1. Re: Another Suggestion (Was: My Bits are Flipping (or, "Why Can't I Be= Negative?")
- Posted by Travis Beaty <eucoder at travisbeaty.us>
Aug 27, 2004
-
Last edited Aug 28, 2004
Hello WingZero.
Yes, that would be very nice. I had expected Euphoria to recognize the fact
that the negative bit was flagged, and the number was therefore negative;
however, in retrospect, I can see why it doesn't. There needs to be some
specific way to allow Euphoria to get negative values from a C shared library
-- otherwise, there are going to be so many hacks and workarounds it'll
boggle the mind (not to mention bloat the code).
This is almost a necessity to do what I'm trying to do.
Crap.
Travis.
On Friday 27 August 2004 02:38 pm, Elliott Sales de Andrade wrote:
> Unfotunately, parameters are only interpreted one way. Wouldn't it be=
> =20
> nice if you could do something like this:=20
> call_back({routine_id("MouseCallback"), {C_INT, C_DOUBLE, C_DOUBLE},=20
> C_DOUBLE}) --{mouse button, x-position, y-position} return: I don't know,=
> =20
> but something that needed to be a double. We wouldn't need to use any
> weird= =20
> workarounds for signed numbers, or strange stuff for doubles/floats.=20
> Obviously, this would be quite complicated to implement. As it stands, the=
> =20
> interpreter just needs to copy a set of routines depending on how many=20
> parameters there are for the callback. It would probably need some nice=20
> assembler or C trickery to get this advanced method to work. With this=20
> somewhat more complex method, though, we could create call-back
> procedures,= =20
> too (something that's moderately annoying.)
--
It's no use crying over spilt milk -- it only makes it salty for the cat.
