MOO-cows Mailing List Archive
[Prev][Next][Index][Thread]

Re: 1.8.0a5 binary string suggestions

James C Deikun writes:
> >    that isn't an ASCII printing character, the space character, or the TAB
> >    character will be represented as the four-character substring "~NNN",
> >    where NNN is the octal representation of the byte; the input character
> >    `~' is represented by the two-character substring "~~".  Call this
> 
> It might be nice to represent '~' with '~<octal code for ~>' rather than 
> '~~', since this would make things like a true length function simpler 
> (just subtract 3 times the number of tildes from length()).

Good point.  I'll make this change in alpha6.

> More importantly, now that we have binary i/o, there's no reason other 
> than compatibility to treat TAB as a printable character anymore, is 
> there?  I seem to recall that it was only done in the first place to 
> support gopher.  Since the gopher code uses normal i/o, there is then no 
> reason not to translate tabs when reading binary strings.

Compatibility is the only reason.  If I'm reading your note correctly, you're
just suggesting that the binary I/O facilities treat TAB as non-printing and
convert it on input into `~011', right?  The facilities will already accept
`~011' on output and convert it into TAB.  This seems like a good suggestion to
me; I'll put it into alpha6 as well.

Thanks!
	Pavel

PS- James was the next of several people to discover that you can't send mail
    to MOO-Cows-Announce from outside of PARC.  I'm sorry for the confusion
    this causes; I'm trying to remember now to add a `Reply-To: MOO-Cows' line
    to each of my announcements.
Home | Subject Index | Thread Index