On Sun, Dec 19, 1999 at 16:14 -0500, Adam Di Carlo wrote:
> Marcel, aren't you working with Shaleh?! Why the redundancy and
> wasted duplicate effort?
Because I am working on my thesis. In my tree net-fetch is integrated
into dbootstrap now, but I haven't tested the integration yet. I've
changed parts of extract_*.c and choose_medium.c. I will not commit
untested code to the boot-floppies CVS. Net-fetch is a generic install
method that is able to use http, ftp, snarf, etc. to fetch files. I only
implemented a http module (http/1.1 client incl. proxy support) to go with
net-fetch. I have no interest in ftp or snarf, but support for them can
be added easily (I defined interfaces between net-fetch and its modules).
Maybe I can test it during the weekend in between family visits.
> --
> From: "Sean 'Shaleh' Perry" <shaleh@valinux.com>
> To: Adam Di Carlo <adam@onshore.com>
> Subject: Re: recent changelogs, FYI
> Cc: debian-boot@lists.debian.org
>
> Marcel told me he would check in the dbootstrap code and I would have to
> basically make it call my script(s). Well, never saw his code, so I am
> figuring out dbootstrap too (-:
You were going to write support for snarf anyhow, so there shouldn't be
too much duplicate effort. The source code needed in dbootstrap to
support snarf is minimal too btw, especially if you link snarf with
dbootstrap (and use argv[0] to select program behavior). If you don't
link snarf with dbootstrap a fork(2) to start the snarf process combined
with filesize monitoring (so you can display a progress bar) and snarf pid
monitoring (including signal handling) should work too. And there are
probably other solutions too.
Cheers,
Marcel
--
Anyone who gets in between me and my morning coffee should be
insecure. --RFC 2324, (7. Security Considerations)
Attachment:
pgp8yz5b7kD2D.pgp
Description: PGP signature