> printf("Disk size:\t\t%*s\n", width,
> df_pretty_sizes(total_disk, mode));
>
> it would be translated (note the '%*s'):
>
> if (mode == DF_HUMAN_UNIT)
> printf("Disk size:\t\t%*s%s\n", width-2,
> df_pretty_sizes_number(total_disk),
> df_pretty_sizes_unit(total_disk));
> else
> printf("Disk size:\t\t%*lld\n", width, total_disk);
So use mode as the argument in the second case, just like you did in the
first.
printf("Disk dize:\t\t"sz_fmt"\n", sz_arg(total_disk, mode, width));
#define sz_fmt "%*llu%s"
#define sz_fmt(v,m,w) width(w, m), scaled(v, m), units(v, m)
And again, the reason we jump through these mildly distasteful hoops is
that *it gets rid of allocated strings entirely*.
That bug in df_pretty_sizes() where it allocates a 20 byte buffer to
store a string that can be 21 bytes long with its null? It'd just
vanish. Not needed.
The only code that is *sure* to be bug free is the code that doesn't
exist.
> To avoid all this mess, we should add another conversion specifier with
> the function register_printf_function(). Unfortunately I was not able to
> find the equivalent one for sprintf().
Doesn't that prevent compile-time verification that the types of the
format specifiers match the arguments?
- z
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
