1. wi32lib: RFC for w32string()
Defined in w32utils.e, w32string() is an user defined type which validates
sequence of integers in the 0 to 255 range.
So, for instance, w32string({5,0,255,7}) is 1, but try naming anything this way,
or printing it. Good luck.
This is why I suggest changing its meaning slightly to something probably more
useful:
1/ create a new String_CT w32CType() class, initially identical to Printable_CT.
2/ change w32string() to validate a sequence of characters in the String_CT
class.
This would reject the garbage above, and enable anyone to add accented or double
byte (UTF-16) characters to the range of what w32strung() accepts.
A simpler variant would be to directly tie w32string() to Printable_CT. I don't
know about all uses of this char class by Eu code, so uncoupling the two issues
is 100% safe, at the cost of a new char class.
Any thoughts?
CChris
2. Re: wi32lib: RFC for w32string()
CChris wrote:
>
>
> Defined in w32utils.e, w32string() is an user defined type which validates
> sequence
> of integers in the 0 to 255 range.
<snip>
> Any thoughts?
Yes, w32string is an eu sequence, calling it a string is a misnomer.
Kat
3. Re: wi32lib: RFC for w32string()
- Posted by CChris <christian.cuvier at agr?culture?gouv.fr>
May 16, 2008
-
Last edited May 17, 2008
Kat wrote:
>
> CChris wrote:
> >
> >
> > Defined in w32utils.e, w32string() is an user defined type which validates
> > sequence
> > of integers in the 0 to 255 range.
>
> <snip>
>
> > Any thoughts?
>
> Yes, w32string is an eu sequence, calling it a string is a misnomer.
>
> Kat
I'd only half agree here. It's a plain Eu sequence, representing (usually) a
printable string.
Oh, and after all I didn't coin that name ;)
CChris
