Re: New draft version of o65 file format

From: Spiro Trikaliotis (ml-cbmhackers_at_trikaliotis.net)
Date: 2005-03-30 14:16:11

Hello André,

* On Tue, Mar 29, 2005 at 09:22:49PM +0200 fachat wrote:
 
> Maybe you and the all others that use the file format could send me some
> text to add that describes their use of the file format.

Hm, well, yes... Sending... ;-)


opencbm floppy modules:
The opencbm API (cbm4win, cbm4linux) will(*) use the o65 file format for
modules to load into floppy. The API itself loads o65 files to account
for the differences between the various floppy drives (1540/1541,
1570/1571, 1581), and to present the API user transfer protocols for its
use. The API user loads additional modules into the floppy drives which
then use late binding to connect to the various support routines of the
floppy ROM, or of the transfer protocols. Name encoding is ASCII, name
length are restricted to 1024 characters currently (will be removed most
probably in the near future). Currently, the simple file format is not
required (but there is no definitive decision about this yet).
Currently, o65 files for the opencbm API are restricted to a total of
64KB (65536 byte) in length.


(*) remember: This is work in progress, so the details are subject to
change.

 
> > 2. While I wrote my own parser, I stumbled across the definition of the
> >    relocation tables (2.6.4.). IMHO, it is not always clear which (and
> >    how many) bytes follow after the different formats. @@@SRT

> If you have any problems, ask me, probably in private mail. We can work
> out more examples for the last section if necessary.

Sorry. This mail was not meant to be sent yet but I wanted to postpone
it. The @@@SRT mark is a sure indicator for this. ;-) It seems I pressed
the wrong keys.

I will have to look up again my source code to ask exactly about my
problems via PM.

> > 3. 2.6.3) undefined references list and 2.6.5) exported globals list:
> >    I want to discuss if it makes sense to define an maximum name length
> 
> I have added a comment, which also comes from Uz' opinion that the target
> platform should just be able to interpret it.

Ok, this is good. Anyway, I might remove the restriction to 1024 byte
soon as I changed my file handling code in the meantime.

Regards,
   Spiro.

-- 
Spiro R. Trikaliotis
http://www.trikaliotis.net/
http://cbm4win.sf.net/

       Message was sent through the cbm-hackers mailing list

Archive generated by hypermail pre-2.1.8.