Filetypes (was Re: Xerox)

From: D. Peschel <dpeschel_at_u.washington.edu>
Date: Wed Dec 16 19:49:54 1998

> Wow, I only knew about T, I, A and B. S, R and A2/B2 I've never heard of
> before. Could be undocumented filetypes ...

In _Beneath Apple DOS_, *everything* is documented. Well, practically
everything. There's a fairly specific breakdown of the DOS machine code
(with entry points, the purpose of each routine, etc.) which was made by
disassembling the code. It's good enough that you would have a chance of
understanding the code if you disassembled it yourself. There are also
sample listings for a small disk editor, individual track formatter, etc.

> Replace with U for USR. The Commodore manual says that USR files are "user
> defined file structures" but there's no way to define that structure. As
> the above proves, they're all written to and read from in the same way. You
> can even use DEL files (the DELeted file type, very exotic) like this, I
> think it's ,L for those.

Hmm. The Commodore OS is more powerful than I thought. I still think the
syntax and implementation is rather clunky. I'll admit that Apple DOS is
crude, but in some ways (mostly control of the disk drive) it's very
flexible since it's all done in software! And of course the Apple drive
combines its amazing simplicity with respectable speed, unlike the
shamefully slow Commodore drive.

> data stream in SEQ/USR/PRG). In the light you present it, the Apple system
> starts to make a little more sense.

What light? The "look at the data types and actions" approach? I guess I
would call that the object-oriented viewpoint. It sounds like you're saying
you understand more about the reasons why DOS is the way it is... is that
right?

-- Derek
Received on Wed Dec 16 1998 - 19:49:54 GMT