On Fri, Oct 12, 2012 at 9:02 PM, Gordon Messmer <yinyang@eburg.com> wrote:
> On 10/12/2012 01:56 PM, Les Mikesell wrote:
>> On Fri, Oct 12, 2012 at 3:44 PM, Nux! <nux@li.nux.ro> wrote:
>>>
>>> Yep, exactly right. People in #openssh confirmed -i HAS to be a real
>>> path to a file.
>>
>> Not very unix-like behavior...
>
> Yes, it is. The alternative is for -i to take a file or a key as an
> argument, and that leads to ambiguous behavior.
How so? What's wrong with what behaves like a named pipe? That is,
why does scp need to seek instead of just reading the contents?
> All the same, I keep my keys in
> an encrypted volume because they grant me access to my customer's
> systems. The idea of writing them to a filesystem that's not encrypted
> is just creepy.
Sure, but who else can open your /dev/fd/##'s?
--
Les Mikesell
lesmikesell@gmail.com
_______________________________________________
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
10-14-2012, 10:41 PM
Gordon Messmer
scp scripting question
On 10/12/2012 07:20 PM, Les Mikesell wrote:
>
> How so? What's wrong with what behaves like a named pipe? That is,
> why does scp need to seek instead of just reading the contents?
Whether it seeks or not is rather irrelevant to the question of whether
or not it is unix-like behavior. seek() is quite normal behavior.
>> All the same, I keep my keys in
>> an encrypted volume because they grant me access to my customer's
>> systems. The idea of writing them to a filesystem that's not encrypted
>> is just creepy.
>
> Sure, but who else can open your /dev/fd/##'s?
I'm not sure what you're talking about, since that's not what zsh does.
When you use "=(command)", zsh will run "command" with its output
directed to a file in /tmp. The path to that file will be substituted
where "=()" was used in the command-line.
_______________________________________________
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos