OpenEuphoria

Re: Initializing variables on declarations

• Posted by Pete Lomax <petelomax at blueyonder.co.uk> Apr 07, 2007
• 593 views

>  > Current syntax:
> > > [global] integer x  x = 5
> > > 
> > > Desired syntax:
> > > [global] integer x = 5
> 
FWIW, I'm for this. I think it could be quite fair to say that only short
literal integers are allowed, though the special case of {} would be welcome.

I am in agreement with most that object x=f(5,g[7]) etc is overkill.

I must admit that I wanted to stand back and watch how this thread panned out,
and am left with the desire to call for an old Pascal-style ruling that we can
(straying off topic a bit here) say eg:

procedure x()
constant a=1,b=2
integer x,y

but NOT

procedure x()
integer x
constant a=1
integer y
constant b=2

I must admit though I am in two minds. I see this as a good thing in many ways
and also as a possibly open running sore of disappointment and perhaps confusion
as to why certain expressions are not allowed. But a firm hand documentationwise
could fix that, surely?

Rob Craig wrote:
>why not privates (inside a routine) as well?
Personally, that is mostly where I want it most.

There is a minor syntactic sugar with more complex expressions at top level:

<type> var=<expr>

is exactly the same as:

<type> var var=<expr>

in disguise. It is quite natural to me now to key var<Ctrl W><Ctrl
V><space><Ctrl V>; however the former is plain easier to read.

As above and general consensus K.I.S.S. for for private routine vars tho.

Regards,
Pete