OpenEuphoria

Re: Euphoria 2.5

• Posted by "Igor Kachan" <kinz at peterlink.ru> Aug 01, 2004
• 579 views

Hi, Ricardo.

You wrote:

----------
> From: Ricardo Forno <rforno at uyuyuy.com>
> To: EUforum at topica.com
> Subject: RE: Euphoria 2.5
> Sent: 31 jul 2004 y. 5:01
> 
> Hi, Igor.
> What are the non-documented side efects 
>  you are speaking of?
> Regards.

[snip]

> > And I am afraid that some time Rob will eliminate
> > the existing non-documented side effect as it already
> > was once with the translator  in "without type_check"
>  >  mode.

On Euphoria manual, [type ... end type] functions serve just 
for testing if the single parameter is of needed value.
But really you can use these functions for any an additional 
job as procedures on each assignment to typed variable too.
Maybe, this feature is not a 'side effect' in usual sense,
but it is not used in the official RDS programs, as far as I know,
is not described in the official RDS documentation and it seems
to be discovered by one of this list members. 
I myself know about this feature from the list and use it in
some my programs. This side effect is very useful, I think.

Really, why not use this thing if it does work well?

integer A
type sequense_with_length(sequence a)
  A=length(a)
    return 1
end type

There is no any user-defined tests of sequence a,
but this is the Euphoria [type...end type] construct,
which will prepare the new length(N) for you on any
new assignment to sequense_with_length N.

Is it a side effect or is it a very useful EU feature -
this is an issue of terms, I think.

Sure, It Is Not a Bug!       

Regards,
Igor Kachan
kinz at peterlink.ru