| Re: [AD] standard path updates |
[ Thread Index |
Date Index
| More lists.liballeg.org/allegro-developers Archives
]
On January 6, 2011, Matthew Leverton wrote:
> On Thu, Jan 6, 2011 at 5:02 PM, Thomas Fjellstrom
>
> <tfjellstrom@xxxxxxxxxx> wrote:
> > I thought it was possible to mix physfs with the standard fs if you set
> > it up right. Don't hang anything directly on / or it might hide
> > everything (or so I'm guessing). PhysFS would be pretty useless if it
> > always hid the regular fs.
>
> I think it will be less confusing if it always uses stdio, as the
> programmer shouldn't even be aware that Allegro is parsing a config
> file to get the directory path. Even if it works with PhysFS, there
> may be some other interface set up that would be incompatible.
>
> Now if somebody wants to take the returned path and feed it through
> PhysFS, then that's his own choice.
That kind of defeats the entire purpose behind the swapable vfs setup. If
you're using a separate VFS you really aught to make sure stuff is visible. Its
something you should do consciously.
Otherwise we might as well drop the whole idea, and let people use physfs and
things like it separately from allegro.
Really, the whole idea was so people could init physfs (properly), and even
allegro's config file could be loaded through physfs. I'm not sure that ever
worked though, but that was the idea. Complete and total replacement, mostly
transparently (in physfs's case you need to setup some mounts to be able to
see anything, but thats just the nature of the beast).
> --
> Matthew Leverton
>
> ---------------------------------------------------------------------------
> --- Learn how Oracle Real Application Clusters (RAC) One Node allows
> customers to consolidate database storage, standardize their database
> environment, and, should the need arise, upgrade to a full multi-node
> Oracle RAC database without downtime or disruption
> http://p.sf.net/sfu/oracle-sfdevnl
--
Thomas Fjellstrom
tfjellstrom@xxxxxxxxxx