function foo redefined in...

[lynch@ils.nwu.edu]

Another annoying, stupid, little thing I've just been reminded of:
MCL 2.0 seems to give warnings about functions being redefined in
files...when it's the same file.

I have seen things like:

 original:  "HD:Applications:MCL:Lynch Lib:foo"
redefined:  "HD:Applications:MCL:Lynch Lib:foo"

and,

 original:  "HD:Applications:MCL:Lynch Lib:foo"
redefined:  #4P"ccl;Lynch Lib:foo"

and

 original:  Unknown
redefined:  "HD:Applications:MCL:Lynch Lib:foo"


In all cases, we are talking about evaluating from a FRED window using the
enter key during development.

I can't even begin to guess how this first one could occur, but I seem to
remember [WARNING:  Lynch memory.  HAH!] that it happened from a saved
application.

The middle one is understandable:  somebody made a quick (but pretty
bone-headed IMHO) test to decide if the original and current files match.

The last is more or less understandable, but I'm not sure why it would have
forgotten the original.  I'm fairly certain that it usually remembers.  I
know I don't alter this variable in my code.  Code that I'm using but
didn't write *MAY* have altered it, but I sincerely doubt it.

So, what variables affect the rememberance of symbol [function, macro,
type, class, structure] definitions?

Semi-related thoughts:

I heard somewhere that MCL 2.0 final would contain source code to *SOME* of
the CLOS objects (YEA!) such as windows and dialog-items.  Is this wishful
memory or a "true rumor"?

Would it be possible/useful for APDA to distribute two images of MCL?  One
that had all these globals turned on when it was created (for development)
and one that had them all off (for delivery).
"TANSTAAFL" Rich lynch@ils.nwu.edu