Narrowing down errors

[tlm70]

 More often than not I can figure out what caused an error in a character using clues from the error message (i.e. statedef, ***.cns, clsn, sprite, trigger, etc.) but in this instance I don't have any error message keywords for reference.

Library error message:
Data: 96,2,1,
Types: trig,int,oper,

Can anyone provide some insight as to how to use this error message to track down the problem?

[Cyanide]

Check the log, it might have something in it. Alternatively, is this something you were coding? What was the last thing you did before it showed up? Normally when this occurs and you're the one doing it you backtrack.

That error is indeed nondescript. Library normally refers to sprite problems but that definitely looks like code.

[tlm70]

 Actually it's a char that seems to be made for 1.0 and I thought I'd take a shot a getting it to run on DOS. Ever since the first Winmugen version came out years ago I learned (somewhat successfully) how to modify Win chars and hi-res stages to DOS. It actually turned out to be as interesting as playing/watching mugen. It's quite a satisfying feeling after hours of modifying the cns cmd and sff files of a char to see these wonderful creations with sometimes ingenious if not astonishing AI running in DOS mugen. Back in the day I was hoping to learn how to make chars that were Win and DOS compatible but I may have come to the end of the line with 1.0.
Anyway, I ended up doing what you said and backtracked to the changes that started those types of errors. So far, the only changes that I've made are the standard inguarddist modification:

P2BodyDist X < 30

It lacks the flexibility of the Win trigger but it works rather well. After that mod I started to get that type of error message again. It's a bit different this time:

Library error message:
Data: 50,343,
Types: int,trig,

I think the 50, int might be referring to lines similar to this in a cns file:

var(50)=enemynear(0),ID

var(50)=ifelse(EnemyNear(0),alive,enemynear(0),ID,enemynear(1),ID)

trigger1 = fvar(20) != playerID(var(50)),const(movement.yaccel)

I've successfully modified a char with lines similarto this before by changing
playerID(var(50)),
to
enemynear,

but in all honesty I don't always understand why particular modified lines work.

Any insight into this would be welcomed and I certainly hope no one is offended by my ongoing attempts to try mods to DOS.

[Cyanide]

dos is amazingly limited in terms of what we have now. The problem for you is it doesn't even recognise what it's failing on.

You're going to have to pull up the DOS docs to see if some of the things you're using are actually valid. I can't remember much about DOS limitations these days. The code you have looks valid. IIRC dos had all those functions.

You may find rather than trying to make them work, your first resort should be to comment things out with ;. So take a punt, comment out lines (or whole blocks) and test, if it fails, re-enable and do to the next block.

50 or 343 could potentially be the line of code.