On 11/22/2010 02:36 AM, Pádraig Brady wrote:
> On 22/11/10 02:54, Paul Eggert wrote:
>> --- a/src/copy.c
>> +++ b/src/copy.c
>> @@ -603,6 +603,12 @@ copy_reg (char const *src_name, char const *dst_name,
>> }
>> }
>> }
>> +
>> + /* Improve quality of diagnostic when a nonexistent dst_name
>> + ends in a slash and open fails with errno == EISDIR. */
>> + if (dest_desc < 0 && dest_errno == EISDIR
>> + && *dst_name && dst_name[strlen (dst_name) - 1] == '/')
>> + dest_errno = ENOTDIR;
>
> I like this because it probably aligns more with other systems.
>http://git.sv.gnu.org/gitweb/?p=coreutils.git;a=commitdiff;h=a16bee4f
>
> I wonder should the Linux kernel also change?
> I.E. should creat("nosuch/") and ftruncate("nosuch/")
> return ENOTDIR rather than EISDIR ?
> There are arguments for and against I suppose.
The strongest arguments for changing the kernel would be bringing it
into compliance with POSIX.
--
Eric Blake address@hidden +1-801-349-2682
Libvirt virtualization library http://libvirt.org
signature.asc Description: OpenPGP digital signature