Re: Ada Java question => clarification

[dewar@merv.cs.nyu.edu (Robert Dewar)]

Brad said

<<,The discrete choice is java.awt.event.UP. [note: In the source code one
can see that java.awt.event.UP is declared as a constant integer. The big
question is what's wrong with using a java.awt.event.UP in this context]
When I check out 3.8.1(8), it tells me just what the message said: that
the choice must be static. No indication here why a constant integer isn't
static. Let's move on... I check out 5.4(5) and am told the same thing>>

you miss my point. For *you* the problem with the message is that you
knew this was a case choice, you knew that the choice had to be static,
what you didn't know was why the expression you wrote was non-static.

Sure the message is incomplete, to be complete, it would have to have the
whole reference manual there, or at least a transitive closure of all
parts of it that could be relevant (right down to the syntax of identifiers)

Imagine another user saying, "well of course I know the function call
is non-static, anyone knows that, but this is the first I knew that
case choices had to be static, where does it say that -- and sure enough
the RM reference given is reasonable in this case".

Or imagine another user who says, well of course I know that case choices
had to be static, but I thought choices only meant the case where you used
the | character and had multiple choices in a single case ...


Get the idea? There is no such thing as a wrong reference here, you can
write many more scenarios like the above.

There is a most likely misunderstanding, and I would tend to agree with
you that your misunderstanding was the most likely, but you cannot prove
that the RM reference is inappropriate or wrong by starting with the
assumption that obviously your misunderstanding is the same as any one
elses in this particular situation.