freedesktop.org/libreoffice/core/commit/?id=719618f28e003c545c20a8fc94d348b667c8376a Remove bogus dependency of extension on comphelper. Despite their names, those MSF variables were already references to XComponentContext, so comphelper::ComponentContext(X).getUNOContext() was a complicated way to just write X. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: regcomp under windows, registering components
On 06/04/2012 08:14 PM, Olivier Hallot wrote: since LibreOffice 3.?, some of these libs moved elsewhere and dll's have new names, specifically: what is now deploymentguimi.uno.dll ? (I have deployment.dll, deploymentgui.dll and deploymentmisclo.dll, so which one?) should be deploymentgui.dll now (for one, there was a move to replace per-platform suffixes like mi or li with a common lo, and for another, during gbuild'ification some lib names got changed slightly, too). \Basis\program\services.rdb seems to have moved to \program\services\services.rdb, my best guess so far... Right? Yes (for one, the three-layer layout got unwound again, so Basis is gone now, and for another, service rdbs were moved into a sub-directory of their own, program\services). Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: 2 commits - autogen.sh config_host.mk.in
autogen.sh|6 +++--- config_host.mk.in |1 - 2 files changed, 3 insertions(+), 4 deletions(-) New commits: commit 1dfb956306a1ec63e1d2f683c3414fffeb07b2e8 Author: Stephan Bergmann sberg...@redhat.com Date: Tue Jun 5 11:57:31 2012 +0200 Indentation clean up Change-Id: I1e3633f5bbf6b3ad0cdb2005a2a5608f3265ffdb diff --git a/autogen.sh b/autogen.sh index 67d462c..663a4ed 100755 --- a/autogen.sh +++ b/autogen.sh @@ -111,10 +111,10 @@ for my $arg (@cmdline_args) { my $config = distro-configs/$1.conf; if (! -f $config) { invalid_distro ($config, $1); -} else { -push @args, read_args ($config); -} } else { +push @args, read_args ($config); +} +} else { push @args, $arg; } } commit 484b8cdf001d1305cf32ec96db0fe36ee0c3a25e Author: Stephan Bergmann sberg...@redhat.com Date: Tue Jun 5 11:45:53 2012 +0200 LO_PATH unused outside configure.in and config_host.mk.in Change-Id: Ib349d9bf93ad59c4c171b87f10c4fe608cecb5fd diff --git a/config_host.mk.in b/config_host.mk.in index 76e398c..116c4f9 100644 --- a/config_host.mk.in +++ b/config_host.mk.in @@ -285,7 +285,6 @@ export LIBXML_LIBS=@LIBXML_LIBS@ export LIBXSLT_CFLAGS=@LIBXSLT_CFLAGS@ export LIBXSLT_LIBS=@LIBXSLT_LIBS@ export LINK_X64_BINARY=@LINK_X64_BINARY@ -export LO_PATH=@LO_PATH@ @x_Cygwin@ export LS=@WIN_LS@ export MANDIR=@MANDIR@ export MACOSX_DEPLOYMENT_TARGET=@MACOSX_DEPLOYMENT_TARGET@ ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: configure.in
configure.in |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) New commits: commit 9616b9440106a40058b7c1f1134417fdbfc0cdf6 Author: Stephan Bergmann sberg...@redhat.com Date: Tue Jun 5 14:53:46 2012 +0200 Bump AC_PACKAGE_VERSION, too Change-Id: I8ec106cc0d8bb09e5a1947aacaf12ad6a0ddd8ef diff --git a/configure.in b/configure.in index ec4461d..bb3eed0 100644 --- a/configure.in +++ b/configure.in @@ -2,7 +2,7 @@ dnl -*- Mode: Autoconf; tab-width: 4; indent-tabs-mode: nil -*- dnl configure.in serves as input for the GNU autoconf package dnl in order to create a configure script. -AC_INIT([LibreOffice],[3.6],[],[],[http://documentfoundation.org/]) +AC_INIT([LibreOffice],[3.7],[],[],[http://documentfoundation.org/]) AC_PREREQ([2.59]) save_CC=$CC ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: [PATCH] [PYTHON] DialogBox class for Python
On 06/05/2012 07:37 PM, Joost Eekhoorn wrote: Where is the place for such a demo ? The LO SDK includes a number of examples, see odk/examples/python/. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [PUSHED] Re: fdo#46808, Adapt UNO services to new style, Part 7, updating ::create
On 06/06/2012 11:28 AM, Noel Grandin wrote: In this patch, shouldn't it be uno::Reference frame::XDispatchProvider xDispatchProvider(rFrame, uno::UNO_QUERY_THROW ); because you are not confirming the result with an is() call. diff --git a/svtools/source/uno/contextmenuhelper.cxx b/svtools/source/uno/contextmenuhelper.cxx index c105204..541b3c9 100644 --- a/svtools/source/uno/contextmenuhelper.cxx +++ b/svtools/source/uno/contextmenuhelper.cxx @@ -346,27 +346,24 @@ ContextMenuHelper::dispatchCommand( } util::URL aTargetURL; +aTargetURL.Complete = aCommandURL; +m_xURLTransformer-parseStrict( aTargetURL ); + uno::Reference frame::XDispatch xDispatch; -if ( m_xURLTransformer.is() ) +uno::Reference frame::XDispatchProvider xDispatchProvider( +rFrame, uno::UNO_QUERY ); +if ( xDispatchProvider.is() ) ^^^ the is() call is still there (this patch just removed the outer if ( m_xURLTransformer.is() ) and looks bigger than it actually is due to the consequential indentation changes) Stephan { -aTargetURL.Complete = aCommandURL; -m_xURLTransformer-parseStrict( aTargetURL ); - -uno::Reference frame::XDispatchProvider xDispatchProvider( -rFrame, uno::UNO_QUERY ); -if ( xDispatchProvider.is() ) ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: ucb/source
ucb/source/core/ucb.cxx |4 +- ucb/source/inc/regexpmap.hxx |2 + ucb/source/inc/regexpmap.tpt | 59 +-- 3 files changed, 55 insertions(+), 10 deletions(-) New commits: commit 9631f96acae6bd179d40ee343ce986296f490741 Author: Stephan Bergmann sberg...@redhat.com Date: Wed Jun 6 12:41:41 2012 +0200 Do not copy etc. singular iterator values. This also reverts previous attempts at fixing this, commits 33839f90e6f8275a584b483827585a6da05aab75 ucb: try to fix weird STL assertion on tinderbox and 6506af86b5e20a860c8d2d1dd578a22914df5f53 ucb: second try to fix weird STL assertion on tinderbox. Change-Id: I89d0eb87fbd164c0a4cf24d60f225767cb2bfc1a diff --git a/ucb/source/core/ucb.cxx b/ucb/source/core/ucb.cxx index fec9972..64a412b 100644 --- a/ucb/source/core/ucb.cxx +++ b/ucb/source/core/ucb.cxx @@ -399,7 +399,7 @@ UniversalContentBroker::registerContentProvider( { osl::MutexGuard aGuard(m_aMutex); -ProviderMap_Impl::iterator aIt(m_aProviders.end()); +ProviderMap_Impl::iterator aIt; try { aIt = m_aProviders.find(Scheme); @@ -445,7 +445,7 @@ void SAL_CALL UniversalContentBroker::deregisterContentProvider( { osl::MutexGuard aGuard(m_aMutex); -ProviderMap_Impl::iterator aMapIt(m_aProviders.end()); +ProviderMap_Impl::iterator aMapIt; try { aMapIt = m_aProviders.find(Scheme); diff --git a/ucb/source/inc/regexpmap.hxx b/ucb/source/inc/regexpmap.hxx index e023633..717a077 100644 --- a/ucb/source/inc/regexpmap.hxx +++ b/ucb/source/inc/regexpmap.hxx @@ -101,6 +101,8 @@ class RegexpMapIter: public RegexpMapConstIter Val friend class RegexpMap Val ; // to access ctor public: +RegexpMapIter() {} + RegexpMapIter operator ++(); RegexpMapIter operator ++(int); diff --git a/ucb/source/inc/regexpmap.tpt b/ucb/source/inc/regexpmap.tpt index 64abc37..b4e274a 100644 --- a/ucb/source/inc/regexpmap.tpt +++ b/ucb/source/inc/regexpmap.tpt @@ -74,11 +74,19 @@ public: typedef RegexpMapImpl Val MapImpl; typedef typename List Val ::iterator ListIterator; + // Solaris needs these for the ctor... + + inline RegexpMapIterImpl(); + inline RegexpMapIterImpl(MapImpl * pTheMap, int nTheList, ListIterator aTheIndex); RegexpMapIterImpl(RegexpMapImpl Val * pTheMap, bool bBegin); + RegexpMapIterImpl(RegexpMapIterImpl const rOther); + + RegexpMapIterImpl operator =(RegexpMapIterImpl const rOther); + bool operator ==(RegexpMapIterImpl const rOther) const; RegexpMapImpl Val const * getMap() const { return m_pMap; } @@ -92,8 +100,6 @@ public: RegexpMapEntry Val get(); private: -RegexpMapIterImpl(); // not implemented - mutable RegexpMapEntry Val m_aEntry; typename List Val ::iterator m_aIndex; RegexpMapImpl Val * m_pMap; @@ -106,6 +112,14 @@ private: } template typename Val +inline RegexpMapIterImpl Val ::RegexpMapIterImpl(): + m_aEntry(rtl::OUString(), 0), + m_pMap(0), + m_nList(-1), + m_bEntrySet(false) +{} + +template typename Val inline RegexpMapIterImpl Val ::RegexpMapIterImpl(MapImpl * pTheMap, int nTheList, ListIterator aTheIndex): @@ -134,11 +148,10 @@ void RegexpMapIterImpl Val ::setEntry() const // template typename Val RegexpMapIterImpl Val ::RegexpMapIterImpl(RegexpMapImpl Val * pTheMap, - bool bBegin) -: m_aEntry(rtl::OUString(), 0) -, m_aIndex(pTheMap-m_aList[Regexp::KIND_DOMAIN].end()) -, m_pMap(pTheMap) -, m_bEntrySet(false) + bool bBegin): + m_aEntry(rtl::OUString(), 0), + m_pMap(pTheMap), + m_bEntrySet(false) { if (bBegin) { @@ -149,17 +162,47 @@ RegexpMapIterImpl Val ::RegexpMapIterImpl(RegexpMapImpl Val * pTheMap, else { m_nList = Regexp::KIND_DOMAIN; + m_aIndex = m_pMap-m_aList[Regexp::KIND_DOMAIN].end(); } } // template typename Val +RegexpMapIterImpl Val ::RegexpMapIterImpl(RegexpMapIterImpl const rOther): +m_aEntry(rOther.m_aEntry), m_pMap(rOther.m_pMap), m_nList(rOther.m_nList), +m_bEntrySet(rOther.m_bEntrySet) +{ +if (m_nList != -1) +m_aIndex = rOther.m_aIndex; +} + +// +template typename Val
[Libreoffice-commits] .: 4 commits - dbaccess/source extensions/source moz/makefile.mk moz/patches reportdesign/source
dbaccess/source/core/api/RowSetBase.cxx |2 extensions/source/plugin/aqua/sysplug.mm | 22 +- moz/makefile.mk |1 moz/patches/seamonkey-1.1.14.source-macosx10.6and7.patch | 33 +++ reportdesign/source/core/sdr/UndoEnv.cxx | 18 5 files changed, 45 insertions(+), 31 deletions(-) New commits: commit 075271062ac719a2790713b7f301a4a4ac7505c0 Author: Stephan Bergmann sberg...@redhat.com Date: Wed Jun 6 09:31:30 2012 +0200 -Werror,-Wself-assign (Clang) Change-Id: Ic541f6539a34ea20b19613edc4fe8670b1a69dee diff --git a/reportdesign/source/core/sdr/UndoEnv.cxx b/reportdesign/source/core/sdr/UndoEnv.cxx index 487bbaf..46b8986 100644 --- a/reportdesign/source/core/sdr/UndoEnv.cxx +++ b/reportdesign/source/core/sdr/UndoEnv.cxx @@ -176,20 +176,6 @@ void OXUndoEnvironment::Clear(const Accessor /*_r*/) { OUndoEnvLock aLock(*this); -#if OSL_DEBUG_LEVEL 0 -// TODO: LLA-OJ please describe what you are doing in this code fragment. -PropertySetInfoCache::iterator aIter = m_pImpl-m_aPropertySetCache.begin(); -PropertySetInfoCache::iterator aEnd = m_pImpl-m_aPropertySetCache.end(); -int ndbg_len = m_pImpl-m_aPropertySetCache.size(); -ndbg_len = ndbg_len; -for (int idbg_ = 0; aIter != aEnd; ++aIter,++idbg_) -{ -uno::Referencebeans::XPropertySet xProp(aIter-first,uno::UNO_QUERY); -xProp-getPropertySetInfo(); -int nlen = aIter-second.aProperties.size(); -nlen = nlen; -} -#endif m_pImpl-m_aPropertySetCache.clear(); sal_uInt16 nCount = m_pImpl-m_rModel.GetPageCount(); @@ -267,10 +253,6 @@ void SAL_CALL OXUndoEnvironment::propertyChange( const PropertyChangeEvent _rEv // no Undo for transient and readonly props. // let's see if we know something about the set -#if OSL_DEBUG_LEVEL 0 -int nlen = m_pImpl-m_aPropertySetCache.size(); -nlen = nlen; -#endif PropertySetInfoCache::iterator objectPos = m_pImpl-m_aPropertySetCache.find(xSet); if (objectPos == m_pImpl-m_aPropertySetCache.end()) { commit a519cacd61624784eff0646c8a472349373b2993 Author: Stephan Bergmann sberg...@redhat.com Date: Tue Jun 5 23:32:45 2012 +0200 -Werror,-Wself-assign (Clang) Change-Id: I92f1271ad36e4ae1221182a3a446f36cf770e003 diff --git a/dbaccess/source/core/api/RowSetBase.cxx b/dbaccess/source/core/api/RowSetBase.cxx index 728934b..758a715 100644 --- a/dbaccess/source/core/api/RowSetBase.cxx +++ b/dbaccess/source/core/api/RowSetBase.cxx @@ -268,8 +268,6 @@ const ORowSetValue ORowSetBase::impl_getValue(sal_Int32 columnIndex) aCacheEnd = m_pCache-getEnd(); aCurrentRow = m_aCurrentRow; ORowSetCacheMap::iterator aCacheIter = m_aCurrentRow.getIter(); -sal_Int32 n = aCacheIter-first; -n = n; ORowSetCacheIterator_Helper aHelper = aCacheIter-second; ORowSetMatrix::iterator k = aHelper.aIterator; for (; k != m_pCache-getEnd(); ++k) commit 39ec7832d165d81385b340bc6c0ca9cf5f194eb5 Author: Stephan Bergmann sberg...@redhat.com Date: Tue Jun 5 23:22:48 2012 +0200 Adapt to --with-macosx-version-min-required=10.6 and Xcode 4 ...signature of TransitionWindowWithOptions apparently changed slightly in 10.6, and /Developer/Tools/Rez can more reliably be found in /usr/bin. Change-Id: Iba55478f3b6de1b4695b686095f4aa0bc0af8e7b diff --git a/moz/makefile.mk b/moz/makefile.mk index 13db91f..76296ad 100644 --- a/moz/makefile.mk +++ b/moz/makefile.mk @@ -88,6 +88,7 @@ PATCH_FILES = \ patches/nss-linux3.patch \ patches/clang_add_nsCaseInsensitiveStringComparator_default_constructor.patch \ patches/clang_missing_this_pointers.patch \ +patches/seamonkey-1.1.14.source-macosx10.6and7.patch # This file is needed for the W32 build when BUILD_MOZAB is set # (currently only vc8/vs2005 is supported when BUILD_MOZAB is set) diff --git a/moz/patches/seamonkey-1.1.14.source-macosx10.6and7.patch b/moz/patches/seamonkey-1.1.14.source-macosx10.6and7.patch new file mode 100644 index 000..e74c4e4 --- /dev/null +++ b/moz/patches/seamonkey-1.1.14.source-macosx10.6and7.patch @@ -0,0 +1,33 @@ +--- misc/mozilla/widget/src/mac/Makefile.in2006-08-04 20:52:25.0 +0200 misc/build/mozilla/widget/src/mac/Makefile.in 2012-06-04 23:26:18.0 +0200 +@@ -130,7 +130,7 @@ + RES_SRC = $(srcdir)/nsMacWidget.r + + $(RES_DEST): $(RES_SRC) +- /Developer/Tools/Rez -i /Developer/Headers/FlatCarbon -useDF $(RES_SRC) -o $(RES_DEST) ++ Rez -i $(MACOS_SDK_DIR)/Developer/Headers/FlatCarbon -useDF $(RES_SRC) -o $(RES_DEST) + + libs:: $(RES_DEST) + +--- misc/mozilla/widget/src/mac/nsMacWindow.cpp2007-11-19 21:40:08.0 +0100 misc/build/mozilla/widget/src/mac/nsMacWindow.cpp 2012-06-04 22:26:29.0 +0200 +@@ -1209,7 +1209,7
[Libreoffice-commits] .: binfilter/bf_basic binfilter/bf_forms binfilter/bf_sc binfilter/bf_sd binfilter/bf_starmath binfilter/bf_svtools binfilter/bf_svx binfilter/bf_sw binfilter/inc
binfilter/bf_basic/source/basmgr/basmgr.cxx|2 binfilter/bf_basic/source/classes/sbxmod.cxx | 18 --- binfilter/bf_forms/source/component/Image.hxx |1 binfilter/bf_sc/source/filter/xml/sc_XMLChangeTrackingImportHelper.cxx |2 binfilter/bf_sc/source/ui/docshell/sc_docfunc.cxx |4 binfilter/bf_sd/source/core/sd_drawdoc2.cxx|4 binfilter/bf_starmath/source/starmath_node.cxx |1 binfilter/bf_starmath/source/starmath_smdll.cxx|7 - binfilter/bf_svtools/source/config/svt_moduleoptions.cxx |4 binfilter/bf_svtools/source/filter.vcl/ixbm/svt_xbmread.cxx| 60 +- binfilter/bf_svx/source/editeng/svx_impedit3.cxx |2 binfilter/bf_svx/source/engine3d/svx_obj3d.cxx |1 binfilter/bf_sw/source/core/doc/sw_docfmt.cxx |2 binfilter/bf_sw/source/core/doc/sw_docftn.cxx |7 - binfilter/bf_sw/source/core/doc/sw_doctxm.cxx |2 binfilter/bf_sw/source/core/layout/sw_tabfrm.cxx |1 binfilter/inc/bf_basic/sbxfac.hxx |5 17 files changed, 42 insertions(+), 81 deletions(-) New commits: commit 5fb512c849e4bbe5918465113b095c793c0252f2 Author: Stephan Bergmann sberg...@redhat.com Date: Wed Jun 6 10:37:57 2012 +0200 Fixes for various Clang (on Mac OS X) warnings Change-Id: I06df07f6d4133658fad90d5622762a8d1cdf04af diff --git a/binfilter/bf_basic/source/basmgr/basmgr.cxx b/binfilter/bf_basic/source/basmgr/basmgr.cxx index efda967..5e3c517 100644 --- a/binfilter/bf_basic/source/basmgr/basmgr.cxx +++ b/binfilter/bf_basic/source/basmgr/basmgr.cxx @@ -2075,7 +2075,7 @@ Any LibraryContainer_Impl::getByName( const OUString aName ) Reference XNameContainer xModuleContainer = (XNameContainer*)new ModuleContainer_Impl( pLib ); -Reference XNameContainer xDialogContainer; +Reference XNameContainer xDialogContainer = (XNameContainer*)new DialogContainer_Impl( pLib ); BasicLibInfo* pLibInfo = mpMgr-FindLibInfo( pLib ); diff --git a/binfilter/bf_basic/source/classes/sbxmod.cxx b/binfilter/bf_basic/source/classes/sbxmod.cxx index f5b6be5..1ffa8e6 100644 --- a/binfilter/bf_basic/source/classes/sbxmod.cxx +++ b/binfilter/bf_basic/source/classes/sbxmod.cxx @@ -356,24 +356,6 @@ SbMethod* SbModule::GetFunctionForLine( USHORT nLine ) return NULL; } -// Ausstrahlen eines Hints an alle Basics - -static void _SendHint( SbxObject* pObj, ULONG nId, SbMethod* p ) -{ -// Selbst ein BASIC? -if( pObj-IsA( TYPE(StarBASIC) ) pObj-IsBroadcaster() ) -pObj-GetBroadcaster().Broadcast( SbxHint( nId, p ) ); -// Dann die Unterobjekte fragen -SbxArray* pObjs = pObj-GetObjects(); -for( USHORT i = 0; i pObjs-Count(); i++ ) -{ -SbxVariable* pVar = pObjs-Get( i ); -if( pVar-IsA( TYPE(SbxObject) ) ) -_SendHint( PTR_CAST(SbxObject,pVar), nId, p ); -} -} - - // #57841 Uno-Objekte, die in RTL-Funktionen gehalten werden, // beim Programm-Ende freigeben, damit nichts gehalten wird. void ClearUnoObjectsInRTL_Impl_Rek( StarBASIC* pBasic ) diff --git a/binfilter/bf_forms/source/component/Image.hxx b/binfilter/bf_forms/source/component/Image.hxx index 8bdda6f..09d86fa 100644 --- a/binfilter/bf_forms/source/component/Image.hxx +++ b/binfilter/bf_forms/source/component/Image.hxx @@ -170,6 +170,7 @@ public: // OComponentHelper virtual void SAL_CALL disposing(); +using OImageControl_BASE::disposing; // ::com::sun::star::form::XApproveActionBroadcaster virtual void SAL_CALL addApproveActionListener(const ::com::sun::star::uno::Reference ::com::sun::star::form::XApproveActionListener _rxListener) diff --git a/binfilter/bf_sc/source/filter/xml/sc_XMLChangeTrackingImportHelper.cxx b/binfilter/bf_sc/source/filter/xml/sc_XMLChangeTrackingImportHelper.cxx index 74085bb..ddd4695 100644 --- a/binfilter/bf_sc/source/filter/xml/sc_XMLChangeTrackingImportHelper.cxx +++ b/binfilter/bf_sc/source/filter/xml/sc_XMLChangeTrackingImportHelper.cxx @@ -516,7 +516,7 @@ void ScXMLChangeTrackingImportHelper::CreateGeneratedActions(ScMyGeneratedList ScMyGeneratedList::iterator aItr = rList.begin(); while (aItr != rList.end()) { -if (((*aItr)-nID == 0)) +if ((*aItr)-nID == 0) { ScBaseCell* pCell = NULL; if ((*aItr)-pCellInfo) diff --git a/binfilter/bf_sc/source/ui/docshell/sc_docfunc.cxx b/binfilter/bf_sc/source/ui/docshell/sc_docfunc.cxx index 876ce46..cdba6bf 100644 --- a/binfilter/bf_sc/source/ui/docshell/sc_docfunc.cxx +++ b/binfilter/bf_sc/source/ui/docshell/sc_docfunc.cxx @@ -976,9 +976,7
Re: Draft document-rendering tiling API to be called from viewer apps
On 06/06/2012 01:43 PM, Tor Lillqvist wrote: Now I am having second thoughts, though, but still, a first, in progress, draft API is in UNOIDL form is in touch/idl. I see no need for a DocumentRenderCallback service. XDocumentRenderCallback will be implemented by UNO objects specific to the various callers of XDocument.render; there is no reasonable generic implementation that could be offered as a service. (Also, there is a subtle difference between an explicit zero-parameter constructor and an implicit default constructor for UNO services, relating to whether the constructor internally calls createInstanceWithContext or createInstanceWithArgumentsAndContext.) But, that's the *specification* and *use*. *Implementing* UNO services in C++, especially if you don't want to just copy-paste existing code as a start, and then modify, without understanding what the parts you don't touch is exactly doing, is still nightmarish with loads of boilerplate-ish rubbish. I would go the UNO approach nevertheless. If you want to be able to call this from Java, don't underestimate the work you save by reusing UNO here. Your impression might vary, but I would say this vastly compensates for the mild amount of boilerplate you need on the C++ implementation side. (I might of course be routine-blinded, but in my eyes the necessary boilerplate boils down to not that much really.) Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Draft document-rendering tiling API to be called from viewer apps
On 06/06/2012 03:11 PM, Tor Lillqvist wrote: I would go the UNO approach nevertheless. It's a hard decision to make... With the current state of touch/idl, what you would want to do in C++ is an implementation of the Document service, plus potentially an implementation of an XDocumentRenderCallback object (if you want to call the service's XDocument.render from C++ code). The boilerplate for the former boils down to a derivation of cppu::WeakImplHelper1XDocument, plus a component_getFactory function (just calling cppu::component_getFactoryHelper) if this is the first service implementation in a given library. The boilerplate for the latter boils down to a derivation of cppu::WeakImplHelper1XDocumentRenderCallback. Let me know if you need skeleton code (or any other form of help) for that. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: boost/boost_1_44_0-clang-warnings.patch boost/makefile.mk
boost/boost_1_44_0-clang-warnings.patch | 61 boost/makefile.mk |3 + 2 files changed, 64 insertions(+) New commits: commit 48d6c11c6fc6276c45bed56318fc4ac0dc033fdf Author: Stephan Bergmann sberg...@redhat.com Date: Thu Jun 7 08:03:22 2012 +0200 Some boost warnings with Clang ...on Mac OS X, in what looks like only newly used boost headers. Change-Id: I86612a419311fe5a2dfda9eec520889707f9e5db diff --git a/boost/boost_1_44_0-clang-warnings.patch b/boost/boost_1_44_0-clang-warnings.patch new file mode 100644 index 000..c0c2206 --- /dev/null +++ b/boost/boost_1_44_0-clang-warnings.patch @@ -0,0 +1,61 @@ +--- misc/boost_1_44_0/boost/algorithm/string/detail/find_format_store.hpp 2010-07-10 22:29:03.0 +0200 misc/build/boost_1_44_0/boost/algorithm/string/detail/find_format_store.hpp 2012-06-06 23:13:26.0 +0200 +@@ -71,7 +71,7 @@ + }; + + templatetypename InputT, typename FindResultT +-bool check_find_result(InputT Input, FindResultT FindResult) ++bool check_find_result(InputT, FindResultT FindResult) + { + typedef BOOST_STRING_TYPENAME + range_const_iteratorInputT::type input_iterator_type; +--- misc/boost_1_44_0/boost/date_time/dst_rules.hpp2008-02-27 21:00:24.0 +0100 misc/build/boost_1_44_0/boost/date_time/dst_rules.hpp 2012-06-06 23:11:03.0 +0200 +@@ -371,7 +371,7 @@ + return is_not_in_dst; + } + +- static bool is_dst_boundary_day(date_type d) ++ static bool is_dst_boundary_day(date_type) + { + return false; + } +--- misc/boost_1_44_0/boost/date_time/gregorian_calendar.ipp 2010-01-10 20:17:23.0 +0100 misc/build/boost_1_44_0/boost/date_time/gregorian_calendar.ipp 2012-06-06 22:27:07.0 +0200 +@@ -47,7 +47,7 @@ + return week; + } + +-if ((week == 53)) { ++if (week == 53) { + if((day==6) ||(day == 5 is_leap_year(ymd.year))) { + return week; //under these circumstances week == 53. + } else { +--- misc/boost_1_44_0/boost/date_time/period_formatter.hpp 2008-02-27 21:00:24.0 +0100 misc/build/boost_1_44_0/boost/date_time/period_formatter.hpp 2012-06-06 23:12:48.0 +0200 +@@ -114,10 +114,10 @@ + { + m_range_option = option; + } +-void delimiter_strings(const string_type separator, +- const string_type start_delim, +- const string_type open_end_delim, +- const string_type closed_end_delim) ++void delimiter_strings(const string_type, ++ const string_type, ++ const string_type, ++ const string_type) + { + m_period_separator; + m_period_start_delimeter; +--- misc/boost_1_44_0/boost/token_functions.hpp2010-06-12 14:06:28.0 +0200 misc/build/boost_1_44_0/boost/token_functions.hpp 2012-06-06 23:12:27.0 +0200 +@@ -285,7 +285,7 @@ + template + struct assign_or_plus_equalstd::input_iterator_tag { + templateclass Iterator, class Token +-static void assign(Iterator b, Iterator e, Token t) { } ++static void assign(Iterator, Iterator, Token ) { } + templateclass Token, class Value + static void plus_equal(Token t, const Value v) { + t += v; diff --git a/boost/makefile.mk b/boost/makefile.mk index 6c9db56..5b8957f 100644 --- a/boost/makefile.mk +++ b/boost/makefile.mk @@ -75,6 +75,9 @@ PATCH_FILES += boost_1_44_0-logical-op-parentheses.patch # Backporting fixes for the GCC 4.7 -std=c++11 mode from Boost 1.48.0: PATCH_FILES += boost_1_44_0-gcc4.7.patch +# Clang warnings: +PATCH_FILES += boost_1_44_0-clang-warnings.patch + ADDITIONAL_FILES= \ libs/thread/src/win32/makefile.mk \ libs/date_time/src/gregorian/makefile.mk ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Building on Mac OS X 10.7
FYI, attached BUILD-lo script allowed me to build recent LO master on Mac OS X 10.7 with Xcode 4. A few comments: * It builds into ~/Software/lo (which must not yet exist). * It uses MacPorts for the various dependencies, but in a way that does not pollute the machine (by checking out and building a MacPorts installation completely local to the LO build). * The pkgconfig and libidl stuff, and maybe also the grep stuff, is probably only needed for moz (at least, that's the first module where I encounter need for it). * Instead of symlinking doxygen into ~/Software/lo/bin, it could also be included in autogen.lastrun via --with-doxygen=... if it were clarified how to escape content in autogen.lastrun (newlines in HOME, anyone?). (This would of course be avoided if the script would pass the configuration arguments to an explicit call of autogen.sh.) * --with-macosx-version-min-required=10.6 (instead of 10.7) is needed at least for the moz and python modules. * As per SUS an unset or null PATH means implementation-defined behaviour (that cannot easily be added to with PATH=${PATH}:...), the script simply fails on an unset or null PATH. Stephan #!/bin/bash set -ex -o pipefail # On latest Mac OS X 10.7.4 with latest Xcode 4.3.2 (incl. Command Line Tools): mkdir ${HOME?}/Software/lo svn checkout http://svn.macports.org/repository/macports/trunk \ ${HOME?}/Software/lo/mports-src # https instead of http fails to work non-interactively due to: # Error validating server certificate for 'https://svn.macports.org:443': #- The certificate is not issued by a trusted authority. Use the # fingerprint to validate the certificate manually! # Certificate information: #- Hostname: *.macports.org #- Valid: from Tue, 22 Feb 2011 17:29:43 GMT until Tue, 18 Mar 2014 # 23:36:56 GMT #- Issuer: 07969287, http://certificates.godaddy.com/repository, # GoDaddy.com, Inc., Scottsdale, Arizona, US #- Fingerprint: # 4d:ea:4a:77:55:ac:8e:2e:9e:11:8a:59:3d:ec:c7:45:7d:b0:72:19 # (R)eject, accept (t)emporarily or accept (p)ermanently? (cd ${HOME?}/Software/lo/mports-src/base ./configure \ --prefix=${HOME?}/Software/lo/mports --with-no-root-privileges \ --with-applications-dir=${HOME?}/Software/lo/mports-apps \ --with-tclpackage=${HOME?}/Software/lo/mports-tcl make make install) \ || exit 1 ${HOME?}/Software/lo/mports/bin/port selfupdate ${HOME?}/Software/lo/mports/bin/port install libidl +universal ${HOME?}/Software/lo/mports/bin/port install automake doxygen grep pkgconfig mkdir ${HOME?}/Software/lo/bin ln -s ${HOME?}/Software/lo/mports/bin/aclocal ${HOME?}/Software/lo/bin/ ln -s ${HOME?}/Software/lo/mports/bin/autoconf ${HOME?}/Software/lo/bin/ ln -s ${HOME?}/Software/lo/mports/bin/autom4te ${HOME?}/Software/lo/bin/ ln -s ${HOME?}/Software/lo/mports/bin/doxygen ${HOME?}/Software/lo/bin/ mkdir ${HOME?}/Software/lo/pkgconfig ln -s ${HOME?}/Software/lo/mports/lib/pkgconfig/glib-2.0.pc \ ${HOME?}/Software/lo/pkgconfig/ ln -s ${HOME?}/Software/lo/mports/lib/pkgconfig/libIDL-2.0.pc \ ${HOME?}/Software/lo/pkgconfig/ git clone ssh://git.freedesktop.org/git/libreoffice/core \ ${HOME?}/Software/lo/core # moz/unxmacxi/misc/build/mozilla/configure:20225 fails with egrep: Regular # expression too big with default /usr/bin/egrep (egrep (GNU grep) 2.5.1), # and adding this replacement to ${HOME?}/Software/lo/bin (which is added to # PATH) does not help, as config_host.mk contains a rewritten PATH (where # ${HOME?}/Software/lo/bin would always go after /usr/bin): ln -s ${HOME?}/Software/lo/mports/bin/egrep \ ${HOME?}/Software/lo/core/solenv/bin/ # moz/unxmacxi/misc/build/mozilla/gfx/src/mac/nsRegionPool.cpp uses NewRgn, # DisposeRgn, etc. that got removed in 10.7 SDK: cat ${HOME?}/Software/lo/core/autogen.lastrun \EOF --disable-ccache --disable-linkoo --disable-zenity --enable-binfilter --enable-dbgutil --enable-epm --enable-python=internal --enable-werror --with-lang=de en-US --with-macosx-version-min-required=10.6 --with-max-jobs=2 --with-num-cpus=2 CC=clang -m32 CXX=clang++ -m32 EOF ulimit -c unlimited cd ${HOME?}/Software/lo/core printf '%s' ${HOME?} | grep -Fqv : || exit 1 PATH=${PATH:?}:${HOME?}/Software/lo/bin \ PKG_CONFIG=${HOME?}/Software/lo/mports/bin/pkg-config \ PKG_CONFIG_LIBDIR=${HOME?}/Software/lo/pkgconfig make check ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: binfilter/bf_forms
binfilter/bf_forms/source/component/Image.hxx |2 -- 1 file changed, 2 deletions(-) New commits: commit 4630cdd4990593733872a264e8abf031fa144e1f Author: Stephan Bergmann sberg...@redhat.com Date: Thu Jun 7 09:24:45 2012 +0200 Disentangle disposing using Change-Id: I4ad806521063c9e6699488cdb52cec349fe9dfe6 diff --git a/binfilter/bf_forms/source/component/Image.hxx b/binfilter/bf_forms/source/component/Image.hxx index 09d86fa..2863013 100644 --- a/binfilter/bf_forms/source/component/Image.hxx +++ b/binfilter/bf_forms/source/component/Image.hxx @@ -185,8 +185,6 @@ public: // ::com::sun::star::beans::XPropertyChangeListener virtual void SAL_CALL propertyChange( const ::com::sun::star::beans::PropertyChangeEvent ) throw(::com::sun::star::uno::RuntimeException); -using OControl::disposing; - protected: ::com::sun::star::uno::Reference ::com::sun::star::frame::XModel getXModel(const InterfaceRef xIface) const; ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: Is it ok to remove visual navigator comments?
On 06/06/2012 05:00 PM, Michael Stahl wrote: it is more useful if it separates a chunk of related stuff from another chunk of related stuff, e.g. if there are multiple classes implemented in one file it can be nice to have a separator between the methods of different classes, but this should usually result in a number of separators in a file that can be counted on one hand. The only thing that never looks right is a rule. There is not in existence a page with a rule on it that cannot be instantly and obviously improved by taking the rule out. — George Bernard Shaw, in The Dolphin (1940) [according to Knuth] :) Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: codemaker/source
codemaker/source/cppumaker/cpputype.cxx |1 + 1 file changed, 1 insertion(+) New commits: commit b67445306634571cd7d2a62942b00c0edb82b215 Author: Stephan Bergmann sberg...@redhat.com Date: Thu Jun 7 14:36:05 2012 +0200 XInterfaceTypeDescription missing from isBootstrapType ...it is base of XInterfaceTypeDescription2 (included in isBootstrapType), which ultimately caused uno-skeletonmaker to crash. Change-Id: I17421f58efd9edd4112532a3221125865cc5560e diff --git a/codemaker/source/cppumaker/cpputype.cxx b/codemaker/source/cppumaker/cpputype.cxx index 7fc720a..628b4c6 100644 --- a/codemaker/source/cppumaker/cpputype.cxx +++ b/codemaker/source/cppumaker/cpputype.cxx @@ -123,6 +123,7 @@ bool isBootstrapType(rtl::OString const name) { com/sun/star/reflection/XInterfaceAttributeTypeDescription2, com/sun/star/reflection/XInterfaceMemberTypeDescription, com/sun/star/reflection/XInterfaceMethodTypeDescription, +com/sun/star/reflection/XInterfaceTypeDescription, com/sun/star/reflection/XInterfaceTypeDescription2, com/sun/star/reflection/XMethodParameter, com/sun/star/reflection/XStructTypeDescription, ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: Branch 'libreoffice-3-6' - codemaker/source
codemaker/source/cppumaker/cpputype.cxx |1 + 1 file changed, 1 insertion(+) New commits: commit 916158a42d925c5aaa629a0c1a01698ed6ea17f5 Author: Stephan Bergmann sberg...@redhat.com Date: Thu Jun 7 14:36:05 2012 +0200 XInterfaceTypeDescription missing from isBootstrapType ...it is base of XInterfaceTypeDescription2 (included in isBootstrapType), which ultimately caused uno-skeletonmaker to crash. Change-Id: I17421f58efd9edd4112532a3221125865cc5560e (cherry picked from commit b67445306634571cd7d2a62942b00c0edb82b215) diff --git a/codemaker/source/cppumaker/cpputype.cxx b/codemaker/source/cppumaker/cpputype.cxx index b4eac78..02370f2 100644 --- a/codemaker/source/cppumaker/cpputype.cxx +++ b/codemaker/source/cppumaker/cpputype.cxx @@ -123,6 +123,7 @@ bool isBootstrapType(rtl::OString const name) { com/sun/star/reflection/XInterfaceAttributeTypeDescription2, com/sun/star/reflection/XInterfaceMemberTypeDescription, com/sun/star/reflection/XInterfaceMethodTypeDescription, +com/sun/star/reflection/XInterfaceTypeDescription, com/sun/star/reflection/XInterfaceTypeDescription2, com/sun/star/reflection/XMethodParameter, com/sun/star/reflection/XStructTypeDescription, ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: Branch 'libreoffice-3-5' - scp2/source
scp2/source/ooo/windowscustomaction_ooo.scp | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) New commits: commit 265242f06dbdd21dade56dc6fc6c1789f33ee966 Author: Andras Timar ati...@suse.com Date: Tue Jun 5 16:57:53 2012 +0200 fdo#43989 let unopkg.exe run with elevated privileges during install ... so it can do the pre-registration of bundled extensions in INSTALLLOCATION/share/prereg/bundle. The meaning of Type of custom action is the following: msidbCustomActionTypeDll + msidbCustomActionTypeBinaryData + msidbCustomActionTypeInScript + msidbCustomActionTypeNoImpersonate + msidbCustomActionTypeCommit = 3585 So it is a binary in a dll from the MSI itself, it runs deferred with system privileges and after commit, because msvcrt90.dll is available only after commit. The same type was chosen for the removing part, it works. Change-Id: I14c347b5cbc47789eb0484769a676703361c2193 (cherry picked from commit 8783ead70cc2bc2a83bf473b0dfb51f3ee10b6da) Signed-off-by: Stephan Bergmann sberg...@redhat.com diff --git a/scp2/source/ooo/windowscustomaction_ooo.scp b/scp2/source/ooo/windowscustomaction_ooo.scp index c111e17..256a924 100644 --- a/scp2/source/ooo/windowscustomaction_ooo.scp +++ b/scp2/source/ooo/windowscustomaction_ooo.scp @@ -214,21 +214,21 @@ End WindowsCustomAction gid_Customaction_Register_Extensions Name = RegisterExtensions; - Typ = 1; + Typ = 3585; Source = shlxtmsi.dll; Target = RegisterExtensions; Inbinarytable = 1; - Assignment1 = (InstallExecuteSequence, Not REMOVE=\ALL\, behind_InstallFinalize); - Assignment2 = (AdminExecuteSequence, Not REMOVE=\ALL\, behind_InstallFinalize); + Assignment1 = (InstallExecuteSequence, Not REMOVE=\ALL\, InstallFinalize); + Assignment2 = (AdminExecuteSequence, Not REMOVE=\ALL\, InstallFinalize); End WindowsCustomAction gid_Customaction_Remove_Extensions Name = RemoveExtensions; - Typ = 65; + Typ = 3585; Source = shlxtmsi.dll; Target = RemoveExtensions; Inbinarytable = 1; - Assignment1 = (InstallExecuteSequence, REMOVE=\ALL\ And Not PATCH, FileCost); + Assignment1 = (InstallExecuteSequence, REMOVE=\ALL\ And Not PATCH, InstallFinalize); End WindowsCustomAction gid_Customaction_Set_Admininstall ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: [PUSHED-3-5] Bug 43989 - After installation first launch terminates after splash screen
On 06/05/2012 05:36 PM, Andras Timar wrote: It was a MAB, please review and cherry-pick to libreoffice-3-5. done ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: officecfg/registry
officecfg/registry/data/org/openoffice/ucb/Configuration.xcu |4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) New commits: commit d4924b747eb6abd439a6313e338384ae50cbdc8e Author: Stephan Bergmann sberg...@redhat.com Date: Thu Jun 7 17:47:11 2012 +0200 Revert accidental changes from fd32788092fcb29526ae73ae3bae6767dbb1e490 Change-Id: I1b3ceda236f26be2dcdc7e4d2afa0ef8d0053691 diff --git a/officecfg/registry/data/org/openoffice/ucb/Configuration.xcu b/officecfg/registry/data/org/openoffice/ucb/Configuration.xcu index 198a660..def1b2c 100644 --- a/officecfg/registry/data/org/openoffice/ucb/Configuration.xcu +++ b/officecfg/registry/data/org/openoffice/ucb/Configuration.xcu @@ -26,7 +26,7 @@ * -- !DOCTYPE oor:component-data SYSTEM ../../../../component-update.dtd -oor:component-data xmlns:oor=http://openoffice.org/2001/registry; xmlns:xs=http://www.w3.org/2001/XMLSchema; xmlns:xsi=http://www.w3.org/2001/XMLSchema-instance; oor:name=Configuration oor:package=org.openoffice.ucb +oor:component-data oor:name=Configuration oor:package=org.openoffice.ucb xmlns:oor=http://openoffice.org/2001/registry; xmlns:xs=http://www.w3.org/2001/XMLSchema; xmlns:xsi=http://www.w3.org/2001/XMLSchema-instance; node oor:name=ContentProviders node oor:name=Local oor:op=replace node oor:name=SecondaryKeys @@ -209,7 +209,7 @@ /prop /node !-- We want the Provider to be the final fallback provider -- -node oor:name=Provider999 oor:op=replace module=gio +node oor:name=Provider999 oor:op=replace install:module=gio prop oor:name=ServiceName valuecom.sun.star.ucb.GIOContentProvider/value /prop ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: Branch 'libreoffice-3-6' - officecfg/registry
officecfg/registry/data/org/openoffice/ucb/Configuration.xcu |4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) New commits: commit 4b462e9ef9c5bb5ddcc7beeaa4c727e239464377 Author: Stephan Bergmann sberg...@redhat.com Date: Thu Jun 7 17:47:11 2012 +0200 Revert accidental changes from fd32788092fcb29526ae73ae3bae6767dbb1e490 Change-Id: I1b3ceda236f26be2dcdc7e4d2afa0ef8d0053691 (cherry picked from commit d4924b747eb6abd439a6313e338384ae50cbdc8e) diff --git a/officecfg/registry/data/org/openoffice/ucb/Configuration.xcu b/officecfg/registry/data/org/openoffice/ucb/Configuration.xcu index 198a660..def1b2c 100644 --- a/officecfg/registry/data/org/openoffice/ucb/Configuration.xcu +++ b/officecfg/registry/data/org/openoffice/ucb/Configuration.xcu @@ -26,7 +26,7 @@ * -- !DOCTYPE oor:component-data SYSTEM ../../../../component-update.dtd -oor:component-data xmlns:oor=http://openoffice.org/2001/registry; xmlns:xs=http://www.w3.org/2001/XMLSchema; xmlns:xsi=http://www.w3.org/2001/XMLSchema-instance; oor:name=Configuration oor:package=org.openoffice.ucb +oor:component-data oor:name=Configuration oor:package=org.openoffice.ucb xmlns:oor=http://openoffice.org/2001/registry; xmlns:xs=http://www.w3.org/2001/XMLSchema; xmlns:xsi=http://www.w3.org/2001/XMLSchema-instance; node oor:name=ContentProviders node oor:name=Local oor:op=replace node oor:name=SecondaryKeys @@ -209,7 +209,7 @@ /prop /node !-- We want the Provider to be the final fallback provider -- -node oor:name=Provider999 oor:op=replace module=gio +node oor:name=Provider999 oor:op=replace install:module=gio prop oor:name=ServiceName valuecom.sun.star.ucb.GIOContentProvider/value /prop ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: officecfg/Configuration_officecfg.mk officecfg/registry postprocess/packregistry
officecfg/Configuration_officecfg.mk |2 + officecfg/registry/data/org/openoffice/ucb/Configuration.xcu | 14 +-- postprocess/packregistry/makefile.mk |6 +++- 3 files changed, 14 insertions(+), 8 deletions(-) New commits: commit 459e482c7a9ba9aff5b5a505cc54247493202b46 Author: Stephan Bergmann sberg...@redhat.com Date: Thu Jun 7 17:54:43 2012 +0200 Make webdav and odma ucp configuration properly conditional Change-Id: I9a8eb6a1e9af051d458a5462c01357540c3d9277 diff --git a/officecfg/Configuration_officecfg.mk b/officecfg/Configuration_officecfg.mk index 0b0a8b0..37372c9 100644 --- a/officecfg/Configuration_officecfg.mk +++ b/officecfg/Configuration_officecfg.mk @@ -206,6 +206,8 @@ $(eval $(call gb_Configuration_add_spool_modules,registry,officecfg/registry/dat org/openoffice/TypeDetection/UISort-impress.xcu \ org/openoffice/TypeDetection/UISort-math.xcu \ org/openoffice/ucb/Configuration-gio.xcu \ + org/openoffice/ucb/Configuration-neon.xcu \ + org/openoffice/ucb/Configuration-win.xcu \ )) # perhaps this file should be moved 2 levels up? diff --git a/officecfg/registry/data/org/openoffice/ucb/Configuration.xcu b/officecfg/registry/data/org/openoffice/ucb/Configuration.xcu index def1b2c..847f786 100644 --- a/officecfg/registry/data/org/openoffice/ucb/Configuration.xcu +++ b/officecfg/registry/data/org/openoffice/ucb/Configuration.xcu @@ -26,7 +26,7 @@ * -- !DOCTYPE oor:component-data SYSTEM ../../../../component-update.dtd -oor:component-data oor:name=Configuration oor:package=org.openoffice.ucb xmlns:oor=http://openoffice.org/2001/registry; xmlns:xs=http://www.w3.org/2001/XMLSchema; xmlns:xsi=http://www.w3.org/2001/XMLSchema-instance; +oor:component-data oor:name=Configuration oor:package=org.openoffice.ucb xmlns:install=http://openoffice.org/2004/installation; xmlns:oor=http://openoffice.org/2001/registry; xmlns:xs=http://www.w3.org/2001/XMLSchema; xmlns:xsi=http://www.w3.org/2001/XMLSchema-instance; node oor:name=ContentProviders node oor:name=Local oor:op=replace node oor:name=SecondaryKeys @@ -54,7 +54,7 @@ value/ /prop /node -node oor:name=Provider3 oor:op=replace +node oor:name=Provider3 oor:op=replace install:module=neon prop oor:name=ServiceName valuecom.sun.star.ucb.WebDAVContentProvider/value /prop @@ -65,7 +65,7 @@ value/ /prop /node -node oor:name=Provider4 oor:op=replace +node oor:name=Provider4 oor:op=replace install:module=neon prop oor:name=ServiceName valuecom.sun.star.ucb.WebDAVContentProvider/value /prop @@ -98,7 +98,7 @@ value/ /prop /node -node oor:name=Provider7 oor:op=replace +node oor:name=Provider7 oor:op=replace install:module=neon prop oor:name=ServiceName valuecom.sun.star.ucb.WebDAVContentProvider/value /prop @@ -153,7 +153,7 @@ value/ /prop /node -node oor:name=Provider12 oor:op=replace +node oor:name=Provider12 oor:op=replace install:module=neon prop oor:name=ServiceName valuecom.sun.star.ucb.WebDAVContentProvider/value /prop @@ -164,7 +164,7 @@ value/ /prop /node -node oor:name=Provider13 oor:op=replace +node oor:name=Provider13 oor:op=replace install:module=neon prop oor:name=ServiceName valuecom.sun.star.ucb.WebDAVContentProvider/value /prop @@ -186,7 +186,7 @@ value/ /prop /node -node oor:name=Provider42 oor:op=replace +node oor:name=Provider42 oor:op=replace install:module=win prop oor:name=ServiceName valuecom.sun.star.ucb.ODMAContentProvider/value /prop diff --git a/postprocess/packregistry/makefile.mk b/postprocess/packregistry/makefile.mk index 2d1d396..1749b5f 100644 --- a/postprocess/packregistry/makefile.mk +++ b/postprocess/packregistry/makefile.mk @@ -316,7 +316,8 @@ MY_FILES_main += \ $(MY_MOD)/org/openoffice/Inet-wnt.xcu \ $(MY_MOD)/org/openoffice/Office/Accelerators-unxwnt.xcu \ $(MY_MOD)/org/openoffice/Office/Common-wnt.xcu \ -$(MY_MOD)/org/openoffice/Office/Paths-unxwnt.xcu +$(MY_MOD)/org/openoffice/Office/Paths-unxwnt.xcu \ +$(MY_MOD)/org/openoffice/ucb/Configuration-win.xcu # Inet-wnt.xcu must come after Inet.xcu MY_DRIVERS += ado .ELIF $(GUIBASE) == cocoatouch @@ -328,6 +329,9 @@ MY_DRIVERS += ado .ELSE
[Libreoffice-commits] .: sc/source
sc/source/ui/inc/tabvwsh.hxx |3 +-- sc/source/ui/view/formatsh.cxx |2 +- sc/source/ui/view/tabvwsh5.cxx |1 - sc/source/ui/view/tabvwsha.cxx |2 +- 4 files changed, 3 insertions(+), 5 deletions(-) New commits: commit b27475db2e8a72d16fca3c812ba803073bb6ef6c Author: Stephan Bergmann sberg...@redhat.com Date: Fri Jun 8 08:43:40 2012 +0200 -Werror=unused-parameter Change-Id: I02d764a6647fa8f0aaeba9088831f3bbe680c8ab diff --git a/sc/source/ui/inc/tabvwsh.hxx b/sc/source/ui/inc/tabvwsh.hxx index 5852065..ecf3ebe 100644 --- a/sc/source/ui/inc/tabvwsh.hxx +++ b/sc/source/ui/inc/tabvwsh.hxx @@ -403,8 +403,7 @@ public: ScViewData*pViewData, SvxNumberInfoItem**ppItem ); -voidUpdateNumberFormatter ( ScDocument* pDoc, - const SvxNumberInfoItem rInfoItem ); +voidUpdateNumberFormatter ( const SvxNumberInfoItem rInfoItem ); voidExecuteCellFormatDlg( SfxRequest rReq, sal_uInt16 nTabPage = 0x ); diff --git a/sc/source/ui/view/formatsh.cxx b/sc/source/ui/view/formatsh.cxx index 98ac665..66266af 100644 --- a/sc/source/ui/view/formatsh.cxx +++ b/sc/source/ui/view/formatsh.cxx @@ -853,7 +853,7 @@ void ScFormatShell::ExecuteStyle( SfxRequest rReq ) if ( SFX_STYLE_FAMILY_PARA == eFam ) { -pTabViewShell-UpdateNumberFormatter( pDoc, +pTabViewShell-UpdateNumberFormatter( (const SvxNumberInfoItem) *(pDocSh-GetItem(SID_ATTR_NUMBERFORMAT_INFO)) ); diff --git a/sc/source/ui/view/tabvwsh5.cxx b/sc/source/ui/view/tabvwsh5.cxx index 7acaa96..5f1d645 100644 --- a/sc/source/ui/view/tabvwsh5.cxx +++ b/sc/source/ui/view/tabvwsh5.cxx @@ -413,7 +413,6 @@ void ScTabViewShell::MakeNumberInfoItem( ScDocument* pDoc, //-- void ScTabViewShell::UpdateNumberFormatter( -ScDocument* pDoc, const SvxNumberInfoItem rInfoItem ) { const sal_uInt32 nDelCount = rInfoItem.GetDelCount(); diff --git a/sc/source/ui/view/tabvwsha.cxx b/sc/source/ui/view/tabvwsha.cxx index d64de03..4efd093 100644 --- a/sc/source/ui/view/tabvwsha.cxx +++ b/sc/source/ui/view/tabvwsha.cxx @@ -527,7 +527,7 @@ void ScTabViewShell::ExecuteCellFormatDlg( SfxRequest rReq, sal_uInt16 nTabPage if(pOutSet-GetItemState(SID_ATTR_NUMBERFORMAT_INFO,sal_True,pItem)==SFX_ITEM_SET) { -UpdateNumberFormatter( pDoc,(const SvxNumberInfoItem)*pItem); +UpdateNumberFormatter((const SvxNumberInfoItem)*pItem); } ApplyAttributes( pOutSet, pOldSet ); ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: [PATCH] [PUSHED] dead code: remove ScDocument::DeleteNumberFormat
On 06/07/2012 07:44 PM, Thomas Arnhold wrote: Thanks! Pushed. Hi Thomas, Can I nudge you to use --enable-werror? Esp. when pruning code, it can help detect further stuff that is becoming unused (in some cases even leading to nice avalanches). Thanks anyway, Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [PATCH] WaE in ScTabViewShell::UpdateNumberFormatter
On 06/08/2012 12:05 AM, David Ostrovsky wrote: this patch solves WaE unused pDoc parameter in ScTabViewShell::UpdateNumberFormatter. I have no idea, why I'm getting it now, git blame says that the last changes in this file were years ago ... Oops, sorry, only discovered this thread now that I have already pushed a fix myself (and nudged Thomas towards --enable-werror in the original patch's thread). If I had seen David's patch earlier, I would of course have pushed it with his authorship (as it happens, our two patches turn out to be exactly the same, anyway). Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [GERRIT] gbuild migration: pyuno module
On 06/08/2012 09:49 AM, Stephan Bergmann wrote: But what source do I need exactly for this? Is it still a feature branch? And https://gerrit.libreoffice.org/#/c/179/ from above only gives me a Not Found. Ah, found it now. The wonderful and frightening world of gerrit... ;) Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Can connectivity/source/inc/odbc/ODefs3.hxx file be dropped ?
On 06/07/2012 08:17 PM, julien2412 wrote: Then I searched with Opengrok about ODefs3, it seems unused. Did I miss something or could I remove it ? Looks like it can be removed, indeed. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [GERRIT] gbuild migration: pyuno module
On 06/08/2012 10:51 AM, Bjoern Michaelsen wrote: On Fri, Jun 08, 2012 at 09:49:06AM +0200, Stephan Bergmann wrote: on wnt: native python executable wrapper is built in pyuno/Executable_python.mk now and is delivered by default to bin/python.exe. But then we have a collision with native python executable artifact which get build in python module. How can i force on gbuild land (RepositoryFixes.mk doesn't handle executables, only libs so far) to create a python.exe not into /bin, but to bin/pyuno/python.exe (Windows isn't tested at all, though) Good question. Hopefully one of the gbuild wizards can step in? Cant that executable be simply called /bin/python_wrapper.exe and then adjusted to be renamed upon installing in scp2? Or do we use that thingie during the build already? I can't remember whether scp2 actually does or does not allow to change the names of files. If it does, that should be a working solution (as we do not call our python wrapper from the build, as far as I know). Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: minutes of ESC call ...
On 06/08/2012 11:09 AM, Michael Meeks wrote: + 3.6 blocker: Java not detected on Windows (fixed thanks to Stephan) Credits actually need to go to David Ostrovsky here, not me. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [GERRIT] gbuild migration: pyuno module
On 06/08/2012 09:49 AM, Stephan Bergmann wrote: On 06/08/2012 09:22 AM, David Ostrovsky wrote: 2. on Linux I'm testing the clean build with --enable-python=internal, and it seems not to work: opening Tools=Macros=Organize Macros=Python and trying to open Hello World python Macor. I do not see any macros available but see this warning in console [david@wizball program (master)]$ ./soffice.bin --writer 'import site' failed; use -v for traceback Any ideas how to proceed? I can have a look. Attached pyuno.patch addresses three problems that I discovered: * The sed call in pyuno/CustomTarget_python_shell.mk used wrong variable names, so that expansions in the python wrapper script were empty. ** The pyuno/zipcore/python.sh also contains NOMACSECTION and MACSECTION blocks, and the original pyuno/zipcore/makefile.mk made sure to only include one of them. This still needs to be fixed. * SAL_DLLPUBLIC_EXPORT was missing from pyuno/source/module/pyuno_dlopenwrapper.c, so that the pyuno.so wrapper did not export initpyuno. * The program/python-core-2.6.1/ tree was missing from the installation set. Getting this back was a bit tricky, as the old system zipped together a temporary tree with some python-core-2.6.1/lib/ structure, that ended up with that hierarchy included in the zip, and scp2 specified to unzip it into the program directory. I changed that to zip together the flat content of $(OUTDIR)/lib/python, and instead explicitly create the python-core-2.6.1/lib hierarchy in scp2 into which to then unzip the zip file. ** Ideally, the lib directory could be removed from the hierarchy completely, but that would require changes to all the places that set up PYTHONPATH etc. ** The original pyuno/zipcore/makefile.mk called strip on the files that went into the zip. From the recent general discussion whether or not to strip when building LO, it indeed seems acceptable to just drop that. ** We should also think about build dependencies, so that the zip file gets recreated whenever its content would change. Maybe it would be best to move creation of the zip file to the python module? Then, both executing .../program/python -c 'import uno' in the installation set and running the Hello World macro from Tools - Macros... worked fine for me on Linux (the former of course only by luck, printing a NONMACSECTION: command not found message; see above). Stephan diff --git a/pyuno/CustomTarget_python_shell.mk b/pyuno/CustomTarget_python_shell.mk index ea87a9b..9dce854 100644 --- a/pyuno/CustomTarget_python_shell.mk +++ b/pyuno/CustomTarget_python_shell.mk @@ -42,8 +42,8 @@ endif $(call gb_CustomTarget_get_workdir,pyuno/python_shell)/python.sh : $(SRCDIR)/pyuno/zipcore/python.sh | \ $(call gb_CustomTarget_get_workdir,pyuno/python_shell)/.dir $(call gb_Output_announce,$(subst $(WORKDIR)/,,$@),$(true),SED,1) - sed -e s/%%PYVERSION%%/$(PYTHON_SHELL_VERSION)/g -e \ -s/%%OOO_LIBRARY_PATH_VAR%%/$(OOO_LIBRARY_PATH_VAR)/g \ + sed -e s/%%PYVERSION%%/$(pyuno_PYTHON_SHELL_VERSION)/g -e \ +s/%%OOO_LIBRARY_PATH_VAR%%/$(gb_Helper_LIBRARY_PATH_VAR)/g \ $(PYTHON_SHELL_STRIPRULE) $? $@ chmod +x $@ diff --git a/pyuno/CustomTarget_zipcore.mk b/pyuno/CustomTarget_zipcore.mk index 3024355..171e3d2 100644 --- a/pyuno/CustomTarget_zipcore.mk +++ b/pyuno/CustomTarget_zipcore.mk @@ -38,7 +38,7 @@ $(call gb_CustomTarget_get_target,pyuno/zipcore) : \ $(call gb_CustomTarget_get_workdir,pyuno/zipcore)/$(pyuno_PYTHON_ARCHIVE_NAME) :| \ $(call gb_CustomTarget_get_workdir,pyuno/zipcore)/.dir $(call gb_Output_announce,$(subst $(WORKDIR)/,,$@),$(true),ZIP,1) - cd $(OUTDIR)/lib/ zip $@ `$(FIND) python -type f | $(GREP) -v \.pyc | $(GREP) -v \.py~ | $(GREP) -v .orig | $(GREP) -v _failed` + cd $(OUTDIR)/lib/python zip $@ `$(FIND) . -type f | $(GREP) -v \.pyc | $(GREP) -v \.py~ | $(GREP) -v .orig | $(GREP) -v _failed` # TODO: implement this if diff --git a/pyuno/source/module/pyuno_dlopenwrapper.c b/pyuno/source/module/pyuno_dlopenwrapper.c index 3bdc912..cb17c51 100644 --- a/pyuno/source/module/pyuno_dlopenwrapper.c +++ b/pyuno/source/module/pyuno_dlopenwrapper.c @@ -84,14 +84,14 @@ static void * load(void * address, char const * symbol) { #if PY_MAJOR_VERSION = 3 -PyObject * PyInit_pyuno(void) { +SAL_DLLPUBLIC_EXPORT PyObject * PyInit_pyuno(void) { return ((PyObject * (*)(void)) load((void *) PyInit_pyuno, PyInit_pyuno))(); } #else -void initpyuno(void) { +SAL_DLLPUBLIC_EXPORT void initpyuno(void) { ((void (*)(void)) load((void *) initpyuno, initpyuno))(); } diff --git a/scp2/source/python/file_python.scp b/scp2/source/python/file_python.scp index 194394d..e21f0a1 100644 --- a/scp2/source/python/file_python.scp +++ b/scp2/source/python/file_python.scp @@ -73,13 +73,26 @@ File gid_File_Py_Pythonloader End #ifndef SYSTEM_PYTHON +#ifndef MACOSX +Directory gid_Dir_Py_PythonCore +ParentID = gid_Brand_Dir_Program; +HostName = STRING
[Libreoffice-commits] .: shell/source
shell/source/backends/wininetbe/wininetbackend.cxx | 66 ++--- 1 file changed, 44 insertions(+), 22 deletions(-) New commits: commit 0dded0d18a5945ed5a38623068ba7aa93da39df0 Author: Stephan Bergmann sberg...@redhat.com Date: Fri Jun 8 16:54:18 2012 +0200 fdo#47044: Adapt to different Windows versions' InternetQueryOption behavior Change-Id: Ia4d1d8f903872e5eefae2d9687866243b9055a13 diff --git a/shell/source/backends/wininetbe/wininetbackend.cxx b/shell/source/backends/wininetbe/wininetbackend.cxx old mode 100644 new mode 100755 index 4f364f7..ac6829e --- a/shell/source/backends/wininetbe/wininetbackend.cxx +++ b/shell/source/backends/wininetbe/wininetbackend.cxx @@ -123,32 +123,54 @@ WinInetBackend::WinInetBackend() GetProcAddress( hWinInetDll.module, InternetQueryOptionA ) ); if (lpfnInternetQueryOption) { -LPINTERNET_PROXY_INFO lpi = NULL; - -// query for the neccessary space -DWORD dwLength = 0; -lpfnInternetQueryOption( -NULL, -INTERNET_OPTION_PROXY, -(LPVOID)lpi, -dwLength ); - -// allocate sufficient space on the heap -// insufficient space on the heap results -// in a stack overflow exception, we assume -// this never happens, because of the relatively -// small amount of memory we need -// alloca is nice because it is fast and we don't -// have to free the allocated memory, it will be -// automatically done -lpi = reinterpret_cast LPINTERNET_PROXY_INFO ( -alloca( dwLength ) ); - -lpfnInternetQueryOption( +// Some Windows versions would fail the InternetQueryOption call +// with ERROR_OUTOFMEMORY when the initial dwLength were zero (and +// are apparently fine with the initial sizeof (INTERNET_PROXY_INFO) +// and need no reallocation), while other versions fail with +// ERROR_INSUFFICIENT_BUFFER upon that initial dwLength and need a +// reallocation: +INTERNET_PROXY_INFO pi; +LPINTERNET_PROXY_INFO lpi = pi; +DWORD dwLength = sizeof (INTERNET_PROXY_INFO); +BOOL ok = lpfnInternetQueryOption( NULL, INTERNET_OPTION_PROXY, (LPVOID)lpi, dwLength ); +if (!ok) +{ +DWORD err = GetLastError(); +if (err = ERROR_INSUFFICIENT_BUFFER) +{ +// allocate sufficient space on the heap +// insufficient space on the heap results +// in a stack overflow exception, we assume +// this never happens, because of the relatively +// small amount of memory we need +// alloca is nice because it is fast and we don't +// have to free the allocated memory, it will be +// automatically done +lpi = reinterpret_cast LPINTERNET_PROXY_INFO ( +alloca( dwLength ) ); +ok = lpfnInternetQueryOption( +NULL, +INTERNET_OPTION_PROXY, +(LPVOID)lpi, +dwLength ); +if (!ok) +{ +err = GetLastError(); +} +} +if (!ok) +{ +SAL_WARN( +shell, +InternetQueryOption INTERNET_OPTION_PROXY + GetLastError= err); +return; +} +} // if a proxy is disabled, InternetQueryOption returns // an empty proxy list, so we don't have to check if ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: [GERRIT] gbuild migration: pyuno module
On 06/11/2012 12:17 AM, David Ostrovsky wrote: On 08.06.2012 15:21, Stephan Bergmann wrote: * SAL_DLLPUBLIC_EXPORT was missing from pyuno/source/module/pyuno_dlopenwrapper.c, so that the pyuno.so wrapper did not export initpyuno. Well it was done intentionally, my understanding was that it is not needed. That definition of initpyuno is not preceded by any declaration of it that would already carry a SAL_DLLPUBLIC_EXPORT, so it is needed here to make the function exported at all. ** We should also think about build dependencies, so that the zip file gets recreated whenever its content would change. Yes, it was my fault: it was lost during migration. Added it again. Still can not see, why somebody would build pyuno, change internal python files in python module and then would build pyuno again?! But still we shouldn't break it. ./g pull make could be broken by such missing dependencies. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: sc/source
sc/source/ui/condformat/condformatdlg.cxx |4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) New commits: commit 0d6e3233a7062c895dd6e00121ab977bde6dea7d Author: Stephan Bergmann sberg...@redhat.com Date: Mon Jun 11 09:29:36 2012 +0200 eMode may be used uninitialized in this function Change-Id: Ic5d17c360be7f3cb3d5004327d2a6d76dce40c52 diff --git a/sc/source/ui/condformat/condformatdlg.cxx b/sc/source/ui/condformat/condformatdlg.cxx index 27127e9..fdfdb1a 100644 --- a/sc/source/ui/condformat/condformatdlg.cxx +++ b/sc/source/ui/condformat/condformatdlg.cxx @@ -51,6 +51,7 @@ #include globstr.hrc +#include cassert #include iostream namespace { @@ -654,7 +655,8 @@ ScFormatEntry* ScCondFrmtEntry::createConditionEntry() const eMode = SC_COND_NOTDUPLICATE; break; default: - break; +assert(false); // this cannot happen +return NULL; } rtl::OUString aExpr1 = maEdVal1.GetText(); ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: ure/source
ure/source/startup.sh |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) New commits: commit fa414fcf975b8eb572ce3679a0b91e2e16ca7f2e Author: Olivier Hallot olivier.hal...@alta.org.br Date: Sat Jun 9 13:31:52 2012 -0300 Fix typo in /ure/bin/startup.sh Fix typo in ure/bin/starttup.sh for AIX LIBPATH Change-Id: I321fe67c48c3eafd81a6f271422272795c27bd20 diff --git a/ure/source/startup.sh b/ure/source/startup.sh index 1b80e16..d3eeebf 100644 --- a/ure/source/startup.sh +++ b/ure/source/startup.sh @@ -42,7 +42,7 @@ if [ -x ${epath}/javaldx ] ; then sd_platform=`uname -s` case $sd_platform in AIX) -LIBPAT=${jpath}${LIBPATH:+:${LIBPATH}} +LIBPATH=${jpath}${LIBPATH:+:${LIBPATH}} export LIBPATH ;; *) ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: [PUSHED] Re: LIBPAT under AIX?
On 06/09/2012 06:35 PM, Olivier Hallot wrote: Patch attached. Pushed, thanks. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Build error: bf_migratefilter.component
On 06/09/2012 08:58 PM, Kálmán „KAMI” Szalai wrote: dmake: Error: -- `/home/libo64/libo/solver/unxlngx6.pro/xml/bf_migratefilter.component' not found, and can't be made core and binfilter repos run out of sync (i.e., does ./g pull help)? otherwise, maybe a problem with changing --enable-binfilter between rebuilds? Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [PATCH] disable mozilla per default on all platforms (was Re: --disable mozilla necessary for Ubuntu)
On 06/08/2012 09:59 PM, David Ostrovsky wrote: On 08.06.2012 12:52, Rene Engelhard wrote: On Thu, Jun 07, 2012 at 09:01:31PM -0700, Joel Madero wrote: It's been a long time since the default install has worked for me on any Ubuntu or Ubuntu based distro. I always have to add --disable-mozilla in Sane option. Anyone serious wants that. :) no, not with this patch any more ;-) I'm not sure we want this patch (see the other branch of this mail thread for fixing the build on Ubuntu instead). But if we do apply it, we should also adapt AC_ARG_ENABLE(mozilla, AS_HELP_STRING([--disable-mozilla], [LibreOffice usually includes a strangely hacked up Mozilla binary for your platform, to build without this version, use this option.]) ) Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Using same User Profile with various users and / or LibO Versions
On 06/10/2012 10:08 AM, Rainer Bielefeld wrote: it would be great if an expert could add some brief notes on https://wiki.documentfoundation.org/UserProfile concerning following questions: - Is it allowed/possible/useful to use the same User Profile with more than 1 User? - Is it allowed/useful/ to use the same User Profile with different LibO Versions? see https://wiki.documentfoundation.org/index.php?title=UserProfileoldid=51149 Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice-qa] New Bugzilla Version Picker items – 2
On 06/08/2012 03:45 PM, Petr Mladek wrote: The problem might be the tilda '~'. It is substituted to $HOME in linux shell. ...but only at the beginning of a word. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: libcdr/libcdr-0.0.8-msc.patch libcdr/makefile.mk
libcdr/libcdr-0.0.8-msc.patch | 10 ++ libcdr/makefile.mk|2 ++ 2 files changed, 12 insertions(+) New commits: commit d6576f9dd1d872b646d2cdc56a17d73f1d57dc13 Author: Stephan Bergmann sberg...@redhat.com Date: Mon Jun 11 15:17:04 2012 +0200 MSC fix Change-Id: I12c8ab5779d952eb640298478da7084b5f643568 diff --git a/libcdr/libcdr-0.0.8-msc.patch b/libcdr/libcdr-0.0.8-msc.patch new file mode 100644 index 000..6a44e7d --- /dev/null +++ b/libcdr/libcdr-0.0.8-msc.patch @@ -0,0 +1,10 @@ +--- misc/libcdr-0.0.8/src/lib/libcdr_utils.h 2012-06-06 13:42:03.0 +0200 misc/build/libcdr-0.0.8/src/lib/libcdr_utils.h 2012-06-11 15:09:43.305289500 +0200 +@@ -50,6 +50,7 @@ + typedef unsigned char uint8_t; + typedef unsigned short uint16_t; + typedef unsigned uint32_t; ++typedef short int16_t; + typedef int int32_t; + typedef unsigned __int64 uint64_t; + diff --git a/libcdr/makefile.mk b/libcdr/makefile.mk old mode 100644 new mode 100755 index c4636d0..771ac46 --- a/libcdr/makefile.mk +++ b/libcdr/makefile.mk @@ -62,6 +62,8 @@ INCPRE+=$(SOLARVER)$/$(INPATH)$/inc$/lcms2 TARFILE_NAME=libcdr-0.0.8 TARFILE_MD5=ce5a1def34578b75959ac31210f031f6 +PATCH_FILES = libcdr-0.0.8-msc.patch + BUILD_ACTION=dmake $(MFLAGS) $(CALLMACROS) BUILD_DIR=src$/lib ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: Build Error
On 06/11/2012 08:23 AM, Rob Snelders wrote: When i tried a build with debugging then I ran into the error as is in the file build-error. The attached patch solved the error. But that patch didn't solve the problem for me (GetFontFileName is declared to return rtl::OString* after all, but happens to remain undefined). I fixed that now with http://cgit.freedesktop.org/libreoffice/core/commit/?id=776abbb83e867ac5336d9da2097d91ad8152ef55 Adapt gcach_layout.cxx to SAL_INFO. Thanks anyway, Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: regcomp under windows, registering components
On 06/11/2012 04:55 PM, Olivier Hallot wrote: Is there a tool to handle services.rdb, beside pure xml text edition? Nothing besides (XML-aware) text editors. regview etc. no longer work on it. (Which I considered a minor problem when changing it, as the information is already textual now. But of course, any scripts etc. built around it would now need to be adapted.) How easy it is to have the extension manager GUI as a installable module, or bundled optional extension? Is it something that can fit into an easy hack? (I am seeking to clone the update manager feature as starting point). If you want to make the extension manager GUI into an optional module (so that, e.g., end users have no easy way to mess with it), that shouldn't be too difficult. Things that come to mind: * The code should mostly be separated into a deploymentgui dynamic library already (in module desktop). * Hooks that call into that code are Tools - Extension Manager... and trying to open an .oxt file (e.g., passing it to LO on command line). * It might be necessary to make the code for only update of extensions conditional on whether the GUI is available. * There might be more... Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: vcl/source
vcl/source/gdi/pdfwriter_impl.cxx | 22 -- 1 file changed, 20 insertions(+), 2 deletions(-) New commits: commit 4eeaa94ae07f45f0e9f1f05c474fa75e955d4e70 Author: Gökçen Eraslan gokcen.eras...@gmail.com Date: Mon Jun 11 17:18:10 2012 +0200 Make DO_TEST_PDF code work again Change-Id: I3b3e5ab68421e25e6bf153f11ca5e538aa482874 diff --git a/vcl/source/gdi/pdfwriter_impl.cxx b/vcl/source/gdi/pdfwriter_impl.cxx index df7d15e..9c1e6bb 100644 --- a/vcl/source/gdi/pdfwriter_impl.cxx +++ b/vcl/source/gdi/pdfwriter_impl.cxx @@ -135,7 +135,8 @@ void doTestCode() aContext.DocumentInfo.Title = OUString( PDF export test document ); aContext.DocumentInfo.Producer = OUString( VCL ); -PDFWriter aWriter( aContext ); +com::sun::star::uno::Reference com::sun::star::beans::XMaterialHolder xEnc; +PDFWriter aWriter( aContext, xEnc ); aWriter.NewPage( 595, 842 ); aWriter.BeginStructureElement( PDFWriter::Document ); // set duration of 3 sec for first page @@ -180,7 +181,7 @@ void doTestCode() aWriter.SetActualText( String( RTL_CONSTASCII_USTRINGPARAM( It was the best of PDF, it was the worst of PDF ... or so. This is a pretty nonsensical text to denote a paragraph. I suggest you stop reading it. Because if you read on you might get bored. So continue on your on risk. Hey, you're still here ? Why do you continue to read this as it is of no use at all ? OK, it's your time, but still... . Woah, i even get bored writing this, so let's end this here and now. ) ) ); aWriter.SetAlternateText( String( RTL_CONSTASCII_USTRINGPARAM( This paragraph contains some lengthy nonsense to test structural element emission of PDFWriter. ) ) ); aWriter.EndStructureElement(); -sal_Int32 nLongPara = aWriter.BeginStructureElement( PDFWriter::Paragraph ); +aWriter.BeginStructureElement( PDFWriter::Paragraph ); aWriter.SetStructureAttribute( PDFWriter::WritingMode, PDFWriter::LrTb ); aWriter.DrawText( Rectangle( Point( 4500, 19000 ), Size( 12000, 1000 ) ), String( RTL_CONSTASCII_USTRINGPARAM( This paragraph is nothing special either but ends on the next page structurewise ) ), @@ -216,13 +217,19 @@ void doTestCode() sal_Int32 nFirstDest = aWriter.CreateDest( aTargetRect ); // enable structure aWriter.EndStructureElement(); + // add something to the long paragraph as an afterthought +/* PDFWriter::aWriter.GetCurrentStructureElement removed as an unusedcode.easy item: + http://cgit.freedesktop.org/libreoffice/core/commit/?id=09279fe3dad24ab58121e4f0a9564d252b64d81a + sal_Int32 nSaveStruct = aWriter.GetCurrentStructureElement(); aWriter.SetCurrentStructureElement( nLongPara ); aWriter.DrawText( Rectangle( Point( 4500,4500 ), Size( 12000, 1000 ) ), String( RTL_CONSTASCII_USTRINGPARAM( Add something to the longish paragraph above. ) ), TEXT_DRAW_MULTILINE | TEXT_DRAW_WORDBREAK ); aWriter.SetCurrentStructureElement( nSaveStruct ); +*/ + aWriter.EndStructureElement(); aWriter.EndStructureElement(); aWriter.BeginStructureElement( PDFWriter::Figure ); @@ -288,6 +295,11 @@ void doTestCode() aTranspRect = Rectangle( Point( 1500, 16500 ), Size( 4800, 3000 ) ); aWriter.SetFillColor( Color( COL_LIGHTRED ) ); aWriter.DrawRect( aTranspRect ); + +/* +EndTransparencyGroup( const Rectangle rBoundRect, const Bitmap rAlphaMask ) is removed as an unusedcode.easy item: + http://cgit.freedesktop.org/libreoffice/core/commit/?id=581e7d7057afa87036d84e42c0e0a8a7368e20c7 + aWriter.BeginTransparencyGroup(); aWriter.SetFillColor( Color( COL_LIGHTGREEN ) ); aWriter.DrawEllipse( aTranspRect ); @@ -296,6 +308,7 @@ void doTestCode() String( RTL_CONSTASCII_USTRINGPARAM( Some transparent text ) ), TEXT_DRAW_CENTER | TEXT_DRAW_VCENTER | TEXT_DRAW_MULTILINE | TEXT_DRAW_WORDBREAK ); aWriter.EndTransparencyGroup( aTranspRect, aTransMask ); +*/ Bitmap aImageBmp( Size( 256, 256 ), 24 ); pAcc = aImageBmp.AcquireWriteAccess(); @@ -333,6 +346,10 @@ void doTestCode() aWall.SetStyle( WALLPAPER_TILE ); aWriter.DrawWallpaper( Rectangle( Point( 4400, 4200 ), Size( 10200, 6300 ) ), aWall ); +/* +BeginPattern/EndPattern is removed as an unusedcode.easy item: + http://cgit.freedesktop.org/libreoffice/core/commit/?id=581e7d7057afa87036d84e42c0e0a8a7368e20c7 + aWriter.Push( PUSH_ALL ); aWriter.BeginPattern(Rectangle(Point(0,0),Size(2000,1000))); aWriter.SetFillColor( Color( COL_RED ) ); @@ -350,6 +367,7 @@ void doTestCode() aWriter.SetFillColor(); aWriter.SetLineColor( Color( COL_LIGHTBLUE ) ); aWriter.DrawRect( aPolyRect ); +*/ aWriter.NewPage( 595, 842 ); aWriter.SetMapMode( MapMode( MAP_100TH_MM ) ); ___
Re: Removed parts of PDFWriter and doTestCode method
On 06/10/2012 01:13 PM, Gökçen Eraslan wrote: I have commented out outdated parts and added a com::sun::star::uno::Reference com::sun::star::beans::XMaterialHolder to make it work. Patch is attached. In IRC, Tor said that writing a unit test is a better way. Maybe I can write one. I pushed the patch now, as http://cgit.freedesktop.org/libreoffice/core/commit/?id=4eeaa94ae07f45f0e9f1f05c474fa75e955d4e70 Make DO_TEST_PDF code work again. (In the future, please provide patches created with git format-patch as they are easier to handle.) For the functionality removed through cleanups, you can either leave it commented out for now, or---if it would be useful for your work---get the relevant functions back and wrap their declarations and definitions in #ifdef DO_TEST_PDF. Replacing this inline test code with a unit test would surely be desirable. However, I would consider that of minor importance for your GSoC project and would advise against spending any more time than necessary on this right now. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: Using same User Profile with various users and / or LibO Versions
[putting libreoffice ML back on cc] On 06/11/2012 05:18 PM, Rainer Bielefeld wrote: The facts you mentioned cause some problems for me: I often try various Versions to find out where the Bug appeared, some of them intended with the same profile to unify test conditions. I do not believe that it will be a hard restriction to use all versions with their own profile, but you never know ... In practice, this will likely work just fine most of the time, I assume. An other problem I see if someone decides to downgrade. Do we have to recommend to do a backup of the existing User Profile before an update because there might be problems after the downgrade? * If users do not force the user profile to a non-default location, LO should largely be able to handle a downgrade just fine. ** If the relevant versions use different default user profile locations, nothing bad can happen. (Just note that once a given user profile has been marked as migrated, it will not be used for migration again. So any changes the user makes to the downgraded version's user profile will be lost should the user later upgrade again.) ** If the relevant versions use the same default user profile location, there *is* a chance that data stored there has changed in a forward-compatible but not backward-compatible way. (I doubt anybody takes care to prevent that; or am I mistaken?) In such a case, it might indeed be advisable to make a backup prior to downgrade. * If a user explicitly forces the user profile to a non-default location and reuses it across the downgrade, then chances are things will not work as intended, and it is definitely advisable to make a backup (and even more advisable to not reuse in the first place). Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [REVIEW] fdo#43989 first launch terminates after splash
On 06/11/2012 05:47 PM, Korrawit Pruegsanusak wrote: I'd like to propose Stephan's patch into stable branch: * http://cgit.freedesktop.org/libreoffice/core/commit/?id=55836370ffb70a34b888f81cdacdfede8fee29cf This fixed fdo#43989, which is 3.5 MAB, but we already have Andras' 8783ead70cc2bc2a83bf473b0dfb51f3ee10b6da backported, so I'm not sure whether we should have Stephan's patch backported too. Also, from the commit message, Stephan committed new fix for bnc#524250 at b5ad981518f4407243f89507b2561dd0b1ef28ee. Should this patch backported as well? From #libreoffice-dev backlogs: Jun 07 11:22:35 sberg timar, tml_, I'm unsure whether we want 55836370ffb70a34b888f81cdacdfede8fee29cf (and maybe b5ad981518f4407243f89507b2561dd0b1ef28ee, too) in -3-6 or even -3-5; without them, we'll keep the behavior on Windows that there is no automatic restart (is this by now considered a feature, or rather a bug?) but also that LO only starts upon second attempt sporadically (i.e., when syncing with bundled/shared extensions needs to be done, after Jun 07 11:22:35 sberg those have changed in some way) Jun 07 11:22:37 IZBot no match found - might try http://cgit.freedesktop.org/libreoffice/core/commit/?id=55836370ffb70a34b888f81cdacdfede8fee29cf in a couple of minutes Jun 07 11:23:03 sberg s/no automatic restart/no automatic restart *after crash*/ Jun 07 11:42:46 timar sberg: and what about, if we restarted only after E_NORMAL_RESTART, and not after crash? Jun 07 11:49:06 sberg timar, one could fiddle something like that into -3-6 etc., sure; but I would keep it simple -- either backport wholesale, or live with the current situation on -3-6 etc., which should be ~ok after your fix for fdo#43989 Jun 07 11:49:10 IZBot LibreOffice-UI normal/medium RESOLVED FIXED After installation first launch terminates after splash screen https://bugs.freedesktop.org/show_bug.cgi?id=43989 My take is that b5ad981518f4407243f89507b2561dd0b1ef28ee isn't entirely trivial (there is that curious distinction between bExistsRecoveryData and bExistsSessionData in that code...), but backporting 55836370ffb70a34b888f81cdacdfede8fee29cf without b5ad981518f4407243f89507b2561dd0b1ef28ee wouldn't be ideal, either. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: sc/inc
sc/inc/conditio.hxx |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) New commits: commit a61b43cbf9c798f553e5d81c60f59a29e47d240f Author: Stephan Bergmann sberg...@redhat.com Date: Tue Jun 12 08:36:34 2012 +0200 -Werror,-Wmismatched-tags Change-Id: If2b4118481dd265a81641d5cdf14f5095b8064e9 diff --git a/sc/inc/conditio.hxx b/sc/inc/conditio.hxx index c68875f..2938ca7 100644 --- a/sc/inc/conditio.hxx +++ b/sc/inc/conditio.hxx @@ -70,7 +70,7 @@ enum ScConditionMode }; class ScConditionalFormat; -class ScDataBarInfo; +struct ScDataBarInfo; namespace condformat { ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: pyuno is broken on mingw on current master
On 06/12/2012 09:17 AM, David Ostrovsky wrote: What do you mean here? Personally, for an experimental platform like mingw, and as you have demonstrated that it does not work right now, I would not bother to make it work prior to gbuild'ification. Has pyuno on mingw ever been known to work? Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: Branch 'libreoffice-3-6' - shell/source
shell/source/backends/wininetbe/wininetbackend.cxx | 66 ++--- 1 file changed, 44 insertions(+), 22 deletions(-) New commits: commit 4c07fd7f899b75edd61aa0d16c0f3dff0ebb94ad Author: Stephan Bergmann sberg...@redhat.com Date: Fri Jun 8 16:54:18 2012 +0200 fdo#47044: Adapt to different Windows versions' InternetQueryOption behavior Change-Id: Ia4d1d8f903872e5eefae2d9687866243b9055a13 (cherry picked from commit 0dded0d18a5945ed5a38623068ba7aa93da39df0) diff --git a/shell/source/backends/wininetbe/wininetbackend.cxx b/shell/source/backends/wininetbe/wininetbackend.cxx old mode 100644 new mode 100755 index 4f364f7..ac6829e --- a/shell/source/backends/wininetbe/wininetbackend.cxx +++ b/shell/source/backends/wininetbe/wininetbackend.cxx @@ -123,32 +123,54 @@ WinInetBackend::WinInetBackend() GetProcAddress( hWinInetDll.module, InternetQueryOptionA ) ); if (lpfnInternetQueryOption) { -LPINTERNET_PROXY_INFO lpi = NULL; - -// query for the neccessary space -DWORD dwLength = 0; -lpfnInternetQueryOption( -NULL, -INTERNET_OPTION_PROXY, -(LPVOID)lpi, -dwLength ); - -// allocate sufficient space on the heap -// insufficient space on the heap results -// in a stack overflow exception, we assume -// this never happens, because of the relatively -// small amount of memory we need -// alloca is nice because it is fast and we don't -// have to free the allocated memory, it will be -// automatically done -lpi = reinterpret_cast LPINTERNET_PROXY_INFO ( -alloca( dwLength ) ); - -lpfnInternetQueryOption( +// Some Windows versions would fail the InternetQueryOption call +// with ERROR_OUTOFMEMORY when the initial dwLength were zero (and +// are apparently fine with the initial sizeof (INTERNET_PROXY_INFO) +// and need no reallocation), while other versions fail with +// ERROR_INSUFFICIENT_BUFFER upon that initial dwLength and need a +// reallocation: +INTERNET_PROXY_INFO pi; +LPINTERNET_PROXY_INFO lpi = pi; +DWORD dwLength = sizeof (INTERNET_PROXY_INFO); +BOOL ok = lpfnInternetQueryOption( NULL, INTERNET_OPTION_PROXY, (LPVOID)lpi, dwLength ); +if (!ok) +{ +DWORD err = GetLastError(); +if (err = ERROR_INSUFFICIENT_BUFFER) +{ +// allocate sufficient space on the heap +// insufficient space on the heap results +// in a stack overflow exception, we assume +// this never happens, because of the relatively +// small amount of memory we need +// alloca is nice because it is fast and we don't +// have to free the allocated memory, it will be +// automatically done +lpi = reinterpret_cast LPINTERNET_PROXY_INFO ( +alloca( dwLength ) ); +ok = lpfnInternetQueryOption( +NULL, +INTERNET_OPTION_PROXY, +(LPVOID)lpi, +dwLength ); +if (!ok) +{ +err = GetLastError(); +} +} +if (!ok) +{ +SAL_WARN( +shell, +InternetQueryOption INTERNET_OPTION_PROXY + GetLastError= err); +return; +} +} // if a proxy is disabled, InternetQueryOption returns // an empty proxy list, so we don't have to check if ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
[Libreoffice-commits] .: sw/CppunitTest_sw_subsequent_ww8tok.mk
sw/CppunitTest_sw_subsequent_ww8tok.mk |1 + 1 file changed, 1 insertion(+) New commits: commit 770e990dabcd4f0415da65ef3ace4fc5a3784278 Author: Stephan Bergmann sberg...@redhat.com Date: Tue Jun 12 10:31:35 2012 +0200 Missing dependency Change-Id: Iffbd7722b7ede8a2824e59f0bfc0241730de2486 diff --git a/sw/CppunitTest_sw_subsequent_ww8tok.mk b/sw/CppunitTest_sw_subsequent_ww8tok.mk index 893f4a4..d95bfa6 100644 --- a/sw/CppunitTest_sw_subsequent_ww8tok.mk +++ b/sw/CppunitTest_sw_subsequent_ww8tok.mk @@ -36,6 +36,7 @@ $(eval $(call gb_CppunitTest_add_exception_objects,sw_subsequent_ww8tok, \ $(eval $(call gb_CppunitTest_use_libraries,sw_subsequent_ww8tok, \ cppu \ +cppuhelper \ sal \ test \ unotest \ ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits
Re: extraordinary stoc/ thrash ...
On 04/05/2012 02:23 PM, Stephan Bergmann wrote: On 04/03/2012 02:27 PM, Stephan Bergmann wrote: I have a vague idea of placing yet another cppuhelper bootstrap mechanism next to the existing ones, which will internally use completely different (read: cheap) mechanisms to set up a component context and associated service manager. That way, I would avoid most of the hassle of having to whack improvements into a rotten framework. I'll toy around with that in the next days/weeks. Hang on... The attached defaultbootstrap.patch is work in progress along that way. It adds a new cppu::defaultBootstrap() to replace cppu::defaultBootstrap_initialComponentContext(), and hooks that up in desktop/source/app/appinit.cxx. Internally, defaultBootstrap tries to be slim and streamlined and avoid to reach out into the overly generic world of UNO infrastructure services as much as possible. It is still completely unfinished, though: Just to tie up the loose ends: This has been properly fixed for quite a while now, starting with http://cgit.freedesktop.org/libreoffice/core/commit/?id=c2fd9b533cfad18735df212cc7fd61443628dc0c New cppu::defaultBootstrap_InitialComponentContext implementation. Stephan ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice-commits] .: moz/extractfiles.mk moz/zipped nss/makefile.mk
moz/extractfiles.mk|7 ++- moz/zipped/makefile.mk | 12 nss/makefile.mk| 12 3 files changed, 26 insertions(+), 5 deletions(-) New commits: commit dc82cf021f76ea83ff7a4bcb2d7525f2e111f0cc Author: Stephan Bergmann sberg...@redhat.com Date: Tue Jun 12 13:44:28 2012 +0200 Make PyUNO work --with-macox-version-min-required=10.6 ...by using system libsqlite3.dylib instead of nss one (see comment in nss/makefile.mk for details). (dmake knows neither nor ! in conditional expressions, that is the reason for the strange .IF .ELSE ... .END constructs.) Change-Id: I5abe0f74ad55aac06b91163751bb8145b759b9ca diff --git a/moz/extractfiles.mk b/moz/extractfiles.mk index e4e40b8..6035d5f 100644 --- a/moz/extractfiles.mk +++ b/moz/extractfiles.mk @@ -63,9 +63,14 @@ NSS_MODULE_RUNTIME_LIST:= \ plds4 \ smime3 \ softokn3 \ -sqlite/sqlite3 \ ssl3 +# For Mac OS X = 10.6 the system lib is used instead (see nss/makefile.mk): +.IF $(OS) == MACOSX $(MAC_OS_X_VERSION_MIN_REQUIRED) = 1060 +.ELSE +NSS_MODULE_RUNTIME_LIST += sqlite/sqlite3 +.END + BIN_RUNTIMELIST= \ xpcom \ xpcom_core \ diff --git a/moz/zipped/makefile.mk b/moz/zipped/makefile.mk index 0f1fe2c..a08561e 100644 --- a/moz/zipped/makefile.mk +++ b/moz/zipped/makefile.mk @@ -170,9 +170,6 @@ FREEBL=freebl3 FREEBL=freebl3 .ENDIF # $(OS) == SOLARIS - -#On Linux/Unix sqlite is delivered to $(SOLARLIBDIR)/sqlite/libsqlite3.so -#See readme.txt in module nss NSS_MODULE_RUNTIME_LIST:= \ $(FREEBL) \ nspr4 \ @@ -184,9 +181,16 @@ NSS_MODULE_RUNTIME_LIST:= \ plds4 \ smime3 \ softokn3 \ -sqlite/sqlite3 \ ssl3 +# On Linux/Unix sqlite is delivered to $(SOLARLIBDIR)/sqlite/libsqlite3.so (see +# nss/README) and for Mac OS X = 10.6 the system lib is used instead (see +# nss/makefile.mk): +.IF $(OS) == MACOSX $(MAC_OS_X_VERSION_MIN_REQUIRED) = 1060 +.ELSE +NSS_MODULE_RUNTIME_LIST += sqlite/sqlite3 +.END + # Remove the nss libs build in moz and those build in the nss module $(MISC)$/replace_old_nss_libs : $(MISC)$/unpacked_$(TARGET)_lib \ $(MISC)$/unpacked_$(TARGET)_inc $(BIN)$/mozruntime.zip diff --git a/nss/makefile.mk b/nss/makefile.mk index 4e8dd9d..4338857 100644 --- a/nss/makefile.mk +++ b/nss/makefile.mk @@ -106,6 +106,18 @@ BUILD_ACTION+=FREEBL_NO_DEPEND=1 BUILD_ACTION+=NS_USE_GCC=1 .ENDIF +# Otherwise, LibreOffice.app/Contents/MacOS/python -c 'import uno' fails to +# dlopen LibreOffice.app/Contents/MacOS/libpyuno.dylib (at +# pyuno/source/module/pyuno_dlopenwrapper.c:73) with dlerror Symbol not found: +# _sqlite3_wal_checkpoint; Referenced from: /System/Library/Frameworks/ +# CoreServices.framework/Versions/A/Frameworks/CFNetwork.framework/Versions/A/ +# CFNetwork; Expected in: [...]/LibreOffice.app/Contents/MacOS/libsqlite3.dylib; +# in /System/Library/Frameworks/CoreServices.framework/Versions/A/Frameworks/ +# CFNetwork.framework/Versions/A/CFNetwork: +.IF $(OS) == MACOSX $(MAC_OS_X_VERSION_MIN_REQUIRED) = 1060 +BUILD_ACTION += NSS_USE_SYSTEM_SQLITE=1 +.END + .ENDIF # $(GUI)==UNX ___ Libreoffice-commits mailing list libreoffice-comm...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice-commits