Updated note about why we #include the miniz.c file.

This commit is contained in:
Daviw Williams 2013-03-01 16:33:31 +01:00
parent a7e49a1394
commit d12db9906f

View File

@ -13,9 +13,13 @@
#include "PolyVoxCore/Impl/ErrorHandling.h" #include "PolyVoxCore/Impl/ErrorHandling.h"
// For some unknown reason the miniz library is supplied only as a // The miniz library is supplied only as a single .c file without a header. The examples just include the .c file
// single .c file without a header. Apparently the only way to use // directly which is also what we do here. Actually is is possible to define 'MINIZ_HEADER_FILE_ONLY' to treat
// it is then to #include it directly which is what the examples do. // the .c file as a header, but this seems messy in terms of our project and CMake as we keep the headers and source
// files in seperate folders. We could create our own header for miniz (based on the stuff between the MINIZ_HEADER_FILE_ONLY
// directives) but the other problem is that we are using #pragma GCC system_header to supress warnings which would
// then be in the .c part of the code. If we ever update GCC on the CDash machine so that it properly supports '#pragma
// GCC diagnosic ignored' (or so that it doesn't warn in the first place) then we can reconsider spliting miniz.c in two.
#include "PolyVoxCore/Impl/miniz.c" #include "PolyVoxCore/Impl/miniz.c"