summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2011-01-24WaE: gcc 4.6.0 various warningsCaolán McNamara
2011-01-24WaE: gcc 4.6.0 various warningsCaolán McNamara
2011-01-24WaE: gcc 4.6.0 various warningsCaolán McNamara
2011-01-24WaE: gcc 4.6.0 various warningsCaolán McNamara
2011-01-24WaE: gcc 4.6.0 various warningsCaolán McNamara
2011-01-24WaE: 4.6.0 unused variablesCaolán McNamara
2011-01-23Remove DECLARE_LIST( RscInconsList, RscInconsistent* )Joseph Powers
2011-01-23Remove DECLARE_LIST( FileList, ByteString* )Joseph Powers
2011-01-23Remove DECLARE_LIST( XMLStringList, XMLElement* )Joseph Powers
2011-01-23Remove DECLARE_LIST( XMLChildNodeList, XMLChildNode* )Joseph Powers
2011-01-23Remove DECLARE_LIST( XMLAttributeList, XMLAttribute * )Joseph Powers
2011-01-24Kill Win9x and NT4 codeTor Lillqvist
2011-01-24Kill Win9x and NT4 codeTor Lillqvist
2011-01-24Make parameters to Max() match in typeTor Lillqvist
2011-01-24Hide ULONG while including more Win32 headersTor Lillqvist
2011-01-24Hide and reveal also ULONG in presys.h / postsys.hTor Lillqvist
Although, I fail to understand why the duplication of work done in prewin.h/postwin.h vs. presys.h/postsys.h is needed.
2011-01-24Fix compilation on 64-bit WindowsTor Lillqvist
2011-01-24Fix compilation error on 64-bit WindowsTor Lillqvist
2011-01-24Fix compilation on 64-bit WindowsTor Lillqvist
No need to force use of 32-bit time_t (which doesn't exist for 64-bit Windows), just use defaults.
2011-01-23Remove DECLARE_LIST() Impl_ParserMessageList & TokenListImplJoseph Powers
2011-01-23Remove DECLARE_LIST( LngLineList, ByteString* )Joseph Powers
2011-01-23Remove DECLARE_LIST( GSIBlock_Impl, GSILine *)Joseph Powers
2011-01-22Remove DECLARE_LIST( ResStack, ResData * )Joseph Powers
2011-01-22Remove DECLARE_LIST( ExportListBase, ExportListEntry* )Joseph Powers
2011-01-22Remove DECLARE_LIST( CfgStackList, CfgStackData* )Joseph Powers
2011-01-23Make GetFileHandle's return type match declarationTor Lillqvist
But as the return type is just 32 bits, it is actually not enough for a HANDLE anyway on a 64-bit OS. But on the other hand, this method is private and not used on Windows at all. So just make it abort() then.
2011-01-23Define ULONG as WIN_ULONG while including shlobj.hTor Lillqvist
To match what prewin.h/postwin.h do.
2011-01-23Hide also ULONG while including <windows.h>Tor Lillqvist
LibreOffice's deprecated ULONG is sal_uIntPtr, and thus 64 bits on 64-bit Windows, unlike Windows's ULONG which is 32 bits on both 32- and 64-bit Windows. So we must hide ULONG, too, to avoid compilation errors on 64-bit Windows.
2011-01-22Remove some dead code.Thomas Arnhold
2011-01-22Delete apparently unused SYSTEM_* definitionsTor Lillqvist
2011-01-22Add SYSTEM_* and __DLLEXTENSION values for WNT && X86_64Tor Lillqvist
2011-01-22Remove empty description and dates.Thomas Arnhold
2011-01-21fix KDE3 library search order (fdo#32797)Luboš Luňák
Use LINKFLAGS instead of STDLIBS, as the latter comes way too late in the link command and may result in some stray -L/usr/lib getting before -L$KDEDIR/lib. Since KDE4 libs often tend to be installed under /usr while KDE3 libs are often somewhere else such as /opt/kde3/lib, this could cause using wrong libraries.
2011-01-21Remove obsolete RCS lines.Thomas Arnhold
2011-01-21Remove ShowHelpStatusText and HideHelpStatusText and associated cruftTor Lillqvist
These methods were not implemented by any subclass, and the implementations in the base class were empty. Still, the quite heavy (especially in our multi-lingual builds) GetHelpText() mechanism was called to get a string to pass to ShowHelpStatusText(). So just remove all of it. See see fdo#33088 and fdo#33315.
2011-01-21Remove old RCS lines.Thomas Arnhold
2011-01-21Replace suitable equalsAscii calls with equalsAsciiL.Thomas Arnhold
2011-01-20cppcheck: unused variableCaolán McNamara
2011-01-20Replace suitable equalsAscii calls with equalsAsciiL.Thomas Arnhold
Done with sed -i 's%\(\.equalsAscii\)(\(\s\?"[^"]\+"\)\(\s\?\))%\1L(\3RTL_CONSTASCII_STRINGPARAM(\2\3)\3)%g'.
2011-01-19don't leak PPDParsersCaolán McNamara
2011-01-19cppcheck: object destroyed immediatelyCaolán McNamara
2011-01-18forcing Qt native paint system is not necessary anymoreLuboš Luňák
bf4b61c7b7f2ca1d9641ab0cf5f88a68bc3930d0 fixed the problem.
2011-01-17paint toolbar handle positioned properly (fdo#32558)Luboš Luňák
The problem was that we passed the whole widget geomtry to QStyle but it expected only geometry of the handle and e.g. Oxygen used center of the geometry.
2011-01-17change the adjust argument to more versatile QRectLuboš Luňák
2011-01-17fix after struct member renaminge (pXRenderFormat)Pierre-André Jacquod
the right one, this time.... sorry.
2011-01-17fix after variable renaming (pXrenderformat)Pierre-André Jacquod
2011-01-17fix cases of using screen visual for a drawableLuboš Luňák
But I fear the assumption that there is just one visual per screen is quite spread in the code :-/.
2011-01-17add const to GetBitCount()Luboš Luňák
2011-01-17and fix the 'XRenderPictFormat* pVisualFormat' misnomer tooLuboš Luňák
2011-01-17do not mix unrelated X11 Visuals (fdo#33108)Luboš Luňák
Do not assume that there is just one generic Visual, as today's XServers are ARGB-capable and cases of both the default depth and 32bit visuals can happen.