OpenEuphoria

RE: Confusing error message

• Posted by Al Getz <Xaxo at aol.com> Aug 16, 2003
• 425 views

Robert Craig wrote:
> 
> 
> Juergen Luethje wrote:
> > a) I suggest for Eu 2.5, that at least in an error message, concerning
> >    a type_check failure of a variable that is declared as integer, the
> >    current value of that variable is shown with maximum available
> >    precision.
> 
> OK, I'll think about that.
> It happened to me once too.
> (A f.p. value was an integer to 10 digits,
> but not beyond, so the type_check message was hard to understand.)
> 
> > b) Also
> >    - ?
> >    - pretty_print()
> >    - print()
> >    - printf(a, "%f", {x})
> >    - printf(a, "%g", {x})
> >    - sprintf(  "%f", {x})
> >    - sprintf(  "%g", {x})
> >    IMHO should use maximum available precision by default.
> 
> I don't think I'll change that.
> Did you know that several years ago it was 6 digits
> as the default? Now it's 10.
> 
> Besides disturbing existing code,
> people generally do not want to see 16 digits
> printed, especially when the last few are
> almost always meaningless garbage.
> 
> Regards,
>     Rob Craig
>     Rapid Deployment Software
>     http://www.RapidEuphoria.com
> 
> 
Hi Rob,

Were you also aware of the strange behavior when you
try to call a procedure with a return statement:
"return ProcedureName(params)"

The procedure name never gets called, and you dont get any
indication that something might be wrong until you reach
the end of the debug session, when you get a warning about
not being able to execute the next line after 'return'.

Take care for now,
Al