David (and other larceny-users)- Hope you don't mind that I put the users list back in the CC field to include it in the discussion. . .
The one Windows FFI gotcha that comes to my mind is that there are two different calling conventions for Windows libraries; the C calling convention (which is chosen via the 'cdecl flag in the FFI) and the Pascal calling convention (which is chosen via the 'stdcall flag in the FFI). This is not documented well (if at all), in part because it is a very recent addition to the FFI. You can look at lib/Standard/file- system.sch for example invocations of the foreign-procedure function that use the 'stdcall flag to hook into Windows API functions that were declared with stdcall in the C code. -Felix On Jul 21, 2007, at 4:38 PM, David Rush wrote: > oops...and are there any Windows-specific FFI gotchas to keep in mind? > > On 7/21/07, David Rush <[EMAIL PROTECTED]> wrote: >> On 7/21/07, Felix Klock <[EMAIL PROTECTED]> wrote: >> > On Jul 21, 2007, at 12:53 PM, David Rush wrote: >> > >> > > But is there support in Larceny on Windows for TCP/socket >> > Not yet, ... >> > I believe the main issue is lack of familiarity with the Windows >> TCP/ >> > socket API; the fundamental support at the runtime and FFI level >> > should all be in place. So if you already know the Windows socket >> > API, you might be able to hack in the support yourself. >> >> Well that is better than I expected. I *don't* know all about the >> Winsock API, but if the FFI is stable on Windows I'm motivated just >> now and I have a need, so I may give it a go. I presume that you want >> to keep it the same as the API that is in lib/Standard/socket.sch? >> >> david rush >> -- >> Once you label me, you negate me >> - Soren Kierkegaard >> > > > -- > Once you label me, you negate me > - Soren Kierkegaard _______________________________________________ Larceny-users mailing list [email protected] https://lists.ccs.neu.edu/bin/listinfo/larceny-users
