Excellent!
Why was the Warning thing added to the IDE in the first place, may I ask?
I find it very distracting.
Mainly as a compromise amongst the folks who work the most on developing QB64. I brought up the idea that we really didn't need to toss errors when a CONST was set multiple times inside a program. (Such as having CONST TRUE = -1, FALSE = 0 inside multiple library files which used to toss an error for multiple occurances.) I figured, "Who cares if a CONST is declared over and over, as long as the VALUE remains the same? We only really need an error when the values don't match."
After all, why would this toss an error?
CONST True = -1
CONST True = -1
CONST True = -1
True starts out -1 when we first type it. It stays -1. So why the heck do we actually need to stop compilation of the program and claim it's an error? (Especially when "True" itself never appears anywhere in the C-code we generate, as we substitute the value -1 everywhere it appears in our code.)
I thought we should just ignore multiple instances, as long as the values match. Fellippe didn't want to break from the tradition of BASIC and wanted to keep the error message to make the programmer delete one of those values. In the end, we settled on a middle ground compromise -- toss a warning to let folks know that they have more than one CONST with the same name and value, but don't actually stop compilation of the program.
And, once the system was in place, it's usage has expanded to now toss warnings for various other things as well (such as for unused variables.)
But, like you, I just find the system to be a distraction. There's even edge cases where I get warnings for "variables not in use", when the are, in fact, in use, (I think its a case where you define a variable, but only reference it via _MEM and not directly, where I found this happening,) which either requires writing drivel code somewhere to shut up the warning system (foo = foo), or it means it's always going to be a warning message down in the bottom of the IDE to drive you batty...
So, a solution came to me: Place a toggle in the IDE and let folks freely ignore those warnings, or not, as whichever method suits their preference. (You can certainly bet that my personal version has it toggled on all the time now. ;D)
IF nobody finds or reports a glitch in how it makes their programs work, I'll push the changes into the development build for widespread testing and usage there, and then we'll see that option full time in the next stable release which comes out (whenever that happens).
