Bug#497525: libc6: fgetws treats C locale wrongly

[Jonathan David Amery <jdamery@ysolde.ucam.org>]

Package: libc6
Version: 2.3.6.ds1-13etch7
Severity: normal

 In the C locale all byte sequences are valid, however using the
following test programme:  

#include <stdio.h>
#include <wchar.h>
#include <errno.h>
#include <stdlib.h>
#include <locale.h>

int main() {
  size_t buflen = 512;
  wchar_t *p = malloc(buflen*sizeof(wchar_t));

  setlocale(LC_ALL, "");

  fgetws(p, buflen, stdin);

  if (ferror(stdin)) {
    printf("%s\n", strerror(errno));
    return(1);
  }
  else {
    printf("OK\n");
    return(0);
  }
}

we get:

: ; perl -e 'printf "%c", 0xa1' | env - LC_ALL=C ~/test
Invalid or incomplete multibyte or wide character

0xa1 is of course neither a multibyte or wide character in the C
locale, and fgetws should just read it in the appropriate manner to
allow propagation of it in the same form throughout the programme.

-- System Information:
Debian Release: 4.0
  APT prefers stable
  APT policy: (500, 'stable')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.18-6-686
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)

Versions of packages libc6 depends on:
ii  tzdata                      2007k-1etch1 Time Zone and Daylight Saving Time

libc6 recommends no packages.

-- no debconf information