OpenEuphoria

Re: Rob: Suggestions for Euphoria 2.6

• Posted by Don <eunexus at yahoo.com> Nov 22, 2004
• 646 views

Vincent wrote:
> 
> Possible improvements and new features for Euphoria 2.6
> and other products
> -------------------------------------------------------
> 
> * Assignment on declare ability (exp: integer foo = 345)
> * 32 Bit integer data type (same as a LONG in C) (not 31 bits)
> * Boolean data type (exp: boolean foo)
> * Improvement(s) with namespace feature
> * Case & switch feature /w break statement <(*)>
> * Increase nested include limit to 50 (from 30) <(**)>
> * Significant speed improvement in Euphoria front-end
> * Minor speed improvements in C back-end
> * Minor speed improvements on certain standard library ruitines
> * Minor speed/code generation improvements on translators and binder
> * Lots of bug fixes for everything (of course)
> * A greatly improved Windows installer (or no installer at all,
>   just maybe a zip file with directions on setting up envi. var.)

Just pointing a couple of things out...
> * 32 Bit integer data type (same as a LONG in C) (not 31 bits)

This is (mostly) already done as an atom type.  From the docs:

Performance Note: 
Does this mean that all atoms are stored in memory as 8-byte
floating-point numbers? No. The Euphoria interpreter usually stores
integer-valued atoms as machine integers (4 bytes) to save space and
improve execution speed. When fractional results occur or numbers get
too big, conversion to floating-point happens automatically.

Since the conversion happens automatically should you need it,
Just use 32 bit integer values and they should only take four bytes
(which by definition is what your asking for).

> * Boolean data type (exp: boolean foo)
You can already do this by declaring your own type...


Don Phillips - aka Graebel
     National Instruments
     mailto: eunexus @ yahoo.com