Privilege level for $ extended attributes? Re: Extended attribute interfaces

[Casey Schaufler]

Robert Watson wrote:
> 
> Andreas,
> 
> Boris and others have raised with me that they are concerned about the
> choice of the symbol '$' as the prefix for system attributes.  In
> particular, because that symbol is widely used as a special character in
> shell scripts and scripting languages.

It's pretty tough to identify a special character which isn't
significant to interpreters, although '$' is among the worst
in that regard. My 2-seconds though would be to use '+', as
it indicates "more", and does not tend to be used in names
very often.
 

> One possibility in my mind as been to assume a name-space segmented by the
> "." or "/" characters, and prefix "system." or ".system." in front of the
> attribute name.

You could do that, but before you do, consider the way Sun does
(or at least did) multilevel directories redirection avoidance.
If one used "/tmp/foo", one gets redirected, while if one uses
"/.No_Hid.tmp/foo" one doesn't. You can do it, but it's ugly.

> The SGI solution of simply maintaining
> two namespaces also makes some amount of sense, but a unified namespace
> leads to a unified namespace solution, whereas two namespaces would
> require applications to distinguish between them (such as backup tools)
> themselves, offloading the problem to be solved in numerous different ways
> by different authors.

Having lived in the multiple namespace world, I agree that one
would be easier to deal with. I prefer the one character special
name marker.

-- 

Casey Schaufler				Manager, Trust Technology, SGI
casey at sgi.com				voice: 650.933.1634
casey_p at pager.sgi.com			Pager: 888.220.0607
To Unsubscribe: send mail to majordomo at cyrus.watson.org
with "unsubscribe posix1e" in the body of the message