summaryrefslogtreecommitdiff
path: root/src/os
AgeCommit message (Collapse)Author
2009-10-04(svn r17711) -Codechange: [OSX] Don't use deprecated methods to construct a ↵michi_cc
Cocoa string.
2009-10-04(svn r17708) -Feature [FS#2053]: [OSX] Implement clipboard support for OS X.michi_cc
2009-10-04(svn r17706) -Codechange: [OSX] Rework the crash handling to use the common ↵michi_cc
CrashLog infrastructure.
2009-10-04(svn r17705) -Fix: [OSX] Re-enable signal handling on OSX 10.3.9. Trying to ↵michi_cc
link with an undefined symbols that lives in the system library seems to confuse the loader on 10.3.9. Use a different function to circumvent it.
2009-10-04(svn r17704) -Codechange: [OSX] Improve detection of OS X version. (planetmaker)michi_cc
2009-10-04(svn r17702) -Change: [OSX] Assure that the minimal OSX version is defined ↵michi_cc
in all cases. -Codechange: [OSX] Improve conditional defines for OS version dependant code.
2009-10-04(svn r17693) -Cleanup: remove some unneeded includesrubidium
2009-09-09(svn r17487) -Fix: silence some MSVC warningsrubidium
2009-09-08(svn r17479) -Fix: silence compile warning that's only triggered when you're ↵rubidium
using a broken iconv.
2009-09-08(svn r17475) -Codechange: show path to crash.log and crash.dmp in crash ↵glx
window. Also tells the user where crash.sav is when saving succeeded.
2009-09-07(svn r17463) -Change (r17453): Retrigger the abort signal after writing the ↵frosch
crashlog, so the next one can catch it again (e.g. to write a core dump).
2009-09-07(svn r17453) -Add: crash logger for all Unixy OSes; should work for all, but ↵rubidium
I don't have all exotic machines so it might break compilation on the more exotic configurations
2009-09-07(svn r17452) -Fix (r17450): emergency savegames weren't rejected by the ↵rubidium
crash logger anymore
2009-09-07(svn r17450) -Codechange: refactor the windows crash logger so parts of it ↵rubidium
can be reused by other crash logger implementations
2009-09-04(svn r17416) -Codechange: move the build date/time 'determination' to ↵rubidium
rev.cpp as that's the file most likely to be regularly recompiled
2009-09-03(svn r17410) -Codechange: use the same define for determining whether ↵rubidium
windows does crash reports instead of using several that aren't necessarily equal
2009-09-03(svn r17409) -Codechange: split the crash log and other windows 'glue' coderubidium
2009-09-03(svn r17408) -Codechange: remove unused win32 stuffglx
2009-08-31(svn r17336) -Codechange: move some os specific files into src/os/rubidium
2009-08-21(svn r17248) -Fix: add GPL license notice where appropriaterubidium
2009-07-14(svn r16828) -Codechange: attempt at making OSX compile with the 10.6 SDK ↵rubidium
too; no guarantees ofcourse...
2009-07-14(svn r16827) -Codechange: make OSX specific files include stdafx.h always as ↵rubidium
the first file. Also unify OSX specific stuff into osx_stdafx.h and remove unused includes.
2009-04-27(svn r16177) -Fix (r15892): [OSX] fixed linking error regarding _select$1050 ↵bjarni
(only present on some systems)
2009-04-01(svn r15909) -Fix (r15892): Apple, PLEASE do not fill your API with #if ↵rubidium
!__LP64__. Especially when another part of your API depends on whether __LP64__ is defined and thus making it non-trivial to silence the warnings from the undefinedness of __LP64__ in #if !__LP64__...
2009-03-29(svn r15892) -Codechange: [OSX] some type fixes so OpenTTD likes more of the ↵rubidium
3.1.x Xcodes.
2009-03-15(svn r15718) -Cleanup: apply some comment coding style on the rest of the ↵rubidium
sources too
2009-03-08(svn r15648) -Fix (r14773): hack around an OSX stupidity in < 10.4 w.r.t. ↵rubidium
signals by not having any signal handling support for OSX < 10.4. Thanks to PearPC and TrueBrain's OSX compiler.
2009-01-10(svn r14949) -Cleanup: pointer coding stylerubidium
2008-08-31(svn r14200) -Fix (r14199): some OSes failed to compile. Furthermore I hate ↵rubidium
wrong comments!
2008-08-01(svn r13913) -Fix (r13893): [OSX] fixed a warning about ↵bjarni
MAC_OS_X_VERSION_10_5 not being defined on all systems
2008-07-31(svn r13893) -Fix: [OSX] solved a deprecated warning specific to 10.5bjarni
2008-06-18(svn r13571) -Codechange: define channels in struct Colour in different ↵smatz
order on LE and BE machines
2008-05-06(svn r12971) -Documentation: add @file in files that missed them and add ↵rubidium
something more than whitespace as description of files that don't have a description.
2008-05-04(svn r12944) -Codechange: use rev.h instead of externs at many placessmatz
2007-12-25(svn r11692) -Codechange: move some functions from 'functions.h' to a more ↵rubidium
logical place and remove about 50% of the includes of 'functions.h'
2007-12-23(svn r11688) -Fix [FS#1581]: Compilation was broken on OS X again.egladil
2007-12-23(svn r11684) -Codechange: split gfx.h in a type and functional header.rubidium
2007-12-21(svn r11678) -Fix (r11674): [OSX] fixed compilationbjarni
2007-12-21(svn r11676) -Fix (r11674): some references to macros.h were leftglx
2007-11-30(svn r11542) -Fix r11540: [OSX] presumed 10.3 style strings when using ↵bjarni
GetCurrentLocale nomatter what OSX the target was
2007-11-30(svn r11540) -Fix: [OSX] Don't try to compile the quartz video driver on OSX ↵bjarni
10.3. It will fail
2007-11-25(svn r11521) -Codechange: [OSX] Check what the running os version is in a ↵egladil
cleaner way.
2007-11-07(svn r11389) -Fix [FS#1386]: Replace calls to deprecated API with newer ones ↵egladil
and handle broken iconv declaration in OSX 10.5.
2007-10-27(svn r11351) -Fix (r11342): [OSX] killed a warningbjarni
2007-10-25(svn r11342) -Fix: [OSX] The cocoa driver incorrectly assumed that the ↵egladil
blitter always was 8bpp. Now both 8bpp and 32bpp blitters can be used. The driver will check the blitter screen depth. In fullscreen it will select a proper video mode for this depth, and in windowed mode it will simply skip doing fake 8bpp.
2007-08-11(svn r10851) -Feature [OSX]: OpenTTD will now pick the same language as ↵bjarni
finder is set to if no config file is found (ln-)
2007-08-03(svn r10761) -Fix (FS#1101, r10216): _pal_last_dirty changed to ↵glx
_pal_count_dirty without updating, so it was of by one
2007-06-19(svn r10216) -Fix: palette animation always redid all palette entries, where ↵truelight
in fact only a few indexes were needed -Codechange: allow blitters to handle palette animation internally or even disable it; 8bpp uses video-backend for palette animation
2007-03-07(svn r9050) -Codechange: Foo(void) -> Foo()rubidium
2007-02-05(svn r8605) -Codechange: [OSX] changed all objective C to objective C++bjarni
This will permanently solve the issue where compilation on OSX broke because C++ code was added to some header files -Note: (important if you develop mac specific code) taken from http://developer.apple.com/releasenotes/Cocoa/Objective-C++.html gdb lacks an integrated C++ with Objective-C parser. This means that gdb won't be able to evaluate expressions that contain both C++ and Objective-C constructs. gdb assumes that the language for ".mm" files is C++. you can change it to objective C by typing: (gdb) set language objc Mixing C++ and objective C has some limitation (see link for all of them)