On Mon, Nov 17, 2003 at 08:19:38PM +0000, Julian Foad wrote:
> Philip Martin wrote:
> >Julian Foad <julianfoad@btopenworld.com> writes:
> >
> >>C. Michael Pilato wrote:
> >>
> >>>IIRC, that information isn't readily available because we throw
> >>>whole chunk of data at iconv(), and it only sez "Something somewhere
> >>>in this big ol' chunk didn't pan out."
> >>
> >>It's a pity if it's not readily available, but we could quite easily
> >>find out (e.g. binary-chop search) if we think it is useful to do so.
> >
> >Easy? What about variable width encodings?
>
> Oh, alright, I give in. Perhaps the appropriate string manipulation
> functions are not available. I don't know, because I haven't looked into
> it. I still feel reasonably confident that I could imlpement an acceptable
> solution within a page of code if I cared, but I don't and so will not
> prove that I can.
>
Well, it could certainly be done in O(N) time: chop the last byte
off until the error goes away.
--ben
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Mon Nov 17 21:28:59 2003