HForm

If I connect a USB-connected SATA drive (via a USB to SATA convertor) to my ARMX6, the USB-connected drive is seen as the boot drive (CMOS setting drive 4). The SATA connected drive is seen as :4 from the desktop and the USB/SATA connected drive as :0. With HForm the SATA connected drive is seen as :0 and the USB/SATA connected drive as :1.

With the (new) SATA connected drive blank and the USB/SATA connected drive populated (but with a broken directory) I need to format the former and copy the data over. It took a bit of guesswork to format drive 0 (appears in desktop as 4) rather than 1 (appears in desktop as 0) in HForm.

Although current hardware has ‘only’ four SATA ports, connecting an extra SATA drive via a USB/SATA connector can give five drives…

Confused what the bug is here?

You do realise that HForm uses device numbers (SCSI IDs for SCSIFS for example, per *DEVICES), not FileCore’s logical drive numbers?

So if you had a SCSI podule with a bus of 7 devices (plus the controller itself) you might reasonably need to select device 6 in HForm, but if you had 6 scanners (?!) after formatting that’d appear as drive 4 (if non removable) or drive 0 (if removable). It especially relevant taking hot plugging into account (as with USB drives) where the numbers get recycled each time you switch them about, HForm has no idea what the landscape looks like.

It’s a little simpler with SDIO memory cards and ADFS, since they have a fixed bus<=>drive mapping.

You do realise that HForm uses device numbers (SCSI IDs for SCSIFS for example, per *DEVICES), not FileCore’s logical drive numbers?

Can HForm not be changed to use the logical drive number RISCOS would assign the drive to save all the confusion? It’s very unintuitive in it’s current state and its all too easy to wipe the wrong drive. We should also consider moving HForm into RISCOS, either in it’s current form (as a BASIC app) or as *FORMAT on the reverent filesystems, with a proper Wimp frontend added off the matching Filers.

We should also consider moving HForm into RISCOS, either in it’s current form (as a BASIC app) or as *FORMAT on the reverent filesystems, with a proper Wimp frontend added off the matching Filers.

I believe that there’s a bounty for a graphical HForm.

Clearly if it is behaving in the expected way it is a feature not a bug.

After saying ‘Format which drive (0-7)?1’ it was receiving the message ‘Are you sure you want to do this to drive SCSI::0’ I found confusing. Whilst I accept that both are logically correct (as you say) I maintain that it is not what a user will expect whilst agonising to make sure he doesn’t format the wrong drive.

I can’t use the HForm without disconnecting all discs other than the one I want to format. I don’t trust that I won’t format the wrong one. A new user will probably be more trusting than I am.

I can’t use the HForm without disconnecting all discs other than the one I want to format

I do that as well!