On 8/14/2014 2:21 PM, Ray Donnelly wrote: > On Thu, Aug 14, 2014 at 6:45 PM, Wayne Stambaugh <[email protected]> > wrote: >> On 8/14/2014 1:13 PM, Alexey Pavlov wrote: >>> >>> 14 авг. 2014 г., в 21:04, Richard Shaw <[email protected]> написал(а): >>> >>>> I found the issue... No idea what it's not a problem on other platforms. >>>> >>>> >>>> diff -Naur cmake-3.0.1.orig/Modules/FindwxWidgets.cmake >>>> cmake-3.0.1.wxfix/Modules/FindwxWidgets.cmake >>>> --- cmake-3.0.1.orig/Modules/FindwxWidgets.cmake 2014-07-31 >>>> 10:03:56.000000000 -0500 >>>> +++ cmake-3.0.1.wxfix/Modules/FindwxWidgets.cmake 2014-08-14 >>>> 12:02:11.604331400 -0500 >>>> @@ -792,6 +792,10 @@ >>>> wxWidgets_CXX_FLAGS "${wxWidgets_CXX_FLAGS}") >>>> string(REPLACE "-I" "" >>>> wxWidgets_INCLUDE_DIRS "${wxWidgets_INCLUDE_DIRS}") >>>> + >>>> + # Flags are a string, not a list, fix it here >>>> + string(REPLACE ";" " " >>>> + wxWidgets_CXX_FLAGS "${wxWidgets_CXX_FLAGS}") >>>> >>>> DBG_MSG_V("wxWidgets_DEFINITIONS=${wxWidgets_DEFINITIONS}") >>>> DBG_MSG_V("wxWidgets_INCLUDE_DIRS=${wxWidgets_INCLUDE_DIRS}") >>>> >>>> Thanks, >>>> Richard >>> >>> Thanks, will rebuild cmake with this patch soon. >>> >>> Regards, >>> Alexey. >> >> If this is part of the CMake source, shouldn't this patch be pushed >> upstream so that it can be fixed on all platforms assuming the CMake >> developers will accept it? At least they can be made aware of the >> issue. I know it's convenient to fix it here and in the short term >> that's fine. These kinds of patches have a way of spiraling out of >> control. If you don't believe me, look at the mess required to build >> KiCad (I'm one of the lead developers) with all of the dependency >> library patching we have to do. It can quickly become a maintenance >> nightmare. > > While I agree wholeheartedly with the sentiment, this patch is a drop > in the ocean: > > $ cd MINGW-packages > $ find . -maxdepth 2 -name "*.patch" | wc -l > 1222 > > .. upstreaming patches is such a massive undertaking and there's only > a small handful of people who work on MSYS2 consistently; seems we > chose to spend our time adding and fixing packages rather than dealing > with sometimes ungrateful, sometimes actually hostile upstream > developers. The best results happen when people already attached to a > project who are trusted and have commit rights discover MSYS2.
I know all about upstream patch issues and share your frustration. That's why the KiCad project has to use a patched version of Boost. It's a terrible waste of developer time. > > If you feel that you could help reduce this number then by all means > please try upstreaming some of them and CC this mailing list when you > do. I barely have the time to get our patches upstreamed so it's unlikely I will have any time to help out. I may take a look at some of the patches we maintain and see if there is any duplication. Thanks to everyone who contributes to MSYS2. I really like the work that is being done. > >> >> Thanks, >> >> Wayne >> >> >> >> ------------------------------------------------------------------------------ >> _______________________________________________ >> Msys2-users mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/msys2-users > ------------------------------------------------------------------------------ _______________________________________________ Msys2-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/msys2-users
