Fixes the truth value for the MacOSX convenience flag

[dkearns]

This a recurring problem throughout the code base that just bit me so I'm testing the waters. Would a PR fixing the boolean handling project-wide be merged?

I have no idea why Mac's get special treatment here or why os$ isn't set to "MAC"? I can't see where adding that would cause a problem.

[flukiluke]

Frankly the whole thing seems janky. Lines like IF os$ = "LNX" THEN extension$ = "" seem to imply os$ = "LNX" holds true on Mac systems?

Doing a brief search through to look at how the values are used, I get the impression os$ = "LNX" is really used to check for 'Unix-like' system, mainly dealing with things like / vs \ in paths. For no good reason, there's also a few calls that ignore os$ entirely and do INSTR(_OS$, ....

Finally there's this block, down around line 12027:

o$ = LCASE$(os$)
win = 0: IF os$ = "WIN" THEN win = 1
lnx = 0: IF os$ = "LNX" THEN lnx = 1
mac = 0: IF MacOSX THEN mac = 1: o$ = "osx"

but their use appears to be limited solely to deciding build dependencies.

I suspect that this is what you get when you keep adding platform as layers of special cases - if I had to put money on it I'd say Linux was done first, then OS X was deemed "close enough" to just use the linux conditional in most places; we now have a lovely mess.

I don't really have strong opinions about how it should be organised, except not how it is currently. I'd be willing to accept a change to the conditional value (assuming it doesn't break anything, of course), but know there's a bigger monster lurking beneath the surface.