Re: Bug#624387: [bug #33198] Incorrect bracket expression when parsing in ru_RU.KOI8-R (Russian locale)

[Jim Meyering]

Paolo Bonzini wrote:

> On Sat, Jun 4, 2011 at 09:48, Jim Meyering <address@hidden> wrote:
>>> The b2 == EOF part is required for the somewhat similar bug I fixed
>>> a month ago:
>>>
>>>     fix a bug whereby echo c|grep '[c]' would fail for any c in 0x80..0xff
>>>     8da41c930e03a8635cbd8c89e3e591374c232c89
>>>
>>> The corresponding test demonstrates the need:
>>>
>>>     tests: exercise bug with 0x80..0xff in [...]
>>>     d98338ebf842ec9b69631837eee50ebdcd543505
>
> [\xff] is not well defined for a UTF-8 locale at all, actually.
> Perhaps FETCH_WC should return wc = EOF in this case (and c = 255),
> and it could be handled on a case-by-case basis elsewhere.
>
> But if wctob returns EOF, and b > UCHAR_MAX, you have introduced an
> out-of-bounds access in setbit.

You must not have seen the guards I added in the comments just
preceding yours, e.g.,

    http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=624387#129