Path: utzoo!mnetor!tmsoft!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!uunet!mcsun!ukc!acorn!pcolmer
From: pcolmer@acorn.co.uk (Philip Colmer)
Newsgroups: comp.sys.acorn
Subject: Re: what about a discussion about standart fileformats when posting to comp.sys.binaries in future ?
Message-ID: <5127@acorn.co.uk>
Date: 13 Feb 91 08:32:09 GMT
References: <1991Feb12.153527.27416@ugle.unit.no>
Sender: pcolmer@acorn.co.uk
Distribution: comp
Organization: Acorn Computers Ltd, Cambridge, England
Lines: 44

In article <1991Feb12.153527.27416@ugle.unit.no> dhmyrdal@lise.unit.no (Dag H}kon Myrdal) writes:

> [Stuff about pros and cons and requirements of file encoding deleted]
>
>The third thing is also the one that has caused problems regarding
>the use of !Spark... David Pilling has chosen a way of coding
>directory-structures and filetypes that is *not* compatible
>with standard 'arc'. But surely, someone should be able to
>do that a better way??? How about storing filetypes the
>same way as one uses the opposite way for extensions? To store
>a filetype in a archive, you could save the filetype as the
>last part of the filename, having an option on the archimedes side
>on whether to treat the files as archimedes or 'standard' files..

My !Submit/!Extract, which use Frank Lancaster's port of tar, get around
this problem by holding a script file in the archive which goes around
setting the load & exec addresses afterwards :-). Cheap, cheerful and it
works. It means that the tar files are then fully UNIX compatible.

>Also, how does Fil's !Submit program develop? Is it ready for
>use?

I haven't done ANYTHING to it since the last release. I was intending to
make the necessary changes to make it work with the latest version of tar,
and also to include some of the suggestions that people of made.

Personally, I feel it is BETTER than !Spark as far as posting large binaries
goes, because it copes with people dropping multiple articles on it. With
!Spark, you need to cut all the headers and footers out first.

However, I won't spend any time progressing this project unless people want
it and are going to adopt it in favour of Spark. It simply isn't worth my
time.

Incidentally, I would suggest that once we *have* got comp.binaries.acorn, a
copy of !SparkPlug (or whatever extractor is required) is posted regularly
(about once a month). This solves the major problem straight away - the
availability of the required extractor. Of course, it needs to be posted in
a form that can be extracted ...

--Fil.

------------------------------------------------------------------------------
Hi, I'm Pisces. I just dive right in ...