Re: [Bug-tar] invalid assumption in src/incremen.c

[Eric Blake]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

According to Sergey Poznyakoff on 8/16/2005 2:46 AM:
>>src/incremen.c makes the invalid assumption that ino_t is no bigger than
>>long.  But since the 1.5.x release of cygwin (keyed by
>>__CYGWIN_USE_BIG_TYPES__ in headers), long is 4 bytes, while ino_t is 8
>>bytes.  This bug appears to be the cause of the breakage of test 20 of
>>tar-1.15.1 on cygwin, since --listed output was being truncated.
> 
> 
> Thanks for reporting. Please try attached patch.

The patch indeed fixes the listed02.at test on cygwin.

> +static uintmax_t 
> +getnum (char *p, char **endp)
> +{
> +  uintmax_t n = 0;
> +
> +  while (*p && isspace (*p))
> +    p++;
> +  
> +  for (;ISDIGIT (*p); p++)
> +    {
> +      uintmax_t n10 = n * 10;
> +      uintmax_t c = n10 + *p - '0';
> +      if (n10 / 10 != n      || c < n10)
> +     {
> +       errno = ERANGE;
> +       n = -1;
> +       break;
> +     }
> +      n = c;
> +    }
> +  *endp = p;
> +  return n;
> +}

Wouldn't POSIX strtoumax (already a gnulib module) be better than rolling
your own?

> +      
>  static FILE *listed_incremental_stream;
>  
>  void
> @@ -344,8 +368,9 @@ read_directory_file (void)
>        char *ebuf;
>        int n;
>        long lineno = 1;
> -      unsigned long u = (errno = 0, strtoul (buf, &ebuf, 10));
> +      uintmax_t u = (errno = 0, strtoul (buf, &ebuf, 10));

strtoul may be to short for uintmax_t, did you mean to use getnum or
strtoumax here?

- --
Life is short - so eat dessert first!

Eric Blake             address@hidden
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (Cygwin)
Comment: Public key at home.comcast.net/~ericblake/eblake.gpg
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFDAy6i84KuGfSFAYARAhQxAJ9ITWVP8rMXBgiAyNoALuNvwTDE5wCfWj6R
1uV0ff4WgwRLqCt1c73kGxw=
=onNb
-----END PGP SIGNATURE-----