summaryrefslogtreecommitdiff
path: root/RepositoryFixes.mk
AgeCommit message (Collapse)Author
2012-04-13setup/teardown default component context once before/after all testsCaolán McNamara
avoids the problems of dangling uno singletons invalidated after the first dispose and the chain of other singletons that don't expect to need to re-initialize, etc. reenable editeng cppunit test inherit i18npool cppunit test from unotest base drop LibreOfficeProtector, do "throwable" work in setUp/tearDown not in ctors/dtors
2012-04-13gbuildize internal zlibCaolán McNamara
and stick headers into the solver in the same hierarchy as is "standard" so we don't have to patch so much, similiarly rename to libz
2012-03-10no i prefix for lcms2.libMatúš Kukan
2012-03-05Change my mind again: Do build a subset sb library even for DISABLE_SCRIPTINGTor Lillqvist
The tables functionality in Writer documents need SbxValue.
2012-03-05Revert "Build libsb with a few dummy classes also for --disable-scripting"Tor Lillqvist
Nah, these dummy classes would turn out quite complex anyway. Better to just use ifdefs elsewhere, the number required is not that large. This reverts commit 6d33801b4498f5c13d2dc75588f88c22a11f455b.
2012-03-01Build libsb with a few dummy classes also for --disable-scriptingTor Lillqvist
Work in progress, the dummy class implementations surely still incomplete and/or might contain methods not actually needed. More dummy class implementations needed, hopefully not too many though. Will add also a few ifdefs for DISABLE_SCRIPTING in some key places in sc and elsewhere to cut down on the need.
2012-02-27s/interpreters/scripting/Tor Lillqvist
2012-02-27Decouple disabling of scripting (BASIC &co) from OS being iOSTor Lillqvist
2012-02-21More --disable-database-connectivity work, still experimentalTor Lillqvist
2012-02-20Shared libraries must in practice be called lib*.so on AndroidTor Lillqvist
Otherwise the package creating code (part of the SDK) won't include them and/or the package installation code (on the OS itself) won't unpack them. (They just silently skip the file.)
2012-02-10unotest: convert to gbuildMatúš Kukan
2012-01-27For iOS, filter out the sb and vbahelper libs that we don't buildTor Lillqvist
2012-01-15ucb: convert to gbuildMatúš Kukan
2012-01-06Merge branch 'feature/gbuild_extensions'Michael Stahl
Conflicts: extensions/source/ole/servreg.cxx extensions/source/plugin/util/makefile.mk postprocess/packcomponents/makefile.mk postprocess/prj/build.lst tail_build/prj/build.lst
2011-12-24Repository: fix Windows name for internal libexttextcatMatúš Kukan
2011-12-24Merge remote-tracking branch 'origin/master' into feature/gbuild_extensionsPeter Foley
Conflicts: Module_tail_build.mk Repository.mk extensions/util/makefile.pmk postprocess/prj/build.lst svx/AllLangResTarget_svx.mk
2011-12-20Abandon attempt to use the debug CRT in a dbgutil buildTor Lillqvist
2011-12-19fix names of libraries for MinGW, also remove onlinecheck.dllMatúš Kukan
2011-12-19sal: convert to gbuildMatúš Kukan
2011-12-19cppu: convert to gbuildMatúš Kukan
Because of unit tests is here custom target where are idl files processed.
2011-12-17extensions: convert nsplugin to gbuildPeter Foley
2011-12-13add libexttextcat to external libsMatúš Kukan
2011-09-30clean-up ghost traces of unicowNorbert Thiebaud
2011-09-28Android now builds up to instsetoo_nativeThorsten Behrens
Last fixes - remove kludge from RepositoryFixes.mk, have redland build w/o threads for the while, and some hackery to exclude pointless code like oosplash from android build.
2011-09-20Kill the explicit linking to salmain.o, use vclmain instead.Jan Holesovsky
2011-09-19windows berkleydb libname fixupCaolán McNamara
2011-09-15Repository.mk: fix conflicing filterconfig changesMichael Stahl
2011-09-15gbuildize desktopDavid Tardon
2011-09-15use correct static lib. name for windowsDavid Tardon
2011-09-15gbuildize automationDavid Tardon
2011-08-26Clean, sort and change RepositoryFixes a little.Matúš Kukan
2011-08-26convert package to gbuildMatúš Kukan
2011-08-23fix windows xmlsecurity linkCaolán McNamara
2011-08-20no reason for changing name of cui library on windowsMatúš Kukan
2011-08-19Fix i18npool build on Windows.Jan Holesovsky
2011-08-18Sort a block of lines that was almost sorted alreadyTor Lillqvist
2011-08-16fix typoCaolán McNamara
2011-08-16this seems to be the normal patternCaolán McNamara
2011-08-16get these names right for windows (I think)Caolán McNamara
2011-08-09convert xmlsecurity to gbuildMatúš Kukan
2011-08-10prefer makefile-gmake-mode to plain makefile-modeTakeshi Abe
2011-08-05get sccomp to link under windowsCaolán McNamara
2011-08-02Add Emacs mode linesTor Lillqvist
2011-07-26Remove superfluous empty lineTor Lillqvist
The comment for gb_Library_ILIBFILENAMES in the WNTGCC case doesn't make much sense for us, though, as I think we want to support WNTGCC only for cross-compilation, and thus we don't have access to the PSDK ("Platform" SDK, with which they mean the Windows SDK). I don't see any point in using MinGW natively on Windows to build LibreOffice. If you build locally, use MSVC.
2011-07-21resyncing to masterBjoern Michaelsen
2011-07-15Bin OLDNAMESLIB which duplicates the existing gb_Library_win32_OLDNAMESTor Lillqvist
2011-07-15The graphite2 library gets built statically with WNTGCC at leastTor Lillqvist
2011-07-14Factor out the oldnames/moldname difference between WNTMSC and WNTGCCTor Lillqvist
2011-07-05Add more libs that semi-unintentionally get built statically for AndroidTor Lillqvist
2011-07-04Some libraries get built as static archives for AndroidTor Lillqvist
We haven't bothered fixing their configury to recognize that shared libraries can be built for the platform. So for them use the .a names in gb_Library_FILENAMES for Android.