I think it's probably better to move this file into dlls/mpr rather than delete it altogether. Apparently it did exist in the Windows 98 DDK.
For what reason, for documentation? Several of its declarations conflict with those in npapi.h, so it's potentially problematic. If it's not part of the PSDK, winelib users shouldn't need it to compile, or should expect to have to install the DDK if they need the prototypes, yes?
If you think it's a better approach I can rename all the conflicting types, and document why the typenames don't match the expected ones. Even so, I'm not convinced a separate header file is preferable to moving the declarations to nps.c (the only user of these declarations.)
--Juan
__________________________________ Do you Yahoo!? Yahoo! Mail - You care about security. So do we. http://promotions.yahoo.com/new_mail
Juan Lang wrote:
For what reason, for documentation? Several of its declarations conflict with those in npapi.h, so it's potentially problematic. If it's not part of the PSDK, winelib users shouldn't need it to compile, or should expect to have to install the DDK if they need the prototypes, yes?
To me, Wine feels like documentation of how Windows works, not just a rehash of the platform SDK. Removing documentation seems to be a step in the wrong direction.
If somebody wanted to use netspi.h again (ie. for non-Wine purposes) they could just pull it from dlls/mpr, rather than extract the definitions from a C source file.
If you think it's a better approach I can rename all the conflicting types, and document why the typenames don't match the expected ones.
I'm not exactly sure why we need to rename the types... if we consider it a private include file, then problems only happen if we include another file with the same definitions, don't they?
Mike
-
Juan Lang -
Mike McCormack