> > This is what is known as 'prototyping'. What this means is you let a > > function or a file know that an external function exists. Alternately, > > you could use: > > > void hunt_victim(); > > I believe using the old style declaration hunt_victim() as a prototype in > spec_procs.c simply to escape warnings is not a good thing. If you screw > up the call of hunt_victim in spec_procs.c (or whereever it's called, not > the file in which it's declared) the compiler can not readily check that > you've passed the correct arguments to hunt_victim since it needs the > prototype to check (although in my experience, it depends on the compiler; > gcc seems to catch most of the screw-ups in cckr declarations). Yeah, it may provide a problem as far as passing the incorrect parameters goes. I haven't tested this insight, since I didn't actually know about it; I assumed the linker would test for things appropriately. Sounds like it's something worth looking into. Generally, I just make sure I pass the right parameters. Something to note in future expansions, in any case. > Levork, master of the run-on sentence.. *8) Took a couple reads to understand what the hell you were saying. >:)
This archive was generated by hypermail 2b30 : 12/18/00 PST