Hi François,
I tried to use winemaker on an app, and I had the following notes about it.
1) Some subdirs were not visited by winemaker, while some others were. Should it visit all of them recursively? Does it stops if there are no source file in a dir (but there are subdirs)? The last point looks like the culprit.
2) Would it be possible to make winemaker understand the i64 postfix and do something about it? I believe the most portable way would be to change from "1234567890i64" to "(((__int64) 12) * ((__int64) 100000000) + ((__int64) 34567890))", as just replacing i64 by LL would be too easy.
3) winemaker changed the name of some .txt files from "a name.txt" to "a_name.txt". Since those are not source code, and their endlines were not unixified, I'm not sure it's such a good idea to touch those files.
4) Would there be a way to transform __try {}/__finally{} blocks to __TRY {}/__FINALLY()? I'm not sure the changes I did are now compilable under MSVC nor MingW.
Sidenote about exception.h: you need to use glibc's setjmp.h to get sigsetjmp(), but the default use of winemaker is to use wine's msvcrt's setjmp.h, which doesn't provide it. I sidestepped this by changing #include <setjmp.h> to #include "setjmp.h" and putting a symlink to /usr/include/setjmp.h in the current dir. It works, but it's far from being perfect.
Other than that, the app compiled without much problems (but a lot of warnings).
Vincent