--- In [email protected], "entropyreduction" <[EMAIL PROTECTED]> wrote:
>
> --- In [email protected], "swzoh" <seanzoh@> wrote:
> >
> > --- In [email protected], "swzoh" <seanzoh@> wrote:
> > > The length of the byteblock "xString" always return 263. I became
> > > curious why it's not 2*2 or 266:
> > > 
> > > 266 = 272 - 1 - 4 - 1
> > > 
> > > where 1-byte header, 4-byte size as int, 1-byte eos.
> > > 
> > > My conjecture is that the header is not 1-byte, but actually 4-byte.
> > > Some kind of Struct Packing arose here, I suppose:
> 
> Yup.  Sean has chastised me for using a CHAR in a struct.  
> 
> I'll revise at some point.
> 
> But see below.
> 
> binary.length, BTW, probably should be called binary.capacity:
> it's always the available space in a byteblock.
> 
> binary.length (the actual number of real bytes in use in a byteblock)
> is neither necessary or feasible, IMO.
> 
>  -- all the API calls I've run into so far that fill a BYTE array 
>     also return a byte-count, or require a byte-count as an argument:
>     so the user ASAIK always has a way of knowing how many bytes 
>     have been returned.
> 
>  -- Furthermore, how can a byteblock know how many of it's bytes are 
>     "real"?  There's no magic terminating null.  A byteblock doesn't 
>     know whether it's been passed through an API call that's filled 
>     it, so it has no idea how manh of it's bytes matter.
>  
>     The only a byteblock can know how many of its bytes are real is
>     if the byteblock's tells it: e.g. some service like
>     binary.set_length.  But doesn't seem much point.
> 
> I'll rename binary.length to binary.capacity at some point.
> 
> Also: I wonder if all this messing about with byteblocks read and
> written straight to variables is a good idea.  
> 
> Should binary instead return and require handles, like unicode?  It
> would mean you wouldn't have remember to use the name of a variable,
> instead of a variable, in calls to binary.  And you could use the
> handle.service syntax.
> 
> Reactions?
>

Alan, swzoh is Sean, AFAIK. :)

I already expressed my opinion on this in the mail, which mostly
accorded with the view here (:so, binary.from_ascii_string() and
binary.from_hex_string() are very exceptional cases).

And, treating byteblock as a handle than a mere variable could be a
good thing, I wonder.

Sean






Attention: PowerPro's Web site has moved: http://www.ppro.org 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/power-pro/

<*> To unsubscribe from this group, send an email to:
    [EMAIL PROTECTED]

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 



Reply via email to