Re: lynx-dev LYLowerCase EBCDIC Crash [PATCH]

[pg]

In a recent note, Thomas E. Dickey said:

> Date: Mon, 22 Oct 2001 09:36:27 -0400 (EDT)
> > >
> > > On Mon, 22 Oct 2001 address@hidden wrote:
> > >
> > > > The problem:  If the last character of a string has the 0x80 bit set,
> 
> then the last character of a string shouldn't have 0x80 rather than 0x00.
> 
Oops.  Clarification (I was not considering the terminating NUL part of
the string -- for example, strlen() doesn't count it):  If the string
character preceding the terminating NUL has the 0x80 bit set ...

The offending code, distilled, is:

    for (i = 0; buffer[i]; i++)
    {
        if (buffer[i] & 0x80) {
            { ... }
            i++;
        } else {
            buffer[i] = UCH(TOLOWER(buffer[i]));
        }
    }

The i++ may skip the NUL if the preceding character has 0x80 set.

-- gil
-- 
StorageTek
INFORMATION made POWERFUL

; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to address@hidden