 Home |
 Content |
| Purchase |
| Ready-to-use samples |
| Online resources |
| Forum |
| Links |
| Happy readers |
Anonymous
Size of HANDLE on OS/2 (by Normunds Kalnberzins, 07/12/2005 02:05:49 PM) |  |
mhmp; dunno, tha values in the table came from spelunking header files. On the OSes I've written ls2capi code for (Win32, Linux/intel, Solaris/sparc, AIX, MacOS, OS X, OS/400), I have validated my assumptions. For the rest -- someone must do it:-)
Now I'm not really sure that this test means what you say it means. In W32 handle is 32-bits, but I think it uses only lowest 16-bits.
The only sure way to test it is to write a little bit of real C code, though I understand that to compile it on each an every platform is royal pain (compilers, settings, etc) and that's one of many reasons why we try to use ls2capi. You may try to cheat the compiler and instead of declaring W32 declare OS2 and see what happens -- you do not need to run a lot of code, just print out sizeof(HANDLE) "assuming OS2". You may notice I'm lazy to try this myself -- you probably are in process of investigating things right now and will have to run similar tests anyway (so keep VC++ handy :-). In fact when in doubt, when everything else fails, writting some C code to see what happens, what are the real structures, to check the assumptions, etc. is a good way to get around "un-debugability" of ls2capi calls.