[Coco] HDBDOS/drivewire discovery

[Aaron Wolfe]

On Sat, Jan 14, 2012 at 2:34 AM, Willard Goosey <goosey at virgo.sdc.org> wrote:
>
> On Sat, Jan 14, 2012 at 05:15:48PM +1000, Bob Devries wrote:
> > My RSDISK programme was written to access the files on a virtual disk image
> > file. At the time, I did not envisage having multiple "disks" in a file.
> >
> Yeah, it was before people started passing around *hard drive* images
> in addition to floppy images.
>
> > Having said that, it should be quite trivial to modify the source (which is
> > included in the archive on RTSI) to ask the user for a disk number (0 to
> > 255), and multiply that by 161280 to get the offset to the wanted disk in
> > the image file.
>
> It certainly doesn't look bad.  Every example of your code that I've
> seen has been clean and readable.
>
> > The only caveat I can think of is that for the above offset to work
> > correctly, the disks in the image *must* be 35 tracks, 18 sectors and 256
> > bytes per sector.
>
> Which RGBDOS/HDBDOS images are.  Pretty much by definition, I believe.

Yep, the file doesn't really have any concept of tracks but each disk
is always 630 sectors, 256 bytes in size which works out the the same
thing.