Re: GnomeVFS depends on Bonobo now
- From: ERDI Gergo <cactus cactus rulez org>
- To: Bastien Nocera <hadess hadess net>
- Cc: Michael Meeks <michael ximian com>, Seth Nickell <snickell stanford edu>, gnome-2-0-list gnome org
- Subject: Re: GnomeVFS depends on Bonobo now
- Date: Fri, 27 Jul 2001 14:26:04 +0200 (CEST)
I am very happy to see this issue resolved -- I remember a less optimistic
state when I asked around at GUADEC about the issue of VFS-based monikers.
On Fri, 27 Jul 2001, Bastien Nocera wrote:
> Maybe that's a stupid question, but I don't understand what the benefit
> is, because I don't understand what this does. Can you explain what this
> is all about for the ignorants like me ?
The point is that currently you have to replicate a lot of code in
monikers from gnome-vfs to do things like gzip or tar support. This is
because if the moniker chain contains non-VFS parts, you can't concatenate
the VFS parts into a single VFS moniker, and you can't reuse the VFS
implementation because they only work on VFS sources.
E.g. let's say you have a compound document that contains some data in one
of its items in gzipped form. The obvious moniker to this would be
file:/home/foo/mydoc!Part 1|Stream 1#gzip:
To do this, you obviously have to implement a standalone gzip: moniker
that can work on other monikers, since there are two Item monikers between
the toplevel VFS moniker and the gzip: moniker. But if the VFS code
depended on Bonobo, you could implement the gzip: moniker inside gnome-vfs
to work on other monikers.
Hope that makes it clear,
Cactus
--
.--= ULLA! =---------------------. `We are not here to give users what
\ http://cactus.rulez.org \ they want' -- RMS, at GUADEC 2001
`---= cactus cactus rulez org =---'
"Dog for sale: eats anything and is fond of children."
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]