Premake

Serge Semashko wrote:

> PS. I have tried to compile the latest version of premake from CVS with 
> the latest mingw (GCC 3.4.0), and it fails to compile it with the 
> following error message:
> 
> In file included from Src/premake.c:15:
> Src/util.h:14: error: syntax error before numeric constant
> make[1]: *** [obj/premake.o] Error 1
> make: *** [Premake] Error 2

Appears that mingw has WIN32 macro predefined and it conflicts with the
following enum:

enum { UNIX, WIN32, NATIVE };

WIN32 item should be renamed to something different to avoid these problems.

Serge Semashko wrote:
> Appears that mingw has WIN32 macro predefined and it conflicts with the
> following enum:
> enum { UNIX, WIN32, NATIVE };

Okay, I renamed WIN32 to WINDOWS.

Jason

Serge Semashko wrote:
> J. Perkins wrote:
>> Okay, I renamed WIN32 to WINDOWS.
> 
> Thanks, there are no compile errors in mingw anymore.
> 
> And what about that crash I have reported, does it reproduce well on 
> your computer or you need anything else to fix it?

I will try to get to that today. I was off the net most of the day 
yesterday and was not able to download the file. I have it now.

> PS. Looks like there is 'monodev.c' file missing from repository.

Doh! I just added it, sorry.

Jason

Serge Semashko wrote:
> I'm trying to experiment with creating premake.lua for LuaPlus.
> Looks like while doing this I have encountered a bug in premake.

Fixed in CVS. I will try to push a new release out shortly.

Jason

> Have not checked the Betas, but at 1.9 if there was an error in premake
> parsing the .lua file (not sure about other types of errors), the exit
> status is still 0; it should be 1 or other non-zero.

This has not changed. I will look into it.

Jason

Michael S. Muegel wrote:
> Have not checked the Betas, but at 1.9 if there was an error in premake
> parsing the .lua file (not sure about other types of errors), the exit
> status is still 0; it should be 1 or other non-zero.

I finally got a chance to look at this, sorry for the delay. However, I 
was not able to reproduce it, in every case I tried premake returns 1 
when the script is broken. If you can send me a case where it doesn't I 
will certainly get it fixed.

Thanks,
Jason

Thomas Harning Jr. wrote:
> Would it be possible to have the generated makefiles check out header 
> dependencies instead of completely relying upon changes in the source files 
> to detect when recompiles are needed?
> It seems there's alot of make-file code involved for a simple thing as 
> compiling a source file if only the source file is what matters.

I'm sorry, but I don't understand what you are asking.

Jason