> I'm working on a web-based filesystem for a class term project. The idea
> is to be able to have web servers show up as directory trees on the local
> machine. Example:
>
> cd /web/www.cs.wisc.edu/ ; cat index.html
>
> would be equivalent to pulling the page down. Right now the HTTP access is
> handled by a user-level daemon, and the kernel filesystem is an
> indirection layer. The FS gets a read request for a file, passes it to the
> daemon which either a) downloads the file or b) retrieves it from on-disk
> cache, passes a pointer (of some sort) back to the kernel, which returns a
> file descriptor.
>
> Now for the questions.
>
> 1) Is this sane?
No. You have no readdir() and no random access to files. You have
no directory hierarchy. _If_ you are going to produce such a monster -
start from learning CODA. But I'ld rather recommend to start with FTP -
web _really_ doesn't provide anything close to filesystem semantics. And
read the RFCs - one of the archives is on ftp.isi.edu/in-notes/
-
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/