Re: (reiserfs) Re: 20 years without semantic innovation is enough

Jamie Lokier (lkd@tantalophile.demon.co.uk)
Fri, 2 Jul 1999 21:10:05 +0200


Bernd Paysan wrote:
> I like HTTP more than e.g. the tar format, since it
> contains attribute: value pairs, and that would allow to make better use
> of a database-like structured filesystem, where extended attributes are
> possible.

I like MIME multipart (with Content-Length) even more since it is
similar, and is also actually meaningful in a file unlike HTTP.

> The only disadvantage is that HTTP has implicit types of
> attributes (making them explicit, e.g. "int64 Content-Length: 1234")
> would help generic parsing a lot.

I don't see what's wrong with "infinite" precision. These are files not
potentially extremely large streams, right?

> BTW: when you start thinking about a query_attributes(2) call, make sure
> not to repeat the XInternAtom flaw: attributes are labeled by strings
> (that's a good idea), but the string isn't used for each command:
> instead, the application assigns a unique number to each attribute name,
> and sends that pair to the filesystem. It subsequently uses it's own
> numbers to name the attributes.

Are you describing the flaw or the way it should have been implemented?

-- Jamie

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/